The Mediterranean Basins: Tertiary Extension within the Alpine Orogen
Geological Society Special Publications Series Editors A. J. FLEET
R. E. HOLDSWORTH A. C. MORTON M. S. STOKER
It is recommended that reference to all or part of this book should be made in one of the following ways.
DURAND, B., JOLIVET, L., HORVATH, E d~; SERANNE,M. (eds) 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156. DOGLIONI, C., GUEGUEN, E., HARABAGLIA,P. ~f~MONGELLI,E 1999. On the origin of west-directed subduction zones and applications to the western Mediterranean. In: DURAND, B., JOLIVET, L., HORVATH, E & SERANNE,M. (eds) 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 541-561.
G E O L O G I C A L S O C I E T Y S P E C I A L P U B L I C A T I O N NO. 156
The Mediterranean Basins: Tertiary Extension within the Alpine Orogen EDITED
BY
BERNARD DURAND Institut Franqais du Petrole, Rueil Malmaison, France LAURENT JOLIVET Universit6 Pierre et Marie Curie, Paris, France FRANK HORVATH Lor~ind E6tv6s University, Budapest, Hungary and MICHEL St~RANNE Universit6 Montpellier 2, France
1999 Published by The Geological Society London
THE G E O L O G I C A L SOCIETY
The Society was founded in 1807 as The Geological Society of London and is the oldest geological society in the world. It received its Royal Charter in 1825 for the purpose of 'investigating the mineral structure of the Earth'. The Society is Britain's national society for geology with a membership of around 8500. It has countrywide coverage and approximately 1500 members reside overseas. The Society is responsible for all aspects of the geological sciences including professional matters. The Society has its own publishing house, which produces the Society's international journals, books and maps, and which acts as the European distributor for publications of the American Association of Petroleum Geologists, SEPM and the Geological Society of America. Fellowship is open to those holding a recognized honours degree in geology or cognate subject and who have at least two years' relevant postgraduate experience, or who have not less than six years' relevant experience in geology or a cognate subject. A Fellow who has not less than five years' relevant postgraduate experience in the practice of geology may apply for validation and, subject to approval, may be able to use the designatory letters C Geol (Chartered Geologist). Further information about the Society is available from the Membership Manager, The Geological Society, Burlington House, Piccadilly, London W l V 0JU, UK. The Society is a Registered Charity, No. 210161. Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN UK (Orders: Tel. 01225 445046 Fax 01225 442836)
Distributors
USA A A P G Bookstore PO Box 979 Tulsa OK 74101-0979 USA (Orders: Tel. (918) 584-2555 Fax (918) 560-2652)
First published 1999
Australia The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. 9 The Geological Society 1999. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London WIP 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/99/$15.00.
Australian Mineral Foundation 63 Conyngham Street Glenside South Australia 5065 Australia (Orders: Tel. (08) 379-0444 Fax (08) 379--4634)
India Affiliated East-West Press PVT Ltd G-1/16 Ansari Road New Delhi 110 002 India (Orders: Tel. (11) 327-9113 Fax (11) 326-0538)
Japan British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 1-86239-033-9 Typeset by Type Study, Scarborough, UK Printed in Great Britain by Cambridge University Press, Cambridge, UK.
Kanda Book Trading Co. Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-Shi Tokyo 206q3034 Japan (Orders: Tel. (0423) 57-7650 Fax (0423) 57-7651)
Contents
Foreword
VII
JOLIVET, L., FRIZON DE LAMOTTE,D., HORVATH,E, MASCLE,A. & SI~RANNE,M. The Mediterranean basins: Tertiary extension within the Alpine Orogen - an introduction Western Mediterranean
St2RANNE,M. The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: an overview
15
CHAMOT-ROOKE, N., GAULIER, J.-M. & JESTIN, E Constraints on Moho depth and crustal thickness in the Liguro-Provenqal basin from a 3D gravity inversion: geodynamic implications
37
VERGIe,S, J. & SABAT,E Constraints on the Neogene Mediterranean kinematic
63
evolution along a 1000 km transect from Iberia to Africa BENEDICTO, A., St~GURET,M. & LABAUME,P. Interaction between faulting, drainage and sedimentation in extensional hanging-wall syncline basins: example of the Oligocene Matelles basin (Gulf of Lion rifted margin, SE France)
81
ZECK, H. P. Alpine plate kinematics in the western Mediterranean: a westward-directed subduction regime followed by slab roll-back and slab detachment
109
MASCLE, A. & VIALLY, R. The petroleum systems of the Southeast Basin and Gulf of Lion (France)
121
WILSON,M. & BIANCHINI, G. Tertiary-Quaternary magmatism within the Mediterranean
141
and surrounding regions
MAUFFRET,A. & CONTRUCCI,I. Crustal structure of the North Tyrrhenian Sea: first
169
result of the multichannel seismic LISA cruise Pannonian Basin
HORVATH, E & TARI, G. IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration
195
TARI, G., DOVt~NYI,P., DUNKL, I., HORVATH,E, LENKEY,L., STEFANESCU,M., SZAFIAN,P. & TOTH, T. Lithospheric structure of the Pannonian Basin derived from seismic, gravity and geothermal data
215
GYORFI, I., CSONTOS,L. & NAGYMAROSY,A. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian Basins
251
GERNER, P., BADA, G., DOVt~NYI,P., MULLER, B., ONCESCU,M. C., CLOETINGH, S. & HORWX~TH,E Recent tectonic stress and crustal deformation in and around the Pannonian Basin: data and models FODOR, L., CSONTOS,L., BADA, G., GYORFI, I. & BENKOVICS,L. Tertiary tectonic evolution of the Pannonian Basin system and neighbouring orogens: a new synthesis of palaeostress data
269 295
vi
CONTENTS
JUHASZ, E., PHILLIPS,L., MOLLER, P., RICKETTS, B., T(3TH-MAKK,/~., LANTOS,M. KovAcs, L. 6. Late Neogene sedimentary facies and sequences in the Pannonian Basin, Hungary
335
SACCHI, M., HORV.~TH, F. & MAGYARI,O. Role of unconformity-bounded units in stratigraphy of continental record: a case study from the Late Miocene of western Pannonian Basin, Hungary
357
VAN BALEN, R. T., LENKEY, L. HORVATH, F. & CLOETINGH,S. A. E L. Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin
391
Eastern Mediterranean
HATZFELD,D. The present-day tectonics of the Aegean as deduced from seismicity
415
JOLIVET,L. & PATRIAT,M. Ductile extension and the formation of the Aegean Sea
427
LIPS, A. L. W., WIJBRANS,J. R. & WHITE,S. H. New insights from 4~ laserprobe dating of white mica fabrics from the Pelion Massif, Pelagonian Zone, Internal Hellenides, Greece: implications for the timing of metamorphic episodes and tectonic events in the Aegean region
457
OKAY,A. I. & TOYsOz, O. Tethyan sutures of northern Turkey
475
General
ZIEGLER,P. A. & ROURE,E Petroleum systems of Alpine-Mediterranean fold belts
517
and basins DOGLIONI, C., GUEGUEN, E., HARABAGLIA,P. • MONGELLI,F. On the origin of west-directed subduction zones and applications to the western Mediterranean
541
FOREWORD This book derives from the Integrated Basin Studies Project (IBS), which ran during the years 1992-1995 with the support of the European Commission DGXII. The papers produced by the IBS group have been complemented by eight papers arising from the conference 'Mediterranean Basins: Tertiary Extension within the Alpine Orogen', held in Cergy-Pontoise, France, 11-13 December 1996. The title of this conference has also been retained as the title of the book. The papers included here collectively cover the majority of the Mediterranean Tertiary extensional basins, with the exception of those in the SW of the region. The book is the second in a series of three that are devoted to studies carried out in European Basins as part of the IBS project. The first volume, Cenozoic Foreland Basins of Western Europe, has already been published by the Geological Society (A. Mascle et al. 1998). A third volume concerned with the Norwegian rifted margin is currently in preparation. The main concept of the IBS project was to develop methods and techniques of modelling in which the main physical phenomena responsible for formation, development and the infilling of sedimentary basins or subbasins are linked together and related to large-scale, deep processes such as the convective movements of the mantle. Critical aspects of such models were as follows. The capacity to realistically link and describe the interactions of deep deformations to the near surface deformations that determine the geometrical evolution of basins and sub-basins. This has to be possible in both extensional and compressional tectonic contexts. The capacity to couple such tectonic models to simple but realistic models of linked erosion and sedimentation processes. The capacity to incorporate a model of fine-grained sediment compaction so as to include the effect of sediment loading and water escape on sediment deformation. The final objective was to obtain practical outputs and deliverables such as accounts of the architecture of basin or sub-basin fills (down to the reservoir scale), their evolution through geological times, fluid pressure regimes and stress histories and informations concerning the thermal evolution. Such methods and techniques are intended to result in better designed reservoir geological models and therefore to contribute to the improvement of field development planning and exploitation. In order to achieve these goals, the IBS teams used the following methodology. (1)
(2)
(3)
Using prototypes proposed mainly by groups from Vrije University in Amsterdam (S. Cloetingh and coworkers), models have been developed based on the existing knowledge concerning the rheology of crust and sediments and from published and unpublished data already available from various thoroughly studied basins in different tectonic settings. At the same time, a group, under the leadership of a team from Newcastle University (A. Aplin and colleagues) re-examined the compaction behaviour of fine-grained sediments and how it may be modelled using theoretical, experimental and observational approaches. A synthesis of this work has been published recently as a special issue of Marine and Petroleum Geology (A. Aplin & G. Vasseur (eds) 1998, Marine and Petroleum Geology, vol. 15, No. 2) A small number of European basins, set in their appropriate tectonic contexts, have been utilized as natural laboratories in which to document the interaction of tectonic and sedimentation processes using extensive syntheses of seismic well and data obtained in the field. This work resulted in a considerable updating of geological knowledge on the chosen basins and for this reason the IBS teams decided to publish these studies in the form of the present series of books.
These basins were chosen as follows. Tertiary rifted basins within the Alpine orogen: the Gulf of Lion in France and the Pannonian basin in Hungary were studied respectively under the leadership of the University of Montpellier (M. S6ranne and colleagues) and of the E6tv6s-Lorand laboratory in Budapest (F. Horvfith and colleagues).
viii
FOREWORD
Foreland basins: the south Pyrenean basins, the Guadalquivir basin in Spain, the molasse basin in Germany and the Barr~me syncline in France were studied respectively by teams from the University of Barcelona (M. Marzo and colleagues), from the Institut de Ciencias de la Terra, Barcelona (M. Fernandez and colleagues), from the University of Ttibingen (H. P. Luterbacher and colleagues), and ETH, Zurich (M. Ford and colleagues), associated under the leadership of the Servei Geol6gic de Catalunya (C. Puigdef~bregas and colleagues). The Norwegian margins: the northern Viking Graben, the Mere basin, the Voring basin and the mid-Norwegian Margin in Norway, studied by teams from the Norwegian universities and Norwegian oil companies under the leadership of Norsk Hydro (A. Nottvedt and colleagues). The cooperation with industry increased during the project and finally, 21 oil companies helped in a significant way. Some, like the Norwegian oil companies were directly involved. In particular, Norsk-Hydro had the leadership of module 3 (Dynamics of the Norwegian Margin). Others contributed less directly by giving documents or helping in the interpretation. The IBS teams acknowledge this help and thank these companies which are listed below (in alphabetical order): Amoco, BEB, BP, Coparex, DEE, EEP, Esso, INA Naftaplin, Norsk-Hydro, MOL, Mobil, OMV, Preussag, Repsol, RWE-DEA, Saga, Shell, Statoil, Total, VVNP, Wintershall. Overall, more than 200 researchers belonging to 38 institutions and 15 countries (eight E U countries, six non-EU European countries and the USA) have participated in the IBS project. The IBS teams have collaborated and participated with groups from the International Lithosphere Programme (Origin of Sedimentary Basins), the network E B R O (European Basin Research Organisation), and the 'Human Capital and Mobility' programme of the European Commission. Through the IBS project, the D G XII of the European Commission has clearly demonstrated its capacity to create a European research space. This capacity was enhanced by the access given to the IBS Program to a European program with Hungary, which resulted in the IBS program on the Pannonian basin. In addition, the cooperation of the Research Council of Norway (NFR) allowed the IBS-DMN program on the Dynamics of the Norwegian Margin to be launched. The Bundesamt for Bildung und Wissenschaft of Switzerland, also helped to launch the IBS-ETH cooperation. We also wish to acknowledge here the role played by these institutions and we thank them for their financial support. Many thanks are also due to DGXII experts and executives and particularly to J. C. Imarisio and J. M. Bemtgen for having helped to make the IBS project realistic and effective. The IBS teams are also indebted to M. Rougeaux, Managing Director of Groupement d'Etudes et de Recherches en Technologie des Hydrocarbures (GERTH) for his help in the management of the project. The European research space which has been created consists in academic teams who voluntarily worked on geological problems of interest for the oil industry, and who now have a solid capacity in this field. Many of these teams are now associated in the Eurobasin School, where a number of European Universities and Research Institutions cooperate, under the auspices of Academia Europea.
Mediterranean Basins: Tertiary Extension within the Alpine Orogen An international workshop on Mediterranean basins was held at the University of Cergy-Pontoise, France, on December 11-13 in 1996. It was co-organized by the University of Cergy-Pontoise and IFP. Funds were provided by the University, the Conseil G6n6ral du Val d'Oise and the Syndicat de l'Agglom6ration Nouvelle de Cergy-Pontoise. Forty-five oral communications were presented on all the major basins from the Alboran Sea to the Pannonian Basin, as well as on surrounding mountain belts. Recently acquired data, in particular those obtained within the framework of the IBS project, were presented and regional syntheses proposed. The coexistence in space and time of growing mountain belts and actively extending basins poses a number of yet unsolved questions in terms of mechanics. This problem is particularly crucial in the Mediterranean region where all Cenozoic basins opened in the internal zones of mountain belts. The Tyrrhenian Sea opened in the backarc region of the Apennines, the Aegean Sea in the backarc domain of the Hellenides and Hellenic arc, the Pannonian Basin behind the Carpathians and the Alboran Sea between the Betics and the Rift In some examples such as the Tyrrhenian Sea and the Aegean Sea, extension is still ongoing while peripheral compression and convergence are active. The Alboran and Pannonian basin are now in a stage of compression. Several models have been proposed to explain this coexistence of compression and extension: slab retreat during subduction process, detachment of a deep lithospheric root under the internal zones leading to radial extension and peripheral compression and slab detachment. The conference
FOREWORD
ix
acted as a forum for interactions between geologists and geophysicists in the study of the complex dynamic problem posed by the Mediterranean region. This volume presents a wealth of new data on various topics centered around the Mediterranean region from the deep mantle structure to the detailed geometry of sedimentary basins. B. D U R A N D Project Leader of IBS L. J O L I V E T Chairman of the Conference on Mediterranean Basins: Tertiary Extension within the Alpine Orogen
References APLIN,A. C. & VASSEUR,G. (eds) 1998. Geological Compaction of Fine Grained Sediments. Marine and Petroleum Geology, 15. CLOETINGH,S., DURAND,B. & PUIGDEFABREGAS,C. (eds) 1995. Integrated Basin Studies. Marine and Petroleum Geology, 12. DURAND, B. & MASCLE,A. 1996. Interest for the European Oil Industry of the Results Obtained by the Integrated Basin Studies JOULE Project no: CT92-120 In: 'The Strategic Importance of Oil and Gas Technology'. Proceedings of the 5th European Union Hydrocarbons Symposium, Edinburgh, 26-28 November 1996, 2, 1151-1167. MASCLE,A., PUIGDEFABREGAS,C., LUTERBACHER,H. P. • FERNANDEZ,M. (eds) 1998. Cenozoic Foreland Basins of Western Europe. Geological Society, London, Special Publications, 134.
The editors thank warmly the following persons for their participation in the review process: D. Avigad, J. M. Azafi6n, B. Biju-Duval, C. Bois, B. Bonin, D. Bonijoly, J. R. Borgomano, R. Caby, B. Colletta, J. M. Daniel, B. de Voogd, J. F. Dewey, R. W. England, C. Faccenna, D. Frizon, J. M. Gaulier, P. Gautier, P. Guennoc, I. Gy0rfy, F. Horv~th, T. Jacquin, A. Jambon, L. Jolivet, M. R. Leeder, I. Lerche, A. Lips, L. Longergan, A. Mascle, G. Matavelli, A. Mauffret, G. Nesen, J. Platt, A. Poisson, A. Richard, E. Roca, L. E. Ricou, F. Roure, M. S6ranne, G. Tari, P. Tremoli6res, J. Verg6s, R. Vially, A. B. Watts, J. Wheeler, T. White, H. Zeck. The following companies are thanked for their generous support of colour printing in the volume: BP, CGG, Elf-EP, Gaz de France, Institut Fran~ais du P6trole, Norsk Hydro and Total-Fina.
FOREWORD This book derives from the Integrated Basin Studies Project (IBS), which ran during the years 1992-1995 with the support of the European Commission DGXII. The papers produced by the IBS group have been complemented by eight papers arising from the conference 'Mediterranean Basins: Tertiary Extension within the Alpine Orogen', held in Cergy-Pontoise, France, 11-13 December 1996. The title of this conference has also been retained as the title of the book. The papers included here collectively cover the majority of the Mediterranean Tertiary extensional basins, with the exception of those in the SW of the region. The book is the second in a series of three that are devoted to studies carried out in European Basins as part of the IBS project. The first volume, Cenozoic Foreland Basins of Western Europe, has already been published by the Geological Society (A. Mascle et al. 1998). A third volume concerned with the Norwegian rifted margin is currently in preparation. The main concept of the IBS project was to develop methods and techniques of modelling in which the main physical phenomena responsible for formation, development and the infilling of sedimentary basins or subbasins are linked together and related to large-scale, deep processes such as the convective movements of the mantle. Critical aspects of such models were as follows. The capacity to realistically link and describe the interactions of deep deformations to the near surface deformations that determine the geometrical evolution of basins and sub-basins. This has to be possible in both extensional and compressional tectonic contexts. The capacity to couple such tectonic models to simple but realistic models of linked erosion and sedimentation processes. The capacity to incorporate a model of fine-grained sediment compaction so as to include the effect of sediment loading and water escape on sediment deformation. The final objective was to obtain practical outputs and deliverables such as accounts of the architecture of basin or sub-basin fills (down to the reservoir scale), their evolution through geological times, fluid pressure regimes and stress histories and informations concerning the thermal evolution. Such methods and techniques are intended to result in better designed reservoir geological models and therefore to contribute to the improvement of field development planning and exploitation. In order to achieve these goals, the IBS teams used the following methodology. (1)
(2)
(3)
Using prototypes proposed mainly by groups from Vrije University in Amsterdam (S. Cloetingh and coworkers), models have been developed based on the existing knowledge concerning the rheology of crust and sediments and from published and unpublished data already available from various thoroughly studied basins in different tectonic settings. At the same time, a group, under the leadership of a team from Newcastle University (A. Aplin and colleagues) re-examined the compaction behaviour of fine-grained sediments and how it may be modelled using theoretical, experimental and observational approaches. A synthesis of this work has been published recently as a special issue of Marine and Petroleum Geology (A. Aplin & G. Vasseur (eds) 1998, Marine and Petroleum Geology, vol. 15, No. 2) A small number of European basins, set in their appropriate tectonic contexts, have been utilized as natural laboratories in which to document the interaction of tectonic and sedimentation processes using extensive syntheses of seismic well and data obtained in the field. This work resulted in a considerable updating of geological knowledge on the chosen basins and for this reason the IBS teams decided to publish these studies in the form of the present series of books.
These basins were chosen as follows. Tertiary rifted basins within the Alpine orogen: the Gulf of Lion in France and the Pannonian basin in Hungary were studied respectively under the leadership of the University of Montpellier (M. S6ranne and colleagues) and of the E6tv6s-Lorand laboratory in Budapest (F. Horvfith and colleagues).
viii
FOREWORD
Foreland basins: the south Pyrenean basins, the Guadalquivir basin in Spain, the molasse basin in Germany and the Barr~me syncline in France were studied respectively by teams from the University of Barcelona (M. Marzo and colleagues), from the Institut de Ciencias de la Terra, Barcelona (M. Fernandez and colleagues), from the University of Ttibingen (H. P. Luterbacher and colleagues), and ETH, Zurich (M. Ford and colleagues), associated under the leadership of the Servei Geol6gic de Catalunya (C. Puigdef~bregas and colleagues). The Norwegian margins: the northern Viking Graben, the Mere basin, the Voring basin and the mid-Norwegian Margin in Norway, studied by teams from the Norwegian universities and Norwegian oil companies under the leadership of Norsk Hydro (A. Nottvedt and colleagues). The cooperation with industry increased during the project and finally, 21 oil companies helped in a significant way. Some, like the Norwegian oil companies were directly involved. In particular, Norsk-Hydro had the leadership of module 3 (Dynamics of the Norwegian Margin). Others contributed less directly by giving documents or helping in the interpretation. The IBS teams acknowledge this help and thank these companies which are listed below (in alphabetical order): Amoco, BEB, BP, Coparex, DEE, EEP, Esso, INA Naftaplin, Norsk-Hydro, MOL, Mobil, OMV, Preussag, Repsol, RWE-DEA, Saga, Shell, Statoil, Total, VVNP, Wintershall. Overall, more than 200 researchers belonging to 38 institutions and 15 countries (eight E U countries, six non-EU European countries and the USA) have participated in the IBS project. The IBS teams have collaborated and participated with groups from the International Lithosphere Programme (Origin of Sedimentary Basins), the network E B R O (European Basin Research Organisation), and the 'Human Capital and Mobility' programme of the European Commission. Through the IBS project, the D G XII of the European Commission has clearly demonstrated its capacity to create a European research space. This capacity was enhanced by the access given to the IBS Program to a European program with Hungary, which resulted in the IBS program on the Pannonian basin. In addition, the cooperation of the Research Council of Norway (NFR) allowed the IBS-DMN program on the Dynamics of the Norwegian Margin to be launched. The Bundesamt for Bildung und Wissenschaft of Switzerland, also helped to launch the IBS-ETH cooperation. We also wish to acknowledge here the role played by these institutions and we thank them for their financial support. Many thanks are also due to DGXII experts and executives and particularly to J. C. Imarisio and J. M. Bemtgen for having helped to make the IBS project realistic and effective. The IBS teams are also indebted to M. Rougeaux, Managing Director of Groupement d'Etudes et de Recherches en Technologie des Hydrocarbures (GERTH) for his help in the management of the project. The European research space which has been created consists in academic teams who voluntarily worked on geological problems of interest for the oil industry, and who now have a solid capacity in this field. Many of these teams are now associated in the Eurobasin School, where a number of European Universities and Research Institutions cooperate, under the auspices of Academia Europea.
Mediterranean Basins: Tertiary Extension within the Alpine Orogen An international workshop on Mediterranean basins was held at the University of Cergy-Pontoise, France, on December 11-13 in 1996. It was co-organized by the University of Cergy-Pontoise and IFP. Funds were provided by the University, the Conseil G6n6ral du Val d'Oise and the Syndicat de l'Agglom6ration Nouvelle de Cergy-Pontoise. Forty-five oral communications were presented on all the major basins from the Alboran Sea to the Pannonian Basin, as well as on surrounding mountain belts. Recently acquired data, in particular those obtained within the framework of the IBS project, were presented and regional syntheses proposed. The coexistence in space and time of growing mountain belts and actively extending basins poses a number of yet unsolved questions in terms of mechanics. This problem is particularly crucial in the Mediterranean region where all Cenozoic basins opened in the internal zones of mountain belts. The Tyrrhenian Sea opened in the backarc region of the Apennines, the Aegean Sea in the backarc domain of the Hellenides and Hellenic arc, the Pannonian Basin behind the Carpathians and the Alboran Sea between the Betics and the Rift In some examples such as the Tyrrhenian Sea and the Aegean Sea, extension is still ongoing while peripheral compression and convergence are active. The Alboran and Pannonian basin are now in a stage of compression. Several models have been proposed to explain this coexistence of compression and extension: slab retreat during subduction process, detachment of a deep lithospheric root under the internal zones leading to radial extension and peripheral compression and slab detachment. The conference
FOREWORD
ix
acted as a forum for interactions between geologists and geophysicists in the study of the complex dynamic problem posed by the Mediterranean region. This volume presents a wealth of new data on various topics centered around the Mediterranean region from the deep mantle structure to the detailed geometry of sedimentary basins. B. D U R A N D Project Leader of IBS L. J O L I V E T Chairman of the Conference on Mediterranean Basins: Tertiary Extension within the Alpine Orogen
References APLIN,A. C. & VASSEUR,G. (eds) 1998. Geological Compaction of Fine Grained Sediments. Marine and Petroleum Geology, 15. CLOETINGH,S., DURAND,B. & PUIGDEFABREGAS,C. (eds) 1995. Integrated Basin Studies. Marine and Petroleum Geology, 12. DURAND, B. & MASCLE,A. 1996. Interest for the European Oil Industry of the Results Obtained by the Integrated Basin Studies JOULE Project no: CT92-120 In: 'The Strategic Importance of Oil and Gas Technology'. Proceedings of the 5th European Union Hydrocarbons Symposium, Edinburgh, 26-28 November 1996, 2, 1151-1167. MASCLE,A., PUIGDEFABREGAS,C., LUTERBACHER,H. P. • FERNANDEZ,M. (eds) 1998. Cenozoic Foreland Basins of Western Europe. Geological Society, London, Special Publications, 134.
The Mediterranean Basins: Tertiary Extension within the Alpine O r o g e n - an introduction J O L I V E T , L. 1'2, D. F R I Z O N D E L A M O T T E 1, A. M A S C L E 3, M. S I ~ R A N N E 4
1DOpartement des Sciences de la Terre, E S A 7072 CNRS, Universitd de Cergy-Pontoise, Avenue du Parc, 8, le Campus, 95011 Cergy-Pontoise cedex, France 2present address: Laboratoire de Tectonique, E S A 7072 CNRS, UniversitO Pierre et Marie Curie, T 26-0 El, case 129, 4 Place Jussieu, 75252 Paris cedex 05, France 3IFP School, 228-232 avenue Napoleon Bonaparte, 92506 Rueil-Malmaison, France 4ISTEEM, Universitd Montpellier II, 34095 Montpellier cedex, France Abstract: The recent evolution of ideas on the Mediterranean region has been triggered by very active data acquisition over the last 15 years. Seismic tomography provides an unique view of mantle heterogeneities, space geodesy leads to precise determinations of the present strain and velocity fields, the combination of structural geology, radiometric dating and metamorphic petrology allows the description of P-T-t-D paths of exhumed metamorphic rocks, and exploration geophysics, onshore and offshore, gives a detailed view of the crustal geometry. Extension started in the Gulf of Lion and propagated eastwards and southwestwards to form the Liguro-Proven~ai basin, Tyrrhenian Sea and the Alboran Sea. It also started, at much the same time, in the Panonnian basin as well as in the Aegean back-arc region. Thus a seminal event occurred some 30 Ma ago that produced a sharp change from overall compression to back-arc extension. Although gravitational forces due to the collapse of a thick crust have affected most basins, it is now almost certain that this event ultimately originated in the mantle, either by slab detachement, slab rollback or both processes acting in sequence.
The reason why, suddenly, some 30 Ma ago, several extensional basins started to form within the overall compressional domain sandwiched between the African and European plates (Fig. 1), is a still open question. This problem was addressed by the IBS (Integrated Basins Studies) program and was the main topic of an international symposium, held at the Universit6 de Cergy-Pontoise (France) in December 1996 and co-organized by the Universit6 de CergyPontoise and by Institut Fran~ais du P6trole (IFP) (Fig. 2). In the past ten years or so, an entirely new view of the kinematic evolution and deformation pattern of the Mediterranean region has emerged, particularly with regards to these processes that may be responsible for the extension during basin formation. Early ideas proposed in the seventies involved mostly crustal dynamics (e.g. Tapponnier 1977), were superceded by m o r e recent models involving postorogenic gravitational collapse (Platt & Vissers 1989; Platt 1993; Gautier & Brun 1994a; Jolivet et al. 1994a), and mantle dynamics (Doglioni 1991; Wortel & Spakman 1992; Spakman et al. 1993; Platt & England 1994). Some of these ideas concerning dynamics were already in the literature
much earlier ( B e r c k h e m e r 1977; Le Pichon 1981), but it is only quite recently that the data sets have emerged to critically test model predictions. At the same time, several detailed 2D reconstructions at various scales were published based on plate kinematic data and geological information (Dercourt et al. 1986; Dewey et al. 1989; Dercourt et al. 1993; Ricou 1994). Such reconstructions have been used as tests of the 3D structure of the mantle obtained from tomography (de Jonge et al. 1993). Recent findings come from new tomographic techniques that provide a detailed view of the mantle heterogeneities, from space geodesy that leads to precise determinations of the present strain and velocity fields, from the combination of structural geology, radiometric dating and m e t a m o r p h i c petrology which allows the description of P-T-t-D paths of exhumed metamorphic rocks, and from exploration geophysics, onshore and offshore.
Lithospheric and crustal structure Seismic t o m o g r a p h y and seismic anisotropy recently changed our understanding of the Mediterranean region. A 3D view of the mantle
JOLWET,L., FRlZON DE LAMOTTE,D. MASCLE,A. & SI~RANNE,M. 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen - an introduction. In: DURAND,B., JOLIVET,L., HORVATH,F. & St~RANNE, M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156,1-14.
2
L. JOLIVET
ETAL.
"--~ 4..a
0
E . ~, .-...
g~z< 4....'
~.~
o~
.
9, ~
.~
~ ~ o=o o
~o
~
~
~-~ o
~a. ~
~-~
r
~:~ ,..~
~::~ . ~
INTRODUCTION
oZ
oz
oZ
3
oZ
oZ
oZ
o
o
~
/ o
~
9
o
o
I.~
9
-' . ,,,,~ 4..,a
.9
f
~
o
"G =
t~
~
8
r
o
oz
~
oZ
oZ
oz
Z
o
'q
4
L. JOLIVET E T A L .
is now available for the whole region with some remaining uncertainties. High seismic velocity anomalies extend seismic slabs at depths down to 1200 km. As shown on Fig. 1, deep slabs are seen everywhere below the actively extending domains though they are not continuous (de Jonge et al. 1993, 1994). Only a small dense body is seen east of the Pannonian Basin. Most slabs imaged by this technique suggested slab break off (de Jonge etal. 1994). Only the Aegean slab appears continuous down to 1000 km or more. More detailed analysis were performed in the Apennine region (Amato et al. 1993, 1998; Chiarraba & Amato 1996; Mele et al. 1997) with sometimes different results. They suggest the presence of a continuous high velocity anomaly in the upper mantle below the northern Apennines, contrasting with the absence of such anomaly below the central Apennines. The presence of the vertical cold slab below the northern Apennines is consistent with the occurrence of deep earthquakes at mantle depth at this latitude. A low velocity anomaly is observed at lower crustal depth below the most elevated part of the central Apennines (Chiarraba & Amato 1996). This suggests high temperature in the lower crust right above the cold slab recognized before where deep earthquakes occur. A similar stratification has been documented along a N-S transect across the Alboran Sea with an intermediate aseismic domain separating two seismic zones in the upper crust and upper mantle (Seber et al. 1996). The distribution of seismicity in the upper crust shows important variations from the frontal zones of subduction zones to back-arc regions. The seismic-aseismic transition is deeper close to the slab than in the back-arc domain. This has been observed across the northern Apennines and Tyrrhenian Sea as well as across the Hellenic arc (Amato et al. 1993; Hatzfeld et al. 1993). Two seismic anisotropy profiles were obtained across the northern and central Apennines (Margheriti et al. 1996; Amato et al. 1998). They reveal a progressive change in the detailed structure of the lithosphere from the foreland domain to the backarc region. The fast direction is parallel to the NW-trending belt in the frontal domain and is progressively rotated toward a more E - W strike in the back-arc region, from Tuscany to Elba island and Corsica. This E-W direction is parallel to the strike of stretching lineations in the exhumed metamorphic rocks along the same transect (Fig. 1). Lithospheric thickness below the Pannonian Basin reflects an Early to Mid-Miocene
extensional event rather than the present-day dynamics which has been compressional since the Pliocene (Tari et aL this volume). Tertiary oceanic crust is present in the LiguroProvencal Basin (Burrus 1984) and southern Tyrrhenian Sea (Kastens et al. 1988). Crustal thickness in the Liguro-Provenqal Basin is only about 5 km on average which is thinner than typical oceanic crust. This suggests a mantle temperature lower than normal and a slow spreading rate (Chamot-Rooke et al. this volume). Wide-angle refraction and the ECORS reflection profiles, as well as inversion of gravity data (De Voogd et al. 1991; Pascal et al. 1993) show the progressive attenuation of crustal thickness across the Gulf of Lion margin. If moderate extension, mostly thin-skinned, is seen in the upper margin, a classical geometry of tilted blocks and intracrustal detachments as well as ductile thinning of the lower crust are observed in the offshore domain. These data provide a detailed image of the lithosphere and the crust. High-temperature anomalies are often present in the lower crust and upper mantle below the actively extending regions suggesting a sharp change in the thermal regime after the formation of a crustal scale thrust wedge. The extension seen at the surface is parallel to the fast direction of seismic waves in the deeper portions of the lithosphere suggesting that the whole lithosphere is deformed in a similar way during extension and that the field observations at the surface are significant at lithospheric scale.
Present-day kinematic framework Global kinematic data (De Mets et al. 1990) show a slow convergence between Africa and Eurasia in a NNW-SSE direction. The rate of convergence increases from less than 5 mm a-1 in the west to approximately 1 cm a -I in the east. Space geodesy provides kinematic data within the Mediterranean deformed region. GPS and SLR measurement in the Aegean show the almost rigid extrusion of the Anatolian block delimited to the north by the North Anatolian Fault, to the west by a series a grabens (such as the Corinth Gulf) and to the south by the Hellenic subduction (Le Pichon et al. 1994; Kahle et al. 1995, 1996; Reilinger et al. 1995; Noomen et al. 1996). This pattern is very similar to the earlier model proposed by McKenzie (1978). The Anatolian block rotates counterclockwise relative to Eurasia about a pole located north of Sinai. The instantaneous rate of westward expulsion of Anatolia is much higher than the rate of convergence between Eurasia and Africa so that
INTRODUCTION the velocity of subduction in the Hellenic trench is mostly controlled by the Anatolia-Africa relative motion which amounts to more than 3 cm a -1. N-S extension in the Gulf of Corinth results from the accomodation of the same motion with respect to stable Eurasia. More than 1.5 cm a -1 of extension are calculated across the Gulf from the regional data set, and approximately the same amount is measured directly across the Gulf (Rigo 1994). Velocities are much slower in the western Mediterranean. Measurement around the Tyrrhenian Sea reveals a slow N-S shortening in the whole region except in the SE (southern Apennines) where a strong E - W extension is measured (Ward 1994). Studies of active faults, focal mechanisms and breakout analysis show stress and strain patterns in the Aegean and Tyrrhenian regions (Armijo et al. 1992, 1996; Frepoli & Amato 1997; Hatzfeld et al. 1990, 1993; Jackson 1994; Montone et al. 1997; Hatzfeid this volume). Radial compression is active at the periphery of extensional basins. Compression is active along the Hellenic trench and its strike changes from E N E - W S W in the west to NNE-SSW in the east (Hatzfeld et al. 1993, 1997; Hatzfeld this volume). Radial compression is also observed around the northern Apennines (Frepoli & Amato 1997). A radial pattern of the compressional stress axis is also recorded in the Pannonian Basin (Gerner et a L this volume). E - W extension is predominant in the northern Apennines west of the crest line. Most of the southern Apennines are under an extensional regime. N-S extension predominates in the Aegean Sea, and is mostly active along the western margin of the Anatolian block as in the Corinth Gulf, whereas arc-parallel extension is recorded in Crete and Peloponnese (Hatzfeld et al. 1993, 1997). The present-day kinematics of the Aegean region is strongly influenced by the westward extrusion of the Anatolian block and the most active extension is recorded in the Gulf of Corinth (Le Pichon et al. 1994). In this example extrusion tectonics strongly shapes the extensional region (McKenzie 1972). Elsewhere the present-day kinematic pattern shows radial extension in the internal zones and compression in the external zones.
Major structures and kinematic evolution Apart from the North Anatolian Fault, which results from the extrusion of the Anatolian block out of the Arabia-Africa collision zone (McKenzie 1972, 1978) and the Mid-Hungarian Shear Zone (Gyfrfi et aL this volume, van Balen
5
et aL this volume), there are no large-scale strike-slip faults in the Mediterranean domain during the Tertiary. Some minor strike-slip component is recorded locally along some normal faults in the Apennines. The Kefallonia fault transfers the convergence from the front of the Mediterranean ridge to the Ionian islands (Le Pichon et al. 1997). A strike-slip component is also postulated along some NE-trending faults in the eastern Betic Cordillera and in the eastern Rif (Frizon de Lamotte et al. 1991; Leblanc & Olivier 1984). Contractional structures are often highly reworked by late orogenic extension. A description of the pre-extension structure and tectonic history is provided by Okay & Tiiysiiz (this volume) from an example in Western Turkey. Radial compression is the rule around major basins and mountain belts with the notable exception of the Atlas and the Pyr6n6es. All belts show an outward migration of the foreland domain (Malinverno & Ryan 1986; Patacca & Scandone 1989; D'Offizi et al. 1994; Jolivet et al. 1994b, 1998; Linzer 1996; Lonergan & White 1997; Fodor et aL this volume; Gy6rfi et al. this
volume; Jolivet and Patriat this volume; Verges & S~bat this volume; Zeck this volume). Lithospheric extension is predominantly E - W in the Western Mediterranean, N-S in the Eastern Mediterranean and E - W in the Pannonian basin. The first large-scale structure to form is the West European Rift System which runs from the east Alboran basin to the Rhine Graben (Brunet al. 1992, S~ranne this volume). This Oligocene structure appears to have formed independently from the backarc basins which developed afterward (S~ranne this volume). It is oblique to the alpine thrust front which is clearly cut in Provence in the eastern Alboran region (Doglioni et al. 1997). The Liguro-Proven~al basin and the Valencia Trough developed mainly on normal thickness continental crust (Mauffret et al. 1995). However its northern part (Gulf of Lion and Ligurian Sea) formed on a continental crust which had been previously thickened during the formation of the Pyr6n6es, Languedoc and Provence mountain chains and during the formation of the Alps (Corsica). The width of the margins varies greatly and the pre-extension crustal thickness might explain these variations (S~ranne this volume). The Liguro-Proven~al basin can be considered to display the characteristics of a normal oceanic basin (Gueguen 1995). Large-scale palaeomagnetic rotations are documented in and around all basins. They were first discovered in the Aegean region (Laj et al.
6
L. JOLIVET E T A L .
1982; Kissel & Laj 1988) and have been subsequently found in the Southern Apennines (Sagnotti 1992; Scheepers et al. 1993; Mattei et al. 1995), in the Pannonian Basin (Gy6rti et a L this volume; van Balen et ai. this volume) and the Betic Cordillera (Allerton et al. 1993; Allerton 1994). If a complex pattern of rotations is seen in the Alboran domain, the Aegean, Tyrrhenian and Pannonian Basins show rather simple outward rotations forming the present-day arcuate geometry. The Alboran, Tyrrhenian and Aegean Seas and the Pannonian Basin formed as postorogenic basins superimposed on compressional structures (Dewey 1988; Faccenna et al. 1997; Mauffret & Contrucci, this volume; Tari et ai. this volume). Extension is distributed within large domains and is associated with the exhumation of high temperature metamorphic core complexes (Lister et al. 1984; Dinter & Royden 1993; Sokoutis et al., 1993; Gautier & Brun 1994b; Jolivet et al. 1998; Platt & England 1994; Jolivet & Patriat this volume). The direction of extension recorded in these deep-seated core complexes is parallel to the present-day extension vector. As a whole, vertical-axis rotations have not perturbed significantly the fossil strain field during extension, it should be noted, however, that the direction of extension seen in Alpine Corsica should be rotated back to a more NW-SE direction before the opening of the Liguro-Provenqal basin (Montigny et al. 1981; Vigliotti & Kent 1990). The observed parallelism of ancient and modern extension directions is quite easy to understand in the northern Tyrrhenian margin where palaeomagnetic data show that no rotation has occurred (Mattei et al. 1996). It is more puzzling in the Aegean where clockwise rotations are documented onland and in several islands of the Cyclades archipelago (Kissel & Laj 1988). Some of the NE-trending lineations of the Cyclades have probably been rotated by 10-20 ~ clockwise after their exhumation (Avigad et al. 1998; Morris & Anderson 1996). This raises the question of the thickness of rotating blocks, whether the whole crust is affected by the rotations or only its upper part (Allerton 1993, 1994). During the most important stage of extension (Miocene) N-S or NE-SW extension is observed in the whole Aegean Sea from Crete to the northern Cyclades, while E - W extension predominates in the Tyrrhenian and Alboran seas (Jolivet et al. 1994b, 1998; Vissers et al. 1995). Extension is achieved by large-scale, steeply dipping, normal faults, which usually localize the largest earthquakes. These faults control the topography of the Aegean region in the Peloponese and
southeastern continental Greece (Jackson & White 1989; Armijo et al. 1992, 1996; Jackson 1994) and in the internal Apennines until the maximum of topography (D'Agostino et al. 1998). The most important topographic gradient of Corsica is a large, east-dipping, normal fault (Daniel et al. 1996). The internal domain of the Apennines, the northern Tyrrhenian sea, the Alboran sea, the eastern Betics and the Cyclade archipelago all display a basin and range topography controled by normal faults. Extension is also achieved along flat-lying normal faults and extensional shear zones. The sense of shear is constant over large regions (100-200 km wide). The hanging wall is always displaced toward the external zones in the northern Tyrrhenian Sea and internal Apennines (Jolivet et al. 1998) as well as in the BeticRif orogen (Vissers et al. 1995). It is instead displaced toward the internal zones in the Aegean Sea (Gautier & Brun 1994a; Jolivet & Patriat this volume). A continuum is observed from ductile to brittle extension. The most spectacular examples occur in the Aegean Sea where shallow, north-dipping ductile shear zones active in the Cyclades during the Miocene are relayed by more recent north-dipping, steep normal faults which root into shallow, north-dipping d6collements within the brittle-ductile transition zone (Rigo et al. 1996; Jolivet & Patriat this volume). Extension reactivates earlier thrusts and basins can develop above ramp and flat decollements such as seen, for example, along the rifted margin of the Gulf of Lion (Benedicto et al. this volume). A continuum is also observed from the deformation in the frontal zones near the subduction slab where high-pressure and low-temperature core complexes are exhumed below detachments (Crete) (Fassoulas et al. 1994; Jolivet et al. 1996, Jolivet & Patriat this volume) and in the back-arc region where high-temperature and low-pressure core complexes come to the surface below detachements with the same geometry (Cyclades) (Buick 1991; Gautier et al. 1993). This continuum is associated with a migration of extension and compression toward the external regions (Malinverno & Ryan 1986; Lonergan & White 1997; Jolivet et al. 1998; Jolivet & Patriat this volume) and by a coeval migration of magmatism, at least in the Tyrrhenian and Aegean regions (Serri et al. 1993) and the Pannonian Basin (Linzer 1996). The migration of deformation regimes and styles, and from the Miocene to the present are significant observations that should be included in lithospheric-scale models.
INTRODUCTION
Rifting history Inception of extension
7
(GyOrfi et aL this volume; van Balen et al. this volume). They were deposited during fast
Subduction is thought to be the major driving mechanism for compression and extension in the Mediterranean and its history can be controlled by the related magmatism (Wageman et al. 1970; Fytikas et al. 1984; Serri et al. 1993;
outward rotations of the Carpathian arc, counterclockwise in the north, clockwise in the south. Back-arc extension and eastward migration of the thrust front is then recorded during an eastward slab retreat and progressive locking of the northern border of the basin.
Sacchi et aL this volume; Wilson & Bianchiui this volume). The history of rifting is summar-
35 Ma. Oceanic lithosphere (Fig. 3a) subducts
ized in four figures spanning the recent evolution from the Oligocene to the Present (Fig. 3). They are modified from Dercourt et al. (1993, 1986). Slab rollback seemingly starts to produce back-arc extension around 30 Ma in the whole Mediterranean region. In the Western Mediterranean the opening of the Liguro-Provenqal, Alboran and Tyrrhenian basins can be described as a continuum starting some 30 Ma ago in the Gulf of Lion. The Late Oligocene and Early Miocene extensional structures in Alpine Corsica (Jolivet et al. 1998) formed in the time gap that exists between the Liguro-Provenqal and Tyrrhenian rifting episodes. Extension in Corsica started approximately 30 Ma ago and lasted until the Mid-Miocene at least. ChamotRooke et aL (this volume) also argue for a continuum of extension based on a new compilation of palaeomagnetic data and crustal structure. S~ranne (this volume) in his overview on the Gulf of Lion and Mascle & Vially (this volume) conclude that the back-arc rifting episode should be differentiated from the earlier formation of the West European Rift which started earlier. Extension in the Alboran Sea started in the Aquitanian (23-20 Ma) (Comas et al. 1992, 1996; Lonergan & White 1997) as a propagation toward the southwest of the back-arc basin initiated in the Gulf of Lion.
Zeck (this volume), Verges & S~bat (this volume) and Doglioni et al. (this volume) describe kinematic models which take into account this progressive extensional history. The timing of extension in the Aegean Sea is less well constrained. The first marine basins were formed in the Cyclades in the Aquitanian and the oldest radiometric dates in high-temperature metamorphic core complexes are around 25 Ma (Altherr et al. 1982; Gautier & Brun 1994a; Wijbrans & McDougall 1988). Extension should have started early enough to allow the exhumation of metamorphic rocks and a significant decrease of crustal thickness. The migration of subduction-related volcanism suggests a rollback from 30 Ma to the present (Fytikas et al. 1984; Jolivet et al. 1998). Early syn-rift deposits in the Pannonian Basin date back to the Early Miocene or Late Oligocene
below the Aegean domain and the Western Mediterranean region. Tomographic data suggest that the Hellenic subduction was initiated at least 40 Ma ago (Spakman et al. 1993) which significantly differs from classical interpretations which suggest that subduction started only in the Neogene. The subducting oceanic lithosphere is of Mesozoic age, probably Cretaceous (Truffert-Luxey 1992) and can thus subduct easily below the southern margin of Eurasia. Slab rollback starts soon after the initiation of subduction. The first extensional event is the formation of the West European Rift
(S~ranne this volume). 20 Ma. In the early Miocene (Fig. 3b), extension
is widespread in the A e g e a n Sea and the Western Mediterranean. Oceanic crust is forming in the Liguro-Provenqal basin. Intracontinental extension shapes the Aegean Sea and the west Alboran Basin. The first marine basins form in Corsica after the collapse of the Alpine belt and the formation of the Tyrrhenian Sea has started. Syn-rift sediments are deposited in the Pannonian Basin and the Carpathian arc rotates outward. Crustal collapse toward active subduction zones appears to control the geometry of extension. 10 M a to Present. The Tyrrhenian Sea (Fig. 3c,
d) is actively rifting between the Apennines and the Corsica-Sardinia block. Oceanic crust eventually forms in the southeastern part in the Pliocene. Extension in the Aegean progressively localizes along the western limit of the Anatolian block while the North Anatolian Fault propagates toward the west and the trench. The extrusion of the Anatolian block toward the low-stress boundary of the Hellenic subduction controls the kinematic and deformation regimes in the Aegean region. Tectonic inversion occurs in the Pannonian Basin (van Balen et al. this volume) progressively from 10 to 5 Ma (Juhasz et al. this volume; Sacchi et aL this volume). The recent compression explains the peripheral upilft and central subsidence of the Pannonian Basin (van Balen et all. this
volume).
8
L. JOLIVET E T A L .
Fig. 3. Tectonic scenario from the Oligocene to the present modified from Dercourt et al. (1993, 1986). Light shade represents the submerged continental crust, and darker shade Mesozoic oceanic crust. Cenozoic oceanic crust is shown in black.
The dynamics of extension Three mechanisms are generally proposed to explain the observed extension in the Mediterranean region. We do not include here models which relate the formation of extensional basins to strike-slip faulting because although such features may influence the recent evolution of the Aegean Sea (Armijo et al. 1996), they cannot account for the major part of extension which predates the formation of the North Anatolian Fault (Gautier et al. 1998). Collapse o f an overthickened crust. The widespread distribution of extension in the Aegean and Alboran Sea, together with the presence of metamorphic core complexes with high-temperature metamorphic overprint is consistent with a weak continental crust (Platt & Vissers 1989; Platt 1993; Gautier & Brun 1994a; Jolivet et al. 1994a; Vissers et al. 1995). The progressive heating of a thick crust following thickening will alter its rheology and lower its overall resistance leading to distributed extension similar to that
observed in the Basin and Range Province (Wernicke 1992). Similar behaviour can also be observed in the northern Tyrrhenian Sea where lower crustal melts and high temperature metamorphic core complexes have also been described (Jolivet et al. 1994b). In this case the extensional domain is narrower at any given time period compared to the sudden general collapse of the Aegean or the Betic-Rif orogen. The three basins show a progressive evolution from a cold toward a warm geotherm associated to crustal collapse and extension. The deep crust below the Apennines is apparently also warmer than expected (Chiarraba & A m a t o 1996) and the whole belt seems now to be collapsing. Crustal collapse is likely to be one important cause of extension. It is believed, however, that it is other processes such as slab rollback, slab detachment, or convective removal that are ultimately responsible for extension and which increase the heat flow and the potentiality of the continental crust to spread laterally.
INTRODUCTION Slab rollback, slab detachment and convective removal Distinguishing between slab rollback
and convective removal (detachment of the dense lithospheric root (Fleitout & Froidevaux 1982; Houseman et al. 1981)) is difficult in terms of thermal and mechanical consequences in the crust. Both processes will lead to the replacement of cold lithospheric material by hot asthenosphere below the mountain belt. Similar consequences in terms of crustal resistance, magmatism and extension are to be expected. The only real difference is the outward migration of magmatism and extension in the case of slab rollback. Convective removal is thought by some authors to be the major process responsible for the formation of the Alboran Sea (Platt & England 1994; Vissers et al. 1995; Seber et al. 1996), although slab rollback has also been proposed to explain the westward translation of the Alboran domain (Frizon de Lamotte et al. 1991; Royden 1993; Lonergan & White 1997). In the case of the northern Tyrrhenian Sea, slab rollback is obviously an active and important process, based on the observed migration of magmatism, extension and frontal compression processes (Jolivet et al. 1998; Malinverno & Ryan 1986; Patacca et al. 1990; Serri et al. 1993). In the case of the Aegean Sea a migration of volcanism is also seen from north to south (Fytikas et al. 1984) but the extensional event seems more widespread. The crustal collapse component may be more important in this region. Numerical and analogue experiments have shown how slab rollback and gravitational forces cooperate to control the extensional process (Faccenna et al. 1996; Giunchi et al. 1994). The slab rollback component can only increase with time, except when slab detachement occurs, while forces due to gravitational collapse can only decrease through time. Slab detachement may have some consequences that are easier to recognize such as a possible isostatic rebound of the upper and lower plates, although this has been challenged by numerical modelling studies (Giunchi et al. 1994, 1996). Slab breakoff has otherwise consequences quite similar to those of convective removal and slab retreat in terms of thermal evolution and magmatism (von Blanckenburg & Davies 1995). Another consequence is the along-strike migration of the foreland basin and the progressive concentration of the slab-pull force on the still attached region during progressive tear. This has been applied in the case of the Tyrrhenian Sea (Wortel & Spakman 1992). Multiple slab detachment events can also be envisaged for the evolution of the western Mediterranean, each trigerring a pulse of extension (Carminati et al. 1998).
9
Thus, across-strike migration of magmatic and extensional events suggest slab rollback or progressive delamination of the lithospheric mantle (and part of the lower crust), whilst along strike migration of depocenters may be used to infer slab detachement (van der Meulen et al. 1998). In any case, there may be several ways of testing deep processes by looking at the surface geology. Sedimentary basins and petroleum exploration.
Thick and quite diverse sedimentary basins are present in the Mediterranean area. The onshore basins have been explored for hydrocarbons over a long time period and some are petroleum provinces of quite significant economic importance (Panonian basin, the Apennines, Northern Africa). The offshore area has been little explored and, apart from some successes in the southern Adriatic Sea and Nile Delta, most of the Mediterranean basins and margins can be regarded as underexplored. Water depths have been a serious limitation in the past but the polyphase geological history and structural complexity have prevented a proper evaluation of their petroleum potential. The papers in this volume provide the necessary geodynamic and structural framework for the future assessment of these basins (Ziegler & Roure this volume). Broadly speaking, the Mediterranean basins fall into four categories. (1) The basins belonging to the African foreland to the south (Macgregor et al. 1998) have been only moderately affected by Tertiary compressional and extensional events. They include Palaeozoic basins of the northern African craton infilled with thick continental deposits with marine influences developing to the north. These basins are presently the focus of an extensive and successfull hydrocarbon exploration in the Algerian Sahara, following a first phase of field discoveries and developments in the 1960s. The Palaeozoic basins are overlain by a Mesozoic cover sequence that was laid down on the southern continental margin of the Tethys ocean. These basins are well developed onshore in northern Africa from Algeria to Egypt. Active exploration of Mesozoic sequences is currently active in Tunisia, Libya and Egypt, with spectacular success in the Western Desert of Egypt. The Tethyan Mesozoic basins are also well preserved in some offshore segments on the northern African margin such as the Gulf of Gabes, the Pelagian Sea, the Gulf of Sirte (with Tethyan oceanic crust possibly being preserved in the deepest parts of the Ionian Sea), below the
10
L. JOLIVET E T A L .
Nile delta and in the Adriatic Sea. Shallow water domains have been explored with only a limited amount of success (Gulf of Gabes). Little attention has focused on the deeper offshore area apart from the southern Adriatic Sea where the Aquila oil field has been developped in water depth in excess of 500 m. (2) Offshore accretionary wedges are developing and propagating southward as the last remaining fragments of the Tethyan oceanic crust are subducted northward below the southern edge of the Alpine orogen. These wedges form the Ionian ridge, the Mediterranean ridge and the Cyprus arc. They typically develop over a d6collement hosted in Messinian salt at their southern external edge, but deeper decollement are expected to occur below their northern inner edge. These wedges are poorly imaged by the seismic reflection data due to structural complexities and, locally, to mud volcanism. As a result they have been little explored and petroleum plays have still to be defined and tested. (3) Tertiary extension within the Alpine orogen has led to the opening of young oceanic basins (Western Mediterranean, Tyrrhenian Sea) or to the subsidence of basins floored by actively thinning continental crust (Aegean Sea and Pannonian Basin). These basins are the main focus of this volume and structural characteristics have been discussed above. Late stage alpine compressions have led locally to tectonic inversion sometimes with quite significant thrusting, as documented in the southern Alboran Sea (Chalouan et al. 1997). Petroleum exploration has only been successfull in the Pannonian Basin (Horvfith & Tari this volume), the other basins being probably devoid of source rocks of regional extent, although some local petroleum systems have been recognized (southern France, Masde & ViaUy this volume), or even large oil fields discovered and developped (Valencia Gulf). (4) Deltas and deep sea fans have developped in Neogene times in different structural settings: the Ebro and Rhone delta at the edge of the Western Mediterranean Neogene oceanic basin, and the Nile Delta superimposed over the Mesozoic Tethyan margin. Only the Nile Delta has proven so far to be an important petroleum province, with hydrocarbons originating from Oligocene source rocks and trapped within turbidites of Late Miocene to Pliocene age.
Conclusion Although some uncertainties remain concerning the nature of the seminal event that produced a sharp change from overall compression to backarc extension and compression some 30 Ma ago in the Mediterranean region, it is now certain
that this event originates in the mantle, either by slab detachement or slab rollback or both acting in sequence. Gravitational collapse of a thick crust may have influenced the development of most basins, and strike-slip faults, such as the North Anatolian Fault have also affected basin evolution recently. If the shape and kinematics of individual basin-mountain belt pairs is controlled by the local geometry of plate boundaries and the local history of crustal thickening, the late Oligocene event results from a change at a larger scale. Thus the 30 Ma event requires a change in mechanical or kinematic boundary conditions on the scale of the Mediterranean region. The subduction regime has changed somehow at this time. It is furthermore possible that one (or several) event(s) of slab detachment trigerred extension after the late Oligocene major reorganization.
References ALLERTON, S. 1994. Vertical-axis rotation associated
with folding and thrusting: an example from the eastern subbetic zone of Southern Spain. Geology, 22, 1039-1042. --,
LONERGAN,L., PLATT, J. P., PLATZMAN,E. S. &
MCCLELLAND, E. 1993. Paleomagnetic rotations in the eastern Betic cordillera, southern Spain. Earth and Planetary Science Letters, 119, 225-241. ALTHERR, R., KREUZER, H., WENDT, I., LENZ, H., WAGNER, G. A., KELLER,J., HARRE, W. & HOHN-
DORF,A. 1982. A Late Oligocene/Early Miocene high temperature belt in the anti-cycladic crystalline complex (SE Pelagonian, Greece). Geologische Jahrbuch, 23, 97-164. AMATO, A., ALESSANDRINI,B., CIMINI, G., FREPOLI,A. & SELVAGGI,G. 1993. Active and remanant sub-
ducted slabs beneath Italy: evidence from seismic tomography and seismicity. Annali di Geofisica, 36, 201-214. --,
MARGHERITI, L., ARAZZA, R., BASILI, A., CHIARABBA, C., CIACCO, M. G., CIMINI, G. B., D1 BONA, M., LUCENTE, E P., NOSTRO, C. & SEL-
VAGGI, G. 1998. Passive seismology and deep structure in central Italy. In: Geodynamics of the Lithosphere and the Earth's Mantle Pageoph. Special volume, in press. ARMIJO, R., LYON-CAEN,H. & PAPANIKOLAOU,D. 1992. East-west extension and Holocene normal fault scarps in the Hellenic arc. Geology, 20, 491-494. - - - , MEYER, B., KING, G. C. P., RIGO, A. & PAPANASTASSIOU, D. 1996. Quaternary evolution of the Corinth Rift and its implications for the Late Cenozoic evolution of the Aegean. Geophysical Journal International, 126, 11-53. AVlGAD,D., BAER,G. & HEIMANN,A. 1998. Block rotations and continental extension in the Central Aegean Sea: paleomagnetic and structural evidence from Tinos and Mykonos. Earth and Planetary Science Letters, in press. BENEDICTO, A., SI~GURET, M. & LABAUME, P. 1999.
Interaction between faulting, drainage and
INTRODUCTION sedimentation in extensional hangingwall syncline basins: example of the Oligocene Matelles basin (Gulf of Lions rifted margin, SE France). This volume. BERCKHEMER, H. 1977. Some aspects of the evolution of marginal seas deduced from observations in the Aegean region. In: (eds) International Symposium on Structural History of the Mediterranean Basins. Technip, Paris, 303-314. BRUN, J. P., GUTSCHER, M. A. & TEAMS, D.-E. 1992. Deep crustal structure of the Rhine Graben from D E K O R P - E C O R S seismic reflection data: a summary. Tectonophysics, 208, 139-147. BUICK, I. S. 1991. Mylonite fabric development on Naxos, Greece. Journal of Structural Geology, 13, 643-655. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provenqal basin (north-western Mediterranean). Marine Geology, 55, 247-269. CARMINATI,E.,WORTEL, M. J. R., SPAKMAN,W. & SABADINI, R. 1998. The role of slab detachment processes in the opening of the western-central Mediterranean basins: some geological and geophysical evidence. Earth and Planetary Science Letters, 160, 651-665. CHALOUAN,A., SAJI, R., MICHARD,A. & BALL5",A. W. 1997. Neogene tectonic evolution of the Southwestern Alboran Basin as inferred from seismic data off Morocco. A A P G Bulletin, 81, 1161-1184. CHAMOT-ROOKE, N., GAULIER,J.-M. & JESTIN,E 1999. Constraints on Moho depth and crustal thickness in the Liguro-Proven~al basin from a 3D gravity inversion: geodynamic implications. This volume. CHIARRABA, C. ~ AMATO, A. 1996. Crustal velocity structure of the Apennines (Italy) from P-wave travel time tomography. Annali di Geofisica, 39,
1133-1148. COMAS, M. C., GARCfA-DUE/qAS,W. & JURADO, M. J. 1992. Neogene tectonic evolution of the Alboran Sea from MCS data. GeoMarine Letters, 12, 157-164. COMAS, M. C., ZANH, R. & KLAUS A. ETAL. 1996. Leg 161 - Western Mediterranean. Initial Reports of the Ocean Drilling Program, 161. D'AGOSTINO, N., CHAMOT-ROOKE,N., FUNICIELLO,R., JOLIVET,L. & SPERANZA,F. 1998. The role of preexisting thrust faults and topography on the styles of extension in the Gran Sasso range (Central Italy). Tectonophysics, 292, 229-254. D'OFFIZI, S., MINELLI, G. & PIALLI, G. 1994. Foredeeps and thrust systems in the northern Apennines. Bolletino di Geofisica Teorica ed Applicata, 36, 141-144. DANIEL, J. M., JOLIVET, L., GOFFI~, B. & POINSSOT, C. 1996. Crustal-scale strain partitionning: footwall deformation below the Alpine Corsica OligoMiocene detachment. Journal of Structural Geology, 18, 41-59. DE JONGE, M. R.,WORTEL, M. J. R. & SPAKMAN,W. 1993. From tectonic reconstruction to upper mantle model: an application to the Alpine-Mediterranean region. Tectonophysics, 223, 53-65. --, -& -1994. Regional scale tectonic evolution and the seimic velocity structure of the lithosphere and upper mantle: the Mediterranean
11
region. Journal of Geophysical Research, 99, 12091-12108. DE METS, C., GORDON, R. G., ARGUS, D. E & STEIN, S. 1990. Current plate motions. Geophysical Journal International, 101, 425-478. DE VOOGD, B., NICOL1CH,R., OLIVET,J. L. ETAL. 1991. First deep seismic reflection transect from the Gulf of Lions to Sardinia (ECORS-CROP profiles in Western Mediterranean). In: MEISSNER,R., BROWN, L., DURBAUM,H. J., FUCHS, K. & SEIFER, E (eds) Continental lithosphere: deep seismic reflections. American Geophysical Union, Geodynamic Series, 22, 265-274. DERCOURT,J., RJCOU,L. E. & VRIELINCK,B. 1993. Atlas Tethys Palaeo environmental Maps. Paris, Gauthier-Villars. --, ZONENSHAIN, L. E, RICOU, L. E. ET AL. 1986. Geological evolution of the Tethys belt from the Atlantic to the Pamir since the Lias. Tectonophysics, 123, 241-315. DEWEY, J. F. 1988. Extensional collapse of orogens. Tectonics', 7, 1123-1139. - - , HELMAN, M. L., TORCO, E., HUTTON, D. H. W. & KNOTT, S. D. 1989. Kinematics of the Western Mediterranean. In: COWARD,M. E, DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. DINTER, D. A. & ROYDEN, L. 1993. Late Cenozoic extension in northeastern Greece: Strymon valley detachment system and Rhodope metamorphic core complex. Geology, 21, 45-48. DOGLIONI, C. 1991. A proposal for the kinematic modelling of W-dipping subductions-possible applications of the Tyrrhenian-Apennines system. Terra Nova, 3, 423-434. --, GUEGUEN, E., HARABAGLIA,P. & MONGELLI, F. 1999. On the origin of W-directed subduction zones and applications to the Western Mediterranean. This volume. --, SABAT, E ~; FERNANDEZ, M. 1997. The Western Mediterranean extensional basins and the Alpine orogen. Terra Nova, 9, 109-112. FACCENNA, C., DAVY, P., BRUN, J. P., FUNICIELLO, R., GIARDINI, D., MATTEI, M. & NALPAS,Y. 1996. The dynamic of backarc basins: an experimental approach to the opening of the Tyrrhenian Sea. Geophysical Journal International, 126, 781-795. , MATTEI, M., FUNICIELLO,R. & JOLIVET,L. 1997. Styles of back-arc extension in the Central Mediterranean. Terra Nova, 9, 126-130. FASSOULAS, C., KILIAS, A. & MOUNTRAKIS, D. 1994. Postnappe stacking extension and exhumation of high-pressure/low-temperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. FLEITOUT, L. & FROIDEVAUX, C. 1982. Tectonics and topography for a lithosphere containing density heterogeneities. Tectonics, 1, 21-56. FODOR, L., CSONTOS, L., BADA, G., GYORFI, Y. d~; BENKOVICS,L. 1999. Tertiary tectonic evolution of the Pannonian basin system and neighbouring orogens: a new synthesis of palaeostress data. This volume. FREPOLI, A. & AMATO, A. 1997. Contemporaneous extension and compression in the Northern
12
L. JOLIVET E T A L .
Apennines from earthquakes plane solutions. Geophysical Journal International, 129, 368-388. FRIZON DE LAMOTTE, D., ANDRIEUX, J. & GuI~zou, J. C. 1991. CinGmatique des chevauchements NGogGnes dans l'arc bGtico-Rifains, discussion sur les modGles gGodynamiques. Bulletin de la Socidt~ Gdologique de France, 4, 611-626. , POISSON, A., AUBOURG, C. & TEMIZ, H. 1995. Chevauchements post-tortoniens vers l'ouest puis vers le sud au coeur de l'angle d'Isparta (Taurus, Turquie). ConsGquences gGodynamiques. Bulletin de la Socidtd Gdologique de France, 166, 59-67. FYTIKAS, M., INNOCENTI, E, MANETTI, P., MAZZUOLI, R., PECCERILLO,A. & VILLARI,L. 1984. Tertiary to Quaternary evolution of volcanism in the Aegean region. In: ROBERTSON,A. H. E & DIXON, J. E. (eds) Geological Evolution of the Eastern Mediteranean. Geological Society, London, Special Publications, 17, 687-699. GAUTIER, E & BRUN, J. P. 1994a. Crustal-scale geometry and kinematics of late-orogenic extension in the central Aegean (Cyclades and Evvia island). Tectonophysics, 238, 399-424. - & -1994b. Ductile crust exhumation and extensional detachments in the central Aegean (Cycladcs and E w i a islands). Geodinamica Acta, 7, 57-85. - & JOLIVET,L. 1993. Structure and kinematics of upper Cenozoic extensional detachement on Naxos and Paros (Cyclades Islands, Greece). Tectonics, 12, 1180-1194. - - , MORICEAU,R., SOKOUTIS,D., MARTINOD,J. JOLIVET, L. 1999. Aegean extension: timing, kinematics and cause. Tectonophysics, in press. GERNER, P., BADA, G., DOVI~NYI, P., MI3LLER, B., ONCESCU, M. C., CLOETINGH, S. & HORVATH, E 1999. Recent tectonic stress and crustal deformation in and around the Pannonian basin: data and models. This volume. GIUNCHI, C., GASPERINI, P., SABADINI, R. & D'AGOSTtNO, G. 1994. The role of subduction on the horizontal motions in the Tyrrhenian basin: a numerical model. Geophysical Research Letters, 21,529-532. --, SABADINI, R., BOSCHI, E. ~r GASPERINI, P. 1996. Dynamic models of subduction: geophysical and geological evidence in the Tyrrhenian Sea. Geophysical Journal International, 126, 555-578. GUEGUEN, E. 1995. Le bassin Liguro-Provenqal: un vdritable ocdan. Exemple de segmentation des marges et de hiatus cindmatiques. Implications sur le processus d'amincissement crustal. Thesis, Universit6 de Bretagne Occidentale. GYORFI, I., CSONTOS, L. & NAGYMAROSY, A. 1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian basin. This volume. HATZFELD, D. 1999. The present-day tectonics of the Aegean as deduced from seismicity. , BESNARD, M., MAKROPOULOS,K. ~% HATZIDIMITRIOU, P. 1993. Microearthquake seismicity and fault-plane solutions in the southern Aegean and its geodynamic implications. Geophysical Journal International, 115, 799-818. , MARTINOD,J., BASTET,G. & GAUTIER,P. 1997. An
analog experiment for the Aegean to descibe the contribution of gravitational potential energy. Journal of Geophysical Research. 102, 649-659. - - , PEDOTTI, G., HATZIDIMITRIOU,P. ~; MAKROPOULOS, K. 1990. The strain pattern in the western Hellenic arc deduced from a microearthquake survey. Geophysical Journal International, 101, 181-202. HORVATH, E & TARI, G. 1999. IBS Pannonian Basin Project: a review of its main results and their bearing on hydrocarbon exploration. This volume. HOUSEMAN, G. A., MCKENZIE, D. E & MOLNAR, P. 1981. Convective instability of a thickened boundary layer and its relevance for the thermal evolution of the continental crust. Journal of Geophysical Research, 86, 6115-6132. JACKSON, J. 1994. Active tectonics of the Aegean region. Annual Review of Earth and Planetary Science, 22, 239-271. & WHITE, N. J. 1989. Normal fauting in the upper continental crust: observations from regions of active extension. Journal of Structural Geology, 11, 15-36. JOLIVET, L. & PATRIAT,M. 1999. Ductile extension and the formation of the Aegean Sea. This volume. , BRUN, J. P., GAUTIER, P., LALLEMANT, S. & PATRIAT,M. 1994a. 3-D kinematics of extension in the Aegean from the Early Miocene to the Present, insight from the ductile crust. Bulletin de la SocidtO Gdologique de France, 165, 195-209. - - , DANIEL, J. M., TRUFFERT, C. d~. GOFFI~,B. 1994b. Exhumation of deep crustal metamorphic rocks and crustal extension in back-arc regions. Lithos, 33, 3-30. - - , FACCENNA,C., GOFFI~,B., MATTEI,M., ROSSETTI, E, BRUNET,C., STORTI,E, FUNICIELLO,R., CADET, J. P. & PARRA,T. 1998. Mid-crustal shear zones in post-orogenic extension: the northern Tyrrhenian Sea case. Journal of Geophysical Research, 103, 12 123-12 160. --, GOFFI~, B., MONII~, P., TRUFFERT-LUXEY, C., PATmAT, M. & BONNEAU, M. 1996. Miocene detachment in Crete and exhumation P-T-t paths of high pressure metamorphic rocks. Tectonics, 15,
1129-1153.
JUHASZ, E., PHIIJJPS, L., M()I,LER, P., RICKETTS, B., TOTH-MAKK, g,., LANTOS, M. & KovAcs, L. O. 1999. Late Neogene sedimentary facies and sequences in the Pannonian basin, Hungary. This volume. KAHLE, H. G., MULLER, M. V., GEIGFR, A., DANUSER, G., MUELLER, S., VEIS, G., BUJJRIS, H. ~ PARADIsis, D. 1995. The strain field in NW Greece and the Ionian Islands: results inferred from GPS measurements. Tectonophysics, 249, 41-52. , & VElS, G. 1996. Trajectories of crustal deformation of western Greece from GPS observations 1989-1994. Geophysical Research Letters, 23, 677-680. KASTENS, K., MASCLE, J., AUROUX, C., BONATTI, E., BROGLIA, C., CHANNELL, J., CURZI, P., EMEIS, K. C., GLAGON, G., HASEGAWA, S., HIEKE, W., MASCLE, G., F McCoY, E, MCKENZlE, J., MENDELSON, J., MOLLER, C., RI~HAULT, J. P., ROBERTSON, A., SARTORI, R., SPROVIER, R. &
INTRODUCTION TORIt, M. 1988. O D P Leg 107 in the Tyrrhenian Sea: insight into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KISSEL, C. & LAJ, C. 1988. The Tertiary geodynamic evolution of the Aegean arc: a paleomagnetic reconstruction. Tectonophysics, 146, 183-201. LAJ, C., JAMET, M., SOREL, D. & VALENTE, J. P. 1982. First paleomagnetic results from Mio-Pliocene series of the Hellenic sedimentary arc. Tectonophysics, 86, 45-67. LE PICHON, X. 1981. Land-locked oceanic basins and continental collision, the eastern Mediterranean as a case example. In: HsO, K. J. (ed.) Mountain Building Processes. Academic Press, London, 201-211. - - , CHAMOT-ROOKE,N., LALLEMANT,S. L., NOOMEN, R. & VEIS, G. 1994. Geodetic determination of the kinematics of Central Greece with respect to Europe: implications for eastern Mediterranean tectonics. Journal of Geophysical Research, 100, 12675-12690. - - , LALLEMANT,S., CHAMOT-ROOKE,N., LEMEUR,D. & PASCAL, G. 1999. The Mediterranean ridge backstop: a consequence of the Hellenic nappes seaward migration. Marine Geology, in press. LEBLANC, D. & OLIVIER, R 1984. Role of strike-slip faults in the Betic-Rifian orogeny. Tectonophysics, 101, 345-355. LINZER, H. G. 1996. Kinematics of retreating subduction along the Carpathian arc, Romania. Geology, 24, 167-170. LISTER, G. S., BANGA, G. & FEENSTRA,A. 1984. Metamorphic core complexes of cordilleran type in the Cyclades, Aegean Sea, Greece. Geology, 12, 221-225. LONERGAN, L. & WHITE, N. 1997. Origin of the BeticRif mountain belt. Tectonics, 16, 504-522. MACGREGOR, D. S., MOODY, R. J. & CLARK-LOWES,D. D. (eds) 1998. Petroleum Geology of North Africa. Geological Society, London, Special Publications, 132. MALINVERNO, A. & RYAN, W. 1986. Extension in the Tyrrhenian sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere. Tectonics, 5, 227-245. MARGHERITI, L., NOSTRO, C., COCCO, M. & AMATO,A. 1996. Seismic anisotropy beneath the Northern Apennines (Italy) and its tectonic implications. Geophysical Research Letters, 23, 2721-2724. MASCLE,A. & VIALLY,R. 1999. The petroleum system of the Southeast Basin and Gulf of Lions (France). This volume. MATTEI, M., FUNICIELLO,R. & KISSEL, C. 1995. Paleomagnetic and structural evidence for Neogene block rotations in the Central Apennines, Italy. Journal of Geophysical Research, 100, 17863-17883. MATTEI, M., K1SSEL, C., SAGNOTTI,L., FUNICIELLO,R. & FACCENNA, C. 1996. Lack of Late Miocene to Present rotation of the northern Tyrrhenian margin (Italy): a constraint on geodynamic evolution. In: MORRIS,A. & TARLING,D. H. (eds)
Palaeomagnetism and Tectonics of the Mediterranean Region. Geological Society, London, Special Publications, 105, 141-146.
13
MAUFFRET,A. & CONTRUCCI,I. 1999. Crustal structure of the North Tyrrhenian Sea: first results of the multichannel seismic LISA cruise. This volume. , PASCAL, G., MAILLARD, A. & GORINI, C. 1995. Tectonics and deep structure of the north-western Mediterranean basin. Marine and Petroleum Geology, 12, 645-666. MCKENZIE, D. 1972. Active tectonics in the Mediterranean region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. - 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions.
Geophysical Journal of the Royal Astronomical Society, 55, 217-254. MELE, G., ROVELLI, A., SEBER, D. & BARAZANGI, M. 1997. Shear wave attenuation in the lithosphere beneath Italy and surrounding regions: tectonic implications. Journal of Geophysical Research, 102, 11863-11875. MONTIGNY, R., EDEL, J. B. & THU1ZAT,R. 1981. OligoMiocene rotation of Sardinia: K-Ar ages and paleomagnetism data of Tertiary volcanics. Earth and Planetary Science Letters, 54, 261-271. MONTONE, E, AMATO,A., FREPOLI,A., MARIUCCI,M. T. & CESARO,M. 1997. Crustal stress regime in Italy. Annali di Geofisica, 4 0 , 741-757. MORmS, A. & ANDERSON, A. 1996. First paleaomagnetic results from the Cyclaic Massif, Greece, and their implications for Miocene extension directions and tectonic models in the Aegean. Earth and Planetary Science Letters, 142, 397-408. NOOMEN, R., SPRINGER, T. A., AMBROSIUS, B. A. T., HERZBERGER, K., KUIJPER, D. C., METS, G. J., OVERGAAUW, B. & WAKKER, K. E 1996. Crustal deformations in the Mediterranean are computed from SLR and GPS observations. Journal of Geodynamics, 21, 73-96. OKAY,A. & TOYSIDZ,0.1999. Tethyan sutures of northern Turkey. This volume. PASCAL, G. P., MAUFFRET,A. & PATRIAT,P. 1993. The ocean-continent boundary in the Gulf of Lion from analysis of expanding spread profiles and gravity modelling. Geophysical Journal International, 113, 701-726. PATACCA, E., SARTORI, R. & SCANDONE, P. 1990. Tyrrhenian basin and Apenninic arcs: kinematic relations since late Tortonian times. Memorie della Societd Geologica Italiana, 45, 425-451. - & SCANDONE, S. 1989. Post-Tortonian mountain building in the Apennines. The role of the passive sinking of a relic lithospheric slab. In: BORIANI, A., BONAFEDE, M., PICCARDO, G. B., VAI, G. B. (eds) The lithosphere in Italy. Advances in science research. Accademia Nazionale dei Lincei, Rome, 157-176. PLATY,J. P. 1993. Exhumation of high-pressure rocks: a review of concept and processes. Terra Nova, 5, 119-133. -& ENGLAND, P. 1994. Convective removal of lithosphere beneath moutain belt: thermal and mechanical consequences. American Journal of Science, 294, 307-336. & VISSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: A working hypothesis for the Alboran Sea and Gibraltar arc. Geology, 17, 540-543.
14
L. J O L I V E T E T A L .
REILINGER, R., MCCLUSKY, S., KAHLE, H. O., MULLE, M. V., STRAUB, C., KASTENS, K., GILBERT, L., HURST, K., VEIS, M., PARADIS1S,D. & BARKA, A. 1995. GPS evidence for westward continuation of dextral strike-slip motion from the North Anatolian Fault Zone across the North Aegean and North-Central Greece. Abstract. American Geophysical Union (AGU). EOS, Supplement, Nov. 7, F620-F621. RIcou, L. E. 1994. Tethys reconstructed: plates, continental fragments and their boundaries since 260 Ma from Central America to south-eastern Asie. Geodinamica Acta, 7, 169-218. RIGO, A. 1994. Etude sismotectonique et gOoddsique du golfe de Corinthe. Thesis, Universit6 Denis Diderot, Paris. --, LYON-CAEN, H., ARMIJO, R., DESCHAMPS, A., HATZFELD, D., MAKROPOULOS, K., PAPADIMITRIOU,P. & KASSARAS,I. 1996. A microseismicity study in the western part of the Gulf of Corinth (Greece): implications for large-scale normal faulting mechanisms. Geophysical Journal International, 126, 663-688. ROYDEN, L. H. 1993. Evolution of retreating subduction boundaries formed during continental collision. Tectonics, 12, 629-638. SACCHI,M., HORVATH,E & MAGYARI,O. 1999. Role of unconformity-bounded units in stratigraphy of continental record: a case study from the Late Miocene of western Pannonian basin, Hungary. This volume. SAGNOrrI, L. 1992. Paleomagnetic evidence for a Plio-Pleistocene counterclockwise rotation of the Sant'Arcangelo basin, Southern Italy. Geophysical Research Letters, 19, 135-138. SCHEEPERS, P. J. J., LANGEREIS, C. G. & HILGEN, E J. 1993. Counter-clockwise rotations in the southern Apennines during the Pleistocene: paleomagnetic evidences from the Matera area. Tectonophysics, 225, 379-410. SEBER, D., BARAZANGI, M., IBENBRAHIM, A. & DEMNATI, A. 1996. Geophysical evidence for lithospheric delamination beneath the Alboran Sea and Rif-Betic mountains. Nature, 379, 785-790. SI~RANNE, M. 1999. The Gulf of Lions continental margin (NW Mediterranean) revisited by IBS: an overview. SERRI, G., INNOCENTI, E & MANETTI, P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the NeogeneQuaternary magmatsim of central Italy. Tectonophysics, 223, 117-147. SOKOUTIS, D., BRUN, J. P., DRIESSCHE, J. V. D. & PAVLIDES, S. 1993. A major Oligo-Miocene detachment in southern Rhodope controlling north Aegean extension. Journal of the Geological Society, London, 150, 243-246. SPAKMAN,W., VAN DER LEE, S. & VAN DER HILST, R. 1993. Travel-time tomography of the EuropeanMediterranean mantle. Physics of the Earth and Planetary Interiors, 79, 3-74. TAPPONNIER, P. 1977. Evolution du syst6me alpin en M6diterran6e: poinconnement et 6crasement
rigide-plastique. Bulletin de la Socidt~ Geologique de France, 7 (19), 437-460. TARI, G., DOVI~NYI, P., HORV,~TH, E, DUNKL, I., LENKEY, L., STEFANESCU,M., SZAFIAN,P. & TOTH, T. 1999. Lithospheric structure of the Pannonian basin derived from seismic, gravity and geothermal data. This' volume. TRUFFERT-LUXEY,C. 1992. De la compression de la ride mOditerrandenne d l'extension en Mer Egde: gdodynamique de la MdditerranOe orientale. Thbse, Universit6 Pierre et Marie Curie. VAN BALEN, R. T., LENKEY, L., HORVA,TH, E & CLOETINGH, S. A. P. L. 1999. Towdimensional modeling of stratigraphy and compaction driven fluid flow in the Pannonian basin. This volume. VANDER MEULEN,M. J., MEULENKAMP,J. E. & WORTEL, M. J. R. 1998. Lateral shifts of Apenninic foredeep depocentres reflecting detachment of subducted lithosphere. Earth and Planetary Science Letters, 154, 201-218. VERGES, J. & SABAT,F. 1999. Contraints on the western Mediterranean kinematic evolution along a 1000 km transect, from Iberia to Africa. This volume. VIGLIOrrl, L. & KENT, D. V. 1990. Paleomagnetic results of Tertiary sediments from Corsica: evidence for post-Eocene rotation. Physics of the Earth and Planetary Interiors, 62, 97-108. VISSERS, R. L. M., PLATr, J. P. & VAN DER WAL, D. 1995. Late orogenic extension of the Betic Cordillera and the Alboran domain: a lithospheric view. Tectonics, 14, 786-803. VON BLANCKENBURG,E & Huw DAVIES,J. 1995. Slab breakoff: a model for syncollisional magmatism and tectonics in the Alps. Tectonics, 14, 120-132. WAGEMAN, J. M., HILDE, T. W. C. & EMERY, K. O. 1970. Structural framework of the of the East China Sea and Yellow Sea. American Association of Petroleum Geologists Bulletin, 54, 1611-1643. WARD, S. N. 1994. Constrains on the seismotectonics of the central Mediterranean from Very Long Baseline Interferometry. Geophysical Journal International, 117, 441-452. WERNICKE, B. 1992. Cenozoic extensional tectonics of the U.S. cordillera. In: BURCHFIEL,B. C., LIPMAN, P W. & ZOBACK, M. L. (eds) The Cordilleran Orogen: Conterminous U.S. Geological Society of America, Geology of North America, G-3, 553-581. WIJBRANS, J. R. & McDOUGALL, I. 1988. Metamorphic evolution of the Attic Cycladic Metamorphic Belt on Naxos (Cyclades, Greece) utilizing 4~ age spectrum measurements. Journal of Metamorphic Geology, 6, 571-594. WILSON, M. & BIANCHINI, G. 1999. Tertiary-Quaternary magmatism within the Mediterranean and surrounding regions. This" volume. WORTEL, M. J. R. & SPAKMAN,W. 1992. Structure and dynamic of subducted lithosphere in the Mediterranean. Procedings of the Koninks Nederlandse Akadamie voor Wetenschappen, 95, 325-347. ZECK, H. P. 1999. Alpine kinematic evolution in the W Mediterranean: a westward directed subduction regime followed by slab roll-back and slab detachment. This volume.
The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: an overview MICHEL
St~RANNE
GOophysique Tectonique SOdimentologie, Universit( Montpellier 2, cc. 060, 34095 Montpellier cedex 05, France (e-mail." seranne@dstu, univ-montp2.fr) Abstract: The Gulf of Lion margin is one of the Tertiary extensional basins of the western
Mediterranean that opened during convergence of Africa and Europe. This Oligocene-Aquitanian rifted margin and associated Burdigalian oceanic basin have been used as case study for stretching models of 'Atlantic-type' margins. However, when the Integrated Basin Study (IBS) project was initiated, several outstanding questions remained about the present structure and the geodynamic setting of the margin within the Western Mediterranean. IBS-Gulf of Lion research was based on the existing onshore and offshore, industrial and academic data, which were heterogeneous and unevenly distributed. Compilation of the stratigraphic correlations on a regional scale allowed precise calculation of the timing of rifting, and clarification of the relationships with Alpine and Mediterranean geodynamics. Reprocessing of the existing ECORS deep seismic reflection profiles shed new light on the extensional structure and mechanisms of extension of the continental margin. Structural and sedimentological studies onshore led to the definition of new tectonostratigraphic models for extensional basins. Results of structural analyses showed a partitioning of the extensional deformation processes across the continental margin. 3D gravity modelling of the margin and basin area led to the production of a new map of the Moho depth by inversion, and testing several hypotheses for the origin of the present day subsidence. Although the Gulf of Lion margin displays structural and stratigraphic features similar to 'Atlantic-type' margins, its structure and evolution corresponds to that of a rifted margin of a large continent formed during the opening of a marginal basin. Integration of the new results of IBS-Gulf of Lion within the geodynamic evolution of the western Mediterranean suggests that the Oligocene rifting of the Gulf of Lion represents the initial stage of a succession of rifting events and back-arc basin formation, due to continuously retreating subduction during convergence of Africa and Europe.
'Atlantic-type' divergent passive margins are a prime site for hydrocarbon accumulation, because of optimum geological conditions generated during their evolution (e.g. Edwards & Santogrossi 1989). As new models of basin formation and basin-fill architecture were developed, it appeared that 'Atlantic-type' models did not account for the evolution of many extensional margins. For example, a number of extensional sedimentary basins are formed during or shortly after m o u n t a i n building (e.g. S6ranne & Malavieille 1994; Cloetingh et al. 1995); in SE Asia and in the Mediterranean/Alpine region, continental extension leads to oceanic crust accretion, whilst the bordering lithospheric plates converge at rate of several centimetres per year (e.g. D e w e y et al. 1989; Rangin et al. 1990). The geodynamics of such extensional basins imply specific geological conditions controlling oil accumulation, which need to be understood in order to produce successful exploration models. One of the modules of the Integrated Basin Studies program addressed the problems of the
formulation of extensional basins in convergent settings. T h e Gulf of Lion in the Western Mediterranean was chosen as a natural laboratory where the structure and evolution of a fully developed rifted continental margin was formed during convergence of the African and Eurasian plates. The Pannonian Basin is the other natural laboratory chosen by IBS where continental extension occurred within the Alps (Horvfith & Tari this volume). The Gulf of Lion (Fig. 1) is located between the Valencia Trough to the west and the Provenqal margin of the Ligurian Sea to the east. These basins represent the rifted continental margin of SW Eurasia, whose conjugate margins have drifted southward or southeastwards, above the northwest-dipping subduction of the African and related plates (Apulia). This contribution aims to give an overview of the structure and development of the Gulf of Lion within the tectonic f r a m e w o r k of the Western Mediterranean, in the light of the new results reached by IBS. The tectonics of the Gulf of Lion are then compared with examples of
SERANNE,M. 1999. The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: An overview.
In: DURAND,B., JOLIVET,L., HoRvATN, E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 15-36.
16
M. SI~RANNE
Fig. 1. Present-day seismic-tectonic map of the Mediterranean-Alpine region (modified from Reba'f et al. 1992), showing the location of the Oligocene-Recent extensional basins formed during N-S convergence of Africa and Eurasia. classic 'Atlantic-type' margins and of back-arc basins.
The western Mediterranean setting The present-day structure of the Mediterranean region (Fig. 1) results from the overall convergence of Africa and Eurasia which involved several local and successive rifting and collision episodes during the Tertiary (Le Pichon et al. 1971; Auzende e t al. 1973; Tapponnier 1977; Biju-Duval et al. 1978; R6hault et al. 1984; Dercourt el at. 1985; Dewey et al. 1989). In the eastern Mediterranean, part of the convergence is still accommodated by northward subduction of Mesozoic Tethyan oceanic crust. On the other hand, the western Mediterranean is floored by oceanic crust formed during Neogene time as a
consequence of rifting and drifting of small (100 km wide) continental blocks. Formation of such extensional basins occurred simultaneously with thrusting and mountain building in the surrounding areas, during continued northward motion of Africa with respect to Eurasia. The exact kinematics and chronology of the events resulting in the present-day western Mediterranean is still intensely discussed; the number of unsolved questions increases as one goes back further in time (for example see the contrasting E o c e n e reconstructions by Biju-Duval e t al. 1978; Boillot et al. 1984; Dercourt et al. 1985; Dewey et al. 1989; Bois 1993). However there is a consensus on the main features and on the succession of tectonic stages which makes it possible to sketch out the Tertiary evolution of the western Mediterranean (Fig. 2).
Fig. 2. Sketch of the geodynamic evolution of the NW Mediterranean since Eocene. Compiled from different sources, including Bellon (1976); Coulon (1977); Girod & Girod (1977); Wildi (1983); Boillot et al. (1984); R6hault et al. (1984); Bergerat (1985); Dercourt et al. (1985); Bouillin (1986); Dewey et al. (1989); Sartori (1990); Bartrina et al. (1992); Mart/et al. (1992); Reba/et al. (1992); Roca & Guimer~t (1992); Ziegler (1992); Keller et al. (1994); Thompson (1994); Tricart et al. (1994); Mauffret et al. (1995); S6ranne et al. (1995); Olivet (1996); Saadallah & Caby (1996).
G U L F OF L I O N O V E R V I E W
17
18
M. SI2RANNE
Until late Eocene time (Fig. 2a, b), the Iberian plate was associated with several continental blocks characterized by Hercynian basement: the Balearics, Corsica, Sardinia, Calabria and the Kabylies blocks. To the SE, Iberia and the Hercynian blocks were bordered by a NWdipping subduction, south of which extended the African plate and derived continental blocks such as Apulia (Dercourt et al. 1985). To the north, Iberia and the Hercynian blocks were bounded by the North Pyrenean Fault Zone, along which they had been translated eastwards during the late Cretaceous opening of the Bay of Biscay (Olivet 1996). The northward motion of Iberia (and associated blocks) resulted in collision with the European plate, and the formation of the Pyrenean thrust and fold belt (Olivet 1996). The Languedoc-Provenqal segment of the Pyrenees results from the collision of Sardinia-Corsica with Provence. The kinematics east of this zone and the linkage with the Alps-Apennines are not clear, mostly due to the complex 3D geometry of the continental subductions converging in the 'Ligurian knot' (Laubscher et al. 1992). During latest Eocene (Priabonian, Fig. 2b), shortening proceeded between Iberia and Europe while E-W extension initiated intra-continental rift basins in western Europe (e.g. Rhine Graben, Bergerat 1985). The intermediate zone of the future Gulf of Lion, was affected by left-lateral strike-slip tectonics, involving formation of oblique extensional basins in releasing bends. From Rupelian time, the Hercynian blocks began to drift apart in several stages. During the first stage, in late Rupelian (Fig. 2c), continental rifting occurred between southern France and Corsica-Sardinia-Calabria, while the Balearics and Kabylies, located SW of the North Balearic Transfer Zone (Mauffret et al. 1995) or 'Accident Paul Fallot' (Durand-Delga & Fontbot6 1980) remained stable. In the Valencia Trough, rifting was delayed until latest Oligocene-Aquitanian times (Fig. 2d) (Batrina et al. 1992). Subduction-related calc-alkaline volcanism in Provence, Sardinia and the Valencia Trough (Bellon 1976; Coulon 1977; Girod & Girod 1977) indicate NW-dipping subduction of the Apulian plate beneath Calabria-SardiniaCorsica during Chattian to Burdigalian. The subduction system consisted of an accretionary prism located in Calabria (e.g. Thompson 1994), and a volcanic arc extending from Sardinia to Provence through the western margin of Corsica (Bellon 1976; Girod & Girod 1977). Intra-continental rifting in the Gulf of Lion-Ligurian basins was therefore in a backarc setting (R6hault et al. 1984). On this
transect, the onset of oceanic accretion is dated at latest Aquitanian-early Burdigalian (Fig. 2e) (Burrus 1984)and it is marked in the stratigraphic record by a break-up unconformity (Gorini et al. 1993). Oceanic accretion in the basin allowed the anticlockwise rotation of Corsica-Sardinia-Calabria. The pole and amount of rotation are still under discussion (see review in Vially & Tr6moli6res 1996). Palaeomagnetic data suggest a rotation of some 30 ~ which would have taken place between 21 and 19 Ma (Montigny et al. 1981; Burrus 1984). However, there is a growing body of evidence for a longer period of rotation and oceanic accretion lasting until latest Burdigalian-early Langhian (Fig. 2f) (Vigliotti & Kent 1990; Vigliotti & Langenheim 1995; Chamot-Rooke et al. this volume). Continental stretching on the margins and accretion of some 200 km of oceanic crust is accommodated by equivalent southeastward retreat of the subduction zone with respect to Europe, while between 35.5 Ma and 19.5 Ma Africa (at the longitude of Algiers) had a northwards motion of 160 km (Dewey et al. 1989). Southwest of the North Balearic Transform Zone, subduction occurred south of the Kabylies blocks. Calc-alkaline volcanism started in the Valencia Trough in the Aquitanian (Fig. 2d), i.e. with the same delay as the onset of extension with respect to the Gulf of Lion and lasted throughout the Burdigalian (Fig. 2e) (Mart/et al. 1992; Verg6s & Sabat this volume). Continental break-up occurred between the Balearics and the Kabylies during Langhian (Bartrina et al. 1992), leaving the Valencia Trough as an aborted intra-continental rift. The Kabylies blocks drifted southwards and were thrust over northern Africa (Wildi 1983; Bouillin 1986). The southward retreat of the subduction is confirmed by migration of the calcalkaline volcanic arc from the Valencia Trough to the Kabylies around Langhian time. Late Miocene to Recent magmatism in the Valencia Trough (Fig. 2f-h) is characterized by intraplate alkaline magmatism (Mart/et al. 1992). The second stage of evolution started during the mid-Miocene. After the end of oceanic accretion in the Gulf of Lion and Ligurian basins, and cessation of the rotation of Corsica-Sardinia, rifting occurred between the latter and Calabria (Fig. 2g, h). Continental extension gave way to oceanic accretion in the Tyrrhenian during late Miocene, both extension and axis of accretion migrated southeastward during Messinian to Pliocene (Kastens et al. 1988; Sartori 1990). By Tortonian, calc-alkaline volcanism in Sardinia had ceased and was
GULF OF LION OVERVIEW replaced by intraplate alkaline volcanism (Fig. 2g) (Bellon 1976; Coulon 1977). Subduction was still active, but the calc-alkaline volcanic arc had jumped to the Tyrrhenian Sea. The Calabrian accretionary wedge collided with the continental Apulian plate during this interval. In the northern Tyrrhenian, rifting did not reach continental break-up. Extension migrated eastward, so did compression in the accretionary wedge represented by the Apennines (Sartori 1990). Outward migration of the subduction system or subduction retreat is recognized throughout the Mediterranean (Jolivet et al. 1994) and it is explained by sinking of the dense lithospheric footwall of the subduction in the asthenosphere (Malinverno & Ryan 1986). Doglioni (1991) has suggested that such process is enhanced by westward-dipping subduction, due to the generalized eastward flow of the mantle with respect to lithospheric plates.
19
were located on structural highs, thus 'missing' the syn-rift sedimentary record (Cravatte et aL 1974); the refraction experiment covered the Ligurian Sea and stopped along the eastern side of the Gulf of Lion margin (Le Douaran et al. 1984); ECORS deep seismic profile, located along these ESPs, left unexplored the deepest rift basins and the corresponding deep crustal structure; besides, the Italian equivalent CROP profile off Sardinia is not across the exact conjugate margin (de Voogd et aL 1991). As a result, the structure of the Gulf of Lion margin remained a poorly understood segment of the NW Mediterranean compared to neighbouring Valencia Trough and Ligurian Sea, as illustrated by the blanks in structural maps of Mauffret et al. (1995).
A g e o f rifting: W e s t E u r o p e a n
rifts v e r s u s
back-arc basin
The recent acquisitions: IBS-Gulf of Lion The objectives of IBS-Gulf of Lion were to: (1) establish the 2D and 3D tectonics and kinematics of extension, (2) define tectono-sedimentary models of extensional basins, (3)investigate thermal and mechanical processes controlling lithospheric extension, in such complex settings. Along with other independent projects (e.g. Gorini 1993; Guennoc et al. 1994; Mauffret et al. 1995) one of the aims of the IBS-Gulf of Lion project, was to re-evaluate, develop and synthesize existing data. No data acquisition was planned except selective field observations. Instead, it was decided: (i) to compile and homogenize available data (bathymetry, topography, gravimetry, heat flow) in the Gulf of Lion and associated Alg6ro-Proven~al oceanic basin; (ii) to use recent synthetic structural maps based on commercial seismic data to be published by BRGM, (iii) to re-process the ECORS-Gulf of Lion deep seismic profile shot in 1988 which was still largely unexploited, (iv) to update the structural study of the onshore part of the margin, integrating the newly released seismic surveys. In spite of many studies undertaken during the 1970s and 80s, including thousands of kilometres of industrial and academic seismic profiling, industrial drillings, which turned the Gulf of Lion into a natural laboratory where subsidence (e.g. Steckler & Watts 1980), and basin models (e.g. Burrus & Audebert 1990) were tested, the formation and evolution of this continental margin was still largely unknown at the beginning of the 1990s. Many of the operations undertaken fell short of the expectations of the community. For example, all exploratory wells
The Gulf of Lion is seen either as a southern extension of the Tertiary West European Rift system (e.g. Bergerat 1985; Ziegler 1992) or as one of the back-arc basins developed above a NW-dipping subduction in the western Mediterranean (e.g. Auzende et al. 1973; Biju-Duval et al. 1978; R6hault et al. 1984; Mauffret et al. 1995). One way to distinguish between the two models is to document precisely the chronology of events. Stratigraphic correlations of the different onshore basins of the southern margin of Eurasia from Valle Penedes to Northern Vat half-grabens (Fig. 3) show remarkable correlation and contemporaneous unconformities related to otherwise identified geodynamic events. The West European Rift system initiated in the Rhine Graben (Fig. 4) during late Eocene time and extension migrated northward and southward (Ziegler 1992). The mid-Eocene tectonics of this area is characterized by left-lateral motion along the NNW-striking bordering faults, consistent with N-S 'Pyrenean' contraction (Bergerat 1985; Larroque 1987). East-west extension started during Priabonian time in the Rhine Graben (Larroque 1987; Maurin 1995), the Limagne (Morange et al. 1971; Burg et al. 1982; Bl6s et at,. 1989;Busson 1992), and the Bresse and Valence grabens (Debrand-Passard & Courbouleix 1984; Busson 1992). The earliest syn-rift sediments of the Camargue basin, only known from industrial boreholes, are not dated. However, these evaporites and shales have been correlated with the evaporite series of the Valence, Bresse and Rhine basins spanning the Priabonian-early Rupelian periods (DebrandPassard & Courbouleix 1984; Busson 1992). In
20
M. SI~RANNE
GULF OF LION OVERVIEW
Fig. 3. (a) Tectonostratigraphic correlations of major onshore rift basins of the northern margin of the Alg6ro-Provenqal basin, compared with the Sardinia conjugate margin, and the West European Rift system. Location of the basins is indicated in Fig. 4. Time scale from Cande & Kent (1992). Data compiled from different sources, including Morange et al. (1971); Cherchi & Montadert (1982); DebrandPassard & Courbouleix (1984); Valette (1991); Bartrina et al. (1992); Biondi et al. (1992); Busson (1992); Roca & Guimerh (1992); Ziegler (1992); Maerten & S6ranne (1995); Benedicto (1996). (b) Key. the Manosque-Forcalquier and A16s basins onset of syntectonic sedimentation along NNEand NE-trending faults respectively, occurred in late Priabonian time, following Pyrenean syntectonic sedimentation (Bartonian) (DebrandPassard & Courbouleix 1984; Bergerat 1985; Biondi et al. 1992). Finally, in the Campidano basin of Sardinia, which represents the Gulf of Lion conjugate margin, the earliest sediments date from the Priabonian (Cherchi & Montadert 1982). When Sardinia is replaced in a pre-rift position, Campidano basin-bounding faults strike NNE, similarly to the rifts on the European margin. E - W extensional stress along NEto NNE-trending faults involves oblique extension with a component of left-lateral strike-slip. Late Rupelian continental syntectonic sediments were deposited, after a hiatus, above an angular unconformity in small continental basins and in earlier basins formed during Eocene (Debrand-Passard & Courbouleix 1984) (Fig. 3). The late Rupelian unconformity can be correlated from southern France to Sardinia; however it is not documented in SW France and Catalunya, where syntectonic sedimentation related to extension started in late Chattian-early Aquitanian times (DebrandPassard & Courbouleix 1984; Bartrina et al. 1992). As a result of N W - S E extension during late Rupelian, northeast-trending fault-bounded
21
basins were formed, and earlier Priabonian basins were reactivated (Fig. 4). From these correlations it can be gathered that the West European Rift system was initiated in Priabonian time, during a phase of E - W extension. The large basins (Al6s, ManosqueForealquier, and probably Camargue) in Southern France were formed during this time by oblique extension. A second phase of rifting due to N W - S E extension started in late Rupelian in the Gulf of Lion area (from southern France to Sardinia). This later phase of extension is contemporaneous with intraplate alkaline volcanism in Languedoc and onset of calc-alkaline subduction-related volcanism in Sardinia (Bellon & Brousse 1977). It is concluded that rifting of the Gulf of Lion margin occurred in late Rupelian time, in relation to back-arc extension, as a geodynamical event distinct from the West European Rift. Spatial overlap of the two events in southern France resulted in the reactivation of Priabonian basins. S t r u c t u r e o f the m a r g i n a n d m o d e s o f crustal s t r e t c h i n g
The Gulf of Lion margin extends SE of the C6vennes Fault (Fig. 5). Extensional structures are predominantly oriented NE, and overprint the Pyrenean thrust and fold belt. The interaction of inherited Pyrenean structures with extension has been analysed by S6ranne et al. (1995). The margin is segmented by several transfer fault zones sub-parallel to extension. The most representative segment which is analysed in the following sections is located between the Arl6sienne and S6toise transfer zones (Fig. 5). Moho depth data (wide-angle refraction and the ECORS deep seismic reflection profiles) are mostly restricted to a dip section across the Gulf of Lion. The lateral crustal variability introduced by the along-strike segmentation of the margin could only be addressed by a 3D approach. Gravity data can be used to obtain reliable geometry of the Moho. Compilations and homogenization of gravity data over the Liguro-Proven~al basin resulted in the preparation of maps, presented in a companion paper (Chamot-Rooke et al. this volume). A totally new Moho depth map was derived by inversion of 3D gravity data. This map was successfully tested against the few available Moho depth measurements. A section across the continental margin (Fig. 6) allows to distinguish: (1) the poorly deformed upper-margin, located mostly onshore, (2)the continental shelf, (3) the slope and (4) the basin floored by intermediate or oceanic crust.
22
M. SF,R A N N E
Fig. 4. Structural map of Priabonian to Aquitanian syn-rift basins from the Rhine Graben to the Valencia Trough. The Gulf of Lion margin is located at the junction of the West European Rift system and the rifts of the NW Mediterranean. Although these basins are often associated in the literature, they differ by the age of rifting onset (Priabonian on one hand, and late Rupelian to Aquitanian on the other hand), structural trend (N-S and NE-SW), and orientation of extensional stress (E-W, and NW-SE). The two systems overlap in the onshore part of the Gulf of Lion, where the Camargue, Albs and Manosque-Forcalquier basins were initiated during Priabonian under E - W extensional stress, and were reactivated during Oligocene-Aquitanian times under NW-SE extensional stress. Compiled from R6hault et al. (1984); Bergerat (1985); Larroque (1987); Bartrina et al. (1992); S6ranne et al. (1995); Benedicto (1996). Bar, Barcelona; Mtp, Montpellier; Mar, Marseille.
GULF OF LION OVERVIEW
23
Fig. 5. Structural map of the Gulf of Lion margin. The rift basins are controlled by NE-trending extensional faults overprinting the E-W Pyrenean thrusts in Provence and Languedoc. Thin-skinned extension (light shading) characterizes the onshore part of the margin, whereas the shelf and the slope area are deformed by basement-involved tectonics (dark shading). The margin is segmented by transfer zones parallel to extension. The ECORS seismic profile and ESP's corresponding to the section Fig. 6 are indicated. Ma, Marseille; Ni, Nimes; Mo, Montpellier; Na, Narbonne; Pc, Perpignan. NPFZ, North Pyrenean Fault Zone; MFB, Manosque-Forcalquier Basin; CB, Camargue Basin; AB, Albs Basin; LMB, Les Matelles Basin; HB, H6rault Basin. Modified after Sdranne et al. (1995).
Upper margin. Landward, extensional deformation is bounded by the C6vennes Fault (Fig. 6). The Mesozoic-Eocene cover is affected by NE-striking, SE-facing normal faults bounding Oligocene or Early Miocene basins. The Jurassic and Neocomian limestones cropping out in Languedoc display n u m e r o u s micro-faults which have yielded extensional stress tensors with a m i n i m u m stress axis oriented N120 ~ (Arthaud et al. 1977), consistent with the basinformation event. Interpretation of industrial seismic reflection profiles indicates that the basin-bounding faults detach above the Palaeozoic basement, in Triassic evaporites and shales (Maerten & S6ranne 1995; Benedicto 1996). The underlying continental crust is therefore not affected by extension in this part of the margin. However, the crust is thinning towards the margin from 30 km beneath the south Massif Central (Sapin & Hirn 1974) to 20 km close to
the shoreline (de Voogd et al. 1991; ChamotR o o k e et al. this volume), suggesting either lower-crustal thinning during Oligocene rifting, or crustal stretching and thinning inherited from the Mesozoic rifting. Immediately east of the considered section, there is a good correlation between the isopaches of Mesozoic series (Debrand-Passard & Courbouleix 1984) and the Moho (Chamot-Rooke et al. this volume) which favours the second alternative. The onshore Gulf of Lion margin is therefore characterized by thin-skinned extensional tectonics of the prerift sedimentary cover, which accommodated only several kilometres extension. Continental shelf. Structurally, the continental shelf extends mainly offshore, SE of the basement-ramp in which thin-skinned extension is transferred. On the section presented (Fig. 6) this ramp is a reactivated E - W - o r i e n t e d
24
M. St~RANNE
Fig. 6. Section of the Gulf of Lion showing the distribution of tectonic style across the margin. Based on field observations and industrial seismic onshore data, and on the ECORS seismic profiled reprocessed during IBS project (Pascal et al. pers. comm.). Moho data from Pascal et al. (1993) and Chamot-Rooke et al. (this volume). Vertical = horizontal scale.
P y r e n n e a n thrust; however the NE-trending N~mes Fault (Fig. 5) corresponds to this ramp for most of the width of the studied area. Formation of the onshore syn-rift Camargue basin was controlled by the N~mes Fault. This basin comprises two half-grabens: the Vistrenque and the Petit Rh6ne grabens, and presents the thickest (>4000 m) and most complete syn-rift succession of the margin (see below). Seismic reflection profiles interpretation of the area undertaken during the course of IBS project has shown that the basinbounding NJmes Fault was a ramp dipping 25-30 ~ SE, into the upper crust; upwards, the fault cuts through the Mesozoic cover at a higher angle (Fig. 6) (Benedicto et al. 1996). Offshore, the structures of the continental shelf are known from the E C O R S deep seismic profile, and from industrial seismic reflection (Gorini 1993; Gorini et al. 1994). Beneath the continental shelf the Moho is 25 to 20 km deep. The major extensional faults dipping 25-30 ~ can be traced in the upper crust and merge with the 5 km thick reflective lower crust (S6ranne et al. 1995) (Fig. 6). They strike parallel to the N~mes Fault (Gorini et al. 1993) and bound thin (<1 km) and discontinuous syn-rift basins. The pre-rift basement consists of Palaeozoic low-grade rocks or late Variscan granite overlain by lower Miocene sediments (Cravatte et al. 1974; Arthaud et al. 1981). It is important to note that (i) high-grade rocks have never been found on the continental shelf, and (ii) the Sirocco granite has not been rejuvenated during either Pyrenean orogeny or Oligocene rifting (whole rock and mineral Rb/Sr age of 280 Ma) (Cravatte et al. 1974). This indicates that the major low-angle normal faults
have not exhumed deep levels of the crust in the continental shelf. The hiatus of the Mesozoic cover accounted by probably initial thin series which were eroded during the Pyrenean orogeny (Arthaud & S6guret 1981). Estimates of horizontal extension across the 110 km wide continental shelf area, computed by the chevron method on the normal faults imaged by the E C O R S profile, range between 15 and 20 km (1.16 < [3 < 1.22). Taking into account possible footwall erosion and faults unresolved by seismic, such figures must be taken as minimal values. Present-day mean crustal thickness (from pre-rift to Moho) of the continental shelf is 18 km, assuming a beta factor of 1.16 to 1.22, leads to a pre-rift crustal thickness of 25.5 to 27 kin. Such low value is not in agreement with the presence of the Pyrenean orogen, constrained by numerous observations (Arthaud & S6guret 1981; S6ranne et al. 1995; Mauffret & Gorini 1996; Vially & Tr6moli6res 1996). This discrepancy could result from a phase of post-orogenic extensional collapse (Gaulier et al. 1994) and crustal attenuation in late Eocene, prior to Oligocene rifting characterized by strike-slip deformation (Mauffret & Gennesseaux 1989). Reflective lower crust within which merge the large extensional faults, as imaged on the E C O R S may be compared to BIRPS profiles around the British Isles (Klemperer & Hobbs 1991), where upper crustal extension by normal faulting decouples in the lower crust subjected to ductile flow (Reston 1990). In that case, a significant part of crustal thinning in the Gulf of Lion continental shelf could be taken up by ductile stretching of the lower crust.
GULF OF LION OVERVIEW
T h e slope. Definition of modes of crustal extension in the slope area relies heavily on interpretation of the E C O R S - G u l f of Lion profiles; however, the profiles had not been processed further than the initial commercial processing provided by the operator CGG, which greatly limited the reliability of the interpretations (de Voogd et al. 1991). In the first stage, stack and post-stack FK migrations were performed (Truffert et al. 1993) using velocity model derived from ESP data reprocessed by (Pascal et al. 1993). A second set of processing consisted in pre-stack depth migration, that focused on the slope area of the margin (Pascal et al. 1994), which gave rise to new interpretations and geodynamic implications, presented in S6ranne et al. (1995). The slope area is characterised by a rapid rise of the Moho between ESP 202 and 203 (Fig. 6), the lack of highly reflective lower crust above the Moho, and tilted blocks separated by low-angle faults merging into a sub-horizontal reflector. The latter is interpreted as a detachment similar to the 'S' reflector in the Bay of Biscay (Le Pichon
25
& Barbier 1987) or in the Galicia margin (Beslier et al. 1993). Geometry of upper crustal faults sug-
gests extension of 12 km to 17 km, depending on the age of the graben in the North Pyrenean Fault zone (see discussion in Gorini et al. 1993; S6ranne et al. 1995; Benedicto et aL 1996) distributed over the 50 km wide slope area. Comparison of such moderate extension ratio (1.32 < [3 < 1.51) with the important crustal stretching (12 km mean crustal thickness over the slope area) points to either very thin pre-rift crustal thickness (16-18 km), or more likely to lower crustal thinning processes. At the base of the slope, a landward dipping detachment ('T' reflector) (Pascal et al. 1993)cuts across the lower crust. This detachment exhumes either lower crustal material from beneath the continental margin (S6ranne et al. 1995) or mantle material and exposes it in the deep basin, similarly to models of lithospheric mantle denudation (Boillot et al. 1988). The former hypothesis accounts better for the extreme crustal stretching in the slope area than the latter.
Fig. 7. Tectonic models of rift basins observed on the Gulf of Lion margin. Their relation with the tectonic style allows to predict their position on the margin. See text for comments.
26
M. SI~RANNE
The basin extends basinwards of the emergence of the 'T' reflector. The <5 km thick crust presents seismologic features intermediate between continental and oceanic or mantelic material (Pascal et al. 1993; Pascal pers. comm.). The continent-ocean boundary has been placed at different locations (Le Douaran et al. 1984; Mauffret et al. 1995). According to velocity analyses associated with IBS reprocessing of an E C O R S profile, oceanic crust is identified beyond ESP 205. This thin (3-4 km) early oceanic crust suggests low ratio of partial melting in the mantle related to a low potential temperature, as discussed by Chamot-Rooke et aL (this volume). The basin basement is therefore an exhumed lower crust or lithospheric mantle north of ESP 205 and accreted oceanic crust beyond. Extensional basins models
Reappraisal of existing geological data, backed by newly released industrial seismic reflection profiles in the landward end of the margin shows that thin-skinned extensional tectonics dominates the onshore margin (S6ranne et al. 1995; Vially & Tr6moliSres 1996). Detailed structural and sedimentological analyses of the onshore basins (Philibert 1992; Maerten 1994; Benedicto 1996; Sanchis 1997) allowed the definition of several types of thin-skinned extensional basins (Fig. 7). D ~ c o l l e m e n t basin. An original d6collement
model for the formation of extensional sedimentary basins has been proposed for the H6rault Basin (Maerten 1994) (Fig. 7). A synrift basin is formed by extensional faulting of the pre-rift cover above a d6collement level located in the Triassic shales and evaporites. The extensional d6collement emerges at surface level above inherited high-angle faults down-faulting the basement. These faults were active during Mesozoic extension, but remained inactive during Oligocene rifting. Unlike other interpretations (Route et al. 1992, 1994; Mascle et al. 1995), accommodation and depocentre migration does not result from folding of a hanging-wall flat stripped off the footwall, into a hanging-wall syncline, but from out-of-sequence extensional faulting of the pre-rift cover. Consequently, offset along the basement-cover interface is reduced (2.7 km for the H6rault Basin instead of 6.5 km in a hanging-wall syncline model), in better agreement with regional sections (Maerten & S6ranne 1995). These d6collement basins are usually located along the margins of the Mesozoic extensional basin, characterized by important offset of the
basement-cover interface (e.g. H6rault Basin, or Manosque-Forcalquier Basin, Benedicto 1996). In these basins, syntectonic clastic sedimentation rapidly passes to lacustrine organicrich limestones and marls which have good hydrocarbon potential. However, clastic sedimentation derived from erosion of Mesozoic carbonates provides rather poor quality reservoirs (Vially & Tr6moli~res 1996; Mascle & Vially this volume). Thin-skinned extensional tectonics favoured the formation of shallow sedimentary basins (depth limited by depth of d6collement) with probable low geothermal potential (the continental crust was not stretched). The thickest syntectonic deposits are expected where the Mesozoic cover was the thickest, and/or where d6collement levels are into the late Palaeozoic sediments. For example, the A16s depocentre, where the Mesozoic section is over 3 km thick and where d6collement occurs in the Triassic or late Palaeozoic sediments (Roure et al. 1992), is presently over 2 km thick and an extra 1 km has been eroded, following inversion (Roy & Tr6molibres 1992). As a result, organic matter in that basin has reached the oil window (Mascle & Vially this volume). Hanging-wall syncline. Most of the small basins
located north of Montpellier (Fig. 5) display a similar architecture (Benedicto 1996). They consist of an asymmetric syncline with the Mesozoic-Eocene sequence on the SE limb and a reduced Neocomian or Eocene steeply dipping NW limb, against a footwall ramp cutting across Jurassic limestones. In the core of the syncline syn-rift sediments present growth structures (Fig. 7). Accommodation results from extension across a ramp in the Mesozoic sequence associated with a shallow flat in the Neocomian or Eocene sequence, that forms the hanging-wall syncline. The basin-fill of Les Matelles Basin (Fig. 5) offers the opportunity to validate the kinematics of hanging-wall syncline formation (Ellis & McClay 1988) by reconstructing the growth structures in association with the evolution of the drainage (Benedicto et al. this volume). The basin fill is very thin and filled with proximal clastics derived from the hanging-wall flat (Mesozoic and Eocene carbonate) and lacustrine limestones (Benedicto et al. this volume). Half-graben on basement ramps. In the domain
of basement involved extension, half-grabens are formed along the major normal faults, and the Camargue basin (Fig. 5) is the most representative example. It was formed by extension along the landwardmost extensional basement
GULF OF LION OVERVIEW
27
o v~ rm
,..-.,,
o
o
&z
~.~ %)
#~
o ~
~
~ o
o ~
o
~
='
~
~ ~
=
~-
.~
Z
28
M. Sl~RANNE
fault of the margin, whose dip-slip offset is increased by the motion along the d6collement of the thin-skinned extension. The Camargue basin represents the thickest syn-rift depocentre of the margin, comprising 4000 m of Early Oligocene and Aquitanian syn-rift deposits. Detailed analyses of industrial seismic reflection profiles and of borehole data (Benedicto et al. 1996) show that the N~mes fault, which bounds the graben to the NW, forms a low-angle basement ramp at depth. The basin fill consists of a transgressive series ranging from continental lacustrine silts and lagoonal evaporites to shallow marine clays (Valette 1991), which presents poor reservoir qualities (Vially & Trdmoli~res 1996). The varying mode of deformation and geometry of the hanging wall controlled the development of gravity-driven thrusts within the evaporitic formation which increased the thickness of halite, presently extracted by Elf-Atochem (Valette & Benedicto 1995). Basinwards of the Camargue basin, syn-rift basins also display diverging sedimentary fill in fault-bounded grabens. However, they are much thinner than the Camargue basin, probably as a result of elevated topography at the onset of rifting (Sdranne et al. 1995). Although the synrift sequence has not been drilled, it is likely that it includes syntectonic clastic sediments interfingering with lacustrine limestones representing a good potential source-rock. Unlike basins located on the thin-skinned extensional domain, siliciclastic sediments derived from the erosion of the surrounding crystalline and metasedimentary rocks of the basement (Fig. 7), have good reservoir potential, which make these grabens a prime target for exploration (Vially & Tr6moli6res 1996).
Atlantic-type
margin versus back-arc
margin
The width and thickness of the extended continental crust, the amount of extension, and total subsidence of the Gulf of Lion margin does not significantly diverge from these of non-volcanic Atlantic-type margins (Table 1). However, rifting is much younger, and the period of rifting (at most 10 Ma) is shorter than in the Central and South Atlantic continental margins. Consequently, rates of extension and of subsidence, averaged over the duration of margin evolution, are higher than for Atlantic margins. Previous subsidence analyses of the Gulf of Lion margin consisted of backstripping of commercial wells (Watts & Ryan 1976; Steckler & Watts 1980), and later, of 2D backstripping of
sections across the margin (Bessis 1986; Burrus & Foucher 1986; Burrus et al. 1987). They all found that the amount of post-rift subsidence was large with respect to the moderate syn-rift subsidence on one hand, and with respect to the young age of the rifting on the other hand. In addition, these studies revealed that uniform stretching calibrated on present-day crustal thickness adequately accounts for subsidence of the shelf, whereas the slope and deep basin records 1 km excess subsidence (Burrus & Audebert 1990). Such different patterns of subsidence for the shelf and the slope cannot be accounted for by a single set of initial pre-rift conditions across the margin (Gaulier et al. 1994). This is in agreement with the recognition that the slope of the margin extends SE of the North Pyrenean Fault (S6ranne et al. 1995) which is a terrane boundary separating the European and the Sardinia-Corsica lithospheres, giving different rheologies. New constraints brought by the reprocessing of the ECORS profile (Pascal et al. 1994) and gravity modelling of the Alg6ro-Provenqal basin (Chamot-Rooke et al. this volume) show that the initial oceanic crust observed at the continent-ocean boundary was significantly thinner than for Atlantic rifted margins (Table 1), which is consistent with small amounts of partial melting of the mantle beneath the stretched zone, and consequently indicate low potential temperature of the mantle (McKenzie & Bickle 1988; Chamot-Rooke et al. this volume). Table 1 also suggests that the thin oceanic crust in the basin off the Gulf of Lion margin and the average fast spreading rate correspond to values measured in back-arc basins, rather than in the Atlantic. Travel-time tomography in the western Mediterranean (Spakman et al. 1993) shows positive velocity anomaly (up to +2%) in the upper 100 km thick interval centred on the Alg6ro-Proven~al basin, that characterizes an upper mantle with temperatures lower than a radially symmetric velocity model of the Earth. Colder mantle may account for the abnormal subsidence of the basin, and could be a result of the >40 Ma long subduction beneath this area. Unlike Atlantic-type margins, in the Gulf of Lion the maximum thickness of syn- and post-rift sediments (7 kin) is located seawards of the continent-ocean boundary. This is due to the large input of terrigenous sediments that prograde across the margin, and reach the closed Alg6roProvenqal oceanic basin. The basin is surrounded by zones of active tectonics and of elevated relief, which induce intense erosion. In addition, the Messinian event induced erosion of pre-Messinian sediments deposited in the upper margin and
GULF OF LION OVERVIEW
29
Fig. 8. Cartoon showing the asymmetry of the margins in a marginal basin formed by subduction retreat. The margin of the stable continent receives large clastic sedimentary influx from wide drainage areas, whereas the margin of the migrating micro-continent is fed by small drainage basins and volcanoes. Oceanic crust in the back-arc basin is formed by passive upwelling of the asthenosphere, generating small amounts of melt, and thus a thin crust. The upwelled asthenospheric cell follows the outward migration of the subduction, which involves faster cooling (and more important post-rift subsidence) of the divergent margin than on the migrating continental margin. High rates of subsidence and of sedimentation on the side of the stable continent lead, within less than 30 Ma, to the formation of a margin displaying features of a mature Atlantictype margin.
their resedimentation in the basin. The sedimentary prism is thicker across the NW margin of the Alg6ro-Proven~al basin (Gulf of Lion margin and Valencia Trough) than across the Sardinian margin, reflecting the larger continental drainage area in the NW, including the Rh6ne and Ebro river, whereas the Sardinian margin is characterized by smaller sedimentation rates, as a result of smaller drainage basins. Figure 8 represents a conceptual model of the Gulf of Lion margin. It developed as the divergent continental margin that remained fixed to a large continent, whereas the west Sardinia margin represents the margin of a small continental block carrying the volcanic arc that drifted away. Asymmetric structure and development of the back-arc basin is related to the outward migration of the Sardinia continent, induced by the subduction retreat. Southeast migration of the subduction involves the parallel displacement of the asthenospheric cell beneath the accretionary ridge in the back-arc basin. This may account for the present-day asymmetric
heat-flow pattern displaying higher values towards the Sardinia margin (Lucazeau & Mailh6 1986). The fast rate of thermal subsidence beneath the divergent Gulf of Lion margin results from low potential temperature of the mantle (Chamot-Rooke et al. this volume). Subsidence is accompanied by large flux of clastic sediments, which suggests that in spite of the young age of the margin, fast burial may be sufficient to promote maturation on the Gulf of Lion margin (Burrus & Audebert 1990). In contrast, the drifted Sardinia margin is characterized by a lower rate of subsidence and sedimentation. Like marginal basins such as the Japan Sea and the South China Sea, the Gulf of Lion margin has the same tectonic setting as the Far East Siberia margin and the Pearl River Basin.
Tectonic evolution in the N W Mediterranean framework Physical experiments of the formation of backarc basins show that tensile stress occurring in
30
M. SI~RANNE
GULF OF LION OVERVIEW the overriding plate is insufficient to break apart the lithosphere if it does not contain a weak zone (Shemenda 1994). In the case of the Gulf of Lion margin, rifting was superimposed onto an orogenically thickened zone. Onset of extension in the area (late Rupelian, 28-30 Ma) post-dates the end of the Pyrenean orogeny by some 10 Ma. Post-orogenic relaxation of the geotherms perturbed during latest Cretaceous-Bartonian (>30 Ma long) thickening event allowed a significant decrease of the integrated strength of the lithosphere, during 10 Ma (Cloetingh et al. 1995), leading to the presence of a weak zone underlying the Pyrenees at the onset of rifting. Roll-back of the subduction in Late Rupelian induced tensile stress in the overriding continental lithosphere, which was expressed by rifting centred on the weak orogenic zone. This zone is distinct from the volcanic arc set in Sardinia. It is also distinct from the area located in the upper margin, which was involved in the earlier rifting of the West European plate, from the Rhine graben to Camargue. However, Late Rupelian back-arc extension in the Gulf of Lion overlaps to the north with the West European rift system, and reactivates grabens, such as the Camargue basin, that were initiated during Priabonian time under E - W extensional stresses. During Late Rupelian to Aquitanian times, the area previously occupied by the Pyrenean range was stretched and thinned. Extension and lithospheric thinning was driven by the subduction retreat and eastward migration of Sardinia-Calabria, corresponding to the conditions of 'passive rifting' (Keen 1985). This is in agreement with (1)the non-volcanic nature of the margins, (2) the >200 km width of the rift and (3) the rapid cooling and thermal contraction away from the rift axis. Continental break-up occurred at the Aquitanian-Burdigalian transition, some 50 km southeast of the axis of the rift, as evidenced by the wider Gulf of Lion margin compared to the West Sardinian Margin. This asymmetry probably reflects the southeastwards retreat of the subduction and migration of Sardinia-Calabria during the 10 Ma long rifting interval. Oceanic accretion in the Alg6ro-Provenqal basin spanned the entire Burdigalian and probably extended into the Langhian, as discussed by Chamot-Rooke et al. (this volume). Subduction
31
retreat proceeded during this time. Continuous location of the calc-alkaline volcanic arc in western and central Sardinia from Late Rupelian through to Serravalian strongly suggests that the subducted slab kept the same dip through time. During this interval the Gulf of Lion margin underwent thermal subsidence over a wide area as witnessed by the Burdigalian marine transgression on the shelf and hinterland (see Fig. 3). During Serravalian-Tortonian times back-arc extension and oceanic accretion in the Alg6roProvencal basin gave way to rifting between Sardinia and Calabria. Concomitantly, the calcalkaline volcanic arc jumped from Sardinia to an area in the present Tyrrhenian sea. As a result, the volcanic arc lies closer to the trench, suggesting a steepening of the subduction angle. Steepening of the subduction must have increased throughout Late Miocene, as presentday seismicity and tomography image a slab dipping up to 70 ~ to the NW beneath the Calabrian arc (Spakman et al. 1993; Selvaggi & Chiarabba 1995). Sinking of the subducted slab in the mantle is accompanied by emplacement of an asthenospheric wedge at the junction of the overriding plate and the subducted lithosphere. Rifting of the Gulf of Lion margin appears as the first stage of a continuous process of subduction retreat over 900 km, in ESE direction, developing from Late Oligocene to Present, while Africa (at the longitude of Algiers) had a northward motion of 350 km.
Conclusion The IBS-Gulf of Lion project allowed us to realise a modern synthesis of data on a European basin within the Alpine-Mediterranean realm, and gave an opportunity to re-evaluate the petroleum potential of this area. The IBS-Gulf of Lion project offered an opportunity to reassess the petroleum evaluation of parts of the margin, and provided new geological constraints on the identification of petroleum systems (Mascle et al. 1996; Vially & Tr6moli~res 1996; Mascle & Vially this volume). Regional reviews (e.g. Deville et aL 1994; Mascle et al. 1994) have stressed that even if the exploration in the area had little success in the past, the full evaluation of the onshore and offshore Gulf of Lion margin still remains to be done.
Fig. 9. Suggested geodynamic evolution of the NW Mediterranean from Priabonian to Present along a section across the present-day Gulf of Lion, Alg6ro-Proven~al basin, Sardinia, Tyrrhenian Sea, Calabria and Ionian Sea. This curved line of section is parallel to the direction of extension of each successive rifted basin. See text for comments.
32
M. SI~RANNE
The study produced a much improved knowledge of the onshore margin, and a decisive advance in the understanding of the geodynamics of the NW Mediterranean. The study characterized a number of new extensional sedimentary basins. Tectonics and kinematics of an extending crust were studied at various scale (hundreds of metres to hundreds of kilometres). Through the development of ECORS data, new insight into the structure and processes of deformation of the lower crust were established, in particular denudation of the lower crust at the ocean-continent transition. Methodology of 3D gravity inversion proved successful in a region where data is very unevenly distributed and of extremely varying quality. There is a number of such regions where the methodology tested in the Gulf of Lion can be applied. Although the Gulf of Lion margin displays structural and stratigraphic features similar to 'Atlantic-type' margins, its structure and evolution corresponds to that of a rifted margin of a large continent, formed during the opening of a marginal basin. Specific features, such as structural partitioning in the margin, thin oceanic crust in the basin, high rates of subsidence and sedimentation in the lower slope and basin, and asymmetry with the conjugate margin can be accounted for by a tectonic model of a back-arc basin developed above a retreating subduction zone. Tectonic evolution of the Gulf of Lion margin corresponds to the early stage of a continuous process of southeastward retreat of subduction or roll back that has taken place in the Alpine-Mediterranean region since Oligocene. For that reason, the results and methodologies of IBS-Gulf of Lion can easily be exported to other Tertiary Alpine-Mediterranean basins. Further afield, the results and methodologies of IBS-Gulf of Lion could be applied to marginal basins of the western Pacific, Central America and the Caribbean. IBS-Gulf of Lion was supported by the European C o m m u n i t y D G X I I (contract J O U L E II - CEC Project no. PL 920287 'Integrated Basin Studies'). BRGM, Elf Aquitaine, EssoRep, Total, Coparex and Elf-Atochem have provided data for this study. This contribution relies heavily on discussions with, and results obtained by, contributors to the IBS-Gulf of Lion project: A. Benedicto, N. Chamot-Rooke, J.-M. Gaulier, F. Jestin, P. Labaume, L. Maerten, A. Mascle, G. Pascal, M. S6guret, C. Truffert, R. Vially. A. Mascle's comments on an earlier version of this paper have been extremely useful. I am grateful to A. Benedicto, M. S6guret, F. Horv~ith and an anonymous referee for their reviews.
References ARTHAUD, F. & SI2GURET, M. 1981. Les structures pyr6n6ennes du Languedoc et du Golfe du Lion (Sud de la France). Bulletin de la SociOtO GOologique de France, 23, 51-63. - - - , Mt~GARD, F. & SI~GURET, M. 1977. Cadre tectonique de quelques bassins s6dimentaires. Bulletin des Centres de Recherche ExplorationProduction Elf Aquitaine, 1, 147-188. , OGLER, M. & SI~GURET, M. 1981. G6ologie et g6ophysique du Golfe du Lion et de sa bordure nord. Bulletin du B R G M , (2), 1, 175-193. AUZENDE, J. M., BONNIN, J. & OLIVET, J. L. 1973. The origin of the Western Mediterranean basin. Journal of the Geological Society, London, 19, 607-620. BARTRINA, M. T., CABRERA, L. L., JURADO, M. L., GU1MERA, J. & ROCA, E. 1992. Cenozoic evolution of the central Catalan margin (Valencia Trough, western Mediterranean). Tectonophysics, 203, 219-248. BELLON, H. 1976. Sdries magmatiques ndogOnes et quaternaires du pourtour de la MOditerrande occidentale, comparOes dans leur cadre g~ochronologique - Implications gdodynamiques. Thbse 3eme Cycle, Paris Sud - Orsay. & BROUSSE, R. 1977. Le magmatisme pdrim6diterran6en occidental. Essai de synth6se. Bulletin de la SociOtO Gdologique de France, 19, 469--480. BENEDICTO, A. 1996. ModOles tectono-sddimentaires de bassins en extension et style structural de la marge passive du Golfe du Lion (SE France). Th~se de Doctorat, Univ. Montpellier 2. - - , LABAUME,P., SI~GURET,M. & SI~RANNE,M. 1996. Low-angle crustal ramp and basin geometry in the Gulf of Lion passive margin: the Oligocene-Aquitanian Vistrenque graben, SE France. Tectonics, 15, 1192-1212. --, St~GURET, M. & LABAUME, P. 1999. Interaction between faulting, drainage and sedimentation in extensional hanging-wall syncline basins: example of the Oligocene Matelles Basin, Gulf of Lion margin (S.E. France). This" volume. BERGERAT, F. 1985. DOformations cassantes et champs de contrainte tertiaires dans la plate-forme europOenne. Th6se d'Etat, Univ. Paris 6. BESLIER, M. O., ASK, M. & BOILLOT, G. 1993. Ocean continent boundary in the Iberia Abyssal Plain from multichannel seismic data. Tectonophysics, 218, 383-394. BESSIS, F. 1986. Some remarks on the study of subsidence of sedimentary basins. Application to the Gulf of Lions margin (Western Mediterranean). Marine and Petroleum Geology, 3, 37-63. BIJU-DUVAL, B., LETOUZET,J. & MONTADERT,L. 1978. Structure and evolution of the Mediterranean basins. In: HsO, K., MONTADERT, L. ET AL. (eds) Initial Reports of the Deep Sea Drilling Project. U.S. Government Printing Office, 42, part 1, 951-984. BIONDI, P., LERAT, 0. & PHILLIPS, J. 1992. Synth~se structural sddimentologique, et ggochimique du
GULF OF LION OVERVIEW
bassin EocOne-OligocOne de ManosqueForcalquier (Alpes de Haute-Provence). Institut Fran~ais du P6trole, unpublished report, 40-151. BLI~S,J. L., BONIJOLY,D., CASTAING,C. d~ GROS,Y. 1989. Successive post-Variscan stress fields in the French Massif Central and its borders (Western Europe plate): comparison with geodynamic data. Tectonophysics, 169, 79-111. BOILLOT, G., GIRARDEAU, J. & KORNPROBST,J. 1988. Rifting of the Galicia Margin: crustal thinning and emplacement of mantle rocks on the seafloor. In: BOILLOT, G. & WINTERER, E. L. (eds) Proceeding of the Ocean Drilling Program Scientific Results. Ocean Drilling Program, 103, 741-756. - - , MONTADERT,L., LEMOINE, M. & BIJU-DUVAL, B. 1984. Les marges continentales actuelles et fossiles autour de la France. Masson, Paris. BOlS, C. 1993. Initiation and evolution of the OligoMiocene rift basins of southwestern Europe: contribution of deep seismic reflection profiling. Tectonophysics, 226, 227-252. BOUILLIN,J.-P. 1986. Le "bassin maghr6bin": une ancienne limite entre l'Europe et l'Afrique ?a l'ouest des Alpes. Bulletin de la Socidtd GOologique de France, 8, 2, 547-558. BURG, J. E, TEYSSIER, C., LESPINASSE,M. & ETCHECOPAR, A. 1982. Direction de contraintes et dynamique du bassin de Saint-Flour - Saint Alban (Massif Central franqais) a l'Oligochne. Comptes rendus de l'AcadOmie des Sciences de Paris, 294, 1021-1024. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provencal basin (North-western Mediterranean). Marine Geology, 55, 247-269. -& AUDEBERT,E 1990. Thermal and compaction processes in a young rifted basin containing evaporite: Gulf of Lions, France. American Association of Petroleum Geologists Bulletin, 74, 1420-1440. - - , BESSIS,E & DOLIGEZ, B. 1987. Heat flow, subsidence and crustal structure of the Gulf of Lions (NW Mediterranean): A quantitative discussion of the classic passive margin model. In: BEAUMONT, C. & TANKART, A. J. (eds) Sedimentary Basins and basin-forming mechanisms. Canadian Society of Petroleum Geologists Memoirs, 12, 1-15. & FOUCHER, J. R 1986. Contribution to the thermal regime of the Provenqal Basin based on flumed heat flow surveys and previous investigations. Tectonophysics, 128, 303-334. BussoN, G. 1992. Bassins pal6og~nes salif~res de l'est de la France (Valence, Bresse et Haute Alsace). GOologie de la France, 1, 15-27. CANDE, S. C. & KENT, D. V. 1992. A new geomagnetic polarity time scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 97, 13917-13951. CHAMOT-ROOKE, N., GAULIER, J. M. & JESSIN, E 1999. Constraints on Moho depth and crustal thickness in the Liguro-Provenqal basin from a 3D gravity inversion: geodynamic implications. This volume. CHERCHI, A. & MONTADERT, L. 1982. The OligoMiocene rift of Sardinia: a model for the early -
-
33
history of the Western Mediterranean Basin. Nature, 298, 736-739. CLOETINGH, S., VANWEES, J. D., VAN DER BECK, R A. & SPADINI,G. 1995. Role of pre-rift rheology in kinematics of extensional basins formation: Constraints from thermomechanical models of Mediterranean and intracratonic basins. Marine and Petroleum Geology, 12, 785-807. COULON, C. 1977. Le volcanisme calco-alcalin cOnozo~que de Sardaigne (Italic) - POtrographie, g~ochimie et genOse des' laves and~sitiques et des ignimbrites - Signification gdodynamique. Th6se d'Etat, Univ. Aix-Marseille 3. CRAVATTE, J., DUFAURE, P., PRIM, M. & ROUAIX, S. 1974. Les forages du Golfe du Lion. Stratigraphie, s6dimentologie. Notes et M~moires, Compagnie Franfaise des P~troles, 11, 209-274. DE VOOGD, B., NICOLICH, R. & 21 OTHERS 1991. First deep seismic reflection transect from the Gulf of Lions to Sardinia (ECORS-CROPprofiles in Western Mediterranean). In: MEISSNER, R., BROWN, L., DURBAUM,H.-J., FUCHS,K. & SEIFERT, E (eds) Continental lithosphere: Deep seismic reflections. American Geophysical Union Geodynamics Series, 22, 265-274. DEBRAND-PASSARD, S. &COURBOULEX, S. 1984. Synthose gOologique du Sud-Est de la France - Stratigraphic et paldogdographie. BRGM, Orldans, France. DERCOURT, J., ZONENSHAIN, L. P. & 15 OTHERS 1985. Pr6sentation de 9 cartes paldog6ographiques au 1/20 000 000e s'6tendant de l'Atlantique au Pamir pour la p6riode du Lias h l'Actuel. Bulletin de la Soci~tO GOologique de France, 8, 637-652. DEVILLE,E., MASCLE,A., LAMIRAUX,C. & LE BRAS,A. 1994. Tectonic styles, reevaluation of plays in southeastern France. Oil & Gas Journal, 53-58. DEWEY, J. E, HELMAN,M. L., TURCO, E., HUTTON,D. H. W. & KNOTT, S.D. 1989. Kinematics of the western Mediterranean. In: COWARD,M. P., DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. DOGLIONI, C. 1991. A proposal for the kinematic modelling of W-dipping subductions - possible applications to the Tyrrhenian-Apennines system. Terra Nova, 3, 423-434. DURAND-DELGA, M. & FONTBOTI~,J. M. 1980. Le cadre structural de la Mdditerran6e Occidentale. In: AUBOUIN, J., DEBELMAS,J. & LATREILLE,M. (eds) GOologie des chafnes alpines issues de la TOthys. B R G M Mdmoires, 115, 67-85. EDWARDS, J. D. & SANTOGROSS1,R A. 1989. Divergent/Passive margin basins - Summary and Conclusions. In: EDWARDS,J. D. & SANTOGROSSI,P. A. (eds) Divergent~Passive margin basins. American Association of Petroleum Geologists Memoirs, 48, 239-248. ELLIS, P. G. & MCCLAY,K. R. 1988. Listric extensional fault systems - results of analogue experiments. Basin Research, 1, 55-70. GAULIER, J. M., CHAMOT-ROOKE,N. & JESTIN,E 1994. Post-orogenic Oligo-Miocene rifting of the thickened lithosphere: a working hypothesis for the
34
M. SI~RANNE
Gulf of Lion margin formation. 6th Conference, European Association of Petroleum Geoscientists & Engineers, Vienna, P802. GIROD, M. & GmOD, N. 1977. Contribution de la p6trologie h la connaissance de l'6volution de la M6diterran6e occidentale depuis l'Oligoc6ne. Bulletin de la SociOtd Gdologique de France, 19, 481-488. GORINI, C. 1993. Gdodynamique d'une marge passive: le Golfe du Lion (MdditerranOe Occidentale). Thbse de Doctorat, Univ. Paul Sabatier - Toulouse. , LE MARREC, A. & MAUFFRET,A. 1993. Contribution to the structural and sedimentary history of the Gulf of Lions (western Mediterranean), from the ECORS profiles, industrial seismic profiles and well data. Bulletin de la SociOtO Gdologique de France, 164, 353-363. , MAUFFRET,A., GUENNOC, P. & LE MARREC, A. 1994. Structure of the Gulf of Lions (Northwest Mediterranean Sea): A review. In: MASCLE, A. (ed.) Hydrocarbon and petroleum geology of France. Springer-Verlag European Association of Petroleum Geology Special Publications, 4, 223-243. GROW, J. A., KLITGORD, K. D. & SCHLEE, J. S. 1988. Structure and evolution of Baltimore Canyon Trough. In: SHERIDAN,R. E. & GROW, J. A. (eds) The Atlantic Continental Margin, U.S. Geological Society of America, The Geology of North America, I-2, 269-290. GUENNOC, P., DEBEGLIA,N., GORIN1, C., LE MARREC, A. & MAUFFRET,A. 1994. Anatomie d'une marge passive jeune (Golfe de Lion - Sud France): Apports des donn6es g6ophysiques. Bulletin des Centres de Recherche Exploration-Production Elf Aquitaine, 18, 33-57. HAYES, D. E., NISSEN, S. S., BUHL, P., DIEBOLD, J., BOCHU, Y., WEUUN, Z. & YONGQ1N, C. 1995. Throughgoing crustal faults along the northern margin of the South China Sea and their role in crustal extension. Journal of Geophysical Research, 100, 22435-22446. HORVATH, E & TARI, G. 1999. IBS Pannonian Basin Project: A review of the main results and their bearing on hydrocarbon exploration. This volume. JOLIVET, L., DANIEL,J.-M., TRUFFERT, C. & GOFFE, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in arc and back-arc regions. Lithos. 33, 3-30. KASTENS,K., MASCLE,J. d~: 19 OTHERS 1988. ODP Leg 107 in the Tyrrhenian Sea: Insight into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KEEN, C. E. 1985. The dynamics of rifting: deformation of the lithosphere by active and passive driving forces. Geophysical Journal of the Royal Astronomical Society, 80, 95-120. KELLER, J. V. A., MINELLI, G. & PIALLI, G. 1994. Anatomy of late orogenic extension: the Northern Apennine case. Tectonophysics, 238, 275-294. KLEMPERER, S. & HOBBS, R. 1991. The BIRPS Atlas Deep seismic reflection profile around the British
Isles. Cambridge University Press, Cambridge, UK. KLITGORD, K. D., HUTCHINSON,D. R. & SCHOUTEN,H. 1988. U.S. Atlantic Continental margin; Structural and tectonic framework. In: SHERIDAN,R. E. & GROW, J. A. (eds) The Atlantic Continental Margin, U.S. Geological Society of America, The Geology of North America, 1-2, 19-55. LARROQUE,J.-M. 1987. Analyse de la ddformation de la sdrie salifOre du bassin potassique de Mulhouse, et dvolution du champ de contrainte dans le sud du fossd rhdnan au Tertiaire et d l'Actuel. Thbse Doctorat, Univ. Montpellier 2. LAUBSCHER, H., BIELLA, G. C., CASS1NIS,R., GELATI, R., LOSEJ, A., SCRASCIA, S. & TABACCO,I. 1992. The collisional knot in Liguria. Geologische Rundschau, 81, 275-289. LE DOUARAN,S., BURRUS,J. ~; AVEDIK,E 1984. Deepstructure of the North-Western Mediterranean basin: Results of a two-ship seismic survey. Marine Geology, 55, 325-345. LE PICHON,X. & BARBIER,E 1987. Passive margin formation by low-angle faulting within the upper crust: The northern Bay of Biscaye margin. Tectonics, 6, 133-150. ~ . , PAUTOT,G.,AUZENDE,J.-M. & OLIVET,J.-L. 1971. La M6diterran6e Occidentale depuis l'Oligoc6ne; sch6ma d'6volution. Earth and Planetary Sciences Letters, 13, 145-152. LVCAZEAU, E & MAILHI~, D. 1986. Heat-flow, heat production and fission-track data from the Hercynian basement around the Provencal basin (Western Mediterranean). Tectonophysics, 128, 335-356. MAERTEN, L. 1994. Structure et stratigraphie sdquentielle du bassin Oligo-MiocOne de l'H&ault (Marge passive du Golfe du Lion). Dipl6me d'Etudes Approfondies, Univ. Montpellier 2. & SERANNE,M. 1995. Extensional tectonics in the Oligo-Miocene H6rault Basin (S. France), Gulf of Lion margin. Bulletin de la SociOtO Gdologique de France, 166, 739-749. MALINVERNO,A. & RYAN,W. B. E 1986. Extension in the Tyrrhenian Sea and shortening in the Apennines as a result of arc migration driven by sinking lithosphere. Tectonics, 5, 227-245. MARTi, J., MITJAVILA,J., ROCA, E. & APARICIO,A. 1992. Cenozoic magmatism in the Valencia Trough (western Mediterranean): relationship between structural evolution and volcanism. Tectonophysics, 203, 155-166. MASCLE,A. & VIALLY,R. 1999. The petroleum systems of the South-East Basin and Gulf of Lions (France). This volume. --, BERTRAND, G. & LAMIRAUX, C. 1994. Exploration for oil and production of oil and gas in France. In: MASCLE, A. (ed.)Hydrocarbon and petroleum geology of France. Springer-Verlag European Association of Petroleum Geology, Special Publications, 4, 3-27. - - , VIALLY,R., DEVILLE,E., BIJU-DUVAL,B. & ROY, J. R 1995. The petroleum evaluation of a tectonically complex area: the western margin of the South East Basin. 7th Conference, European
GULF OF LION OVERVIEW Association of Petroleum Geoscientists & Engineers, Glasgow. - - , VIALLY,R., DEVILLE,E., BIJU-DUVAL,B. & ROY, J. P. 1996. The petroleum evaluation of a tectonically complex area: the western margin of the Southeast Basin (France). Marine and Petroleum Geology, 13, 941-961. MASCLE, J. & R~HAULT, J.-R 1990. A revised seismic stratigraphy of the Tyrrhenian: Implications for the basin evolution. In: KASTENS,K. A., MASCLE, J. & ET AL. (eds) Proceedings of the ODP, Scientific Results. Ocean Drilling Program, 10, 617-636. MAUFFRET, A. & GENNESSEAUX,M. 1989. Compression, ddcrochements et distension sur le pourtour m6diterran6en nord-occidental. Comptes rendus de l'Acaddmie des Sciences de Paris, 308, 9601-967. & GORINI, C. 1996. Structural style and geodynamic evolution of Camargue and Western Provenqal basin, southeastern France. Tectonics, 15, 356-375. --, PASCAL, a., MAILLARD, A. & GORINI, C. 1995. Tectonics and deep structure of the north-western Mediterranean Basin. Marine and Petroleum Geology, 12, 645-666. MAURIN,J.-C. 1995. Drapage et d6collement des s6ries jurassiques sur la faille de d6tachement majeure du Rift Rhdnan sud: implications sur la g6om6trie des ddp6ts syn-rift oligocbnes. Comptes rendus de l'Acaddmie des Sciences de Paris, 321, 1025-1032. MCKENZIE,D. P. & BICKLE,M. J. 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology, 29, 625-679. MEYERS, J. B., ROSENDAHL,B. R. & AUSTIN,J. A. 1996. Deep-penetrating MCS images of the South Gabon Basin: implications for rift tectonics and post-break-up salt remobilization. Basin Research, 8, 65-84. MONTIGNY, R., EDEL, J. B. & THUIZAT,R. 1981. OligoMiocene rotation of Sardinia: K-Ar and paleDmagnetic data of Tertiary volcanics. Earth and Planetary Sciences Letters, 54, 261-271. MORANGE, A., HI~RITIER, E & VILLEMIN, J. 1971. Contribution de l'exploration p6troli6re h la connaissance structurale et s6dimentaire de la Limagne, dans le Massif Central. In: JUNG,J. (ed.) Gdologie, Gdomorphologie et structure profonde du Massif Central Franfais. Plein Air Service Edition, 295-308. NISSEN, S. S., HAYES, D. E., BOCHU, Y., WEIJUN, Z., YONGQUIN, C. & XIAUPIN,N. 1995a. Gravity, heat flow, and seismic constraints on the processes of crustal extension: Northern margin of the South China Sea. Journal of Geophysical Research, 100, 22447-22483. - - , BUHL, P., DIEBOLD, J., BOCHU,Y., WEIJUN, Z. & YONGQIN, C. 1995b. Deep penetration seismic soundings across the northern margin of the South China Sea. Journal of Geophysical Research, 100, 22407-22433. OLIVET,J. L. 1996. Cindmatique de la plaque Ibdrique. Bulletin des Centres de Recherche ExplorationProduction Elf Aquitaine, 20, 131-195. PASCAL, G., TRUFFERT, C., MARQUIS, G. r LABAUME, 19. -
-
35
1994. ECORS-Gulf of Lion deep seismic reflection profiles revisited: geodynamical implications. 6th Conference, European Association of Petroleum Geoscientists & Engineers, Vienna, P803. PASCAL, G. P., MAUFFRET,A. & PATRIAT,P. 1993. The ocean-continent boundary in the Gulf of Lion from analysis of expanding spread profiles and gravity modelling. Geophysical Journal International, 113, 701-726. PHILIBERT, S. 1992. Gdom~trie et cindmatique de l'extension oligoc~ne en Languedoc. Exemple du synclinal des Matelles (Hdrault). Dipl6me d'Etudes Approfondies, Univ. Montpellier 2. RANGIN, C., JOLIVET, L., PUBELLIER, M. & TETHYS PACIFICWORKING GROUP. 1990. A simple model for the tectonic evolution of southeast Asia and Indonesia region for the past 43 m.y. Bulletin de la Socidtd Gdologique de France, 6, 889-905. REBA~,S., PHILIP,H. & TABOADA,A. 1992. Modern tectonic stress field in the Mediterranean region: evidence for variation in stress directions at different scales. Geophysical Journal International, 110, 106-140. RI~HAULT,J. P., BOILLOT,G. & MAUFFRET,A. 1984. The western Mediterranean basin geological evolution. Marine Geology, 55, 447-477. RESTON, W.J. 1990. The lower crust and the extension of the continental lithosphere: kinematic analysis of BIRPS deep seismic data. Tectonics, 9, 1235-1248. ROCA, E. & GUIMERA,J. 1992. The Neogene structure of the eastern Iberian margin: structural constraints on the crustal evolution of the Valencia trough (western Mediterranean). Tectonophysics, 203, 203-218. ROURE, E, BRUN,J. P., COLETTA,B. & VIALLY,R. 1994. Multiple extensional structures, fault reactivation, and petroleum plays in the Alpine foreland basin of southeastern France. In: MASCLE, A. (ed.) Hydrocarbon and petroleum geology of France. Springer-Verlag European Association of Petroleum Geology, Special Publications, 4, 245-268. -& VAN DEN DRIESSCHE, J. 1992. Geometry and kinematics of extensional structures in the Alpine foreland basin of southeastern France. Journal of Structural Geology, 14, 503-519. RoY, J. R & TRI~MOLII~RES,P. 1992. L'inversion du demi-graben oligoc6ne d'Alhs (France): Donn6es de terrain et mod61isation. Comptes rendus de l'Acaddmie des Sciences de Paris, 315, 1777-1781. SAADALLAH, A. & CABY, R. 1996. Alpine extensional detachment tectonics in the Grande Kabylie metamorphic core complex of the Maghrebides (northern Algeria). Tectonophysics, 267, 257-273. SANCHIS,E. 1997. Relation tectonique - s(dimentation dans le bassin tertiaire d'AlOs. Mise en (vidence d'un demi-graben Ludien et d'un synclinal sur rampe OligocOne. Diplome d'Etudes Approfondies, Univ. Montpellier 2. SAPIN, M. & HIRN, A. 1974. Results of explosion seismology in the southern Rh6ne valley. Annales de Geophysique, 30, 181-202.
36
M. S E R A N N E
SARTORI,R. 1990. The main results of the ODP Leg 107 in the frame of Neogene to Recent geology of perityrrhenian areas. In: KASTENS, K. A., MASCLE, J. ET AL. (eds) Proceedings of the ODP, Scientific Results. Ocean Drilling Program, 107, 715-730. SELVAGGI,G. & CHIARABBA,C. 1995. Seismicity and Pwave velocity image of the Southern Tyrrhenian subduction zone. Geophysical Journal International, 121, 818-826. SI~RANNE,M. BENEDICTO,A., TRUFFERT,C., PASCAL,G. & LABAUME, P. 1995. Structural style and evolution of the Gulf of Lion Oligo-Miocene rifting: Role of the Pyrenean orogeny. Marine and Petroleum Geology, 12, 80%820. St~RANNE,M. & MALAVIEILLE,J. (eds) 1994. Late orogenic extension, Tectonophysics, Special Issue, 238. SHEMENDA,A. I. 1994. Subduction: insights from physical modelling. Kluwer Academic Publishers, Dordrecht, The Netherlands, 215. SHIPBOARD SCIENTIFIC PARTY. 1990. Introduction, background, and principal results of Leg 128 of the Ocean Drilling Program, Japan Sea. In: INGLE, J. C. J., SUYEHIRO, K., VON BREYMANN,M. T. ET AL. (eds) Proceedings of the ODP, Initial Reports. Ocean Drilling Program, 128, 5-562. SKOGSEID,J. 1994. Dimensions of the Late CretaceousPaleocene Northeast Atlantic rift derived from Cenozoic subsidence. Tectonophysics, 240, 225-247. SPAKMAN, W., VAN DER LEE, S. & VAN DER HILST, R. 1993. Travel-time tomography of the EuropeanMediterranean mantle down to 1400 kin. Physics of the Earth and Planetary Interiors, 79, 3-74. STECKLER,M. S. & WATTS.A. B. 1980. The Gulf of Lion: Subsidence of a young continental margin. Nature, 287, 425-429. TAPPONNIER,R 1977. Evolution tectonique du syst6me alpin en M6diterran6e: poinqonnement et 6crasement rigide-plastique. Bulletin de la Soci~tO GOologique de France, 19, 437-460. TEISSERENC,P. & VILLEMIN,J. 1990. Sedimentary basin of Gabon - Geology and oil systems. In: EDWARDS,J. D. & SANTOGROSSI,P. A. (eds) Divergent~passive margin basins. American Association of Petroleum Geologists Memoirs, 48, 177-199. THOMPSON, S. N. 1994. Fission-track analysis of the crystalline basement rocks of the Calabria Arc,
southern Italy: evidence of Oligo-Miocene late orogenic extension and erosion. Tectonophysics, 234, 331-352. TRICART, P., TORELLI, L., ARGNANI, A., REKHISS, E & ZITELLINI, N. 1994. Extensional collapse related to compressional uplift in the Alpine Chain off northern Tunisia (Central Mediterranean). Tectonophysics, 238, 317-329. TRUFFERT, C., PASCAL, G. & LABAUME,P. 1993. Structure crustale du Golfe du Lion: retraitement des profils ECORS M6diterrande. G6osciences Marines, Paris, 16-17 Ddcembre. 1993, Socidt6 Gdologique de France. VALETTE, M. 1991. Etude structurale du gisement salifOre oligocOne de Vauvert (Gard). Th6se de Doctorat, Univ. Montpellier 2. & BENEDICTO,A. 1995. Chevauchements gravitaires halotectoniques dans le bassin distensif de Camargue (Marge du Golfe du Lion, SE de la France). Bulletin de la Soci~tO GOologique de France, 166, 137-147. VERGES, J. & S,~BAT, E 1999. Constraints on the Neogene Mediterranean kinematic evolution along a 1000 km transect, from Iberia to Africa. This volume. VIALLY, R. & TREMOLIERES, E 1996. Geodynamics of the Gulf of Lion, Implication for petroleum exploration. In: ZIEGLER, R & HORVATH,E (eds) Peri- Tethys Memoir 2 - Structure and prospects of Alpine basins and forelands. Mdmoires du Musdum National d'Histoire Naturelle, 170, 129-158. VlGuorrI, L. & K~NT, D. V. 1990. Paleomagnetic results of Tertiary sediments from Corsica: Evidence of post-Eocene rotation. Physics of the Earth and Planetary Interiors, 62, 97-108. & LANGENHEIM,g. E. 1995. When did Sardinia stop rotating? New paleomagnetic results. Terra Nova, 7, 424-435. WATTS, A. B. & RYAN, W. B. E 1976. Flexure of the lithosphere and continental margin basins. Tectonophysics, 36, 25-44. WILDI, W. 1983. La cha~ne tello-rifaine (Algdrie, Maroc, Tunisie): structure, stratigraphie et 6volution du Trias du Mioc6ne. Revue GOologie Dynamique et GOolographie Physique, 24, 201-297. ZIEGLER, E A. 1992. European Cenozoic rift system. Tectonophysics, 209, 91-111. -
-
-
-
Constraints on Moho depth and crustal thickness in the LiguroProvencal basin from a 3D gravity inversion: geodynamic implications N. C H A M O T - R O O K E
1, J.-M. G A U L I E R 2 & F. J E S T I N 1
ILaboratoire de G~ologie, C N R S - U R A 1316, Ecole Normale Sup&ieure, 24 rue Lhomond, 75231 Paris Cedex 05, France (e-mail: rooke@sphene, ens.fr) 2Institut Franqais du POtrole, Geology & Geochemistry Division, 1 & 4 av Bois-Pr~au, BP311, 92506 Rueil Malmaison Cedex, France Present address: Beicip Franlab, 232 avenue Napoleon Bonaparte, 92 500 Rueil Malmaison, France Abstract: 3D gravity modelling is combined with seismic refraction and reflection data to constrain a new Moho depth map in the Liguro-Proven~al basin (Western Mediterranean Sea). At seismically controlled points, the misfit between the gravimetric solution and the seismic data is about 2 km for a range of Moho depths between 12 km (deep basin) and 30 km (mainland). The oceanic crust thickness in the deep basin (5 kin) is smaller than the average oceanic crust thickness reported in open oceans (7 km), pointing to a potential mantle temperature 30-50~ below normal and/or very slow oceanic spreading rate. Oceanic crust thickness is decreasing towards the Ligurian Sea and towards the continent-ocean boundary to values as small as 2 km. Poor magma supply is a result of low potential mantle temperature at depth, lateral thermal conduction towards an unextended continental margin, and decrease of the oceanic spreading rate close to the pole of opening in the Ligurian Sea. Re-examination of magnetic data (palaeomagnetic data and magnetic lineations) indicates that opening of the Liguro-Proven~al Basin may have ceased as late as Late Burdigalian time (16.5 Ma) or even Later. The absence of a significant time gap between cessation of opening in the Liguro-Proven~al Basin and rifting of the Tyrrhenian domain favours a continuous extension mechanism since late Oligocene time driven by the African trench retreat. Determination of the deep structure across continental margins has been one of the major goals achieved through deep seismic profiling in recent years. Two kinds of deep-seismic soundings have been obtained so far across the Gulf of Lion margin and adjacent deep basin: multichannel seismics of E C O R S type and wide-angle refraction data using the two-ship expanding spread profile technique. Both techniques are complementary, since m u l t i c h a n n e l profiles image the main structures along a section of the basin, from the surface to Moho depth, while ESPs constrain the 1D velocity structure at some spots along the section. Deep-reflection and refraction techniques were particularly successful in reaching the crust-mantle discontinuity (Moho) in the Gulf of Lion area. However, seismic data remain sparse. Variability of the crustal structure is high, both across the margins (from unstretched continental crust to oceanic crust) and along (for instance from smoothly stretched Gulf of Lion margin to sharply stretched Provenqal margin). The oceanic crust thickness in the deeper part of the basin is also highly variable. The deep structure of the basin
is thus complex, preventing reliable extrapolation from sparse seismic data. Gravity data can be used to obtain a reliable geometry of the Moho discontinuity in regions where no further information is available. The triangular shape of the Liguro-Proven~al Basin requires 3D gravity inversion. T h e use of Fourier transforms to invert for the shape of the Moho is briefly described. Seismic Moho data are then used to determine free parameters in the inversion and to control the validity of the gravimetric solution. The new Moho depth and crustal thickness maps are finally used to discuss some of the geodynamic implications.
Establishing a n e w M o h o depth m a p from 3 D gravity inversion
Seismic Moho depth data Moho depth data are sparsely available in the Gulf of Lion and surrounding areas. Continental refraction data indicate an average Moho depth of 30-32 km on the mainland (southern R h 6 n e valley, Sapin & H i r n 1974; Spanish mainland,
CHAMOT-ROOKE,N., GAULIER,J.-M. & JESTIN,E 1999. Constraints on Moho depth and crustal thickness in the Liguro-Provenqal basin from a 3D gravity inversion: geodynamic implications. In: DURAND,B., JOLIVET,L., HORVATH,E & S]~RANNE,M. (eds) The MediterraneanBasins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 37~1.
38
N. CHAMOT-ROOKE E T A L .
Zeyen et al. 1985; Danobeitia et al. 1992). Moho depth increases to 60-65 km beneath the Pyr6n6es belt (Choukroune & T e a m 1989; Choukroune et al. 1990; Anguy et al. 1991) and to 55-60 km beneath the Alps (Giese et al. 1973). Beneath the Balearic Islands the Moho lies at an average depth of 22 km (18 km under Menorca according to Banda et al. 1980; 24-28 km under Mallorca according to Danobeitia et al. 1992). The Moho is slightly deeper beneath Corsica and Sardinia (average 30 km depth, Egger et al. 1988; Egger 1992). At sea, both refraction and reflection data are available. Based on Expanding Spread Profiles (ESPs) the average Moho depth in the LiguroProvenqal basin and in the Valencia Trough is 15 km (Le Douaran et al. 1984; Gallart et aL 1980; Danobeitia et al. 1992; Mauffret et al. 1992, 1995; Pascal et al. 1992, 1993; Torn6 et al. 1992). The Liguro-Provenqal Basin is floored with thin oceanic crust of early Miocene age (Bayer et al. 1973; Burrus 1984), whereas thin continental crust underlies the Valencia Trough (Watts & Torn6 1992a, b; Pascal et al. 1992). Progressive thinning of the continental crust from the upper margin to the deeper basin is clearly evidenced over the Gulf of Lion margin through subsidence analysis (Steckler & Watts 1980) and multichannel deep seismic profiling (ECORS profiles, de Voogd et al. 1991). The exact position of the ocean-continent boundary is, however, still debated (e.g. Pascal et al. 1993; Mauffret et al. 1995; review in Vially & Tr6moli~res 1996). The non-uniform data distribution precludes the use of standard interpolation to recover a complete Moho depth map. We thus briefly describe in the following section the use of gravity data to constrain better the Moho depth map where no other information is available. 3D gravity inversion procedure
The 3D inversion procedure is based on Fourier transform (FT) methods (Parker 1973), using the first-order theory in which high-order terms in the gravity expansion are being neglected. The gravity anomaly at the Earth's surface due to a periodic undulation h(x,y) of an interface at depth d associated with a constant density contrast Ap is thus given by A g(x,y) = 2 7r G Ap h(x,y) e -kd
(1)
where k is the wavenumber of the undulation (k = 2 "rr/)t, where ~ is the wavelength) and G is the universal gravity constant. Fourier transforming this equation leads to A g*(kx, k y ) = Z r r G
Aph*(kx, ky) e - k a
(2)
where A g*(kx, ky) and h*(kx, ky) denote the 2D Fourier transform of the gravity signal and of the undulations, respectively (k x and ky are the horizontal wavenumbers in x and y directions). h*(kx, ky) can thus be obtained by Fourier transforming the gravity signal as h*(kx, ky) = 1/(2 v G Ap) A g*(kx, ky) e kd
(3)
Inverse FT finally leads to h(x,y). Exponential terms in equation 3 at high wavenumbers (short wavelengths) require filters to smooth the observed data prior to inversion. We tested a number of filters in the spectral domain by using forward and inverse modelling. Best results were obtained using gaussian or cosine filters to attenuate short wavelengths before applying the inverse FT. Such filters also minimizes the side-lobe effects inherent to FT methods. A cosine filter cutting all wavelengths shorter than 60 km was used in the final solutions. The method applies to a single interface with known reference level (d) and density contrast (Ap). We show in the next section how we derived a mantle Bouguer gravity anomaly to perform the inversion on the Moho discontinuity, and how we adjusted the reference level and crust-mantle density contrast.
M a n t l e b o u g u e r in the L i g u r o - P r o v e n q a l Basin
Bathymetry and topography were taken from the International Bathymetric Chart of the Mediterranean (IBCM) and from the ETOPO5 digital data files respectively. A single sedimentary layer from sea bottom to top of basement is used. The basement depth map was compiled from Arthaud et al. (1980) for the Oligocene sedimentary basins offshore, a digital data file from Mauffret (pers. comm.) and Gorini (1993) for the Gulf of Lion, R6hault (1981) for the Ligurian Sea and surrounding areas, Maillard et al. (1992) for the Valencia Trough, Moussat (1983) and R6hault et al. (1987) for the Tyrrhenian Sea, R6hault et al. (1984) for the remaining areas. Data were compiled in seismic two-way travel time. Velocities at ESP reprocessed by Pascal et al. (1993) were then used for time to depth conversions. Gravity maps were derived from the IBCM Bouguer anomaly maps. Since in the oceanic domain a simple Bouguer approximation with a density of 2670 kg m -3 was used in the IBCM charts, we first recalculated the corresponding flee-air anomaly with the bathymetry also provided by IBCM. The obtained flee-air anomaly
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN
39
Fig. 1. Free-air anomaly map in the Liguro-Proven~al basin. On land: Bouguer anomaly from the IBCM charts (Bouguer density 2.67). Contour interval: 20 mgal.
map (Fig. 1) was compared to two other sources of data: individual flee-air gravity profiles extracted from the G E O M A R data base and satellite-derived gravity field of Sandwell (Sandwell & Smith 1995). At wavelengths greater than 15 km the IBCM ship-derived map does not differ significantly from the satellite gravity map. The latter has, however, a much better resolution due to continuous cover as opposed to nonuniform ship-track coverage. The final mantle Bouguer anomaly map was obtained by correcting the free air anomaly for water and sediment, both being replaced by crust (Fig. 2). This was done by calculating their respective gravity contribution using contouring and summing up individual thin (100 m thick) sheet contributions in 3D (Talwani & Ewing 1960). Notice that at this stage, Fourier domain gravity expansion was not used. 3D marine Bouguer anomaly maps (replacing water by sediment) were also produced.
Reference level and crust-mantle density contrast The mantle Bouguer gravity anomaly can be attributed to Moho undulations to a first
approximation. The possible contribution of deeper sources, in particular the effect of laterally variable thermal and crustal density structure, will be discussed in a later section. If we consider topography of the Moho as the only gravitational source remaining, then the previous inversion methodology applies. Free parameters in the gravity inversion are the reference level (depth d) and the density contrast of the crust-mantle boundary (Ag). Uncertainties, apart from the oversimplified sedimentary column, arise from the choice of the mean sediment density and crustal density used to calculate the mantle Bouguer. The best couple of parameters d and A9 was determined by minimizing by least squares fit the difference between the calculated Moho depth and the observed seismic Moho depth. Results for d and dxp best couple are displayed in Fig. 3, and seismic Moho depth and gravimettic Moho depth are compared in Table 1. We tested the influence of sediment and crust density in the ranges 2200-2350 kg m -3 and 2750-2800 kg m -3 respectively. Quality of the fit is rather insensitive to densities, with a standard deviation of 2300 m for all runs at refraction and reflection control points (mean misfit between seismic and gravimetric Moho depths). Standard
40
N. CHAMOT-ROOKE E T A L .
Fig. 2. Mantle Bouguer anomaly map (Model A solution; no thermal correction nor variable crust density correction). Contour interval: 30 mgal.
deviation for ESP points only (oceanic domain) reduces to 1500 m for the same runs. In all cases the best reference level was found close to 27-28 km. Varying the sediment density or the crust density is merely a c c o m m o d a t e d by a modification of the Ap contrast at the crust-mantle interface: Ap is increasing with decreasing sediment density (Fig. 3d) and/or increasing crust density (Fig. 3c). A reasonable Ap contrast of 580 kg m -3 was obtained for a mean sediment density of 2350 kg/m -3 and a crustal density of 2750 kg m -3, thus giving a mantle density of 3330 kg m -3. This model is referred to as Model A in the following discussions (Fig. 4a). Fourier domain expression for gravity requires the reference level to be the median depth of the interface in order to optimize convergence of the series expansion (Parker 1973; Cowie & Karner 1990). The same limitation applies for the firstorder theory used here (non-zero first term only), so that the solution will only be approximate if the
median depth of the interface is different from the reference level. The reference level found through the inversion process (27-28 kin) is actually deeper than the a posteriori calculated mean depth of the interface (25 km). The corresponding error in the calculated gravity is a function of the wavelengths of the anomalies through the exponential term in equation 1, the error decreasing with increasing wavelength. The validity of the final solution was checked by calculating the true 3D Bouguer (using the discrete sum of thin sheet contributions, Talwani & Ewing 1960). The standard deviation of the residual (observed minus modelled, Fig. 5) is 15 regal, or 600 m translated into depth of the Moho. Typical misfit reaches 20 mgal on the margins and 10 mgals in the deeper parts of the basin. The misfit is high towards the sides of the box due to the outside box apodization prior to Fourier inversion. At that point, the obtained solution could easily be improved by re-introducing short wavelengths as small perturbations of the Moho.
Fig. 3. Reference Moho depth (d) and mantle-crust density contrast (Ap) best couple obtained by least squares minimization at seismic control points (standard deviation s.d. in km). (a) Model A solution (no thermal nor variable crust density corrections; d = 27.7 km and Ap _- 580 kg m-3). (b) Model B solution (includes thermal and variable crust density corrections). (c) Effect of crust density increase. (d) Effect of sediment density decrease.
MOHO DEPTH IN THE L I G U R O - P R O V E N ~ A L BASIN
41
42
N. C H A M O T - R O O K E E T A L .
Fig. 4. Results of the 3D gravity inversion (depth to Moho contoured every 2 km). (a) Model A solution. (b) Model B solution. Location of seismic control points are shown as open dots. Sections A, B, C, D, E, F and G run across ESP data points at sea.
M O H O D E P T H IN T H E L I G U R O - P R O V E N ~ A L BASIN
43
Table 1. Comparison between seismic Moho depth and gravimetric Moho depth at ESP control points (for both
Model A and Model B solutions) Lat N
Long E
(~
(3
(~
(')
42 42 42 42 42 41 41 41 41 41 42 42 41 41 41 42 42 42 42 43 43 42 42 42 42 43 42 42 42 42 41
53 43 29 16 2 49 34 22 5 10 23 40 22 34 48 13 26 41 55 6 16 47 3 19 34 22 36 14 16 31 4
4 4 4 5 5 5 5 5 6 6 5 5 6 6 6 7 7 7 8 8 7 6 6 6 6 7 7 7 6 6 3
34 43 55 7 20 31 44 54 8 14 41 33 26 39 52 19 32 46 28 11 54 49 47 42 37 35 11 50 15 8 41
ESP no.
Seismic Moho (km)
Model A (km)
Misfit (km)
Model B (km)
Misfit (km)
ESP201 ESP202 ESP203 ESP204 ESP205 ESP206 ESP207 ESP208 ESP209 ESP210 ESP211 ESP212 ESP215 ESP216 ESP217 ESP219 ESP220 ESP221 ESP222 ESP223 ESP224 ESP225 ESP226 ESP227 ESP228 ESP229 ESP230 ESP232 ESP233 ESP234 ESP002
19.9 19.8 15.1 14.8 14.2 15.1 14.7 15.5 15.3 15.0 14.6 13.3 15.0 15.3 14.0 13.0 12.5 13.0 15.5 14.0 13.0 22.0 14.0 14.5 20.5 15.5 12.0 13.0 13.5 13.0 14.5
22.2 20.2 15.4 13.0 14.3 14.9 15.1 15.9 15.4 15.2 12.6 15.4 15.0 15.1 15.0 13.1 12.0 13.1 19.4 16.0 13.3 19.6 14.4 14.3 16.8 15.1 14.2 13.1 14.3 14.5 16.9
-2.3 -0.4 -0.3 1.8 -0.1 0.2 -0.4 -0.4 -0.1 -0.2 2.0 -2.1 0.0 0.2 -1.0 -0.1 0.5 -0.1 -3.9 -2.0 -0.3 2.4 -0.4 0.2 3.7 0.4 -2.2 -0.1 -0.8 -1.5 -2.4 s.d. 1.52
21.9 20.1 16.1 14.2 15.9 16.5 16.4 17.1 16.7 16.1 13.4 16.2 15.4 15.1 14.5 11.5 10.4 11.9 18.6 15.5 13.7 19.1 13.4 13.2 16.2 16.0 12.5 12.5 14.2 14.8 17.6
-2.0 -0.3 -1.0 0.5 -1.7 -1.4 -1.7 -1.6 -1.4 -1.1 1.2 -2.9 -0.4 0.1 -0.5 1.5 2.1 1.1 -3.1 -1.5 -0.7 2.9 0.6 1.3 4.3 -0.5 -0.5 0.5 -0.7 -1.8 -3.1 s.d. 1.72
Model A: no thermal and no variable crust cdensity corrections. Model B: thermal and variable crust density included. s.d. is standard deviation in km.
Thermal, variable crust density a n d variable s e d i m e n t density corrections T h e o b s e r v e d h e a t flow in t h e L i g u r o - P r o v e n q a l b a s i n is h i g h (Fig. 6a). V a l u e s c o r r e c t e d for thermal blanketing are mostly above 100 m W m -2 a n d r e a c h 140-160 m W m -2 in t h e axial p a r t of t h e basin, b a s e d o n t h e r e c e n t E G T h e a t flow c o m p i l a t i o n ( E u r o p e a n G e o t r a v e r s e , D e l l a V e d o v a et al. 1995). H i g h h e a t f l o w a n o m a l y in t h e b a s i n is a c o n s e q u e n c e of t h e O l i g o c e n e rifting a n d t h e f o l l o w i n g s h o r t p e r i o d of oceanic accretion. The net effect of the p o s i t i v e t h e r m a l a n o m a l y in t h e b a s i n is to r e d u c e t h e m a n t l e d e n s i t y at d e p t h , p r o d u c i n g a
n e g a t i v e c o n t r i b u t i o n of s e v e r a l tens of m g a l to t h e gravity field (Watts & T o r n 6 1992a, b). To i n c l u d e this c o r r e c t i o n in t h e c a l c u l a t i o n of t h e m a n t l e B o u g u e r , w e m o d e l l e d t w o sections across t h e basin: a G u l f of L i o n s e c t i o n ( s e c t i o n B l o c a t e d in Fig. 4) a n d a L i g u r i a n s e c t i o n ( s e c t i o n E l o c a t e d in Fig. 4). T h e t e m p e r a t u r e field was calculated following a pure shear M c K e n z i e - t y p e m o d e l for the rifting p e r i o d (30-23 M a ) , i n c l u d i n g lateral h e a t c o n d u c t i o n a n d finite rifting d u r a t i o n ( A l v a r e z et aL 1984), a n d o c e a n i c a c c r e t i o n for t h e s p r e a d i n g p e r i o d ( 2 3 - 1 9 Ma). T h e r m a l c o o l i n g by c o n d u c t i o n was a s s u m e d f r o m 19 M a to p r e s e n t . A g e s are t a k e n f r o m q u a n t i t a t i v e s u b s i d e n c e analysis of Bessis
44
N. CHAMOT-ROOKE E T A L .
Fig. 5. Residual Bouguer (observed minus modelled) using Model A solution and 3D analytic calculations (thin sheets summation).
(1986) and Burrus (1989). The crustal structure used as input was derived from the Model A solution. Finally, the thermal anomaly was converted to a density anomaly to obtain the corresponding gravity field through 2D gravity modelling (Fig. 7). The result of the 2D thermal/gravity modelling is that the anomalous gravity field can be directly obtained with a reasonable accuracy by simply multiplying in the Fourier domain the anomalous surface heat flow with an exponential function of Fourier gravity type (Fig. 7, top):
Ag*(kx, ky) = A A+*(kx, ky) e -k o
(4)
where Ag*(kx, ky) is the gravity anomaly associated with the anomalous heat flow A+*(kx, ky), and A and D are function parameters. Notice that D has the dimension of a distance and can be viewed as the mean depth of the thermal disturbance. Parameters A and k integrate the geometry and amplitude of the thermal anomaly. Parameters A and D were found to be quite similar for the two sections modelled (A in the range 2.95-3.04 10-2 m 3 S-2 W -1 and D = 50 + 3 km). Mean A and D values were thus applied to the thermal anomaly heat flow map to get a
3D estimate of the gravity contribution. The results are shown in Fig. 6b. The negative thermal contribution to the gravity field is maximum in the axial part of the LiguroProvenqal basin, towards the Ligurian Sea, where the gravity effect reaches -50 mgal. This is significantly less than the gravity effect calculated in 2D (-65 mgal and -110 mgal for sections B and E respectively) indicating that 2D modelling would overestimate the thermal correction. Notice that highest heat flow values were measured in the Ligurian Sea, rather than the central part of the basin as would be predicted. We attribute the low values of the central part of the basin to the blanketing effect of the R h o n e delta deposits. A second type of correction is related to the lateral variations of the density of the crust. The density of the oceanic crust is known to be greater than the mean density of the continental crust. The contribution to the gravity field is opposite to the previous thermal correction since it implies a positive gravity anomaly over the oceanic domain. A n estimation of this contribution was obtained by calculating the difference between a mantle Bouguer at 2750 kg m -3 for crustal density and a mantle Bouguer at
M O H O D E P T H IN T H E L I G U R O - P R O V E N G A L BASIN
Fig. 6 (a) Observed heat flow (open dots: data points, from Della Vedova et al. 1995). Contour interval: 10 mW m -2. (b) Calculated gravity effect of the decreasing mantle density with increasing surface heat flow. See text for details of the calculations. Contour interval: 10 mgal.
45
46
N. C H A M O T - R O O K E
ETAL.
O
o
,~
9,,...,
~
~
i
e..,
r~ e",
o ~
~'~
o
.=. -~
~
~
~
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN 2880 kg m -3. The net gravity effect is reaching +40 mgal in the true oceanic domain. A similar estimate is reached if a simple Bouguer plateau approximation is calculated for a 7 km thick oceanic crust and a 130 (2880-2750) kg m -3 oceanic crust-continental crust density contrast. A last type of correction deals with the gravity effect of sediment compaction. If the density of sediment is increasing with depth, which is likely to be the case in a sedimentary basin, then the uniform density model tends to overestimate sediment density on the margins and correlatively underestimate density towards the centre of the basin. The amplitude of the gravity effect depends on the density-depth function and hence on lithologies. It reaches +50 mgal in the Viking Graben (North Sea) according to Cowie & Karner (1990). Exponential-type porositydepth functions down to about 4000 m below sea-floor are available for the Gulf of Lion margin from few industrial wells (Bessis 1986). However, compaction will not follow a simple exponential density-depth variation in the deeper part of the basin due to the thick accumulation of salt and evaporites during the Messinian. Velocity-depth functions derived from expanding spread profiles indicate high P-wave velocity (5.1-5.3 km s-1) in the sediments lying immediately above the acoustic basement (Pascal et al. 1993). In the absence of any other information, empirical velocity-density conversion would give a 2550 kg m -3 density for the lowermost sediments (Nafe & Drake 1963). Using these constraints, the mean density would range from 2090 kg m -3, estimated at ESP 225 where sediments are the thinnest (1900 m) to 2370 kg m -3, estimated at ESP 204 where sediments are the thickest (8000 m). The net gravity effect, using a simple Bouguer plateau approximation, would reach +30 mgal, equivalent to about I km of Moho topography. These estimations remain, however, highly speculative in the absence of reliable depth-density data, and we did not introduce this correction further. Corrections for thermal and oceanic crust density were applied to derive a new Moho depth map, referred as Model B (Fig. 4b). Notice that since the correction for thermal anomaly and the correction for variable crustal density are opposite in sign in the oceanic domain, Model A and Model B are quite similar there. On the mainland, negative heat flow anomaly (cold areas), such as beneath the Alps or beneath Sardinia, will tend to deepen the Moho in the gravity inversion. However, the way the gravity anomaly was derived from the heat flow is strictly valid in the oceanic domain only since homogeneous stretching followed by oceanic
47
accretion was assumed. The quality of the fit at Moho control points is slightly enhanced in Model B with respect to Model A (standard deviation of 2000 m instead of 2300 m). However, the Model B solution degrades at ESP control points (1700 m versus 1500 m). In particular, the calculated Moho in Model B shows a systematic misfit along the E C O R S profile (Table 1). This misfit is mainly resulting from the low heat-flow values, relative to surroundings, that prevail over the Gulf of Lion area. The conclusion is that Model A (uncorrected) and Model B (corrected for variable heat-flow and variable crust density) give quite similar results. In the next section, only Model A will be considered since this solution is closer to ESP control points in the oceanic domain. We display in Fig. 8 a set of crustal sections crossing the Liguro-Provenqal basin and Ligurian Sea through the refraction and reflection data points (sections are located in Fig. 4). Sections A to F are roughly perpendicular to the main structures. Section G runs obliquely from Mallorca to the Ligurian Sea and the Gulf of Genoe. Mesozoic rifting in southern France was not included in the calculations, so that the gravimetric Moho is several kilometres deeper than the seismic Moho in the Rh6ne graben (Section E). Apart from this discrepancy, and if we allow a standard error of 1 km on the seismically determined Moho (Pascal et al. 1993), the gravimetric solution fits the seismic data well.
Discussion Variability o f the oceanic crust thickness
The mean oceanic crustal thickness is close to 5.0 km. Systematic variability is, however, found both along (parallel to the magnetic lineations) and across (oceanic crust close to ocean-continent boundary) strike. The southernmost part of the Ligurian Sea (section E in Fig. 8) is floored with very thin oceanic crust (Pascal et al. 1993; Mauffret et al. 1995). At ESP 220, the Moho is rising to less than 12 km depth, which is the shallowest point of the studied area. The mean crustal thickness is 3.3 km (ESP 219, 220, 221, 230, 232) which is significantly less than the average 5.0 km thickness. Line G (Fig. 8, bottom; located in Fig. 4) is a cross-section running SW-NE in the deep basin from the island of Mallorca to the Gulf of Genoe. This along-strike section follows more or less the axial Liguro-Provenqal basin along the expected location of the fossil oceanic ridge, except for the northern end of the section. Notice the close agreement between the seismically determined
48
N. CHAMOT-ROOKE E T A L .
Moho and Model A gravimetric solution. Oceanic crustal thickness along the section shows a remarkable gradual decrease towards the Ligurian Sea. At ESP 210 (along the ECORS line) crustal thickness is close to 5.5 kin, whereas it may be as thin as 2 km in the southernmost part of the Ligurian Sea (ESP 220). Following pioneering works (Reid & Jackson 1981; Foucher et aL 1982), considerable progress has been made in the recent years on the calculations of volume of melt generated either during rifting (McKenzie & Bickle 1988; White & McKenzie 1989; Bown & White 1995a,
b) or during oceanic spreading (Scott & Stevenson 1989; White et al. 1992; Cordery & Morgan 1993; Bown & White 1994; Suet al. 1994). Production of oceanic crust is a result of partial melting of the mantle as it rises beneath spreading centres. The thickness of oceanic crust is thus directly related to the amount of partial melting, in turn controlled by asthenospheric mantle temperature and lithospheric temperature field. Under moderate to normal mantle temperature conditions (range of mantle temperatures producing less than 7-8 km thick crust), all models predict a decrease of crustal thickness with
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN
49
Fig. 8. Crustal sections across the Liguro-Proven~al basin deduced from 3D gravity inversion (Model A solution). Open dots: ESP data points at sea (vertical bar is + 1 km). Filled dots: reflection or refraction data points on land. Sections A to G are located in Fig. 4.
decreasing spreading rate and/or decreasing asthenospheric mantle temperature (Reid & Jackson 1981; Bown & White 1994; Su et al. 1994). To explore further crustal thickness variation as a function of position in the basin, we use for each E S P the distance to the inferred Sardinia-Corsica pole of rotation (Fig. 9a & b). Different solutions have been proposed for the kinematics of the rotation (see review in Vially & Tr6moli6res 1996). They range from simple rotation about a single pole for both rifting and spreading to more complicated solutions involving several phases. In the following we use a commonly adopted solution with a pole
of rotation located in the Gulf of G e n o a (N43.5~176 Rdhault et al. 1984). ESP data are located at distances to the pole ranging from 70 km (northernmost Ligurian Sea section) to 350 km (Gulf of Lion, E C O R S section). We show in Fig. 9a the predicted oceanic crust thickness for various potential asthenospheric m a n t l e t e m p e r a t u r e (Bown & White 1994). Crustal thickness is given as a function of distance to the pole. We assume a 9.8 ~ Ma -1 rotation rate which matches the magnetic anomalies interpretation of Burrus (1984). Equivalent halfspreading rates are 30 mm a -1 at a distance of 350 km and 6 mm a -1 at a distance of 70 km. Notice that since the pole of rotation is relatively
50
N. CHAMOT-ROOKE E T A L . DISTANCE 20
,
1 oo i ,
,
,
TO THE
200 i
POLE 300 i
DISTANCE
(km) 400 I ,
20
15
I
F
I
___
.o0.
- . . . . . . . . . . . .
10
POLE
(km)
200
300
400
i
i
i
i
E
7
co co i.s z v
co co i.u z x,.
6
6~
_o 5
5-0
91I-
,'
TO THE
100
g
10_o
.I-- .........
,
co z v
"r I-
I-co
co
or O
or O
tco
38
3
5
..
Oceanic
9
Transitional 0 100
200
DISTANCE
300 TO THE
POLE
400 (km)
0 100
200
DISTANCE
300 TO THE
POLE
400 (km)
E S P data
.....
T = 1300~
z~ and 9
M O D E L A gravirnetric solution
....
T = 1300~
o)=5.2~
v
M O D E L B gravimetric solution
.........
T = 1300~
- (o=3.3~
[]
and 9
and v
- m=9.8~
Fig. 9. (a) Crustal thickness (Model A, Model B and seismic refraction) at ESP points versus distance to the pole of opening. Dashed curves : predicted oceanic crustal thickness for various potential mantle temperature (1280-1400~ from Bown & White 1994). A 30 mm a 1 half-spreading rate is assumed along the ECORS line (350 km away from the pole; spreading rate derived from Burrus 1984). Equivalent rotation rate is 9.8 ~ Ma -1. Filled symbols: oceanic or transitional type crust. Open symbols: continental crust. (b) Range of crustal thickness for oceanic crust (filled circles) and transitional type crust (open diamonds) versus distance to the pole of opening. Dashed curves are model predictions using the same mantle temperature (1300~ but different rotation rates (9.8 ~ Ma -1, 5.2 ~ Ma -l and 3.3 ~ Ma -1, equivalent to 30 mm a -1, 20 mm a i and 10 mm a ~ half-spreading rates along the ECORS line (section A, Fig. 4) and to 15 mm a-1, 8 mm a -1 and 5 mm a 4 halfspreading rates in the Ligurian Sea (section E, Fig. 4).
close, spreading rate is directly p r o p o r t i o n a l to the distance to the pole. A first-order interpretation of Fig. 9a is that all oceanic data points fall close to, or below, the predicted curve for a 1280~ p o t e n t i a l t e m p e r a t u r e . This is a b o u t 30-50~ below n o r m a l t e m p e r a t u r e inferred at spreading centres away from hot spots and fracture zones (White et al. 1992; B o w n & White 1994). T h e same conclusion was r e a c h e d i n d e p e n dently from a 3D analysis of the subsidence in the Liguro-Provenqal basin ( C h a m o t - R o o k e et al. in prep.). A f t e r corrections for s e d i m e n t a r y loading, for a b n o r m a l crustal thickness and for p o t e n t i a l effects of t h e r e c e n t c o m p r e s s i o n regime, subsidence is still about 500 m greater than e x p e c t e d f r o m the age of the basin. A low p o t e n t i a l m a n t l e t e m p e r a t u r e w o u l d thus explain b o t h the a b n o r m a l subsidence of the
d e e p basin and the low m a g m a supply during oceanic accretion, leading to the production of a b n o r m a l l y thin o c e a n i c crust. Similar lowt e m p e r a t u r e b o u n d a r y conditions may have also prevailed in the Tyrrhenian Basin (R6hault et al. 1990). This is not surprising since b o t h the L i g u r o - P r o v e n ~ a l Basin and the T y r r h e n i a n Basin o p e n e d as back-arc basins above the same lithospheric slab n o w subducting off Calabria (Le Pichon 1984; M a l i n v e r n o & Ryan 1986). T h i n crust and high s u b s i d e n c e h a v e b e e n r e p o r t e d in other marginal basins, in particular in the West Pacific. L o w asthenospheric temperature may be the result of a long history of subd u c t i o n of cold material, such as b e l o w the back-arc basins of the Philippine Sea plate for instance. In the W e s t e r n M e d i t e r r a n e a n , the a m o u n t of cold material subducted was probably m u c h smaller, due to the slow c o n v e r g e n c e of
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN Africa towards Europe. Lithospheric thickening coeval with orogenic compression may also be an efficient way to lower temperatures at mantle depth, provided time span between compression and extension is short enough to prevent thermal equilibration. Extensional collapse of a former orogen is now well established for the origin of the Tyrrhenian Basin (Kastens et al. 1990; Jolivet et al. 1994). The potential role of the Pyrenean orogeny has also been discussed for the formation of Liguro-Proven~al Basin (Mauffret et al. 1995), although the continuity of the Pyrenean belt towards the Gulf of Lion margin has not been clearly established so far. A closer examination of Fig. 9a also shows the systematic trend observed along section G: oceanic crust becomes thinner closer to the pole of rotation, as predicted by melting models through spreading rate dependence. According to Bown & White (1994) models, a sharp decrease in melt production occurs at full spreading rates below 15 mm a q, whereas at higher rates the oceanic crust thickness is remarkably uniform (7 _ 1 km average, after White et al. 1992). Small oceanic crust thickness is well documented at the slow spreading Arctic ridge, where the crust is about 2.5 km thick for a 12-15 mm a-1 full rate (Jackson et al. 1982), and close to some of the Atlantic non-volcanic margins (review in Bown & White 1994; Srivastava & Roest 1995). In the Liguro-Proven~al basin, a marked decrease in oceanic crust thickness occurs in the southernmost part of the Ligurian Sea, or 200 km away from the pole of rotation (Fig. 9a & b). The predicted full spreading rate there would be 34 mm a q, assuming a 9.8 ~ Ma -1 rotation rate. This is more than twice the trade-off value of 15 mm aq inferred from modelling and observations. The thin crust of the Ligurian Sea may result from two effects: (1) heat was lost laterally towards the adjacent continents (Provencal margin and Corsica), thus preventing large-scale melting even during oceanic accretion; (2) the rate of opening was half the value generally assumed, and consequently the critical 15 mm a -1 full spreading rate below which magma supply is small was reached at a distance of 200 km from the pole. The latter hypothesis is explored in Fig. 9b where we show the predicted oceanic crust thickness for different opening rates. A good solution is obtained if the opening rate is close to 5 ~ Ma -a. Lateral heat conduction can be effective both during continental rifting and early oceanic spreading. During rifting, heat transfer from a hot extended area to cooler surroundings reduces significantly the amount of melt generated, so
51
that even at potential temperature above normal, little or no melt is produced (Alvarez et al. 1984). This effect, combined with finite (as opposed to instantaneous) duration of rifting, explains the absence of significant amount of melt produced at non-volcanic continental margins (Bown & White 1995a, b). Lateral heat loss can also be efficient in reducing crustal thickness during the early phase of oceanic spreading. The time to establish a permanent thermal regime at the ridge axis is highly dependent on the spreading rate. At high spreading rates, the ridge will rapidly move away from the cooler adjacent continent. Conversely, small transient perturbations at slow spreading ridges will have large effects on the thickness of oceanic crust produced (Su et al. 1994). At high potential temperatures, the relative importance of buoyancy-driven mantle flow with respect to plate-driven mantle flow may even lead to an increase in crustal thickness with decreasing spreading rates (Su et al. 1994). However, at low to normal potential temperatures as those considered here, plate-driven flow will dominate and variations in crustal thickness will basically reflect variations in the temperature field. How far the ridge temperature field is affected by the proximity of the continent depends not only on the oceanic spreading rate but also on the kinematics of stretching. For short rifting duration and high thinning factor, reducing the continental crust to oceanic crust thickness, the stretched continental crust is thermally indistinguishable from oceanic crust. More generally, the duration of rifting and the amount and distribution of thinning, as well as spreading rate, will control the temperature field and thus oceanic crust genesis at the ocean-continent boundary. A convenient way to estimate the time needed to establish a steady-state thermal regime at the ridge axis is to follow the abnormal subsidence induced by lateral conduction towards cooler continent. Section E, located in Fig. 4, crosses the area of abnormally thin oceanic crust at the southernmost end of the Ligurian Sea. A thermal model including continental rifting (from 30 to 23 Ma), oceanic spreading (from 23 to 19 Ma) and post-spreading cooling (from 19 Ma to present) was already presented and discussed in a previous section (see Fig. 7). We are interested now in the very early phase of oceanic spreading. The rifting history and amount of stretching is taken identical to the model of Fig. 7. The steadystate solutions (including lateral conduction) for intermediate (40 mm a-1) and slow (18.5 mm aq) spreading are shown as large dots in Fig. 10. This is equivalent to the ocean-continent boundary being located at an infinite distance from the
52
N. CHAMOT-ROOKE E T A L .
Fig. 10. (a) Predicted subsidence of the oceanic lithosphere as a function of proximity of the oceancontinent boundary. An intermediate full spreading rate of 40 mm a-1 is assumed. The ridge axis remains located on the left of the diagram, whereas the ocean-continent boundary (OCB) is progressively moving away. The 2D steady-state solution (ocean-continent boundary at infinity) is shown by large dots. Thin lines are subsidence curves (below water) from the ridge axis to the OCB, 1 Ma (OCB at 20 km) and 2 Ma (OCB at 40 km) after initiation of spreading (modelled section is section E in Fig. 4). The shaded grey area is the location of maximum melt extraction. (b) Same as (a) for slow spreading (18.5 mm a 1 full rate, equivalent to best value obtained from Fig. 9b).
ridge axis. We then model the transient regime as the ocean-continent boundary moves away from the ridge axis. At high spreading rate, the steadystate regime at the ridge axis is reached in less than 1 Ma (in other words, the proximity of the continent no longer affect the ridge thermal regime). In the slow spreading case, thermal equilibrium at the ridge axis is reached within 3 Ma. Even though melt production rates decrease rapidly with distance from the ridge, melt extraction may not be limited to the ridge axis itself. The lateral dimension of the melting region is still a matter of debate (e.g. Scott & Stevenson 1989; Cordery & Morgan 1993). It may reach several tens of kilometres, in which case the melt would be focused towards the ridge axis (Spiegelman & McKenzie 1987). In terms of volume,
however, melting models seem to indicate that about 80-90% of the melt that will contribute to oceanic crust genesis is produced within 20-30 km of the ridge axis. For the Ligurian Sea section, a steady-state regime over such distances will be reached within 2 Ma at a high spreading rate (40 mm a-t); at a lower spreading rate (18.5 mm a-l), it may take more than 4 Ma to establish a permanent thermal regime (see Fig. 10). The total oceanic basin width would then be 80-100 km in both cases. Above this width, lateral conduction towards continent no longer affects the temperature field in the area where melt extraction is maximum. Notice that we discuss here the area where melt extraction is maximum, and not the total width of the melting region which may reach 80 to 100 km (Cordery & Morgan 1993). The width of the oceanic domain along the modelled Ligurian Sea section is between 60 and 80 km. A first conclusion is that lateral conduction cannot be neglected, in particular if spreading was slow. Part of the anomalous oceanic crustal thickness may be the result of side heat loss towards both continental masses (Provence and Corsica). Northward, lateral cooling during rifting may even have been sufficiently large to prevent large-scale melting and the formation of oceanic crust. For similar stretching factors and duration of rifting, the critical total basin width may be of the order of 70-100 km (Alvarez et al. 1984), which is about the size of the northernmost Ligurian Sea basin. Notice that along section F, crustal thickness is clearly too thick to be compatible with an oceanic origin (Fig. 8). Three coeval effects will act to lower considerably magma supply: (1) a low spreading rate due to the proximity of the pole of opening; (2) a low potential temperature, as discussed previously; (3) high lateral conduction due to the small basin width. Oceanic crust produced, if any, would necessary be very thin. The northernmost Ligurian Sea is thus most probably floored with thinned continental crust. A n i m p o r t a n t implication of our thermal modelling is that thin abnormal oceanic crust, formed during the early stage of oceanic spreading, can be found within 40-50 km of the o c e a n - c o n t i n e n t boundary. Variability of oceanic crust thickness is thus expected to be found not only along strike, as an effect of a narrowing basin towards the Ligurian Sea, but also across strike, close to the o c e a n - c o n t i n e n t boundary. In the following sections, we first discuss estimates of the opening rate of the Liguro-Provenqal Basin, since it is one of the important parameters controlling melting and oceanic crust genesis. We then examine the early
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN
53
closely spaced in time between 16.2 and 17 Ma) rather than anomaly 6. The alternative is that the tristanite ridge erupted after cessation of spreading. Then the positive magnetic anomaly of the ridge, aligned with the positive axial anomaly of the basin, would be fortuitous. Rate and timing of the Corsica-Sardinia Additional constraints are given by the timing block rotation of the rotation of the Sardinia-Corsica block. Estimates of the rate and timing of opening of The amount of Tertiary rotation has been widely the Liguro-Proven~al basin have been obtained debated. Apart from structural evidences based on a common geological history of the Sarfrom different types of data: structural restoration of the position of the Sardinia-Corsica dinia-Corsica block and the Provenqal mainland block, based on pre-rotation markers (e.g. West- (e.g. Westphal et al. 1976), palaeomagnetic phal et al. 1976), seismic stratigraphy of pre-rift, measurements allowed various determinations syn-rift and post-rift horizons (e.g. Mauffret et of the amount and timing of the rotation. Data used range from palaeomagnetic measurements aL 1995), age of volcanism in Sardinia (Bellon et on the Tertiary volcanic rocks of Sardinia al. 1977), palaeomagnetic data in Sardinia (Montigny et al. 1981) and Corsica (Vigliotti & Kent (Bellon et aL 1977; Edel & L6rtscher 1977; Edel 1990), identification of the magnetic anomalies 1979; Montigny et al. 1981), aeromagnetic in the oceanic domain (Burrus 1984), age of measurements over Sardinia (Gald6ano & dredged samples in the basin (R6hault 1981). Ciminale 1987), and finally palaeomagnetic We focus here on evidence for age and rate of study of the Tertiary sediments from Corsica oceanic spreading leading to quantitative esti- (Vigliotti & Kent 1990), since Tertiary calc-alkamates: age of dredged samples, age of magnetic line volcanism is absent there. Different data lineations, age and amount of rotation of Sar- sets consistently point to a 30-40 ~ counterclockdinia and Corsica rocks derived from palaeo- wise rotation of the Sardinia-Corsica block with respect to stable Europe. The timing of the rotamagnetism. The basement is buried below a thick sedi- tion, however, remains uncertain. Dating relies mentary cover over the entire Liguro-Provenqal both on K - A r age determinations of the Sarbasin, with the exception of an exposed ridge in dinia andesitic and ignimbritic suite, compiled in the Ligurian Sea. The southernmost part of the Montigny et al. (1981), and Miocene fossiliferous Ligurian Sea is floored with very thin oceanic limestones of the Neogene basins of Corsica crust. Crustal thickness gradually increases (Vigliotti & Kent 1990). Both sets of data are towards the northeasternmost portion of the apparently conflicting. Montigny et al. argue for Ligurian Sea, which is most probably floored a 20.5-19 Ma (early Burdigalian) rapid rotation. with stretched continental crust (see previous This is the solution retained by Burrus (1984) to section). Along section F (Fig. 8) the crust is thin identify magnetic lineations previously delinat ESP 224 only (5-6 km). Immediately east of it eated by Bayer et al. (1973). Vigliotti & Kent, on (ESP 223) runs an axial N30~ linear the other hand, conclude that a rotation of about ridge cropping out at its northernmost end 20 ~ was still to be completed in post-Burdi(R6hault 1981; R6hault et al. 1984). The ridge galian-Langhian time. They propose that the was sampled by dredging, and tristanite samples rotation did not end before 15 Ma (late of mid-Miocene age were recovered (18 _+ Langhian). Their results are controversial not 0.5 Ma in R6hault et al. 1984). Recent re-dating only because they contradict the Montigny et al. would indicate a younger 17 Ma age (R~hault compilation, but also, as they state themselves, pers. comm. 1996). The tristanite ridge coincides because secondary magnetization cannot be with the central magnetic anomaly high recog- completely ruled out. However, their conclusion nized throughout the basin and may thus repre- is not different from that initially reached by sent the latest magmatic rocks erupted at the Bellon et al. (1977) from their palaeomagnetic end of the spreading phase (R6hault et al. 1984). study of the Tertiary volcanism of Sardinia. Palaeomagnetic data from the Montigny et al. The discrepancy between the age of anomaly 6, proposed as the axial anomaly (Burrus 1984), compilation for Sardinia and the Vigliotti & and the age of the tristanite ridge has already Kent data for Corsica are summarized in Fig. 11. been mentioned in R6hault et al. (1984). Notice that we use all data reported in Montigny Anomaly 6 normal polarity interval lasted i Ma, et al. rather than a selection of the data set. from c. 20.3 to 19.3 Ma (Berggren et aL 1985; Youngest ages, in particular, were systematically Harland et al. 1990). A 17 Ma age would be more omitted in their original plot. Results obtained compatible with anomaly 5C (three triplets from the sedimentary rocks of Corsica by oceanic stage leading to the formation of oceanic crust close to the ocean-continent boundary.
54
N. CHAMOT-ROOKE E T A L .
Vigliotti & Kent are actually not incompatible with Sardinian palaeomagnetic measurements (see also Vigliotti & L a n g e n h e i m 1995). Although scattering for the Sardinia volcanic rocks is high, it is clear that all rocks, regardless of their ages, show a westward oriented magnetization. Taking into account the apparent polar wander path for Eurasia, the Miocene pole position predicts a +7 ~ palaeomagnetic declination at the location of the Sardinia-Corsica block (Vigliotti & Kent 1990). The conclusion is that almost none of the observed sites, in the range 35-15 Ma, remained stable with respect to Eurasia. Although a rapid rotation cannot be completely ruled out, a much slower rotation ending in Langhian or Serravalian time would still be compatible with the palaeomagnetic data. Magnetic anomalies i n t e r p r e t a t i o n in the oceanic domain is rather inconclusive. The welllineated axial positive anomaly is bounded by one additional positive anomaly on both sides,
so that only three anomalies are available for age identification. Considering the set of possible anomalies in the range 25-10 Ma, most of which are more or less regularly spaced in time every 1 Ma, several interpretations are possible. The recognition of the axial anomaly as anomaly 6 (20 Ma) is clearly not compatible with available palaeomagnetic data. If the identification of the Late Burdigalian horizon throughout the basin is correct (Mauffret et al. 1995), then the oceanic crust cannot be younger than 16 Ma. Further, onlapping of this horizon on the basement in the axial part of the basin would still have to be proved. Could this horizon be younger, and the Late Burdigalian be deeper? A possible deeper candidate for the Late Burdigalian horizon is present on the ECORS seismic line (horizon X in de Voogd et al. 1981). This horizon is onlapping the acoustic basement a few tens of kilometres away from the inferred location of the fossil ridge. Stratigraphic correlation of the Late Burdigalian from GLP2 well
Fig. 11. Palaeomagnetic data for the Tertiary volcanic rocks of Sardinia (after Montigny et al. 1981; filled dots, normally magnetized; open dots, reversely magnetized) and for the Miocene sedimentary rocks of Corsica (square symbol; after Vigliotti & Kent 1990). The timing of Western Mediterranean basins opening is shown on the right (dark line, oceanic spreading; grey lines, continental rifting, dashed if not well constrained).
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN (close to ESP 202 on the upper Gulf of Lion margin) to the deep basin would then have to be reconsidered. The exact age of oceanic spreading in the Liguro-Provenqal basin is thus still open to discussion. Cessation of spreading in Late Burdigalian time (16.5 Ma) would be more or less compatible with palaeomagnetism, stratigraphic correlations and age of the tristanite ridge. Palaeomagnetism alone would tend to indicate an even younger Langhian age (14 Ma?). Calcalkaline volcanic rocks erupted in Sardinia until 13 Ma (Bellon et al. 1977), indicating that subduction was still going on beneath, and that the source for volcanism moved away not before that time (possibly in relation to opening of the Tyrrhenian Basin). Notice that the heat flow in the axial part of the basin is well above 120 mW m -z, which is the theoretical value for a 15 Ma aged basin. Even in the more conservative hypothesis, oceanic spreading between 23 Ma (post-syn-rift Early Aquitanian deposits) and 16.5 Ma (pre-Late Burdigalian) requires reconsideration of spreading rate estimates. The width of the oceanic domain along the E C O R S line (section A, Fig. 8) is between 150 km (Mauffret et al. 1995) and 230 km (Burrus 1984 or R6hault et al. 1984). Since no differential rotation was found in the p a l a e o m a g n e t i c data between 30 and 23 Ma, most of the rotation was thus achieved during the oceanic spreading phase (Bellon et al. 1977; Burrus 1984). A small oceanic domain, less than 200 km wide, can then be excluded. This would locate the pole of rotation too close (somewhere in the southern Ligurian Sea) to reach the 40 ~ rotation of the Corsica-Sardinia block. A self-consistent solution is obtained if the oceanic width is 230 + 20 km. The pole of rotation is found 330 km away, somewhere in the Gulf of Genoa, close to the location discussed previously. The mean full spreading rate would then be 30 _+ 5 mm a q, about half the value assumed in Burrus (1984). Plate separation rates during rifting can be estimated from the amount of stretching across extended conjugate margins. Assuming that the volume of continental crustal rocks remained constant during rifting, calculations for section A (Fig. 12) indicate about 150 + 30 km of total continental extension (i.e. displacement of Sardinia; similar estimates are found in Le D o u a r a n et al. 1984). Taking into account the duration of rifting (7-10 Ma), the mean plate separation rate during rifting was then 20 _+7 mm a q, to be compared to 30 _+ 5 mm a -1 during spreading. We thus favour a continuous extension mechanism t h r o u g h o u t the basin history, rather than a sudden pulse of oceanic accretion. Extension in
55
the Liguro-Provenqal basin was then relieved by the opening of the Tyrrhenian basin. Although rifting in the Tyrrhenian basin may have started as early as Early Miocene (post-early-Burdigalian?, Sartori 1990), the entire basin was not rifted until Tortonian (9 Ma, Sartori 1990; Kastens et al. 1990). Extension was, however, already active in Alpine Corsica in Late Oligocene or Early Miocene time, since ductile extension there ended with the deposition of well-dated Burdigalian limestones in Corsica and Langhian sediments in the Corsica Basin (Jolivet et al. 1991, 1994). During Pliocene time, oceanic crust was produced in two small oceanic basins (Kastens et al. 1990), the Vavilov basin (possibly spreading in Early Pliocene time) and Marsili basin (younger than 2 Ma). The recent M6d6e survey revealed that the toe of the Calabrian prism in the Ionian Sea is no longer active (Le Pichon et al. pers. comm.), indicating that extension in the Tyrrhenian Sea may have recently ceased (notice that post-spreading volcanism may be active with little or no extension, as found in many of the back-arc basins of the West Pacific). Drilling results in the Tyrrhenian Sea (Kastens et al. 1990) are compatible with the outward arc migration model of Malinverno & Ryan (1986) which emphasizes the role of 'rollback' (slab retreat) of the African subduction hinge zone, now located in the Ionian Sea. Restoration of the Tyrrhenian margins to their original widths is somewhat speculative, in particular for the north Calabrian margin. According to Malinverno & Ryan (1986), the total extension (continental stretching plus oceanic spreading) across the basin was 330-350 km, uncertainty amounting to several tens of kilometres. Notice that this is close to the amount of extension obtained on Section A across the Gulf of Lion and Sardinia margin (380 e 50 km), thus suggesting some sort of self-regulating mechanism (such as distance from the subduction front to the back-arc accreting ridge). If we assume that rifting in the Tyrrhenian began 9 Ma ago, then the mean opening rate would be 40 mm a -1. This may be regarded as a maximum value. If we include significant preTortonian rifting, the opening rate may be as low as 20 mm a -1 (Malinverno & Ryan 1986). A high oceanic spreading rate was r e p o r t e d in the Marsili basin (70-80 km over a 2 Ma period, 40 e 5 mm a -1, Kastens et al. 1990), but the exact width of the oceanic domain remains uncertain. The total extension in the Liguro-Provenqal and Tyrrhenian basins, taken along a N135 ~ azimuth, is thus of the order of 700 km (el00 km) since Oligocene time. Opening most
56
N. CHAMOT-ROOKE E T A L .
Fig. 12. Crustal section obtained from 3D gravity inversion (Model A solution) along the ECORS profile (section A located in Fig. 4; Fig. 8 top). Top: observed and modelled Bouguer.
probably became effective once the entire Liguro-Provenqal basin was rifted (possibly in Late Oligocene time). Le Pichon (1984) compared this amount of extension with the presentday length of the Calabrian seismic zone. The seismic slab is 650 km long, dipping 70 ~ along a N315 ~ azimuth (Selvaggi & Chiarabba 1995). Convergence of Africa towards Europe since the Burdigalian may be more than 200 km in that part of the Mediterranean Sea (Dewey et al. 1989), or 150 km taken along the azimuth of the slab. The remaining 500 km can thus be directly compared to the amount of extension in the Western Mediterranean basins. The length of the slab is clearly too long compared to the amount of extension in the Tyrrhenian basin only. If the Liguro-Provenqal basin is included, then total extension becomes greater than the length of the seismic slab. Le Pichon (1984) concluded that part of the slab (the last 200 km, subducted in Late Oligocene time) may now be aseismic following reheating at depth. This may be substantiated by tomographic results (e.g. De
Jonge et al. 1994), but resolution becomes poor below 400 km depth. The succession of basins opening thus suggests a continuum of extension since Oligocene time, with progressive southeastward migration of basins formation following rollback of the African hinge zone as proposed by Malinverno & Ryan (1986). We suggest that the time gap between opening of the Liguro-Provenqal Basin and stretching of the Tyrrhenian Basin should be reconsidered. Cessation of spreading in the Liguro-Provenqal basin 19 Ma ago (as generally assumed), and post-Tortonian rifting of the Tyrrhenian Basin (post-10 Ma) would imply high velocity African trench retreat in the early stage (> 60 mm a-I), followed by strike-slip motion for 10 Ma (to accommodate the northward motion of Africa with respect to the Corsica-Sardinia-Calabria block then belonging to Europe), and finally subduction again (> 40 mm a -l) during Tyrrhenian opening. We favour a model in which trench retreat proceeded at a more or less constant velocity
MOHO DEPTH IN THE LIGURO-PROVEN~AL BASIN (30-40 mm a-1) while consuming the old Ionian oceanic lithosphere. Side continental collisions (in the Northern Apennines, between the Kabylies and Africa, in Sicily) may have slowed down the upper plate motion locally, without affecting significantly subduction motion along the remaining oceanic portions of the Ionian lithosphere driven by slab pull (Le Pichon 1982; Le Pichon & Alvarez 1984). Crust at the ocean-continent
boundary
The western portion of section A (Figs 7 and 8) is along the ECORS deep-seismic profile and cut across ESP 201 to 209 mid-points. Eastern portion of section B is close to the CROP deepseismic line. The two sections illustrate the contrast between the wide and smoothly stretched Gulf of Lion margin and the narrow and abruptly stretched Provenqal margin. On both sections a very thin crust (2-3 km) is found at the base of the lower continental margin (ESP 204 on section A, ESP 211 on section B). Our gravity modelling suggests that thin crust should also be found over the conjugate Sardinian margin (Fig. 4), but no further seismic information is available there. On the Gulf of Lion margin, the detailed analysis of Pascal et al. (1993), based on x-t to -r-p transformations and synthetic "r-p seismograms, provides reliable P-wave velocity structure down to Moho depth. The upper crust over the entire basin, including thin continental and oceanic domains, is characterized by a 5.6-5.8 km s -1 velocity layer. The mid-crust velocity increases laterally oceanward from ESP 203 and 212 to ESP 204 and 211 respectively, from 6.2 to 6.6 km s-1. Below the thin crust area is a high-velocity layer with velocities ranging from 7.1 to 7.4 km s-a. The thickness of the high-velocity layer is maximum at ESP 203 (2.6 km) and at ESP 212 (2.9 kin). It reaches 1 km at ESP 211 and 212. As pointed out by Mauffret et al. (1995), the high-velocity layer is apparently not restricted to the very base of the margin, but extends oceanward (0.4 km thick at ESP 205 and 1.4 km thick at ESP 206). At the base of the Gulf of Lion margin, the top of the high P-wave velocity body correlates with the oceanward rising of a shallow bright reflector on the ECORS line ('T' reflector in Pascal et al. 1993). The T reflector comes very close to the top of basement between ESP 211 and 212 (LIGO 4 seismic profile, Pascal et al. 1993). This is also the area of thinnest crust in our gravity inversion. Landward, the T reflector coincides with Moho. The continent-ocean boundary has been
57
placed at various locations along the ECORS line: close to ESP 203 based on magnetic anomalies (Bayer et al. 1973), between ESP 205 and ESP 206 based on the limit of synrift deposits (Burrus 1984; Le Douaran et al. 1984), and more recently as far as ESP 206 based mainly on the lack of the high velocity layer beyond this point (Mauffret et aL 1995), typical oceanic layer 3 velocities (6.8 km s-1) being found at ESP 207. In the latter case, the transitional area (possibly thinned continental crust) would be 100 km wide. Thin crust at the base of the Gulf of Lion shows strong similarities with transitional type crust observed on other non-volcanic margins. Thin crust overlying a high velocity layer is found on the western margin of Iberia (3-4 km thick above a 7.3-7.6 km s-1 layer, Whitmarsh et al. 1990, 1993), in the Tagus abyssal Plain (2 km thick above 7.6 km s-~ layer increasing to 7.9 km s-a towards Moho, Pinheiro et aL 1992) and its conjugate Newfoundland margin off Grand Banks (2-3 km thick above a 7.2-7.7 km s-1 layer, Reid 1994), across Southwest Greenland margin (2.5 km thick above 7.0-7.6 km s-1 layer, Chian & Louden 1994) and its conjugate Labrador margin (1-2 km thick above a 6.4-7.7 km s-1 layer, Chian et al. 1995). On the southern Newfoundland margin the high-velocity body is limited by one or two landward-dipping reflectors rising to basement surface seaward and connecting to Moho landward (Keen & de Voogd 1988; Reid 1994). The similarity with the T reflector of the Gulf of Lion is striking. The origin of the high velocity lower crust on non-volcanic margins (underplating or serpentinite ?) and the nature of the thin overlying crust is still debated (Chian et al. 1995; Srivastava & Roest 1995). Further discussion is beyond the scope of this paper, although much could be learned by integrating the Gulf of Lion data into a much broader review of tectonic processes at non-volcanic margins. A few conclusions can, however, be drawn from the preceding sections. Underplating is very unlikely. Taking into account cooling during rifting (which lasted at least 7 Ma) and low initial mantle temperature (30-50~ below normal), the amount of melt generated during rifting was probably very small as for other non-volcanic margins (Bown & White 1995a). The thin crust flooring the lower margin of the Gulf of Lion as well as the lower margin of Sardinia may be of oceanic origin. Thermal modelling (Fig. 10) demonstrates that the width of abnormal thin oceanic crust may reach 40-50 km, which is about the right amount obtained by gravimetric
58
N. C H A M O T - R O O K E E T A L .
modelling and seismic refraction (see sections in Fig. 8).
Conclusions The deep structure of the Liguro-Provenqal Basin is discussed using gravity as an additional control to existing sparse seismic data. The main results can be summarized as follows. (1) 3D gravity inversion using Fourier transforms is a potential powerful tool to obtain a reliable geometry of the Moho discontinuity in areas where only sparse deep seismic informations are available. Free parameters in the inversion (Moho reference depth and crust-mantle density contrast) can be obtained through least squares minimization of the standard deviation of the gravimetric solution at some given reference points where the Moho is seismically constrained, such as ESP data. The misfit in the Liguro-Provenqal Basin is about 2 km for a 12-30 km Moho depth range. Corrections for variable crust density and variable heat flow distribution can also be included, although in the particular case of the Liguro-Proven~al Basin they do not improve significantly the solution. (2) The oceanic crust thickness in the LiguroProvenqal Basin is smaller than usually reported in open oceans (5 + 1 km instead of 7 e 1 km). We show that the thin oceanic crust is compatible with a 1280~ potential mantle temperature, which is 30-50~ below normal. Low temperature is also responsible for the abnormal subsidence of the deep Liguro-Provenqal Basin. Thin crust and high subsidence have been reported in other marginal basins, in particular in the West Pacific. Subduction of cold material may be responsible for significant cooling at depth. However, thermal conduction is slow and would probably require a long history of subduction. For the same reason (slow conductive process), lithospheric thickening coeval with orogen formation and prior to extension may be a more efficient way to lower temperatures at mantle depth. (3) The decrease of oceanic crust thickness towards the Ligurian Sea to values as small as 2 km indicates a very low magma supply. The temperature field there is affected by two different effects, both of conductive origin : heat is lost laterally towards the adjacent continents (Provenqal margin and Corsica), thus preventing large-scale melting even during oceanic accretion; rate of opening in the Ligurian Sea falls below the critical 15 mm a -1 full spreading rate below which magma supply becomes negligible. An important implication is that in the northern Ligurian Sea, where lateral conduction is maximum and distance to the pole of opening
minimum, the crustal thickness is too thick to be of oceanic origin. (4) The thin crust at the o c e a n - c o n t i n e n t boundary may be oceanic crust produced in the very early stage of oceanic spreading, as proposed for some of the non-volcanic Atlantic margins. Several million years are required to establish a steady-state thermal regime at the ridge axis, due to lateral cooling towards unextended continental areas. The lower margin of the Gulf of Lion has actually many of the characteristics e n c o u n t e r e d on other non-volcanic margins: tilted blocks rooted on a seawarddipping reflector, high P-velocity body bounded to the top by a landward-dipping reflector, and connecting at depth with Moho towards the continent, very thin crust in the transitional region from thinned crust to oceanic crust. (5) A close re-examination of magnetic data (palaeomagnetic data and magnetic lineations) shows that the exact age of oceanic spreading, as well as the opening rate, is still debatable. Cessation of spreading in Late Burdigalian time (16.5 Ma) would be a reasonable compromise between somewhat divergent palaeomagnetism, stratigraphic interpretation and age of dredged samples data. An even younger age cannot be completely ruled out. In any case, we favour a constant and slow opening at a rate of 5 ~ Ma -1, rather than a sudden pulse of oceanic accretion. (6) The new Moho depth map is used to estimate the amount of extension across the LiguroProvenqal Basin. Comparison with the Tyrrhenian Basin and length of the Calabrian slab suggests a continuous extension mechanism since Oligocene time driven by the African trench retreat, following Malinverno & Ryan (1986) model. It is suggested that the process continued u n i n t e r r u p t e d until recent time, unless drastic velocity changes occurred at the subduction zone where the old Ionian oceanic lithosphere was being consumed. This paper is within the framework of the Integrated Basin Studies (IBS) program. We are grateful to G. Pascal for his helpful comments on reflection and refraction data interpretation. Stimulating lectures given by X. Le Pichon at Collbge de France in 1996 helped in clarifying some of the ideas developed here. We thank our colleagues who made available to us some of their unpublished data, in particular A. Mauffret for his digital file of sediment thickness around the Gulf of Lion margin and C. Truffert for some new processed maps of the magnetic field.
References ALVAREZ, F., VIRIEUX, J. & LE PICHON, X. 1984.
Thermal consequences of lithosphere extension
M O H O D E P T H IN T H E L I G U R O - P R O V E N ~ A L BASIN over continental margins : the initial stretching phase. Geophysical Journal of the Royal Astronomical Society, 78, 389-411. ANGUY, Y., DAIVlOTTE,B. & Room~, E 1991. Tirs sismiques latdraux compl6mentaires au profil ECORS-Pyr6ndes : apport a la connaissance de l'architecture profonde de la cha~ne. Cornptes rendus de l'AcadOmie des Sciences, Paris', sot. D, 313, 677-684. ARTHAUD, F., OGIER, M. & SEGURET,M. 1980. G6ologie et Gdophysique du golfe du Lion et de sa bordure nord. Bulletin du Bureau de Recherches g(ologiques et miniOres, 1, 175-193. BANDA, E., ANSORGE, J., BOLOIX, M. & CORDOBA, D. 1980. Structure of the crust and upper mantle beneath the Balearic islands (western Mediterranean). Earth and Planetary Science Letters, 49, 219-230. BAYER, R., LE MOUEL, J.-L. & LE PICHON, X. 1973. Magnetic anomaly pattern in the Western Mediterranean. Earth and Planetary Science Letters, 19, 168-176. BELLON, H., COULON, C. & EDEL, J.-B. 1977. Le ddplacement de la Sardaigne. Synth6se des donndes gdochronologiques, magmatiques et pal6omagndtiques. Bulletin de la SociOtO g(ologique de France, 7, 825-831. BERGGREN, W. A., KENT, D. g., FLYNN, J. J. & VAN COUVERIN6, J. A. 1985. Cenozoic geochronology. Geological Society of America Bulletin, 96, 1407-1418. BESSlS, E 1986. Some remarks on the study of subsidence of sedimentary basins. Application to the Gulf of Lions margin (Western Mediterranean). Marine and Petroleum Geology, 3, 37-63. BowN, J. W. & WHITE, R. S. 1994. Variation with spreading rate of oceanic crustal thickness and geochemistry. Earth and Planetary Science Letters, 121, 435-449. & -1995a. Effect of finite extension rate on melt generation at rifted continental margins. Journal of Geophysical Research, 100, 18011-18029. & 1995b. Finite duration rifting, melting and subsidence at continental margins. In: BANDA,E., TORNI~, M. & TALWANI, M. (eds) Rifted Continent-Ocean Boundaries. Kluwer Academic Press, Netherlands, 31-54. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provenqal Basin (North-Western Mediterranean). Marine Geology, 55, 247-269. /989. Review of geodynamic models for extensional basins; The paradox of stretching in the gulf of Lions (Northwest Mediterranean). Bulletin de la SociOt( gdologique de France, 8, 377-393. CtteN, Y. J. 1992. Oceanic crustal thickness versus spreading rate. Geophysical Research Letters, 19, 753-756. CHIAN, D. & LOUDEN,K. E. 1994. The continent-ocean crustal transition across the southwest Greenland margin. Journal of Geophysical Research, 9117-9136. & REID, I. 1995. Crustal structure of the Labrador Sea conjugate margin and implications
-
-
-
-
-
-
9 9 ,
59
for the formation of non-volcanic continental margins. Journal of Geophysical Research, 100, 24239-24253. CHOUKROUNE, R & TEAM, E. t989. The ECORS pyrenean deep seismic profile reflection data and the overall structure of an orogenic belt. Tectonics, 8, 23-39. --, ROURE, E, PINET, B. & TEAM, E. P. 1990. Main results of the ECORS Pyrenees profile. Tectonophysics, 173, 411-423. CORDERY, M. J. & MORGAN,J. P. 1993. Convection and melting at mid-ocean ridges. Journal of Geophysical Research, 98, 19477-19503. COWIE, P A. & KARNER, G. D. 1990. Gravity effect of sediment compaction : examples from the North Sea and the Rhine Graben. Earth and Planetary Science Letters, 99, 141-153. DANOBEITIA, J. J., ARGUEDAS,M., GALLART,J., BANDA E. & MAKRIS, J. 1992. Deep crustal configuration of the Valencia trough and its Iberian and Balearic borders from extensive refraction and wide-angle reflection seismic profiling. Tectonophysics, 302, 37-55. DE JONGE, M. R., WORTEL,M. J. R. & SPAKMAN,W. 1994. Regional scale tectonic evolution of the seismic velocity structure of the lithosphere and the upper mantle: the Mediterranean region. Journal of Geophysical Research, 89, 12091--12108. DE VOOGD, B., NICOL~CH,R., & 21 OTHERS. 1991. First deep seismic reflection transect from the Gulf of Lions to Sardinia (Ecors-Crop profiles in Western Mediterranean). In: Continental lithosphere: deep-seismic reflections. American Geophysical Union, 265-274. DELLA VEDOVA, B., LUCAZEAU, E, PASQUALE, V., PELLIS, G. & VERDOYA, M. 1995. Heat flow in the tectonic provinces crossed by the southern segment of the European Geotraverse. Tectonophysics, 244, 57-74. DEWEY, J. E, HELMAN, M. L., TURCO, E., HUTYON, D. H. W. & KNOTT, S. D. 1989. Kinematics of the western Mediterranean. In: COWARD,M.R, DIETRICH, D. & PARK, R.G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. EDEL, J.-B. 1979. Paleomagnetic study of the Tertiary volcanics of Sardinia. Journal of Geophysics, 45, 259-280. & LORTSCHER,A. 1977. Pal6omagn6tisme du volcanisme tertiaire de Sardaigne. Nouveaux r6sultats et synth6se. Bulletin de la Soci~t~ g~ologique de France, 7, 815-824. EGGER, A. 1992. Lithospheric structure along a transect from the northern Apennines to Tunisia derived .from seismic refraction data. PhD Thesis, Swiss Federal Institute of Technology, Zarich. DEMARTIN, M., ANSORGE, J., BANDA, E. & MAIESTRELLO,M. 1988. The gross structure of the crust under Corsica and Sardinia. Tectonophysics, 150, 363-389. FOUCHER, J. E, LE PICHON,X. & SIBUET,J. C. 1982. The ocean-continent transition in the uniform lithospheric stretching model; role of partial melting in the mantle. In: KENT, E, BOTT, M. H. P.,
60
N. C H A M O T - R O O K E E T A L .
McKEyzm, D. P. & WmLIAMS, C. A. (eds) Evolution of sedimentary basins. Philosophical Transactions of the Royal Society of London, A305, 27-43. GALDI2ANO, A. & CIMINALE, M. 1987. Aeromagnetic evidence for the rotation of Sardinia (Mediterranean Sea): comparison with the paleomagnetic measurements. Earth and Planetary Science Letters, 82, 193-205. GALLART, J., DAIGNII~RES,M., BANDA, E., SURINACH, E. & HIRN, A. 1980. The eastern Pyrenean domain: lateral variations at crust-mantle level. Annales Geophysicae, 36, 2, 141-158. GmSE, P, MOr~ELLI,C. & STEINMETZ,L. 1973. Main features of crustal structure in western and southern Europe based on data of explosion seismology. Tectonophysics, 20, 367-379. GORINI, C. 1993. G~odynamique d'une marge passive: le Golfe du Lion (MOditerranOe Occidentale). Th6se de 3~me cycle, Toulouse. HARLAND,W. B., ARMSTRONG,R. L., Cox, A. V., CRAIG, L., SMITH,A. G. & SMITH, D. G. 1990. A geologic time-scale 1989. Cambridge University Press, New York. JACKSON, H. R., REID, I. & FALCONER, R. K. H. 1982. Crustal structure near the Arctic mid-ocean ridge. Journal of Geophysical Research, 87, 1773-1783. JOLIVET, L., DANIEL, J.-M. & FOURNmR, M. 1991. Geometry and kinematics of ductile extension in alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. - - , TRUEFEm, C. & GOFFE, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in arc and back-arc regions. Lithos, 33, 3-30. KASTENS, K. & MASCLE, J. & ET AL. 1990. The geological evolution of the Tyrrhenian Sea: an introduction to the scientific results of ODP Leg 107. In: KASTENS,K., MAS('LE, J. ETAL. (eds) Proceedings of the Ocean Drilling Program, Scientific Results. College Station, TX (Ocean Drilling Program), 107, 3-26. KEEN, C. E. & DE VOOC;D,B. 1988. The continent-ocean boundary at the rifted margin off eastern Canada : new results from deep seismic reflection studies. Tectonics, 7, 107-124. LE DOUARAN, S., BURRUS,J. & AVEDIK, E 1984. Deep structure of the north-western Mediterranean Basin: Results of a two-ship seismic survey. Marine Geology, 55, 325-345. LE PICHON, X. 1982. Land-locked oceanic basins and continental collision: the Eastern Mediterranean as a case example. In: Hs0, K. (ed.) MountainBuilding Processes. Academic Press, London, 201-211. -1984. The Mediterranean Seas. In: Origin and History of Marginal and Inland Seas. Proceedings of the 27th International Geological Congress, V N U Science Press, Amsterdam, 23, 189-222. - & ALVAREZ,E 1984. From stretching to subduction in back-arc regions: dynamic considerations. Tectonophysics, 102, 343-357.
MCKENZIE, D. 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters, 40, 25-32. - & BICKLE, M. J. 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology, 29, 625-679. MAILLARD,A., MAUFFRET,A.,WATTS,A. B., TORNI~,M., PASCAL, G., BUHL, P. & P1NET, B. 1992. Tertiary sedimentary history and structure of the Valencia trough (western Mediterranean). Tectonophysics, 203, 57-75. MALINVERNO,A. & RYAN,W. B. E 1986. Extension in the Tyrrhenian Sea and shortening in the Appenines as result of arc migration driven by sinking of the lithosphere. Tectonics, 5, 227-245. MAUFFRET,A., MAILLARD,A., PASCAL,G., TORNI~,M., BUHL, E & P1NET, B. 1992. Long-listening multichannel profiles in the Valencia trough (Valsis 2) and the Gulf of Lions (ECORS): a comparison. Tectonophysics, 203, 285-304. --, PASCAL, G., MAILLARD, A. & GORINI, C. 1995. Tectonics and deep structure of the north-western Mediterranean basin. Marine and Petroleum Geology, 12, 645~66. MONTIGNY, R., EDEL J.-B. & THUIZAT,R. 1981. OligoMiocene rotation of Sardinia : K-Ar ages and paleomagnetism data of tertiary volcanics. Earth and Planetary Science Letters, 54, 261-271. MOtJSSAT, E. 1983. Evolution de l a m e r tyrrh~nienne centrale et orientale et de ses marges septentrionales en relation avec la nOotectonique de l'arc calabrais. Thbse de 36me cycle, Universit6 Pierre et Marie Curie, Paris. NAFE, J.E. & DRAKE, C. L. 1963. Physical properties of marine sediments. In: HILL, M.N. (ed.) The Sea. Interscience Publishers, 3, 794-815. PARKER, R. L. 1973. The rapid calculation of potential anomalies. Geophysical Journal of the Royal Astronomical Society, 31,447M55. PASCAL,G.,TORNE, M., BtJHL, E, WATVSA. B. & MAUFFRET,A. 1992. Crustal and velocity structure of the Valencia Trough (Western Mediterranean). Part II: detailed interpretation of five expanded spread profiles. In: BANDA, E. & SURINA('H, P. (eds) Geology and Geophysics of the Valencia Trough, Western Mediterranean. Tectonophysics, 203, 285-304. --, MAIJFFRET, A. & PATRIAT,P. 1993. The oceancontinent boundary in the Gulf of Lion from analysis of expanding spread profiles and gravity modelling. Geophysical Journal International, 113, 701-726. PINHEIRO, L. M.,WtlrI'MARSIt, R. B. & MILES,P. R. 1992. The ocean-continent boundary off the western continental margin of iberia, II, Crustal structure in the Tagus Abyssal Plain. Geophysical Journal International, 109, 106-124. REHAULT, J.-E 1981. Evolution tectonique et s~dimentaire du bassin ligure (M~diterranOe occidentale). Doctorat d'Etat, Universit6 Pierre et Marie Curie, Paris. - - - , BOILLOT, G. & MAUFFRET,A. 1984. The Western Mediterranean Basin geological evolution. Marine Geology, 55, 447-477.
M O H O D E P T H IN T H E L I G U R O - P R O V E N ~ A L BASIN , MOUSSAT,E. & FABBRI,A. 1987. Structural evolution of the tyrrhenian back-arc basin. Marine Geology, 74, 123-150. , TISSEAU, C., BRUNET, M.-E & LOUDEN, K. 1990. Subsidence analysis on the Sardinian margin and the central Tyrrhenian Basin: thermal modelling and heat flow control; deep structure implications. In: DANOBEITIA,J. J. & PINEY,B. (eds) Geophysics
of the Mediterranean Basin. Journal of Geodynamics, 269-310. REID, I. 1994. Crustal structure of a non-volcanic rifted margin east of Newfoundland, Journal of Geophysical Research, 99, 15 161-15 180. & JACKSON, H. R. 1981. Oceanic spreading rate and crustal thickness. Marine Geophysical Research, 5, 165-172. SANDWELL, D. T. & SMITH, W. H. E 1995. Marine
-
-
Gravity Anomaly from Satellite Altimetry (Map). Geological Data Center, Scripps Institution of Oceanography. SAPIN, M. & HIRN, A. 1974. Results of explosion seismology in the Southern Rh6ne Valley. Annales Geophysicae, 30, 181-202. SARTORI, R. 1990. The main results of ODP Leg 107 in the frame of Neogene to Recent geology of Perityrrhenian areas. In: KASTENS, K. & MASCLE, J. ET AL. (eds) Proceedings of the ODP, Scientific Results, 107, 7l 5-730. SCOTT,D. R. & STEVENSON,D. J. 1989. A self-consistent model of melting, magma migration and buoyancy-driven circulation beneath mid-ocean ridges. Journal of Geophysical Research, 94, 2973-2988. SELVAGGI,G. & CHIARABBA,C. 1995. Seismicity and Pwave velocity image of the Southern Tyrrhenian subduction zone. Geophysical Journal International, 121, 818-826. SPIEGELMAN, M. & MCKENZIE, D. 1987. Simple 2-D models for melt extraction at mid-ocean ridges and island arcs. Earth and Planetary Science Letters, 83, 137-152. SRIVASTAVA,S. P. • ROEST, W. R. 1995. Nature of thin crust across the southwest Greeland margin and its bearing on the location of the ocean-continent boundary. In: BANDA, E., TORNI~, M. & TALWANI, M. (eds) Rifted Continent-Ocean Boundaries. Kluwer Academic Press, Netherlands, 95-120. STECKLER, M. S. & WATTS, A. B. 1980. The Gulf of Lion: subsidence of a young continental margin. Nature, 287, 425-429. Su, W., MUTTER, C. Z., MUTTER, J. C. & BUCK, W. R. 1994. Some theoretical predictions on the relationships among spreading rate, mantle temperature, and crustal thickness. Journal of Geophysical Research, 99, 3215-3227. TALWANI,M. 8,~EW[NG, M. 1960. Rapid computation of gravitational attraction of three-dimensional bodies of arbitrary shape. Geophysics, 25, 203-225.
61
TORNt~, M., PASCAL, G., BUHL, e., WATTS A. B. & MAUFFRET,A. 1992. Crustal and velocity structure of the Valencia trough (western Mediterranean), Part I. A combined refraction/wide-angle reflection and near-vertical reflection study. Tectonophysics, 203, 1-20. VIALLY, R. & TRI~MOLII~RES,P. 1996. Geodynamics of the Gulf of Lions. Implications for petroleum exploration. In: ZIEGLER, E A. & HORVATH, E (eds) Structure and prospects of Alpine basins and forelands. Peri-Tethys Memoirs, 2. Edition Technip, Paris, 129-158. V|GLIOTTI, L. & KENT, D. V. 1990. Paleomagnetic results of Tertiary sediments from Corsica: evidence of post-Eocene rotation. Physics of the Earth and Planetary Interiors', 62, 97-108. & LANGENHEIM,V. E. 1995. When did Sardinia stop rotating? New paleomagnetic results. Terra Nova, 7, 424-435. WATTS,A. B. & TORNI~, M. 1992a. Subsidence history, crustal structure and thermal evolution of the Valencia Trough: a young extensional basin in the Western Mediterranean. Journal of Geophysical Research, 97, 20021-20041. & 1992b. Crustal structure and the mechanical properties of extended continental lithosphere in the Valencia trough (western Mediterranean). Journal of the Geological Society of London, 149, 813-827. , BUHL, P., MAUFFRET, A., PASCAL, G. & PINET,B. 1990. Evidence for reflectors in the lower continental crust before rifting in the Valencia trough. Nature, 348, 631-634. WESTPHAL, M., ORSINI, J. 8r VELLUTINI, J. 1976. Le microcontinent corsosarde, sa position initiale: donn6es paldomagn6tiques et raccords gdologiques. Tectonophysics, 30, 141-157. WHITE, R. S. & MCKENZIE, D. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research, 94, 7685-7729. - - , MCKENZIE, D. & O'NIONS, R. K. 1992. Oceanic crustal thickness from seismic measurements and rare earth element inversions. Journal of Geophysical Research, 97, 19 683-19 715. WHITMARSH, R. B., MILES, V. R. 8r MAUFFRET,A. 1990. The ocean-continent boundary off the western continental margin of Iberia, I, Crustal structure at 40~ Geophysical Journal International, 103, 509-531. - - , P1NHEIRO,L. M., MILES,P. R., RECQ, M. & SIBUET, J.-C. 1993. Thin crust at the western Iberia oceancontinent transition and ophiolites. Tectonics, 12, 1230-1239. ZEYEN, H. J., BANDA, E., GALLART, J. & ANSORGE, J. 1985. A wide angle seismic reconnaissance survey of the crust and upper mantle in the Celtiberian Chain of eastern Spain. Earth and Planetary Science Letters, 75, 393-402. -
-
-
-
Constraints on the N e o g e n e Mediterranean kinematic evolution along a 1000 km transect from Iberia to Africa J. V E R G I ~ S 1 & F. S A B A T 2
1Institute o f Earth Sciences 'Jaume A l m e r a ; CSIC, Solo i Sabaris s/n, 08028 Barcelona, Spain (e-mail:
[email protected]) 2Dept. de Geodindmica i Geofisica, Univ. de Barcelona, Marti i FranquOs s/n, 08028 Barcelona, Spain Abstract: We present in this paper a simple kinematic model to illustrate the evolution of
the western Mediterranean since upper Oligocene times. The five-step mass balanced reconstructions of the region are constructed along a 1000 km long crustal-scale transect from Iberia to Africa across the Val6ncia and Algerian basins. The model uses constant divergence and convergence rates and fits the geological and geophysical data well along the transect. The amount of convergence and divergence along the studied transect can be resolved by the combination of two tectonic mechanisms: the northern motion of the African plate and the southwards retreat of a subducting Tethyan oceanic slab located between the pre-Neogene Iberian and African margins. The evolution of the western Mediterranean region during the last 30 Ma can be summarized as follows: (a) pre-Neogene (late Oligocene) Iberian and African margins separated by a north-dipping fragment of Tethyan oceanic crust; (b) widespread extension affecting the Iberian margin developing the Valbncia trough and the Algerian basin from upper Oligocene to the top of Langhian; (c) end of the formation of the Val6ncia trough and the collision between the Kabylies and Africa domains at c. 13 Ma; (d) end of the Algerian basin opening and northern Kabylies and Tellian thrust systems coupled tectonic domains in the upper Tortonian (8 Ma); (e) shortening within the whole domain by the northwards motion of Africa after late Tortonian times. The propagation of stress within the western Mediterranean region seems to affect the whole region as far north as the northern side of the Pyrenean range located a distance of 1200 km from the south Atlas front, the southern limit of the system.
In the early 1990s, as part of an extensive Spanish program of deep seismic profiling of the Iberian crust (Santanach 1997), the ESCI-Valencia trough profile was shot across the centre of the Neogene Val6ncia basin (Gallart et al. 1997). This profile (Fig. 1) is continuous with the F r e n c h - S p a n i s h E C O R S profile across the Pyrenees (Choukroune & E C O R S Team 1989). The ESCI-Valencia trough profile is 450 km long with a roughly N W - S E trend, normal to the tectonic structures (50 km long segment located onshore and 400 k m long segment offshore). The profile extends from north to south across the Ebro basin, the Catalan Coastal Range, the onshore and offshore extensional system of faults that represent the northern margin of the Val6ncia trough, the Balearic p r o m o n t o r y and the Algerian basin (Figs 1 and 2). The Valencia trough has been extensively investigated in the last years, especially by French and Spanish teams (e.g. Banda & Santanach 1992; Torn6 et aL 1996). Initial results of the seismic data and geology along the ESCI profile have been published in Gallart et al. (1997), and SSbat et al. (1997).
Although a great effort has been devoted to deciphering the history of the Val6ncia trough region (see Banda & Santanach 1992; S?abat et al. 1997 and Torn6 et al. 1996 for geological and geophysical results) there is no general consensus on the pre-Neogene geometry of the basin and on its evolution through time. The purpose of this paper is three-fold: (a) to reconstruct the pre-Neogene crustal structure of the Valbncia trough assuming conservation of the upper crust extending mass based on two published lithospheric transects, (b) to constrain the timing of deformation based on available geological data and cross-cutting relationships between tectonic structures and stratigraphy along the ESCIVal6ncia trough profile and (c) to build up a simple kinematic m o d e l of the western Mediterranean along the Valencia trough and its southern continuation to the African domain based on the two previous results. We then use shortening and extension estimates along the profile as well as the duration of different tectonic processes to calculate rates at which these processes developed.
VERGt~S,J., & SABAT,E 1999. Constraints on the Neogene Mediterranean kinematic evolution along a !000 km transect from iberia to Africa. In: DURAND,B., JOLIVET,L., HORVATH,E & Ss M. (eds) The Mediterlanean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 63-80.
64
J. VERGES & F. SABAT
Fig. 1. Western Mediterranean tectonic map showing main orogenic belts and foreland basins. The marine basins are separated in floored by continental-transitionalcrust and oceanic crust (shaded). The ECORSPyrenees and the ESCI-Val6ncia trough deep crustal scale seismic profiles are represented by discontinuous thick lines. The studied transect is 1000 km long and crosses from Iberia to Africa.
Pre-Neogene structure of the present Valencia trough and Balearic promontory In this first section we try to estimate the preNeogene structure of the present Val6ncia trough. This pre-Neogene reconstruction is important to compose the Iberian margin before the onset of the extension within the Val6ncia trough. We use two published crustal and lithospheric cross-sections across the trough to infer pre-Neogene crustal thickness and length based on geological data and upper crustal mass conservation during deformation.
The ValOncia trough The Val6ncia trough displays a triangular geometry with a long N E - S W axis of c. 458 km from the Val6ncia coastline and a short NW-SE axis of c. 210 km to the northern coastline of the Balearic islands (e.g. Roca 1992, 1994; Torn6 et al. 1996). The Val6ncia trough is limited to the northwest and west by the Catalan Coastal and Iberian ranges and by the system of normal faults, located onshore that roughly parallel the eastern Iberian peninsula coast (e.g.
Guimer?a 1994). To the southeast, the trough is bounded by the Betics and the Balearic thrust system. The northeastern boundary of the trough is located in the large NW-SE-oriented magnetic anomaly located to the NE of the Menorca Island (Galdeano & Rossignol 1977) and interpreted as the transform boundary between the Valencia basin and the Liguro-Proven~al basin (e.g. Mauffret et al. 1992). The bathymetry mimics the crustal geometry of the Val6ncia trough. The crust is thinnest along the major axis of the trough and increases towards all margins except the northeastern connection with the Liguro-Proven~al basin, where bathymetry increases and crustal thickness decreases (e.g. Torn6 et al. 1996). Depth to the Moho is around 24-26 km in the coastline and less than 16 km in the centre of the basin (Torn6 et al. 1996).
Pre-Neogene Valencia trough crustal reconstruction Assuming conservation of upper crustal mass during extension, we can determine the preextensional length (Lo) of the Val6ncia trough
N E O G E N E KINEMATIC E V O L U T I O N
65
,x:Z
+..o
"~ < , z
,.o
e40"~
9
66
J. VERGI~S & F. SABAT
domain in a given section if we know two important parameters, namely the present area of the upper crust between two selected pin lines (A and B) and the pre-extension thickness of the upper crust. The present upper crustal area along the Val6ncia trough can be determined from a depth-migrated near-vertical and wide angle seismic section along the ESCI-Val6ncia trough profile (Vidal et al. 1997, fig. 8). The lower crustal area is calculated from the same profile, and the lithospheric mantle area is determined from a combined 2D gravity and geoid model (profile PIII, Ayala et al. 1996). The northwestern pin line (A, Fig. 2) is located at the more external onshore normal fault that roughly coincides with the major lower crustal thinning towards the centre of the trough. The southeastern pin line is located near the northern shoreline of the Mallorca Island (B, Fig. 2). The present distance between these two pin lines is 210 km. The thickness of the crust before extension within the Val6ncia trough domain is assumed to be between 30 and 33 km inferred from the preN e o g e n e reconstructed t o p o g r a p h y of the Catalan Coastal ranges during thrusting (Roca 1994; L6pez-Blanco et al. in press). L6pezBlanco et al. calculated a mountain front paleotopography between 700 and 1250 m on the basis of correlating mean mechanical d e n u d a t i o n rates with mean drainage basin elevation of two large middle Eocene fan deltas. These elevations would correspond to higher crustal values than assumed in our calculations, although these values were probably limited to the thrust front, adjacent to the fan delta deposits. Moreover, Eocene shallow-marine and Oligocene continental deposits involved within the Balearic thrust system demonstrate significant crustal thickness prior to the late Oligocene onset of the extension in the Val6ncia trough.
We assume in our reconstruction that 47% of the total crustal thickness is constituted by upper crust and 53% corresponds to lower crust similar to the present ratio between upper and lower crust underneath the non-extended Ebro basin ( 14.1-15.6 km of the 30-33 km pre-Neogene total crustal thickness would correspond to the upper crust) according to data in Vidal et al. (1997). The upper crustal area of the present Val6ncia basin is 1920 km 2 (Fig. 3); thus, assuming an upper crustal thickness of 14-17 kin, the preNeogene initial length of the Val6ncia trough would be of 136-123 km. This implies an extension of 74-87 km corresponding to a mean stretching factor of 1.62 and a mean thinning factor of 50% (Table 1). The same procedure applied to the lower crust shows a mean stretching factor of 2.02 and a mean thinning factor of 52%. Upper mantle mean stretching and thinning factors calculated for profile P-III of Ayala et al. (1996) are 1.94 and 52% respectively (Table 1). These results indicate that the stretching and thinning factors increase with depth from 1.62 for the upper crust to 2.02 and 1.94 for the lower crust and upper mantle. If this differential thinning with depth is tied to extension it would imply a decoupling between upper and lower crust and between lower crust and upper mantle. Different zones characterized by oblique reflectors crossing the lower crust and merging towards both the top and the bottom of this lithospheric layer (Fig. 2), interpreted as ductile shear zones (Gallart et al. 1997; S~bat et al. 1997), could accommodate differential lithospheric extension. Differential extension would match rheological models proposed for the region (Zeyen & Fernandez 1994). Furthermore, the extra extension could also be produced in part after the main period of
Table 1. Reconstruction o f the pre-Neogene ValOncia trough domain with 210 km o f present length
Present area (A 1) Initial length (Lo) Extension (E) Stretching factor ([3) Mean [3 Thinning factor (~/) Mean (~)
Upper Crust*
Lower Crust*
Upper mantle*
1920 km 2
1731 km 2
9323 km 2
136-123 km 74-87 km 1.54-1.70 1.62 48-51% 50%
109-99 km 101-111 km 1.93-2.11 2.02 52-52 % 52%
112-89 km 121-98 km 1.74-2.14 1.94 52-52 % 52%
Assuming 30-33 km and 77-95 km of pre-Neogene crustal and lithospheric mantle thicknesses. The ratios of upper and lower crustal thicknesses are 47% and 53%, based on the present crustal geometry underneath the Ebro basin. * Vidal et al. (1997). * Ayala et al. (1996).
NEOGENE KINEMATIC EVOLUTION
67
Fig. 3. Recent and pre-Neogene reconstruction of the Valbncia trough and Balearic promontory domains. Same pin lines as Fig. 2. The present Valbncia trough and Balearic promontory domains are reconstructed in the pre-Neogene stage as four different blocks labelled as pre-Vai6ncia trough, Mallorca Island, Northern Balearic basement and Balearic thrust system. Dashed lines represent the motion of individual points during coeval extension and compression. The formation of the Valbncia trough during extension moved Mallorca Island and Northern Balearic basement blocks to the south underneath the cover-involved Balearic thrust system (almost fixed at the backstop point). Hence, the amount of extension in Val6ncia trough is balanced by amount of upper crustal shortening in Balearic thrust system.
extension along the Valencia trough. This would match the observation that the present trough is not currently in thermal equilibrium (Janssen et al. 1993). In our next step, we use the value of 74-87 km of extension of the upper crustal layer as the minimum value of extension along the Val6ncia trough. The calculated mean stretching factor of 1.62 is too high when compared with the extension resolved from the displacement on wellimaged normal faults (e.g. Roca & Guimer?a 1992; S~bat et al. 1997), but it is low when compared with the results of backstripping modelling (e.g. Watts & Torn6 1992; Torn6 et al. 1996). To solve the discrepancy between the upper crustal stretching factor and extension deduced from normal faults, S~bat et al. (1997) proposed a geological model (model 'B' in their paper) showing a large, low-dipping, N E - d i r e c t e d normal fault that cuts the complete upper crust and that could accommodate a large amount of upper crustal brittle extension, up to 80 kin. That would accommodate almost all the calculated extension for the whole Valencia trough (between points 1 and 2 in Fig. 2). Moreover, using the upper crustal extension
as a reference (74-87 km), the total clockwise rotation produced around a pole of rotation n e a r to the s o u t h w e s t e r n coastline of the Val6ncia trough would be around 20 ~ These observations agree with the magnetic results obtained in the Mesozoic rocks that constitute the lower outcropping Mallorca thrust sheet that show at least 20 ~ of clockwise rotation (Par6s et al. 1992).
P r e - N e o g e n e Valencia trough a n d Balearic p r o m o n t o r y crustal reconstruction A n important result that can be deduced from the calculated upper crustal amounts of extension (74-87 km) is that they counter-balance the total shortening determined for the Balearic fold-and-thrust belt across the Mallorca Island by Gelabert et al. (1992) and Gelabert (1997). This author determined a minimum amount of 84 km of shortening based on a combination of detailed structural mapping and balanced and restored cross-sections. A l t h o u g h the total shortening involved in the Balearic promontory was probably greater than the 84 kin, calculated
68
J. VERGI~S & F. SABAT
for the Mallorca Island, with a present length of only 70 km compared with the 106 km of length of the present Balearic promontory, we use this well-constrained estimate of shortening. For the pre-Neogene crustal reconstruction of the Balearic promontory we use three different tectonic units equaling the present area (Fig. 3). These three units are the Mallorca Island unit with a length of 70 km, the Balearic thrust system with a length of 84 km and a thickness of 1.5 km, and the northern Balearic basement unit with a length of 36 km, which would form together with the Mallorca Island basement the total length of 106 km of the present Balearic promontory. To the first order, the calculated extension for the Valencia trough is roughly compensated by the shortening in the Balearic promontory in such a way that a hypothetical pin line located in the former southeastern t e r m i n a t i o n of the Balearic cover thrust system corresponding to the later southeastern shoreline of Mallorca Island (C', Fig. 3) moved very little with respect to the Iberian crust (pin line A, Fig. 3) during the synchronous d e v e l o p m e n t of the Valencia trough by extension and the Balearic promontory by shortening. To test this suggestion, the present combined crustal area of the Val6ncia trough and Balearic promontory, a total length (L1) of 316 km (distance between pin lines A and C in cross-sections Figs 2 and 3) should be equal to the combined crustal layer area of the preNeogene Val6ncia trough and the three preNeogene blocks corresponding to the present Balearic p r o m o n t o r y (Fig. 3). The present crustal area is calculated from the two same published crustal transects (Table 2). The pre-extensional thickness of the crust and the ratio between upper and lower crustal layers are the same than in the previous calculations (30-33 km and 47% for the upper crust, Table 1). The Mallorca Island compressive belt comprises a complex thrust system formed by the stack of
Table 2.
several tectonic units involving only Mesozoic and younger rocks above the main d6collement level in Keuper evaporites (e.g. Fallot 1922; S?abat et aL 1988; Gelabert et al. 1992). The average thickness of the Mesozoic section involved in the Mallorca thrust belt is 1.5 km (Gelabert et al. 1992). The basement corresponding to the southeastern side of the Balearic thrust system is not presently involved in the Balearic promontory because it corresponds to the northern part of the Kabylies as will be discussed below. The present area of the Balearic promontory along the transect is 2480 km 2 which includes the combined area of the Mallorca Island (1638 km2), Balearic thrust system (126 km 2) and the northern Balearic basement block (716 km2). The total length of the cover rocks within the Mallorca Island is the sum of the shortening of the Balearic thrust system (84 km) and the present length of the Mallorca Island (70 kin). Assuming 1.5 km thickness of Mesozoic strata (e.g. Gelabert et al. 1992), the thrust system in Mallorca would be 3.3 km thick on average which is in good agreement with the results of regional balanced and restored cross-sections (Gelabert 1997) and gravity modeling across the island (Ayala et al. 1994).
Timing of deformation To evaluate the timing of deformation along the studied transect, we use the available data to define a chronology of deformational events for each structural domain. We start with a summary of the Pyrenean chain to the north and end with the Maghrebides to the south (Fig. 1). We also include the timing of the opening of the Liguro-Provenqal basin, to the northeast of the Valencia trough (Fig. 4). The deformational history of the Pyrenean orogen ended about the time of initiation of the Val6ncia trough development.
Total area o f the upper crust a n d combination o f upper and lower crust
Present area of the Valencia trough and Balearic promontory domains with 316 km of present length.
Present area (A1)
Upper Crust*
Crust*
Crust*
3040 km 2
6122 km 2
5636 km 2
Restored length of the pre-Neogene Val6ncia trough domain Mallorca Island length Shortening of the Mallorca Island thrust system
136-123 km 70 km 84 km
* Vidal et al. (1997) * Ayala et al. (1996). The area calculated in the transect of Ayala et al. (1996) is 8-9% smaller than in Vidal et al. (1997). Errors in these calculations could come from pre-extension thicknesses estimations and from modeling errors.
NEOGENE KINEMATIC EVOLUTION
69
Fig. 4. Timing of tectonic and volcanic events along the studied transect (explained and referenced in the text). Cooling ages in the Maghrebides at 25 and 18-16 Ma are related to extensional tectonics (Moni6 et al. 1996; Saadallah & Caby 1996). Intramontane basins are bounded by normal faults and thus related to extension (Caby et al. 1996; ARe & G61ard 1997).
The Pyrenean
domain
The Val6ncia trough developed between two Iberian plate boundaries, the Pyrenean boundary to the north and the Betics boundary to the south. The northern margin evolved by the subduction of the Iberian plate underneath the European plate developing the Pyrenean thrust system (e.g. ECORS Team 1988; Mufioz 1992) and the intraplate Catalan Coastal and Iberian ranges. The Pyrenean thrust system was active from the lowermost Eocene at 55 Ma to the upper Oligocene at 24.7 Ma in the Artesa de Segre section as dated by magnetic chronostratigraphy (Meigs et al. 1996), although older (latest Cretaceous) and younger (Miocene to Quaternary) deformation are also present (e.g. Verg6s et al. 1995). Minimum shortening was 47 %, at mean rates that decreased from 4.5 mm a -1 during early and mid-Eocene to c. 2 mm a-1 to the end of the compressive deformation (Verg6s et al. 1995). These rates were calculated only from the south-directed cover and basement thrusts.
Concurrent with the cessation of the main compressive deformation on the Pyrenean thrust front, the area remained stable and deformation migrated towards the south and southeast, into the present Val6ncia trough, Balearic promontory and Algerian basin domains. However, the formation of the LiguroProvencal and Valencia basins resulted in normal faults affecting the eastern Pyrenean domain since Aquitanian times in the el Rossell6, Conflent, and la Cerdanya onshore grabens (Guimer?a et al. 1992). All these subbasins are bounded by normal faults displaying the main N E - S W direction characteristic of both the Gulf of Lion and the Val6ncia trough extensional fault system. The activity of these onshore faults ceased during Late Miocene coincident with the onset of a new system of normal faults (Empordta fault system) affecting both the easternmost Pyrenees and the NE termination of the Catalan Coastal ranges. The Late Miocene Empord?a fault system developed perpendicular to the previous system and migrated landwards from
70
J. VERGIffS & F. SABAT
the lower-middle Tortonian to the present day (Saula et al. 1996). Coeval alkaline volcanism flowed regularly along the major faults from 10.6 to 0.011 Ma (e.g. Martf et al. 1992). T h e ValOncia t r o u g h
The onset, end and total duration of the Tertiary rifting episode in the Valencia trough is approximately constrained, although the precise timing is variable depending on the author and the methodology applied to define it. We define the timing of rifting based on the assumption that faulting, crustal thinning and consequent tectonic subsidence developed synchronously. The initiation of rifting is determined by thick synrift deposits localized at the base of the relatively small wavelength, long half-grabens that form the northwestern margin of the Val6ncia trough (Roca & Guimerfa 1992). The upper Oligocene age of these relatively thick continental deposits is poorly constrained, although in the Barcelona C-1 oil well more than 1500 m of alluvial red beds were penetrated below wellestablished Aquitanian Alcanar Formation deposits (Bartrina et al. 1992). Aquitanian conglomerates and breccias together with the middle Miocene Cambrils Group (Burdigalian to upper Serravalian) are interpreted as synrift deposits separated by an erosive unconformity from the upper Miocene terrigenous Castell6n Group (e.g. Clavell & Berfistegui 1991; Maillard et al. 1992; Bartrina et al. 1992). Most of the normal faults along the ESCIVal6ncia trough profile are onlapped by middle Miocene strata that would represent post-rift deposits (Sgabat et al. 1997, fig. 2). However, some of them continued active during the middle Miocene and are sealed by upper Miocene to Quaternary deposits and only few of them are still active at present (Maillard et al. 1992). In this transect, the major fault bounding the E1 Camp graben has been active in recent times as determined by geomorphological studies (Masana & Guimer~t 1992). Subsidence analysis on the Barcelona B-1 and C-1 wells shows rapid tectonic subsidence from the upper Oligocene to middle Miocene, with a total duration of 10-15 Ma, followed by much slower subsidence until present (Bartrina et al. 1992). Local and flexural backstripping results from boreholes located in the Catalan continental slope best fit a finite rifting model (13 = 1.4) with an initial rifting period of 16 Ma followed by a thermal cooling period of about 8 Ma (up to present) in which the subsidence is relatively slow (Roca & Desegaulx 1990; Watts & Torn6 1992).
The onset of extension is defined by the base of the syn-rift, upper Oligocene 1500-2000 m thick, alluvial red beds. To minimize the rate of extension, we associate the base of the thick section to the base of the upper Oligocene at 28.7 Ma. This assumption reduces the rates of sediment accumulation during the late Oligocene to 0.3-0.4 mm a-1 (1.5-2.0 km/c. 5 Ma) and implies that the onset of rifting would overlap with shortening in the frontal south-central Pyrenean structures for about 4 Ma. The end of rifting can be defined by both the end of synrift deposits and the end of persistent normal fault activity occurring at about 11 Ma. The maximum duration of extension is thus 17.5 Ma. This duration is consistent with published subsidence modelling results (Watts & Torn6 1992; Janssen et al. 1993). The mean rate of regional extension during the Tertiary rifting period (74--87 km/17.5 Ma) is 4.2-5.0 mm a-i, although a probable higher rate dominated the earlier part of the rifting period (late Oligocene and early Miocene) and a slower rate dominated the end of the period (midMiocene) as suggested by synrift deposit geometries and fault activity. Two phases of volcanism are recognized within the Val6ncia trough domain. Calc-alkaline volcanism dominated from Aquitanian to mid-Serravalian times whereas alkaline volcanism prevailed since the mid-Miocene to Recent (Martf et al. 1992). The Liguro-Provenqal
basin
The Liguro-Provenqal basin is the northeastern continuation of the Val6ncia trough (Fig. 1). It is separated from the Val6ncia trough by the Paul Faillot transfer fault showing strong magnetic anomalies (Galdeano & Rossignol 1977), produced by volcanic rocks as inferred from seismic reflection data (Mauffret et al. 1992). Synrift continental deposits dated as Late Oligocene and Aquitanian confirm the age of the rifting period (e.g. Arthaud et al. 1980/81; Guennoc et al. 1994) that is sealed by the earliest Burdigalian break-up unconformity (Gorini et al. 1993). These growth strata can be as thick as 4000 m in the Camargue basin (Sdranne et al. 1995). After the main episode of rifting, the present Liguro-Provenqal basin formed by a period of drifting, formation of oceanic crust and subsequent translation of the Corsica and Sardinia blocks towards the southeast (Burrus 1984). The onset of oceanic crust is dated around the Aquitanian-Burdigalian boundary at c. 22 Ma. The end of oceanic crust formation and coeval Corsica and Sardinia block rotation was dated as mid-Burdigalian (19 Ma), although new studies
NEOGENE KINEMATIC EVOLUTION of magnetic data indicate that the end of the rotation could be dated as late Burdigalian at 16.5 Ma (Chamot-Rooke et al. 1996) or even later (Vigliotti & Langenheim 1995). The maximum opening of the Liguro-Proven~al basin occurred near the Paul Fallot transfer fault with a present length of oceanic crust of c. 225 km. Using the data in Chamot-Rooke et al. (1996), the accretion of this oceanic crust took place at a maximum rate of c. 41 mm a -i. T h e Balearic thrust s y s t e m
The oldest syntectonic conglomeratic deposits linked to the Mallorca thrust system, are upper Oligocene in age, and are located in the Serra de Llevant, in the southeastern region of the island of Mallorca (S?abat et al. 1988). Older NWdirected structures could be located within the most internal parts of the thrust system and actually overprinted and cut by the Emile Baudot normal fault system at the northwestern margin of the present Algerian basin. These older structures could be synchronous to the SE-directed thrust system of the Kabylies dated as old as 40 Ma (Saadallah & Caby 1996). However, we only analyse the more external part of the thrust system cropping out on Mallorca Island. The end of compressional deformation on Mallorca Island is well recorded in the Serra de Llevant where structures are sealed by Serravalian and younger deposits (Gelabert et al. 1992), indicating a terminal age younger than 14.9 Ma. The end of deformation in the Serra de Tramuntana could be younger than uppermost Langhian assuming a forward progradation of deformation, but older than late Miocene based on normal faults that cut the thrust system (Gelabert et al. 1992). A new episode of extension developed within the Balearic promontory between 12 and 6.8 Ma (e.g. Benedicto et aL 1993; S?abat et aL 1997). A regional balanced and restored crosssection based on surface and subsurface data across Mallorca Island shows a minimum amount of shortening of 84 km (Gelabert 1997). This shortening developed between late Oligocene (28.7 Ma) and mid-Serravalian times (c. 12 Ma) with a total maximum duration of 16.7 Ma and a minimum rate of shortening of c. 5.0 mm a-1 (slightly modified from Gelabert 1997), a value that is very close to the rate of extension within the Val6ncia trough. T h e A l g e r i a n basin
There is little information about the Algerian basin. From the ESCI-Valbncia trough profile
71
we can deduce that the normal fault system that forms the Emile Baudot scarp cuts the southernmost part of the Balearic thrust system at upper crustal levels. This system of normal faults represents the boundary at depth between the relatively thick continental crust of the Balearic promontory and the uniform and thin crust underneath the Algerian basin interpreted as oceanic crust (Gallart et al. 1997). The sedimentary succession overlying this crust is formed by evaporitic levels related to the Messinian crisis and younger deposits. It is possible that sediments older than Messinian exist, although they are not recognizable on the seismic profile. The Messinian evaporitic deposits subsequently developed diapiric structures all along the profile. The age of the oceanic crust underneath the 90 km transect across the Algerian basin in the ESCI-Val6ncia trough profile has to be older than the evaporitic deposition during the Messinian crisis at 6.4 Ma (Rouchy & Saint Martin 1992) which is the oldest recognized deposit infilling the basin. The initial age of the Algerian basin formation is more difficult to determine. The Alborfin basin, to the west of the study transect, was completely formed by 7.5-8 Ma (Comas 1996; Comas et al. 1997). That seems to be also the end of the Algerian basin formation after the collision between the Kabylies and the African margin. T h e K a b y l i e s a n d the M a g h r e b i d e s
In this section we try to summarize the sequence of tectonic processes that took place within the southern side of the transect, the Kabylies and the Maghrebides fold-and-thrust belt (e.g. Wildi 1983). The Kabylies comprise the internal basement massifs of the south-directed north African thrust system and show a longer and more complex geological history than the external domains. The pre-mid-Miocene sequence of tectonic events that occurred within the internal Kabylies shows the preserved part of the geological history of the ancestral present day Algerian basin. The pre-opening Algerian basin history of the Kabylies is characterized by continuous southdirected thrusting from mid-Eocene at 40 Ma to late Oligocene at 25 Ma (Caby et al. 1996). After 25 Ma cooling ages indicate a rapid phase of tectonic exhumation in the more northern and internal parts (Saadallah & Caby 1996). The more external parts are characterized by Oligocene and early Miocene deposition of the Numidian flysch from c. 23 Ma to 19 Ma, followed by olistostromes and emplacement of
72
J. VERGI~S & F. SABAT
flysch nappes (Caby et al. 1996). A new period of extension is marked by both cooling ages from 18 to 16 Ma (Moni6 et al. 1992, 1996) and by the opening of deep intramontane basins from 16 to 10 Ma (Caby et al. 1996). Current data determine the timing of the collision of the Kabylies at c. 18 Ma (Tricart et al. 1994) although Dewey et al. (1989) determine that deformation within the Kabylies continued through time and that the emplacement of the unit on top of the African continental margin occurred during Tortonian times at around 8 Ma. The African coast is also characterized by volcanic series with initial calc-alkaline, tholeiitic, then alkaline magmatic activity from 15 to 5 Ma (Hernfindez & Bellon 1985; Moni6 et al. 1996; Caby et al. 1996). In the Tell tectonic unit there was a change from calc-alkaline to alkaline volcanic series at 8 Ma (Hernfindez et al. 1987).
Model for the evolution of the western Mediterranean In this last section of the paper we build up a simple model largely based on the previous sections, including pre-rifting reconstructions of the Val6ncia trough and Balearic promontory, and timing and sequence of tectonic events in a long transect, starting at the ESCI-Val6ncia trough deep seismic profile to the north and ending in the stable African plate, south of the south-Atlas thrust front (see location in Fig. 1). This 1000 km long transect is perpendicular to almost all the Miocene to Recent extensional and contractional major structures and parallel to the kinematic indicators (e.g. G e l a b e r t et al. 1992; Saadallah & Caby 1996; A~te & G61ard 1997) and to the present stress direction determined from focal mechanism solutions derived from earthquakes, which allows us to consider that deformation has been within the plane of the transect. The stable Iberian peninsula has been used as
the fixed pin line of the section. Different tectonic units have been represented by simple crustal blocks with averaged crustal thicknesses separated by either normal of thrust faults showing only dip-direction. The two ends of the sections comprise the stable Iberian and African tectonic domains (50 km length for each). The Iberian block represents the fixed pin line throughout evolution. Additional geological information important for the evolution of the transect is depicted in the section, especially the calc-alkaline and alkaline volcanism and subduction zones. The different steps have been constructed using an iterative technique between the present step (0 Ma) and the oldest step at the beginning of the late Oligocene (28.7 Ma). The southern limit of the oldest section is the position of Africa that should be located c. 175 km to the south of its present position during late Oligocene times (Boccaletti et al. 1977; Olivet et al. 1982). The calculated m e a n rates of geodynamic processes have been used to locate different blocks in their presumed position at each chosen time. To simplify we consider a constant 6.1 mm a -1 of n o r t h w a r d s c o n v e r g e n c e of Africa with respect to the fixed Iberian plate from upper Oligocene (175 km/28.7 Ma-1). Vertical separation between sections are scaled to time so changes in the discontinuous lines showing the motion of crustal blocks represent true changes in the rate of these motions. The upwards convergence of these lines represents shortening whereas upwards divergence represents extension. The reconstructed steps are from oldest to youngest: (a) the pre-Neogene reconstruction of the Iberian and African margins at 28.7 Ma; (b) widespread stretching of the Iberian margin at c. 18 Ma; (c) the end of formation of the Val6ncia trough and collision between the Kabylies and Africa domains at c. 13 Ma; (d) the end of formation of the Algerian basin and Tellian front at 8 Ma; and (e) the present day at 0 Ma.
Fig. 5. Kinematic model along a transect assuming mass-balance preservation during evolution as well as estimates of extension and shortening rates. The model comprises five steps from the pre-Neogene reconstruction of both the Iberian and African margins to the present times. Tectonic units are constructed using simple regular blocks with exact length and mean crustal thicknesses. The Kabylies tectonic unit is constructed by an irregular block from 28.7 Ma to 13 Ma to match the down-going oceanic slab geometry. Vertical scale is time dependent. Calc-alkaline volcanoes have ash-clouds to the right whereas alkaline volcanoes have them to the left. Main boundaries between tectonic blocks are depicted by either normal faults or thrusts. The model is fixed in the Iberian block although partially fixed points are also represented by black pin lines. Rates of extension and shortening are also represented by horizontal arrows pinned at the thick and discontinuous lines showing divergence and convergence between tectonic domains (within the more regional convergence between Iberia and Africa).
NEOGENE KINEMATIC EVOLUTION
73
74
J. VERGI~S & F. SABAT
Pre-Neogene reconstruction o f the Iberian and African margins at 28. 7 Ma This section has been constructed using information from the two sides of the section. On the northwestern side, the reconstructed preNeogene Valbncia trough (maximum calculated length of 136 km), the Balearic promontory (present length of 106 km), the reconstructed Balearic thrust system (minimum shortening of 84 km) and the Kabylies (inferred total length of almost 100 km from which 28.5 km correspond to basement underneath the southern segment of the Balearic thrust system) are joined to the undeformed Iberian plate with a represented length of only 50 km (Fig. 5a). There are two main uncertainties in this reconstruction: (a) what is the shape of the pre-Neogene Kabylies domain? and (b) where was this domain located with respect the rest of the Iberian margin? The assumed shape of the Kabylies in the preNeogene reconstruction is highly speculative. We assume a non-regular block adapted to the subduction plane with an area (1000 km 2) smaller than its present-day regular block representation (1150 km 2) to account for the accretion of the infill of the oceanic basin in front of the Kabylies (the Numidian flysch). To minimize the mass of the continental basement that was extended during the onset of the Algerian basin, the Kabylies constitutes part of the basement of the Balearic thrust system. However, another regular block of basement is located between the pre-Neogene Kabylies that will correspond to the present Kabylies and the northern Balearic basement that will form the southern part of the Balearic promontory. We further split this block in two to account for the transitional crust located at the two extremes of the future Algerian basin. This reconstruction in which the Kabylies tectonic unit represents the basement of the cover Balearic thrust system is the best solution to account for oceanic crust along most or all the Algerian basin. The southeastern margin of the upper Oligocene section has been constrained by the reconstructed position of the stable and non deformed African plate, represented in all the sections by a length of 50 km and a mean crustal thickness of 36 km (Giese & Buness 1992). The preNeogene position of the African plate was c. 175 km south of its present position (Boccaletti et al. 1977; Olivet et al. 1982; Archambault 1984). The Tell and Atlas tectonic units were linked to the northern side of the non-deformed African plate to form the southeastern margin of the preMediterranean Sea. The pre-Neogene length of these two units are also deduced from
non-balanced and partial geological cross-sections across northern Africa (Roeder 1992) suggesting minimum values of shortening of 30% for the Tell and 20% for the Atlas thrust belts. So, the pre-Neogene length of the Tell and Atlas tectonic units increased from the present 70 and 170 km to 100 and 212.5 km respectively. If this reconstruction is correct then the gap between the reconstructed Iberian and African margins is as large as 385 km, represented in this paper by Tethyan oceanic crust subducting underneath the southern side of the Oligocene Kabylies wedge linked to the Iberian plate (e.g. de Jonge et al. 1993, 1994; Zeck 1996; Lonergan & White 1997). This will be discussed in the next reconstruction during Burdigalian times. Consequently, the minimum reconstruction of both the Iberian margin (minimum length of the Kabylies) and the African margin (minimum values of shortening) constrain the maximum length of this presently missing tectonic block. Other reconstructions increasing either the Iberian margin or the African margin or both would reduce the original length between the two margins (Tethyan oceanic crust) and would increase the basement mass available to extend and to fill the Algerian basin.
Widespread stretching o f the Iberian margin at c. 18 Ma Two major geodynamic processes occurred between the late Oligocene and the next, midBurdigalian, reconstruction at 18 Ma involving both African and Iberian margins (Fig. 5b). In the southern half of the section, Africa moved northwards at a mean rate of 6.1 mm a -I. In the northern half of the section, widespread extension affected the whole pre-Neogene Iberian margin. Using the calculated mean value of northwards motion of Africa, the present nondeformed African block was located at 118 km to the south of its present position. The African margin has been constructed in this mid-Burdigalian section like the previous late Oligocene one, since no major deformational events have been determined for this period of time. The Iberian margin was affected by a more severe tectonic episode characterized by synchronous extension and compression. The Valencia trough and Balearic promontory lengths and crustal thicknesses have been resolved by extrapolating mean extension and shortening rates (equal and close to 5.0 mm a -1) to determine the position of the two blocks at 18 Ma and by assuming conservation of mass during the evolution as has been discussed
NEOGENE KINEMATIC EVOLUTION earlier in this paper. In this reconstruction, the extension of the Valbncia trough is balanced by cover-involved shortening in the Balearic promontory whereas a 36 km long block, constituted only by the northern basement of the Balearic thrust system unit, rests attached to the southern border of the Balearic promontory. This basement block moved towards the southeast at the same rate than the Val6ncia trough extension rate (maximum of 5 mm a-l). From the kinematic reconstruction, the basement of Mallorca and northern Balearic thrust system moved to the south underneath the thin-skinned Balearic thrust system that remained fixed during thrusting evolution (represented by a backstop at the southern edge of the Balearic thrust system, Fig. 5a and b). To the south of the Balearic promontory, the position of the Kabylies block at 18 Ma has been determined by assuming that the unit moved to the south at a constant rate of 18.5 mm a-1 calculated between its pre-Neogene position linked to the Iberian margin at 28.7 and its collision to the African continental margin at 13 Ma. As the motion of the Kabylies block to the south was faster than the extension within the VaRncia trough (18.5 versus 5 mm a-1) the Algerian extensional basin opened between the Balearic promontory and the southern front of the Kabylies. The age of the opening is in agreement with cooling ages related to the northern internal part of the Kabylies where several periods of extension have been recognized at 25 Ma and between 18 and 16 Ma, followed by the formation of intramontane basins from 16 to 10 Ma. As stated in our earlier reconstruction, during late Oligocene time the southern margin of the Kabylies thrust over a north-dipping oceanic crust subduction zone, as proved by the existence of calc-alkaline volcanic rocks within the Balearic promontory and Valbncia trough at this time (Figs 4 and 5b). The position of both the volcanics and the subduction trench, the length ~)f the reconstructed subducted slab and the approximately 110 km of depth at which this oceanic slab can provide partial melts match with a low-dipping subduction zone (about 20 ~ according to our geometric reconstruction). Southward motion of the Kabylie deformation front sourced the Numidian flysh at c. 23 Ma followed by olistostromes and thrusting after 19 Ma (Caby et al. 1996). These tectonic events could also be linked to the development of a submarine accretionary prism on top of the subducting plate (marked by a small vertical arrow in sections in Fig. 5). Therefore, this stage from the latest Oligocene to the mid-Burdigalian times was characterized
75
by wider extensional processes affecting the whole Iberian margin driven by the retreat of a subduction slab (Royden 1993). This retreat produced the formation of both the Valbncia trough and the proto-Algerian basin to the south. Cooling ages of metamorphic rocks located in the northern side of the Kabylies seem to indicate more complex extensional events related to the subduction retreat processes.
End o f the formation o f the ValOncia trough and collision between the Kabylies and Africa domains at c. 13 Ma At the beginning of the Serravalian, the formation of the VaRncia trough and the Balearic thrust system had almost ended and is represented in the section as blocked (pin line at the southern margin of the Balearic promontory, Fig. 5c). Since late Burdigalian time there was a rapid migration of calc-alkaline volcanism from the centre of the VaRncia trough at 18 Ma to the northern Kabylies at 13 Ma (Fig. 5). The onset of calc-alkaline volcanics of Serravalian age cropping out in the northern side of the present Kabylies constrain the geometry of the subducting slab underneath the African margin. The position of these volcanic rocks together with the depth at which this oceanic slab could provide partial melts suggest that the retreating slab increased its dip (about 60 ~, according to our geometric reconstruction). This assumed increase of dip at the beginning of the Serravalian could be linked to a decrease of the rate of oceanic subduction produced at the end of the subduction and the beginning of the continental collision between the Kabylies and the north of Africa. This age of continental collision at around 13 Ma is in agreement with determined longer-term shortening affecting the accretionary prism located in front of the Kabylies (indicated by a vertical arrow in Fig. 5). The emplacement of the Numidian nappes on top of the African margin could start before the continental collision followed by the emplacement out of sequence of the Kabylies unit similarly to the tectonic evolution determined for the southern Apennines and western Sicily (Lentini et al. 1996). Although more speculative, the increase in dip during subduction retreat could produce a significant additional extension in the lower lithospheric layers but not in the crust (at least in the upper crust) as evidenced in the previous sections of this paper. This upper mantle thinning could be responsible for the onset of
76
J. VERGI~S & F. SABAT
alkaline volcanism in the Valencia trough at this time (Fig. 5c) that was generalized in the next step in late Tortonian times in both the Val6ncia trough and the southern side of the Algerian basin (Fig. 5d). The differential value and timing of lithospheric mantle thinning is also documented by present heat flow analysis from the easternmost Ebro basin (Cabal & Fernandez 1995). These authors demonstrated a lithospheric mantle thinning migration beyond the upper crustal extended region of the Val6ncia trough from mid-Miocene to present. E n d o f the f o r m a t i o n o f the Algerian basin and Tellian f r o n t at 8 M a Deformation migrated to the African margin during late Miocene times. Although the chronology of deformation is less constrained in this margin, the emplacement of the Kabylies front accretionary prism started before 13 Ma, producing folding and thrusting within the Tellian domain. However, the continental collision of the Kabylies on top of the African margin occurred at c. 8 Ma (Dewey et al. 1989) coeval with the end of calc-alkaline volcanics in the Tell domain (Hern~indez et al. 1987). In the late Tortonian reconstruction the opening of the Algerian basin and the shortening within the Kabylies and the Tell domains were almost terminated (pin line in Fig. 5d). Although difficult to prove, at the end of the oceanic lithospheric subduction between 13 and 8 Ma there was an important change with respect to the subducting units, taking in account that oceanic lithosphere ended and that the African continental crust is mostly preserved. These observations suggest that if subduction continued underneath the African plate as interpreted from seismicity distribution and from tomographic studies (Spakman et al. 1993), only the lithospheric mantle should be involved in subduction, although the lower crustal layer of the more external and thin crustal domains could be also involved (Kabylies and Tell units as described in the next section). Present day at 0 Ma. The stable Iberian tectonic unit is composed of a block of 50 km of length and 32 km of crustal thickness. Although out of the plane, we represented mantle-derived volcanism occurring in the Empordh region to complete the picture (Fig. 5e). The Val6ncia trough unit is represented by an homogeneous tectonic block with a length of 210 km and a crustal average thickness
of 19.2 km to account for the present area of the basement and Neogene deposits (4026 kmZ; Fig. 3). The contact with the Iberian crust is a southdipping normal fault. The Balearic promontory is composed of a single block with a length of 106 km and a crustal thickness of 23.4 km (Fig. 3). The ESCI-Val6ncia trough reflection seismic profile covers the northern 90 km of the Algerian basin. The crust underneath the Algerian basin is homogeneous and thin. The total thickness of the crust is c. 9.5 km, from which 6-7 km correspond to the basement underlying the 2.5 km thick upper Miocene to Quaternary sedimentary infill (Vidal et al. 1997). Vidal et al. also pointed out the oceanic character of this crust, corroborating seismic refraction data (Hinz 1972). The northern limit of this thin crust is located at the southern boundary of the Balearic promontory (t~mile Baudot escarpment; Fig. 2), whereas to the west there is a transition from thin Algerian basin crust (oceanic crust) to 11.5 km thick transitional crust (extended continental crust) within the Alborfin basin at longitude l~ (Fig. 1). The northern African domain is separated by three different tectonic units: the Kabylies, the Tell and the Atlas. The approximate length of these units along the studied transect are 50, 70 and 170 km (from the Carte Tectonique de l'Europe et des R6gions Avoisinantes; Khain et al. 1975). The depth of the Moho for these tectonic domains are taken from Giese & Buness (1992). The Moho depth increases from about 22 km on the northern Tunisian shoreline to more than 36 km south of the south Atlas front. In the present section we used 23, 28, 32 and 36 km of mean estimates of Moho depth for the Kabylies, Tell, Atlas and Africa domains respectively, determined from a more easterly Tunisian transect. Using these crustal thicknesses together with minimum estimates of shortening (30% for the Tell and only 20% for the Atlas) it is possible to determine a minimum amount of 72.5 km of shortening involving the complete African margin along the studied transect. The present state has been little modified since late Tortonian times. The northwest motion of African plate with respect to Europe (Dewey et al. 1989) was the major mechanism to produce several tens of kilometres of shortening, especially within the Atlas tectonic domain (Frizon de Lamotte et al. 1996, 1998). These last few millions of years were therefore dominated by the motion of the African plate after the blockage of different coupled tectonic domains that underwent extension synchronous with shortening (Valencia trough-Balearic promontory and Balearic thrust system and Algerian
NEOGENE KINEMATIC EVOLUTION basin-northern African thrust system) in front of a southwards retreating subduction zone. Although the majority of the recent deformation deformed the Atlas domain, a small amount of stress must be transferred to the north to produce inversion of normal faults within the Valbncia trough (Maillard et al. 1992), and the neotectonic seismicity, associated with compressive regime, on both the northern margin of the Val6ncia trough and the Pyrenees (Olivera et al. 1992). Alkaline volcanism migrated towards the Iberian plate, localized in the Empord?a basin from 11 Ma to the present (0.011 Ma, Fig. 4) and towards the African plate in the Tell mountains with an age as young as 1 Ma.
Conclusions Cross-sectional reconstructions presented in this paper along the 1000 km long transect crossing the western M e d i t e r r a n e a n from Iberia to Africa do not differ substantially from earlier map reconstructions for this area, especially from those of (e.g. Dercourt et aL 1986; Gealey 1988; Dewey et al. 1989; Doblas & Oyarzun 1990; Vegas 1992; Ricou 1994; L o n e r g a n & White 1997). However, in this paper we provide a simple quantified model based on a five-step kinematic reconstruction that can be easily modified with the addition of new geological and geophysical data. A n up-to-date set of palinspastic maps would provide a better 3D understanding of the whole western M e d i t e r r a n e a n domain. With the initial assumptions needed for the construction of the model (mass preservation and constant rates of tectonic processes), small modifications in the presented crustal scale reconstructions would not substantially change the results. However, large changes would be difficult within the context of the model and the geological data. Convergence and divergence within the studied transect can be resolved by the combination of two tectonic mechanisms: the northern motion of the African plate and the southwards retreat of a subducting Tethyan oceanic slab located between the pre-Neogene Iberian and African margins. Whereas the northern African motion only contributes to move the African margin together with the oceanic crust 6.1 mm a -1 to the north, the retreat of the oceanic subducting plate together with a possible later increase in the dip of the subduction plane generated a generalized extension affecting the whole Iberian margin allowing the Val6ncia trough and the Algerian basin to open separated by the Balearic promontory. The rate of local
77
extension was c. 5 mm a -a for the Val6ncia trough and c. 13.5 mm a -1 for the Algerian basin (from 28.7 to 13 Ma). The m o t i o n of the Kabylies to the south with respect to the Iberian fixed plate was the sum of extension rates within the Iberian margin which is c. 15 mm a -1. However the rate of subduction along the transect was c. 24.6 mm a -1 that is the sum of the rate of southwards translation of the front of the Kabylies (c. 18.5 mm a -1) and the rate of African margin northwards push (6.1 mm a -1) from 28.7 Ma to the end of oceanic subduction close to 13 Ma. Convergence and divergence within the western Mediterranean region can be explained following this sequence of events: (1) end of formation of the Val6ncia trough and Balearic thrust system coupled and synchronous tectonic domains at the end of Langhian (13 Ma) and the onset of c o n t i n e n t a l collision between the Kabylies and northern Africa; (2) end of the form a t i o n of the A l g e r i a n basin and n o r t h e r n Kabylies and Tellian thrust systems coupled and synchronous tectonic domains at late Tortonian (8 Ma); (3) shortening within the whole domain by the northwards motion of Africa after late Tortonian times. The propagation of stress within the blocked western M e d i t e r r a n e a n region can affect the northern Pyrenean thrust system located as far as 1200 km to the north of the south Atlas front. Although more speculative, the increase of lithospheric extension with depth, documented for the Valbncia trough, can be explained by the increase in dip of the subducting slab between 18 and 13 Ma. If so, part of the mantle lithospheric thinning could be younger than upper crustal extension (up to Recent). Although the comparison with the Albor~n basin and the Betics is not the subject of our contribution, it is interesting to observe that there is an important parallelism of the development of the studied transect and a transect crossing the Betics and Albor~in basin at least until 13 Ma (e.g. Lonergan & White 1997). The different evolution of these two transects was influenced by the lateral variations of both the preNeogene geometry and position of the Iberian and African margins and the amount of convergence of the African plate. This paper benefited from fruitful discussions on the Valbncia trough with the Tectonic Group of the Barcelona University especially with E. Roca as well as with J. Friedmann, C. Lewis, M. Fernandez, M. Torn6, C. Doglioni and T. Minshull. We are also very grateful to D. Frizon de Lamotte, P. Tricart, L. Lonergan and an anonymous referee for their thoughtful reviews.
78
J. VERGI~S & F. SABAT
References A[TE, M. O. & Gt~LARD,J.-P. 1997. Distension nOoghne post-collisionelle sur le transect de GrandeKabylie (Algdrie). Bulletin de la SociOtO geologique de France, 168, 424-436. ARCHAMBAULT, M. E 1984. Evolution cinOmatique post-EocOne de l'Atlantique Nord et Central Implications sur le fonctionnement des Afores et l'Ovolution du domaine MOditerranOen Occidental. Thhse 3hme cycle, Brest. ARTHAUD, E, OGIER, M. & SI~GURET,M. 1980/1981. GOologie et gOophysique du golfe du Lion et de sa bordure nord. Bulletin du BRGM, 3, 175-193. AYALA, C., POUS, J., SA.BAT,E, CASAS,A., RIVERO, L. & GELABERT,B. 1994. Modelizaci6n gravimOtrica de la Isla de Mallorca. Revista Sociedad Geol6gica Espaha, 7, 215-227. , & TORNE, M. 1996. The lithosphereasthenosphere boundary of the Valencia trough (western Mediterranean) deduced from 2D geoid and gravity modelling. Geophysical Research Letters, 23, 3131-3134. BANDA, E. & SANTANACH,P. 1992. Geology and Geophysics of the Valencia trough, Western Mediterranean. Tectonophysics, 203, 1-361. BARTRINA, M. Y., CABRERA, L., JURADO, M. J., GUIMERA, J. & ROCA, E. 1992. Evolution of the central Catalan margin of the Valencia trough (western Mediterranean). Tectonophysics, 203, 219-247. BENEDICTO, A., RAMOS-GUERRERO, E., CASAS, A., SABAT,E & BARON,A. 1993. Estructura y relleno de la cubeta ne6gena de Inca (Mallorca). Revista Sociedad GeolOgica Espaha, 6, 167-176. BOCCALETI'I, M., GUAZZONE, G. & MANETTI, P. 1977. Evoluzione paleogeografica e geodinamica del Mediterraneo; i bacini marginali. Memorie delle Societd Geologica Italiana, 13, 1-39. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provenr Basin (North-Western Mediterranean). Marine Geology, 55, 247-270. CABAL, J. & FERN,a,NDEZ, M. 1995. Heat flow and regional uplift at the Northeastern border of the Ebro basin, NE Spain. Geophysical Journal International, 121, 393-403. CABY, t . , SAADALLAH,A.& HAMMOR,D. 1996. Alpine tectonometamorphic evolution of the Kabylian crystalline massifs, Algeria. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Cergy-Pontoise Workshop Abstracts, Paris, France. CLAVELL, E. & BERASTEGtJI, X. 1991. Petroleum geology of the Gulf of Valencia. In: SPENCER, A. M., (ed.) Generation, Accumldation and Production of Europe's' Hydrocarbons. E A P G Special Publications, Oxford University Press, 335-368. CHAMOT-ROOKE, N., MAILLARD,A., GAULIER, J. M. & PASCAL, G. 1996. Crustal structure of the LiguroProvencal basin from gravity modelling: geodynamic implications. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Cergy-Pontoise Workshop Abstracts, Paris, France.
CHOUKROUNE, R &ECORS TEAM. 1989. The ECORS Pyrenean deep seismic profile reflection data and the overall structure of an orogenic belt. Tectonics, 8, 23-39. COMAS, M. 1996. The Alboran Sea and the Betic-Rif Orogen: Aftermath of Extesnion and Compression in the Alboran Basin. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Cergy-Pontoise Workshop Abstracts, Paris, France. , DANOBEmA, J. J., ALVAREZ-MARRON,J. & SOTO, J. I. 1997. Crustal reflections and structure in the Albor~in Basin: Preliminary Resuls of the ESCIAlborfin Survey. Revista Sociedad Geol6gica Espaha, 8, 530-542. DE JONGE, M. R.,WORTEL, M. J. R. &; SeAKMAN,W.1993. From tectonic reconstruction to upper mantle model: an aplication to the Alpine-Mediterranean region. Tectonophysics, 223, 53-65. , & 1994. Regional scale tectonic evolution and the seismic velocity structure of the lithosphere and upper mantle: The Mediterranean region. Journal of Geophysical Research, 99, B6, 12091-12108. DERCOURT, J., ZONENSSHAIN,L. P. & 17 OTHERS. 1986. Geological Evolution of the Tethys Belt from the Atlantic to Pamirs since the Lias. Tectonophysics, 123, 241-315. DEWEY, J. E, HELMAN,M. L.,TuRco, E., HIJTTON,D. H. W. & KNOTt, S. D. 1989. Kinematics of the western Mediterranean. In: COWARD, M. R, DmTRICIq, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. DOBLAS, M. & OYARZUN,R. 1990. The late OligoceneMiocene opening of the North Balearic Sea (Valencia basin, western Mediterranean): A working hypothesis involving mantle upwelling and extensional detachment tectonics. Marine Geology, 94, 155-163. ECORS PYRENEES TEAM. 1988. The ECORS deep reflection seismic survey across the Pyrenees. Nature, 331,508-511. FALLOT, R 1922. Etude G&)logique de la Sierra de Majorque. ThOse Libr. Polytech. ch. Bdranger, Paris. FR1ZON DE LAMO'ITE,D., GUEzot/, J. C. & MERCIER,E. 1996. The Geodynamics of the Western Mediterranean Regions: an African View. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Cergy-Pontoise Workshop Abstracts, Paris, France. - - , MERCIER, E., OUTI'ANI, E, ADDOUM, B., GHANDRICHE, H., OUAIA,J., BOUAZIZ,S. &; ANDRIEUX,J. 1998. Structural inheritance and kinematics of folding and thrusting along the front of the Eastern Atlas Mountains (Algeria and Tunisia). Peri-Thetys Memoir, 3, Memoires du MusOum National d'Histoire Naturelle, 237-252. GALDEANO, A. & ROSSIGNOL,J. C. 1977. Assemblage h altitude constante des cartes d'anomalies magn6tiques couvrant l'ensemble du bassin occidental de la MOditerrande. Bulletin de la Soci~tO gOologique de France, 19, 461-468.
N E O G E N E KINEMATIC EVOLUTION GALLART,J., VIDAL,N., ESTi~VEZ,A., POUS,J., SABAT,E, SANTISTEBAN,C., SURINACH,E. & GROUP,E.-V. T. 1997. The ESCI-Valencia Trough Vertical Reflection Experiment: A Seismic Image of the Crust From the NE Iberian Peninsula to the Western Mediterranean: Revista Sociedad GeolOgica Espaha, 8, 401-415. GEALEY, W. K. 1988, Plate tectonic evolution of the Mediterranean-Middle East region. Tectonophysics, 155, 285-306. GELABERT, B. 1997. L'Estructura GeolOgica de la Meitat Occidental de l'Illa de Mallorca. Ph.D. Thesis, Barcelona. --, S,~BAT, E & RODRIGUEZ-PEREA, A. 1992. A structural outline of the Serra de Tramuntana of Mallorca (Balearic Islands). Tectonophysics, 203, 167-183. GIESE, P. & BUNESS,n . 1992. Moho Depth. In: BLUNDELL, D., FREEMAN, R. & MUELLER, S. (eds) A continent revealed. The European Geotraverse. Cambridge, Cambridge University Press, Atlas Map 2. GORINI, C., LE MARREC, A. & MAUFFRET, A. 1993. Contribution to the structural and sedimentary history of the Gulf of Lions (Western Mediterranean), from the ECORS profiles, industrial seismic profiles and well data. Bulletin de la Soci~t~ g~ologique de France, 164, 353-363. GUENNOC, P., DEBEGLIA, N., GORINI, C., LE MARREC, A. & MAUFFRET,A. 1994. Anatomie d'une marge passive jeune (Golfe du Lion-Sud France): apports des donn6es g6ophysiques. Bulletin des Centres de Recherches Exploration-Production. Elf Aquitaine, 18, 33-57. GUIMERA, J. 1994. Cenozoic evolution of eastern Iberia: Structural data and dynamic model. Acta Geologica Hispdnica, 29, 57-66. - - , ANADON,P., CABRERA,EL. & ROCA, E. 1992. La distensi6 ne6gena del marge mediterrani. In: HistOria Natural dels Pa~sos Catalans, Geologia (II). Fundaci6 Enciclopedia Catalana, Barcelona, 281-359. HERNANDEZ,J. & BELLON,H. 1985. Chronologie K-Ar du volcanisme miocene du Rif oriental (Maroc): implications tectoniques et magmatologiques. Revues de Gdologie dynamique et de GOographie physique, 26, 85-94. , LAROUZIERE,E D., BOLZE, J. & BORDET,P. 1987. Le magmatisme n6ogene b6tico-rifain et le couloir de decrochement trans-Alboran. Bulletin de la Socidt~ g~ologique de France, 8, 257-267. HINZ, K. 1972. Results of seismic refraction investigations (Project Anna) in the Western Mediterranean, south and north of island of Mallorca. Bulletin des Centres de Recherches de PauSNEAP, 6, 405-426. JANSSEN,M. E.,TORNI~,M., CLOETINGH,S. & BANDA,E. 1993. Pliocene uplift of the eastern Iberian margin: Inferences from quantitative modelling of the Valencia Trough. Earth and Planetary Science Letters, 119, 585-597. KHAIN,V., LEONOV,Y., TCHERNOOK,S., ANISSIMOVA,V., GERASSIMOVA, R., GOSTEVA, A., PAVLENKO,M., RICHTER,V. & SEMOVA,V. 1975. Carte Tectonique
79
de l'Europe et des R~gions Avoisinantes at 1:10,000,000 scale. Direction G6n6rale de G6od6sie et de Cartographie pres le Conseil des Ministres de I'URSS. LENTINI, E, CATALANO,S. d~ CARBONE, S. 1996. The External Thrust System in southern Italy: a target for petroleum exploration. Petroleum Geoscience, 2, 333-342. LONERGAN, L. & WHITE, N. 1997. Origin of the BeticRif mountain belt. Tectonics, 16, 504-522. LOPEZ-BLANCO, M., MARZO, M., BURBANK, D., VERGt~S, J., ROCA, E., ANADON, P. & P[/qA, J. in press. Tectonic and climatic controls on the development of large, foreland fan deltas: Montserrat and Sant Lloren~ del Munt systems (Middle Eocene, Ebro basin, NE Spain). In: MARZO, M. & STEEL, R. (eds) Sedimentology and sequence stratigraphy of the Sant Llorenq del Munt clastic wedges (SE Ebro basin, NE Spain). Sedimentary Geology Special Issue. MAILLARD,A., MAUFFRET,A., WATTS,A. B.,TORNE, M., PASCAL, G., BUHL, P. & PINET, B. 1992. Tertiary sedimentary history and structure of the Valencia trough (western Mediterranean). Tectonophysics, 203, 57-75. MART~, J., MITJAVILLA,J., ROCA, E. & APARICIO, A. 1992. Cenozoic agmatism of the Valencia Trough (Western Mediterranean): Relationship between structural evolution and vulcanism. Tectonophysics, 203, 145-165. MASANA, E. • GUIMERA,J. 1992. Escarpe de falla del Cuaternario reciente en la falla del Camp (Cadenas Costeras Catalanas). Geogaceta, 11, 56-58. MAUFEREI, A., MAILLARD,A., PASCAL,G., TORNI2,M., BUHL, P. & PINET, B. 1992. Long-listening multichannel seismic profiles in the Valencia trough (Valsis 2) and the Gulf of Lions (ECORS): a comparison. Tectonophysics, 203, 285-304. MEIGS, A., VERGI~S, J. & BURBANK, D. W. 1996. Tenmillion-year history of a thrust sheet. Geological Society of America Bulletin, 108, 1608-1625. MONIt~, E, JOLIVET, L., BRUNET, C., TORRES-ROLDAN, R. L., CABu R., GOEEE, B. & DUBOIS, R. 1996. Cooling paths of metamorphic rocks in the Western Mediterranean region and tectonic implications. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. CergyPontoise Workshop Abstracts, Paris, France. - - , MONTIGNY,R. & MALUSKI,H. 1992. Age burdigalien de la tectonique ductile extensive dans le massif de l'Edough (Kabilies, Alg6rie). Donn6es radiom6triques 39Ar-4~ Bulletin de la Soci~t~ Gdologique du France, 163(5), 571-584. Murqoz, J.A. 1992. Evolution of a Continental Collision Belt: ECORS-Pyrenees Crustal Balanced Cross-section. In: MCCLAY, K. (ed.) Thrust Tectonics. Chapman & Hall, London, 235-246. OLIVERA, C., SUSAGNA,T., ROCA, A. & GOULA, X. 1992. Seismicity of the Valencia trough and surrounding areas. Tectonophysics, 203, 99-109. OLIVET,J. L., BONNIN,J., BEUZARD,P. & AUZENDE,J. M. 1982. Cin6matiqne des plaques et pal6ogdogrphie: une revue. Bulletin de la SociOtO gdologique de France, 7, 875-892.
80
J. VERGt~S & F. SABAT
PARES, J. M., FREEMAN, R. & ROCA, E. 1992. Neogene structural development in the Valencia trough margins from palaeomagnetic data. Tectonophysics, 203, 111-124. RICOU, L. E. 1994. Tethys reconstructed: plates, continental fragments and their Boundaries since 260 Ma from Central America to South-eastern Asia. Geodinamica Acta, 7, 169-218. ROCA, E. 1992, L'estructura de la conca CatalanoBalear: paper de la compressi6 i de la distensi6 en la seva gOnesi. Tesi doctorale, Univ. de Barcelona. 1994. La evoluci6n geodin~imica de la Cuenca Catalano-Balear y fireas adyacentes desde el Mesozoico hasta la actualidad. Acta Geologica Hispdnica, 29, 3-25. & DESEGAULX,P. 1990. Geological evolution and vertical movement analysis of the Valencia trough area (western Mediterranean). Marine and Petroleum Geology, 9, 167-185. -& GUIMERA, J. 1992. The Neogene structure of the eastern Iberian margin: structural constraints on the crustal evolution of the Valencia trough (western Mediterranean). Teetonophysics, 203, 203-218. ROEDER, D. 1992. Sardinia Channel and Atlas in Tunisia: Extension and Compression. In: BLUNDELL, U., FREEMAN, R. • MUELLER, S. (eds) A continent revealed. The European Geotraverse. Cambridge, Cambridge University Press, 199-202. ROUCHY, J. M. & SAINT MARTIN, J. P. 1992. Late Miocene events in the Mediterranean as recorded by carbonate-evaporite relations. Geology, 20, 629-632. ROYDEN, L. 1993. The tectonic expression slab pull at continental convergent boundaries. Tectonics, 12, 303-325. SAADALLAH,A. & CABY,R. 1996. Alpine extensional detachment tectonics in the Grande Kabylie matemorphic core complex of the Maghrebides (northern Algeria). Tectonophysics, 267, 257-273. SABA'L E, Murqoz, J. A. & SAN'I'ANA('H,P. 1988. Transversal and oblique structures at the Serres de Llevant thrust belt (Mallorca Island). Geologisches Rtmdschau, 77, 529-538. - - , RO('A, E., Mulqoz, J. A., VERGES,J., SANTANACH, P., MASANA, E., SANS,M., Esr~:vEZ, A. & SANTISTEBAN, C. 1997. Role of extension and compression in the evolution of the eastern margin of lbcria: the ESCI-Valbncia trough seismic profile. Revis'm Sociedad Geol6gica Espaha, 8, 431-448. SANTANA('H,P. (ed.) 1997. ESCI, Estudios Sfsmicos de la Corteza Ib6rica. Revista Sociedad Geol6gica Espa~ta, 8, 303-543. SAUl.A,E., PI('A RI; J., MAT0, E., LLENAS,M., LOSANTOS, M., BERAS'rEGuI, X. & AGUSTL J. 1994. Evoluci6n geodimimica de la rosa del Empord~ y las Sierras Transversales. Acta Geol6gica Hispdnica, 29, 55-75. -
-
-
-
SI~RANNE,M., BENEDICTO,A., LABAUME,P., TRUFFERT, C. & PASCAL, G. 1995. Structural style and evolution of the Gulf of Lion Oligo-Miocene rifting: role of the Pyrenean orogeny. Marine and Petroleum Geology, 12, 809-820. SPAKMAN, W., VAN DER LEE, S. ~ VAN DER HILST, D. 1993. Travel-time tomography of the EuropeanMediterranean mantle down to 1400 km. Physics of the Earth and Planetary Interiors', 79, 3-74. TORNI~, M., BANDA, E. & FERNANDEZ, M. 1996. The Valencia Trough: geological and geophysical constraints on basin formation models. In: ZIEGLER, R A. & HORVATH,E (eds) Peri-Tethys Memoir 2: Structure and Prospects of Alpine Basins and Forelands. M6moires du Museum National d'Histoire Naturelles, Paris, 170, 103-128. TRICART, P., TORELLI, L., ARGNANI, A., REKHISS, E 8,: ZITELUNI, N. 1994. Extensional collapse related to compressional uplift in the Alpine Chain of Northern Tunisia (Central Mediterranean). Tectonophysics, 238, 317-329. VEGAS, R. 1992. The Valencia trough and the origin of the western Mediterranean basins. Tectonophysics, 203, 249-261. VERGI~S, J., MILLAN, H., ROCA, E., MUSOZ, J. A., MARZO, M., CIRI~S, J., DEN BEZEMER, T., ZOETEMEIJER, R. & Ct,OETINGH, S. 1995. Eastern Pyrenees and related foreland basins: Pre-, synand post-collisional crustal-scale cross-sections. Marine and Petroleum Geology, 12, 903-916. VIDAL, N., GALLARL J. & DAlqOBE1TIA, J. J. 1997. Contribution of the ESCI-Val6ncia Trough wideangle ata to a crustal transect in the NE Iberian margin. Revista Sociedad Geol6gica Espaha, 8, 417-429. VIGLIOTI'I, L. & LANGENHEIM,g. E. 1995. When did Sardinia stop rotating? New paleomagnetic results. Terra Nova, 7, 424-435. WATTS, A. B. & TORNE, M. 1992. Subsidence history, crustal structure, and thermal evolution of the Valencia Trough: A young extensional basin in the western Mediterranean. Journal Geophysical Research, 97, 20021-20041. WILDI, W. 1983. La chaine tello-rifaine (Algerie, Maroc, Tunisie): structure, stratigraphie et 6volution du Trias au Miocbne. Revue de Gdologie Dynamique et de GOographie Physique, 24, 201-297. ZEYEN, H. & FERN,~NI)EZ,M. 1994. Integrated lithospheric modeling combining thermal, gravity, and local isostasy analysis: Application to the NE Spanish geotransect. Journal of Geophysical Research, 99, 18089-18102. ZECK, H. R 1996. Betic-Rif orogeny: subduction of Mesozoic Tethys lithosphere under eastward drifting Iberia, slab detachment shortly before 22 Ma, and subsequent uplift and extension tectonics. Tectonophysics, 254, 1-16.
Interaction between faulting, drainage and sedimentation in extensional hanging-wall syncline basins: example of the Oligocene Matelles basin (Gulf of Lion rifted margin, SE France) A. B E N E D I C T O 1, M. S t ~ G U R E T 1 & P. L A B A U M E 2 llnstitut des Sciences de la Terre de l'Eau et de l'Espace, CNRS-Universitd Montpellier II, cc.60, 34095 Montpellier Cedex 5, France 2Laboratoire de GOophysique Interne et Tectonophysique, CNRS-UniversitO J. Fourier, BP53, 38041 Grenoble Cedex 9, France Abstract: A tectono-sedimentary model for hanging-wall syncline basins emphasizing the role of erosion in the basin architecture is proposed, based on the well-exposed Matelles basin. Detailed geological mapping is the base for the analysis of the extensional double ramp-flat Matelles fault system, hanging-wall deformation and relationships between faulting/drainage/sedimentation. Displacement of the hanging-wall above the fault resulted in the formation of (a) upper and lower rollovers associated with the upper and lower rampflat couples, (b) a hanging-wall ramp syncline above the lower ramp and (c) a half graben associated with the emergent ramp and upper flat couple. Only the hanging-wall ramp syncline and the lower rollover are preserved. Ramp syncline basin-fill architecture is characterized by progressive unconformity and faultward migration of the growth synclines 'climbing up' the monoclinal prerift strata corresponding to the ramp-side limb of the ramp syncline. Clast composition analysis of the synrift deposits indicates that sediments were mainly supplied from the hanging-wall flat, and allows to propose palinspastic reconstructions of the hanging wall in order to establish the palaeo-drainage system and its evolution. 'Upper hanging-wall flat catchments' and 'lower rollover catchments' are proposed as main features of the drainage system in extensional hanging-wall syncline basins.
Ramp-flat extensional fault systems comprise a succession of ramp-flat couples. A n a l o g u e modelling has provided insights on the kinematic evolution of simple systems involving two listric ramp-flat couples, i.e., going from the surface downward, an emergent ramp, an upper flat, a lower ramp and a lower (or basal) flat which corresponds to the d6collement level (Fig. 1) (Gibbs 1984; McClay & Ellis 1987a, b; Ellis & McClay 1988; McClay 1990, 1996; McClay & Scott 1991). If the footwall remains undeformed, displacement of the hanging wall on such a fault results in hanging-wall deformation characterized by (1) an upper and a lower rollover associated with the upper and lower ramp-flat couples, respectively, (2) a ramp syncline above the lower ramp and (3) a half graben associated with the emergent ramp. Secondary structures are crestal graben collapse systems within the rollovers, and reverse faults in the hanging-wall flat above the lower ramp (Fig. 1). The origin of natural ramp-flat extensional fault systems has been attributed to two types of mechanisms: (i) deformation o1: a pre-existing fault by compaction or isostatic movements (McClay & Scott 1991; Davison 1987; Xiao & Suppe 1989; Vendeville 1991); and (2) fault flattening into an intermediate d6tachment (Gibbs
1984) or weak overpressured zone (Bruce 1984; Bradshaw & Zoback 1988). However, despite the descriptions of several natural examples (e.g. Allmendinger et al. 1987; Bally et al. 1981; Beach et al. 1987; Gibbs 1987), the detailed geometry and kinematics of such natural systems have been rarely investigated, and implications of specific hanging-wall deformation on drainage (erosion)/sedimentation in the basin remains unexplored. In this paper, we examine a natural example of a basin controlled by a ramp-flat extensional fault system, the Oligocene Matelles basin, located onshore of the Gulf of Lion passive margin (Fig. 2). Our aims are to see what generalizations can be made about interactions between fault geometry, hanging-wall deformation and drainage (erosion)/sedimentation in this particular type of basin. Detailed geological mapping at scale 1/7500 provides the base for the analysis of hanging-wall geometry and basin-fill sediment organization. These data integrated with i n t e r p r e t a t i o n of an industrial seismic retiection profile across the basin (H-S3-J seismic profile, H6rault H-83 Survey, Operator TOTAL, contrasted with others profiles across the area) and of a high-resolution seismic reflection profile (300 m long and 150 m deep,
BENEDICTO,A. ETAL. 1999. Interaction between faulting, drainage and sedimentation in extensional hangingwall syncline basins: example of the Oligocene Matelles basin (Gulf of Lion rifted margin, SE France). In: DURAND,B., JOLIVET,L., HORVATH,F. & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 81-108.
82
A. BENEDICTO E T A L .
half
upper crestal grabencollapse
graben
ramp syncline
lower crestal graben collapse
,.
,//
g-.4
7 reverse
faults I footwall
II
ynrift I
~ -s.@%~l,,~ .... '~ .............
~o "'~
lower
iV" or basal flat
,..
I
prerift ~ ' i '
'
Fig. 1. Ramp-fiat listric fault geometry showing hangingwall deformation features. Modified from Ellis & McClay (1988).
acquired and processed by the Department of Geophysics, University of Leeds) in a key area of the basin, allow us to infer the geometry of the fault at depth. Basin-fill architecture and clast composition analysis show that the distribution of drainage systems and geometry of the basinfill associated with a ramp-flat extensional fault system have specific characteristics, different from those associated with classical steeply dipping planar normal faults or simple listric faults (i.e., comprising only an emergent ramp and a basal flat or low-angle ramp).
Geological setting of the Matelles basin area In the onshore landward part of the Oligocene-Aquitanian Gulf of Lion rifted margin (Auzende et al. 1973; Biju-Duval et al. 1979; Rehault et al. 1984; Burrus 1984), between the C6vennes and Nimes faults (Fig. 2a), the extensional tectonic style is characterized by thinskinned extension of the Mesozoic to Eocene cover above the Palaeozoic basement (S6ranne et at. 1995). The extension resulted in the formation of several half-grabens associated with NE-trending normal faults through the Mesozoic cover with a major ddcollement at the level of the Triassic shales and evaporites. The Matelles basin is one of these small-scale and narrow NEtrending half-grabens (Fig. 2b). A more detailed structural map of the whole Gulf of Lion margin and a detailed description of the regional setting and of the structural style of the margin are found in S6ranne et al. (1995) and Benedicto (1996). The NE-trending half-grabens are filled with Oligocene sediments and are bounded to the
W-NW by NE-trending normal faults. The Matelles basin is bounded by the Matelles fault, the southern continuation of the Corconne fault (Fig. 2b). These faults are traditionally considered as Mesozoic east-facing normal faults reutilized as sinistral strike-slip faults during the Late Cretaceous-Eocene N-vergent Pyrenean compression (Arthaud & Sdguret 1981; Arthaud & Laurent 1995). Two major structures inherited from the Pyrenean compression also characterize the area: (1) to the south, the E-Wtrending Montpellier thrust (Fig. 2b), which displaced the Jurassic succession northwards by at least 6 kin, and (2) to the NW, the E - W trending, N-vergent Pic Saint-Loup anticline (Fig. 2b), which affects the Jurassic succession and is interrupted to the east by the Matelles fault. West of the Matelles fault, the structure is tabular and sub-horizontal between the Montpellier thrust and Pic Saint-Loup anticline, while Pyrenean E-W-trending kilometre-scale folds occur east of the Matelles fault. The synthetic stratigraphy of the area, documented by surface and borehole data (Andrieux et al. 1971; Philip et al. 1979) is represented in the Fig. 3a. The Oligocene continental sediments correspond to the syn-rift fill of the Matelles basin and neighbouring half-grabens. They comprise alluvial fan coarse breccias along the boundary faults (Matelles Breccia, Mattauer 1962), passing eastwards to finer fluvial facies alternating with palustrine limestones. These deposits are attributed to the middle Stampian on the base of Characea and Vertebrate fauna correlated with the P19-P20 foraminiferous zones of Blow (about 34-32 Ma) (Grambast 1962; Crochet 1984).
T E C T O N I C S & S E D I M E N T A T I O N IN R A M P S Y N C L I N E BASINS
83
~zE
~
0
~ tr ,.....~ , ~
~ ~.~.~
~ ~.~ ~ ,,~
r.~ m
~ m
0
0
84
A. BENEDICTO
ETAL.
Fig. 3. (a) Synthetic stratigraphy of the Matelles area. Some major features are: (1) Palaeocene-Lower Eocene continental facies were deposited during the early stages of the Pyrenean compression. At the front of the Montpellier thrust, they correspond to alluvial fan breccia deposits reworking the Jurassic succession of the Montpellier thrust-sheet. Northwards, the alluvial fans pass into fluvial marls and sandstones and palustrine limestones, up to 150 m thick and thinning northwards; (2) Bartonian (Middle Eocene) continental facies were deposited during the late stages of the Pyrenean compression. At the front of the Montpellier thrust and in the St Martin de Londres basin, along the northern limb of the Pic Saint-Loup anticline, the Bartonian deposits correspond to alluvial fans, passing laterally in other sectors to fluvial marls and sandstones; (3) Most of the material of the Upper Eocene (Priabonian) fluvial deposits is derived from the erosion of distant Pyrenean belt located southward. These deposits rest unconformably above most of the Pyrenean folds, except in the Saint Martin de Londres basin, where they are involved in the latest stages of Pyrenean folding. (b) Geological map of the Matelles basin, a to d: shallow cross-sections presented in Fig. 4. 1 to 3: slivers of Lutctian limestones.
Basin structure
Surface geometry of the .fault and footwall
The detailed geological map (Fig. 3b) and crosssections (Fig. 4) show that the Matelles basin consists of a NE-trending syncline formed by Upper Jurassic to U p p e r Eocene sequences corresponding to the prerift sediments, and by u n c o n f o r m a b l e Oligocene s e q u e n c e s corresponding to the synrift basin fill. The syncline is b o u n d e d to the NW, and s e p a r a t e d from a tabular U p p e r Jurassic succession, by a major NE-trending normal fault system.
The outcrop of the extensional fault system is formed by four SE-facing major normal faults (Fig. 3b). These faults are formed by hectometrelong segments striking between N340 and N040, with abrupt changes of direction between them. The Matelles fault, 17 km long, is the major fault of the system. It is comprised of two parts, their linkage being characterized by an abrupt change of strike west of the Matelles village, from NNE for the southern part to NE for the
T E C T O N I C S & S E D I M E N T A T I O N IN R A M P S Y N C L I N E BASINS
85
86
A. BENEDICTO E T A L .
Fig. 4. Shallow geological cross-sections through the Matelles basin showing the hanging-wall syncline geometry (location in Fig. 3b). (a) Northern Matelles syncline. In this transect the Jurassic limestones are involved within the hanging wall. (b) and (c) Southern Matelles syncline. Note the unconformity of the synrift deposits above the steeply dipping NW limb of the syncline, and the Lutetian limestone slides within the Oligocene sediments. (d) The Combaillaux syncline.
TECTONICS & SEDIMENTATION IN RAMP SYNCLINL BASINS northern part. The Matelles fault overlaps the Combaillaux fault to the south, the SaintMathieu fault to the north and links the Corconne fault in the north. The southern part of the Matelles fault dips 60 ~ to 75 ~ (locally 85 ~ to the SE and cuts the sub-horizontal Upper Jurassic carbonates of the footwall, forming a high-angle footwall ramp (Fig. 4b and c). North of Matelles village, the geometry of the northern part of the Matelles fault is similar to that of the southern part. Northeastwards, the fault dip decreases and reaches 30 ~ to the SE west of Saint-Jean de Cucules village, where the fault cuts off at low-angle the gently S-dipping Upper Jurassic carbonates of the footwall (Fig. 4a). In the latter sector, the Matelles fault corresponds to a low-angle footwall ramp. This zone, where the fault has a low-angle dip, coincides with the southern limb of the Pic Saint-Loup anticline. More to the north, NW of Saint-Mathieu village, the Matelles fault is a high-angle fault that cuts the sub-vertical N120-striking beds of the northern limb of the Pic Saint-Loup anticline. Finally, the fault links with the Corconne fault. This northern termination of the Matelles fault also overlaps the Saint-Mathieu fault, a 60~ SE-dipping fault that cuts off in the footwall an E-W-trending folded Upper Jurassic to Lower Cretaceous succession. Therefore, the northernmost part of the Matelles fault and the SaintMathieu fault correspond to high-angle footwall ramps. To the south, the Matelles fault overlaps the Combaillaux fault and disappears southwards into the synrift deposits. The Combaillaux fault dips about 70~ to the SE, cutting off the gently S-dipping Upper Jurassic to Eocene succession of the footwall, forming a high-angle footwall ramp (Fig. 4d). The Combaillaux fault branches to the south into a lateral ramp (or transfer zone) of the E - W trending Montpellier thrust.
Surface hanging-wall geometry The geometry of the hanging wall is characterized by a syncline trending parallel to the basinboundary faults (Fig. 3b). The syncline is asymmetric, with a steeply dipping northwestern limb and a gently dipping southeastern limb. Three NE-trending sub-basins are presently differentiated: the Saint-Mathieu syncline in the north (against the northern termination of the Matelles fault and the Saint-Mathieu fault), the Matelles syncline in the centre (against the Matelles fault), and the Combaillaux syncline in the south (against the southern termination of the Matelles fault and the Combaillaux fault).
87
Geometry of the prerifi. The Matelles syncline displays the best-exposed geometry of the prerift sequence. In the SE limb of the syncline, the prerift dips gently 30 ~ to the NW (i.e. toward the major fault) and displays the complete regional stratigraphy from the Upper Jurassic limestones to the Priabonian fluvial deposits (Fig. 4b and c). Bed dips flatten southeastwards yielding a rollover geometry, affected by a few synthetic and antithetic normal faults. In the NW limb of the Matelles syncline, the prerift dips steeply (60-70 ~ to the SE, i.e. parallel to the Matelles fault, and displays thinner Lower Cretaceous to Lutetian stratigraphic intervals than the SE limb. The NW limb is partially covered by the synrift Oligocene deposits. This limb corresponds to a hanging-wall flat resting against the footwall ramp of the Matelles fault (Fig. 4a, b and c). The stratigraphic position of the base of the hanging-wall flat (i.e., the stratigraphic unit closest to the fault) changes along-strike. From the north to the south, it is located (1) within Upper Jurassic carbonates SE of the Pie Saint-Loup anticline (W of Saint-Jean de Cucules village) where a 350 m thick Upper Jurassic succession is involved in the hangingwall flat (Fig. 4a), (2) within Berriasian limestones N and WSW of Matelles village (Fig. 4b), (3) within Valanginian marls more to the south (Fig. 4c), and within Upper Cretaceous detritics in Combaillaux syncline (Fig. 4d). The Saint-Mathieu syncline is separated from the Matelles syncline by a WNW-ESE-trending Pyrenean fold in the prerift succession (Fig. 3b). In this syncline the geometry of the prerift is similar to that in the Matelles syncline, but the prerift succession is only constituted by Upper Jurassic and Lower Cretaceous limestones. The Combaillaux syncline is also similar to that of the Matelles syncline, but (1) the Lower Cretaceous marly limestones are absent in the prerift succession and the Upper Cretaceous detritics rest directly on the Upper Jurassic; and (2) the NW limb shows a more complex shape with, from the NW to the SE, a narrow rollover geometry close to the Combaillaux fault, a SE gently dipping monocline, and a SE steeply dipping monocline (Fig. 4d). The anticline fold between the two monoclines is in the southern prolongation at depth (not emergent) of the Matelles fault.
Geometry and facies distribution of the synrift. The Saint-Mathieu syncline displays the best evidence of lateral facies changes within the Oligocene synrift deposits, which are well constrained on surface by the geological map (detailed in Fig. 5) and at depth by a transverse high-resolution seismic reflection profile (Fig. 6). The Saint-Mathieu syncline is filled by an
88
A. BENEDICTO ETAL.
Fig. 5. Detailed map of the Saint-Mathieu syncline. Note the progressive unconformities on both limbs of the syncline. For the prerift see Fig. 3b. E, Eocene; J, Jurassic; K, Cretaceous; Q, Quaternary.
asymmetric Oligocene depositional system constituted by (Figs 5 and 6): (1) dominant coarse alluvial fan/fan delta deposits (Matelles Breccia) on the NW steep limb, (2) dominant fluvial-lacustrine marls in the axis of the syncline, and (3) dominant marly shales and lacustrine carbonate deposits on the SE limb. The three main facies show lateral indentations. The coarse breccia passes, from the NW to the SE, from monogenic and framework supported with a small amount of red siltstone matrix to polygenic finer breccia with a marl/carbonate fluvial lacustrine matrix. Oligocene deposits unconformably overly the prerift sequence on both limbs of the syncline: on the eastern limb, Oligocene limestones, dipping 10-30 ~ to the NW, rest above Valanginian marls; on the western limb coarse breccia, dipping 35-60 ~ to the SE, rest unconformably
above steeply SE-dipping Jurassic and Valanginian strata. The internal structure of the syncline fill is characterized by progressive unconformities in both limbs and divergent configuration towards the axis of the syncline (Figs 5 and 6). The high-resolution seismic section suggests a divergent pattern within the Oligocene lacustrine deposits above the prerift (Fig. 6a). In the Matelles syncline, the geometry and facies distribution of the Oligocene deposits are well constrained on surface by the geological map (Fig. 3b). Synrift Oligocene deposits overlie the steeply dipping prerift deposits of the NW limb of the hanging wall with a high-angle unconformity, whereas they are apparently conformable with the Priabonian clastics of the gently dipping SE limb of the syncline. Dips of the synrift deposits in the NW limb decrease up-section, the younger
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS
89
Fig. 6. (a) Depth converted section of the high-resolution seismic line through the Saint-Mathieu syncline (location in Fig. 5). (b) Cross-section of the Saint-Mathieu syncline from field and seismic data (location in Fig. 5). Note: (1) NW gentle dips of the prerift Lower Cretaceous marly limestones and Oligocene lacustrine limestones of the SE limb of the syncline; (2) SE-steep dips of the prerift Upper Jurassic limestones and Oligocene breccia of the NW limb of the syncline, (3) unconformity (erosion) between synrift and prerift sequences in the NW limb of the syncline, (4) progressive unconformities in both limbs within the synrift sediments, (5) migration of synrift depocentres toward the fault, and (6) the lateral facies changes within the synrift deposits.
beds being sub-horizontal and unconformable over the steeply dipping Lutetian and Valanginian strata (Figs 3b and 4b and c). The Oligocene depositional system comprises similar facies to those described in the Saint-Mathieu syncline, but with a more complex distribution. The coarse and angular breccia occur along the NW-limb of the syncline. Their detailed mapping allows differentiation of several individual fan systems, passing distally to d o m i n a n t marl facies with a few lacustrine limestone intercalations (Fig. 7b). Some palaeocurrent measurements show that sediments were mainly supplied from the WNW. Predominantly marl deposits in the SE limb of the syncline give evidence for low sediment supply from the SE. In the Matelles syncline, three slivers of Lutetian limestones (1 to 3 in Figs 3b, 4b and c), some tens of metres thick and up to two kilometres long along-strike, are intercalated within the Oligocene deposits: slides 1 and 2 are slightly unconformable with the Oligocene deposits of the SE limb of the syncline (Fig. 4b and c),
whereas slide 3 has a syncline geometry and rests above the steeply dipping Lutetian limestones of the NW limb of the syncline (Fig. 4c). Deformation structures at the front of the slivers in the underlying Oligocene deposits correspond to SE-verging thrusts. These thrusts are sealed by the Oligocene deposits which cover the slivers, showing that the latter were placed as olistoliths sliding from the NW in the basin during the Oligocene sedimentation (Philibert 1992). In the Cornbaillaux syncline, the lower quality of the outcrops of the Oligocene deposits does not allow mapping and precise characterization of facies distribution. Main facies correspond to the dominant coarse breccia deposits, which are located in two main areas north and south of the syncline (Fig. 7a). These two areas are interpreted as corresponding to two individualized fan systems. D o m i n a n t marly facies are distributed between b o t h fan systems. The synrift deposits are discordant above the prerift Lutetian limestones of the NW limb, and are apparently conformable with the prerift Priabonian conglomerates of the SE limb (Fig. 4c).
90
A. BENEDICTO ETAL.
Fig. 7. Facies distribution maps of the synrift deposits in the (a) Combaillaux and (b) Matelles synclines. Distribution of breccia dominant deposits allows to individualize six fan systems in the Matelles syncline and two in the Combaillaux syncline.
Deep structure and structural interpretation In the Saint-Mathieu and the Matelles synclines, the NW limb of the syncline corresponds to a hanging-wall flat (as indicated by the prerift series parallel to the fault) that implied the existence of an upper fault flat, presently eroded, from which the hanging-wall flat was detached. The rollover structure of the SE limb of the syncline reflects the listric geometry of the footwall ramp at depth.
Although industrial seismic data in the region are scarce and of rather poor quality, they provide useful information about the deep geometry of the Matelles fault and depth of the major d6collement level. The H83-J seismic line crosses the Matelles basin from NW to SE (Fig. 2b). This line illustrates well the SE limb of the hanging-wall Matelles syncline and the rollover geometry displayed by the prerift southeasternwards (Fig. 8a). Truncation of the hanging-wall reflectors towards the NW suggests that the
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS Matelles fault dips steeply near the surface and becomes listric at depth with a low-angle segment between 0.6 and 1 s (twt). This lowangle segment separates NW-dipping reflectors of the rollover from some horizontal reflectors interpreted as belonging to the footwall. Then the fault passes into a major sub-horizontal ddcollement at 1 s (twt). This fault profile is coherent with the rollover geometry of the hangingwall towards the SE. Depth conversion using stack velocities results in a low-angle ramp dipping about 30 ~ and a d6collement at about 2.600 m which is interpreted as corresponding to the Triassic marls and evaporites (Fig. 8b) by lateral correlation with the Castries well (Fig. 2b). Therefore, the Matelles and the SaintMathieu faults are interpreted as double rampflat fault systems comprising (Fig. 8b): (1) a major horizontal lower d6collement level within the Triassic shales and evaporites, (2) a lowermajor ramp which consists in a deep low-angle ramp within the Lower-Middle Jurassic carbonates which gradually changes upwards (listric geometry) to a high-angle ramp within the Middle-Upper Jurassic carbonates, (3) an upper flat that, along-strike, is located within different stratigraphic units (Upper Jurassic to Lower Eocene from north to south), and (4) an emergent ramp within the Cretaceous and Eocene. This fault geometry is responsible for the hanging-wall deformation leading to the formation of (a) the Matelles basin as a hangingwall syncline basin above the high-angle part of the lower-major ramp, and (b) the lower rollover of the SE limb which results from the listric geometry of the fault connecting the deep high-angle and low-angle ramps. The upper fault flat, the emergent ramp of the fault, and the associated upper hanging-wall rollover have been eroded. The more complex shape of the NW limb of the Combaillaux hanging-wall syncline is related with the overlap zone between the Matelles and the Combaillaux faults. The overlap zone is interpreted as a relay ramp (in the sense of Peacock & Sanderson 1991), which gently dips (10 ~ to the SSE. The relay ramp induces an intermediate fault flat gently dipping to the SE, resulting in an intermediate shallow ramp-flat fault couple (Fig. 9). The gently SE-dipping monocline of the NW limb of the syncline (Fig. 4d) corresponds to the portion of the hanging wall above the gently inclined flat of the relay ramp, and the SE steeply dipping monocline (Fig. 4d) corresponds to the portion of the hanging wall above the high-angle deep ramp of the Matelles fault. The anticline fold between
91
the two monoclines corresponds to the hangingwall fold above the transition from the gently dipping relay ramp to the high-angle deep ramp of the Matelles fault. The emergent ramp and the upper fault flat of the Combaillaux fault corresponding to those of the Matelles and the Saint-Mathieu faults are also eroded. The graben located immediately to the SE of the Matelles syncline (between the SaintC16ment and the Prades faults, Fig. 2b), which is not well imaged in the seismic profile, is interpreted as corresponding to a rollover crestal graben collapse (Fig. 8b). The antithetic secondary fault represents the boundary of the deformed part of the hanging wall, i.e. the rollover due to the listric geometry of the Matelles fault, as is classically shown by analogue modelling (see Ellis & McClay 1988). The structural analysis of the Assas basin, illustrated in the regional cross-section east of the Matelles basin (Fig. 8b), is not described in detail here as it is not the objective of this paper (see Benedicto 1996). This basin, mainly filled by lacustrine limestones, is also interpreted as a hanging-wall ramp syncline. However in this case, the syncline was developed above a ramp of the d6collement (without emergent ramp). The d6collement ramp was induced by the different depth of the Triassic d6collement level. Inherited vertical offset of the Triassic (and Palaeozoic basement) resulted from Mesozoic faults (Prades fault). These inherited Mesozoic faults were reactivited only minimally or not at all during the Oligocene extension, but passively transported (and deformed) within the detached cover.
Hanging-wall ramp syncline 'versus' forced fold. The structural interpretation of the Matelles basin as a hanging-wall ramp syncline above a ramp-flat extensional fault system, differs from that of modelled extensional forced folds (Patton 1984; Vendeville 1987; Withjack et al. 1990; Erslev 1991) by the facts that: (a) the Matelles hanging-wall basin is an asymmetric syncline, while forced folds are upward-widening monoclines; (b) strata of the ramp-fault side limb of the Matelles syncline dip parallel to the fault (ramp) implying an upper fault flat, while in forced folds hanging-wall strata are cut through by the fault (in all cases with an angle which depends on the fault dip); and (c) the rollover structure of the opposite limb of the Matelles syncline implies a listric geometry of the ramp fault at depth and the existence of a basal d6collement, while in forced folds the fault-opposite limb does not exist (upwardwidening monoclines) and there is no cover
92
A. B E N E D I C T O E T A L .
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS d6collement. Specifically, the most significant pattern of the hanging-wall ramp syncline is the hanging-wall flat lying on the footwall ramp which is, as in the case of thrust systems, the signature of a ramp-flat fault system. This simple but significant observation points out the doubt about the consistence of some forced fold model interpretations of field data. For example, in the case of the eastern E1-Qa plain (eastern margin of the Gulf of Suez) interpreted as corresponding to a forced fold structure (Withjack et aL 1990), and that we suggest might correspond to a hanging-wall ramp syncline (compare Withjack et al. fig. 13b in 1990, with Fig. 4b of this paper).
Step by step restoration cross-section The step by step restoration cross-section from present to the Priabonian times (Fig. 10) shows a pre-Oligocene syncline geometry. Thickening of the Neocomian, Lower Eocene and Priabonian sequences southeastwards gives evidence for the activity of a pre-Oligocene SE-facing fault. As (1) there is no evidence for an emergent fault controlling thickening of these sequences (no major fault affects the presently outcropping Lower Eocene and Priabonian sequences) and (2) from the restored section, these sequences were not deposited northwestwards, we conclude that the upper flat and the lower ramp of a pre-existing fault were already active during the Neocomian, Early E o c e n e and P r i a b o n i a n times, inducing formation of a pre-Oligocene syncline as a hanging-wall syncline. On the other hand, the fact that the thick Neocomian deposits extend above the whole SE-down lifted comp a r t m e n t indicates that, at least during this period, the pre-Oligocene fault was not a lowangle fault at depth (or a listric fault associated with a d6collement level), but a steeply dipping planar (or sub-planar) fault. In the absence of more data about the geometry at depth of the pre-Oligocene Matelles fault, we propose that the latter was a SE-facing steeply dipping fault involving the Palaeozoic basement at depth. This interpretation agrees with the interpretation of (1) a Mesozoic steeply dipping normal fault involving the Palaeozoic basement and (2) a strike slip Eocene fault (Arthaud & S6guret 1981). The latter controlled the Lower Eocene
93
Fig. 9. Simplified scheme showing the overlap relations between the Combaillaux and the Matelles faults. The overlap zone corresponds to a relay ramp that induces a secondary ramp-flat (crosssection d in Fig. 4).
and Priabonian deposits in a narrow NE-trending syncline close to the fault. The deep listric part of the Oligocene Matelles fault was then newly formed from a ddcollement in the Triassic during the rifting. Upwards, the fault re-utilized the upper ramp and the upper ddcollement of the inherited fault, resulting in a double flatramp listric extensional system.
Interaction between faulting, drainage and sedimentation The well-exposed Matelles and Saint-Mathieu synclines provide good conditions to constrain the relations between drainage, sedimentation and tectonics in an extensional hanging-wall ramp syncline basin.
Conceptual model Extensional hanging-wall fault-bend folding. In a ramp-flat extensional fault system, deformation of prerift strata within the hanging wall results from the fault-bend folding as the hanging wall slides above the irregular fault surface (McClay & Ellis 1987a, b; Ellis & McClay 1988). The mode of deformation of the hanging wall in this type of extensional systems, and the resulting
Fig. 8. (a) Line drawing of the time-migrated H83-J seismic reflection line through the Matelles basin area (location in Fig. 2b). E, Palaeocene to Middle Eocene; K, Lower and Upper Cretaceous; J, Jurassic, T, Triassic; Pz, Palaeozoic. (b) General cross-section of the Matelles basin area showing the extensional fault system. It is characterized by a thin-skinned extension (ddcollement) of the Mesozoic to Eocene cover above the Palaeozoic basement. Differences between (a) and (b) sections are due to their slightly different orientation (see Fig. 2b).
94
A. BENEDICTO E T A L . Fig. 10. Step by step restored cross-section of the Matelles basin. The restoration assumes an inclined simple shear (antithetic simple shear a = 60~ deformation mechanism for the hanging wall (for more details about the restoration technique see Benedicto 1996). (a) Present cross-section (see Fig. 8b); (b) Late Stampien (through-going normal fault); (e) Stampien (The Matelles fault); (d) Early Stampien (crestal graben collapse); Priabonian (preOligocene faulting).
geometry have been largely discussed in the literature (see introduction). The kinematics of fault-bend folding caused by slip of the hanging wall along an extensional fault formed by two ramp-fiat couples is illustrated in the Fig. 11. In this figure, we use a planar fault system for simplicity. Although a listric and a planar fault geometries induce different hangingwall geometries (Faure & Chermette 1989), the kinematic model is equivalent. Our model in Fig. 11 assumes deformation of the hanging wall by inclined shear (White et al. 1986; White 1987; Faure & Chermette 1989; Dula 1991; Xiao & Suppe 1992). We favour an antithetic inclined shear of 60 ~ (from the horizontal), coherent with Coulomb fracture orientation, and within the range of most common shear angles in natural systems following Faure & Chermette (1989),
Fig. 11. Fault-bend folding caused by slip of the hangingwall along an extensional fault system formed by two ramp-flat couples. X, Y, and Z are the active axial fold surfaces locked to the footwall: The hanging-wall material passes through them and folds. X', Y', and Z' are the passive axial fold surfaces: They are locked to the hanging wall and transported with it during extension.
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS Dula (1991), and Haugue & Gray (1996). The effect of compaction on final shear orientation (Davison 1987; Xiao & Suppe 1989; Kerr & White 1992) is not taken into account here, because in the Matelles case, the thickness of the synrift Oligocene sediments is small (<500 m) and most of the hanging wall is constituted by already compacted Jurassic and Valanginian prerift series. Different fold axial surfaces parallel to the shear direction form within the sliding hangingwall (Fig. 11). In Fig. 11a, X, Y and Z correspond to the axial surfaces of fault bends. They are locked to the footwall and act as active axial folding surfaces of the hanging-wall, the rocks of the hangingwall passing through them during sliding (Fig. 11b and c). As the hanging wall flat passes through the active X and Z axial surfaces, it folds as a syncline and then it unfolds, becoming a faultward-dipping monocline. When the hanging-wall flat passes though the active Y axial surface, it folds as an anticline and then it unfolds becoming a monocline parallel to the major ramp. X', Y' and Z' are passive axial surfaces of the hanging-wall folds. They are locked to the hanging wall and move with it during sliding (Fig. 11b and c). X' and Z' correspond to anticline folds of the hanging wall, while Y' corresponds to a syncline fold.
Extensional growth syncline geometry. If the structure described above is fed by sediments, the basin fill in the half-graben associated with the emergent ramp/upper flat fault couple and in the hangingwall syncline associated with the lower ramp/lower flat fault couples, grows up and deforms following the deformation of the prerift rocks within the hanging wall. Figure 12 shows the architecture of the synrift basin fill in double flat-ramp extensional systems. In this figure, the fault is listric and the lower fault fiat has been changed to a low angle ramp in order to better simulate the Matelles fault system (Fig. 8b). We concentrate our attention on the architecture of the hanging-wall ramp syncline as it is the main depocentre of the system, due to the relative size of the different elements forming the Matelles fault system. In Fig. 12a, the potential void created by the fault slip is entirely filled by sediments coming from outside of the plane of the diagram, and there is neither erosion of the footwall nor of the hanging wall. Y'I, Y'2 and Y'3 are the respective passive axial surfaces of the folded successive syn-rift deposits (stages 2 to 4), i.e., they correspond to the respective axes of the successive growth synclines. Note that only the Y ' I axial surface of the syn-rift deposits corresponds
95
to the axial surface of the pre-rift strata. Y'2 and Y'3 form on the hanging-wall flat monocline above the major ramp. In fact, Y'2 and Y'3 (i.e. the growth synclines) form at the same time that the hanging-wall anticline unfolds above the high-angle deep ramp after passing the Y axial active surface. The line which links the axes of the successive growth synclines is the axial plane (S) of the final composite growth syncline. (S) is parallel to the high-angle deep ramp (oL= 13).The axial surface (S) remains parallel to the highangle deep ramp only if the thickness of the hangingwall upper flat sliding above the highangle deep ramp does not change during the extension. In the scheme of Fig. 12b, most of the sediments are supplied by erosion of the hanging wall in the sketch plane. Due to erosion, the hanging-wall flat becomes progressively thinner as the system evolves and consequently the ramp-side border of the basin migrates towards the ramp. If the erosional surface is relatively planar, gently dipping basinward, as is assumed in Fig. 12b, then as the system evolves, this surface is folded as a syncline (hanging-wall ramp syncline) when the hanging-wall anticline above the transition between the upper flat of the fault and the high-angle deep ramp unfolds becoming monoclinal; as a result, the successive youngest growth synclines form above the monocline steeply dipping pre-rift strata of the hanging-wall flat (ramp-side limb of the ramp syncline). The axes of the synclines are closer and closer to the ramp. The axial surface of the composite growth synclines (S) is not parallel to the high-angle deep ramp (oL > 13). This system, resulting from progressive erosion of the prerift series of the hanging-wall flat and basin infill during extension, induces this particular pattern of growth synclines 'climbing up' the monoclinal prerift beds of the steeply dipping ramp-side limb of the ramp syncline. By opposition, synrift sedimentation on the hanging-wall flat before it reaches the high-angle deep ramp would cause thickening of the hanging-wall flat and would lead to a rolloverward displacement of the basin through time. In this case, the axial surface of the final composite growth syncline plane will dip steeper than (or in opposite sense) the major ramp (o~ < 13).
Drainage system and role of erosion. Analogue modelling of ramp-flat extensional fault systems most often considers rates of sedimentation of 100%, i.e., completely filling the potential basin (McClay & Ellis 1987a, b; Ellis & McClay 1988; McClay 1990, 1996; McClay & Scott 1991). The role of different rates of sedimentation has been
96
A. BENEDICTO E T A L .
Fig. 12. Model of sedimentation in a ramp-flat extensional fault system. In order to better simulate the Matelles basin conditions, the lower fault flat has been changed for a low angle ramp. (a) Model without erosion and 100% basin-infilling by material coming from the outside of the plane of the diagram. (b) Similar model, but with erosion of the hanging-wall flat and basin-infilling by material coming from within the plane of the diagram. Note migration of the axis of the successive synrift synclines though time in both cases. (S) is the axial plane of the composite growth syncline, ~ is the dip of the high-angle deep ramp, [3 is the dip of the composite growth syncline axial plane. Y is the active axial fold surface of the flat-ramp couple. Y'I, Y'2 and Y'3 are the equivalent passive axial fold surface for each stage; They indicate the location of the axes of the growth synclines.
investigated in some numerical models (Faure 1990; Schlische 1991), but the effect of the erosion was never considered. The drainage system feeding half-grabens associated with steep planar faults has been recently investigated from field examples in active basins (Leeder & Jackson 1993; Gawthorpe & Hurst 1993; Jackson & L e e d e r 1994; Eliet & G a w t h o r p e 1995) and numerical modelling
(Travis & Nunn 1994). These works show the role of erosion process in the stratigraphic architecture of the basin; but the role of the erosion in ramp-flat extensional fault systems remains unexplored. Our analyis of the interactions between faulting, drainage and sedimentation in double rampflat extensional fault systems, applied to the Matelles basin, is based on the approach of
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS
97
Fig. 13. Sketch illustrating some typical interactions between faulting, drainage and sedimentation in extensional systems. (a) Model proposed by Leeder & Jackson (1993) for steep planar faults. 1, smaller footwall catchments; 2, larger hanging-wall catchments; 3, very large catchment in the offset fault zone. (b) Model proposed in this work for ramp-flat extensional fault systems. Explanation in the text.
Leeder & Jackson (1993) for the case of active basins associated with steeply dipping planar faults and simple hanging-wall tilting. Leeder & Jackson (1993) show that the drainage basins (catchments) play a major role in the evolution of sedimentary basins, providing the sediment flux from tectonic uplands that is dispersed and deposited over the basin floor. They concluded that smaller catchments develop in the uplifted footwall, while larger catchments develop in the tilted hanging wall (Fig. 13a). In the case of double ramp-flat extensional systems, relationships between faulting and drainage may be largely different. In the present study, which refers to ancient basins, the limitation is that catchment size in map view is unknown, because they are not preserved due to the post-rift erosion. In our model (Fig. 13b), which is considered as a closed system in which the volume of material is conserved, the Leeder & Jackson model can only be applied to the half-graben associated with the emergent ramp and the upper rollover. In this case, we refer to the catchments related to the half-graben as footwall and upper rollover catchments. C a t c h m e n t s which supply sediments to the hanging-wall ramp syncline
develop (1) on the upper hanging-wall flat, in its transition zone to the lower ramp, and (2) on the lower rollover. We refer to these catchments as upper hanging-wall catchments and lower rollover catchments, respectively. Clast composition in the basin fill depends on the lithology of the rocks eroded within the catchments. The half-graben related to the 'emergent ramp - upper fault flat couple' is fed by material eroded from the footwall and the hanging wall in classical footwall and hangingwall c a t c h m e n t s in the sense of L e e d e r & Jackson (1993). The hanging-wall ramp syncline f o r m e d above the lower r a m p is, in general, fed by m a t e r i a l e r o d e d f r o m the hanging wall. However, if the level of sedimentation in the hanging-wall syncline is lower than the base of the hanging-wall flat (i.e., the upper fault flat), erosion in the upper hanging-wall flat catchments may reach the footwall (erosion of the complete thickness of the hanging-wall flat) (1 in Fig. 13b). We refer to this particular type of catchment as 'mixed hanging-wall flat-footwall catchments'. N o t e t h a t erosion in the rollover catchments (in both the upper and lower rollovers) c a n n o t reach the footwall because equilibrium b e t w e e n erosion and
98
A. BENEDICTO ETAL.
sedimentation occurs higher than the level of the fault flats. By varying the size of the catchment and the size of the upper fault flat, several intermediate types of catchments are possible. Rapid erosion of shorter hanging-wall flats (related to narrower upper flat) will generate a variant of 'mixed hanging-wall flat-footwall catchments' and footwall catchments with erosion of the half-graben associated with the emergent ramp. Clast composition of the synrift sediments in the hanging-wall ramp syncline will reflect the bedrock composition of the hanging wall. However, care must be taken if the thickness of the hanging-wall flat is small, because in such a case the footwall rocks may be rapidly reached by the erosion process, leading to 'mixed hanging-wall flat-footwall catchments'. Note that the different catchments are related to either passive or active folds of the hanging wall shown in Fig. 11: footwall catchments are related to the X active axial surface; upper rollover catchments are related to the X' passive axial surface; upper hanging-wall flat catchments are related to the Y active axial surface; and lower rollover catchments are related to the Z' passive axial surface. This implicates that upper hanging-wall flat catchments are fixed relative to the footwall fault-bend through time, and the hanging-wall rocks pass through these catchments when sliding along the fault. As a consequence, (1) the basinward eroded part of the hanging-wall flat is progressively dragged in the basin and buried under the basin-fill, while (2) the erosion front in the hanging-wall flat, which tends to migrate away the basin due to regressive erosion, is also displaced basinward by the hanging-wall flat movement. If tectonic movement balances removing of material by erosion, a steady state is reached and a large volume of material may be eroded without enlargement of the catchment. The Matelles ramp-flat m o d e l Extensional growth syncline geometry in the Matelles basin. The accurate bed by bed mapping of the syn-rift deposits in the Matelles and the Saint-Mathieu synclines (Figs 3b, 5 and detail in Fig. 14a), and the high resolution seismic section through the Saint-Mathieu
syncline (Fig. 6) show that each bed forms an asymmetric syncline whose axis has slightly shifted to the NW relative to the previous one. On the steep limb of the syncline, the Oligocene beds unconformably overlies progressively older prerift monocline units (Fig. 14). This situation exactly corresponds to the progressive unconformity resulting from progressive erosion of the prerift series of the hanging-wall flat and basin infill during extension, inducing the pattern of growth synclines 'climbing up' the monoclinal prerift beds of the steeply dipping ramp-side limb of the ramp syncline, as explained in the conceptual model (see above). Clast composition in the synrift breccia deposits and drainage system evolution of the Matelles basin. In order to constrain the relationships between fault activity, hanging-wall syncline development and basin-fill sedimentation in the Matelles basin, we have analysed the clast composition of the synrift Oligocene deposits (Fig. 15). Clast stratigraphy is directly derived from the local stratigraphic sequence. It is composed of seven lithologies: (1) dark oolitic grainstones (Lower Callovian-Oxfordian), (2) grey mudstones (Upper Oxfordian-Kimmeridgian), (3) white peri-recifal packstones/grainstones (Tithonian), (4) white mudstones/wackestones (Berriasian), (5) blue/yellow mudstones/wackestones (Valanginian marly limestones), (6) yellow bioclastic grainstones (Valanginian 'Calcaires Miroitants'), and (7) white lacustrine/palustrine limestones (Lutetian). Some exotic facies clasts (not local stratigraphy) are also present; they correspond to clasts reworked from the Priabonian conglomerates. Clast composition was analysed in 35 sites. Even if rapid lateral changes are related with the boundaries of the individual fans, data may be synthesised by the differentiation of four domains (1 to 3 in Fig. 15 and 4 in Fig. 5). Domain 1, where 100% of the clasts are of Lutetian carbonates, corresponds to the Combaillaux depocentre and to the base of the synrift series in the Matelles syncline. Domain 2 corresponds to the intermediate synrift series in the Matelles syncline. It comprises: domain 2a, characterized by 70(+10)% of Lutetian and 30(+10)% of Upper Jurassic clasts; and corresponds to the southern fans; domain
Fig. 14. Detailed maps of the southern termination of the (a) Saint-Mathieu and (b) Matelles synclines, and (c) block diagram showing migration of the synrift growth depocentres towards the NW, above the eroded NW limb of the prerlft series of the syncline. Note the geometric relationships between the synrift and the prerift strata.
T E C T O N I C S & S E D I M E N T A T I O N IN R A M P S Y N C L I N E BASINS
99
100
A. BENEDICTO E T A L .
Fig. 15. Breccia clast composition of the syn-rift deposits in the (a) Combaillaux and (b) Matelles synclines. (c) Along-strike correlation. 1 to 3, different domains of different clast composition. See text for explanation.
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS
2b, characterized by 70(+10)% of Valanginian marly limestones and 30(+10)% of Upper Jurassic clasts; and corresponds to a small fan between domains 2a and 2c; domain 2c, characterized by 100% of Valanginian marly limestones clasts and corresponds to the northern fans of these intermediate synrift series. Domain 3, characterized by 100% of Valanginian clasts including 1-2% of Upper Valanginian bioclastic grainstones ('Calcaires Miroitants'). This domain corresponds to the youngest deposits of the basin restricted to three small fans. Domain 4, which corresponds to the polygenic breccia in the Saint-Mathieu syncline (Fig. 5). However, in this small sub-basin, the westernmost very coarse breccia overlying the Upper Jurassic limestones are composed by 100% of Upper Jurassic limestones. The lateral and vertical distribution of clast composition of the synrift sediments (Fig. 5 and Fig. 15) allows us to infer three major stages in the development of the Matelles and Combaillaux synclines. To illustrate better the relationships between tectonics and drainage during the evolution, tentative palinspastic maps for stages 1 and 2 of extension and sedimentation have been produced (Figs 16 and 17).
Stage 1, palinspastic map Fig. 16. This map corresponds to the erosional, transport and depositional stage of the deposits at the base of the synrift series (Fig. 15c). The 100% Lutetian limestone clast composition indicates that during this stage the basin was supplied by hangingwall flat catchments (catchments i to 7 in Fig. 16) and erosion was not active enough to reach the underlying Valanginian marly limestones (Fig. 16b). In the Combaillaux syncline, two catchments are inferred from the two individualized fans systems differentiated by the distribution of the outcropping dominant breccia deposits (1 and 2 in Fig. 16). In the Matelles syncline, differentiation of individualised fans is not possible because dominant breccia deposits corresponding to the first stage crop out only in the south of the syncline (3 in Figs 15 and 16). The other catchments represented in Fig. 16 are hypothetical and correspond to the fans systems identified in Fig. 15. Dimensions of the catchments cannot be estimated because there is no relation between size of the catchment and amount of material deposited within the basin, as was pointed out in the conceptual model. In the sense of relative dimensions of Leeder & Jackson (1993), a larger hangingwall catchment probably developed in the overlapping zone between the
101
Combaillaux and the Matelles faults (catchment 2 in Fig. 16).
Stage 2, palinspastic map Fig. 17. The morphodynamic system during stage 2 was more diversified (Fig. 17). In the Combaillaux syncline, two catchments are differentiated for this stage. The 100% Lutetian clast composition argues for hangingwall flat catchments (catchments 1 and 2 in Fig. 17). In the Matelles syncline, four catchments are differentiated for this stage corresponding to the four individualised fan systems differentiated by the distribution of the outcropping dominant breccia deposits (catchments 3 to 7 in Fig. 17). In the southern part of the Matelles syncline, the breccia composition (70(+10)% Lutetian and 30(+10)% Jurassic) indicates that the Valanginian sequence was absent in the stratigraphic succession of the hanging-wall and that the erosion reached the Jurassic series of the footwall. The catchment 4 (Fig. 17) was a 'mixed hangingwall-footwall catchment'. In the intermediate part of the Matelles syncline, the breccia composition (70(+10)% Valanginian, 30(+10)% Jurassic) indicates that the Lutetian was absent in the stratigraphic succession of the hangingwall and that the erosion was deep enough to reach the Jurassic series of the footwall. The catchment 5 (Fig. 17) was a 'mixed hanging-wall-footwall catchment'. In the northern part of the Matelles syncline, breccia composition with 100% of Valanginian clasts correspond to fans fed by hanging-wall catchments reworking only Valanginian deposits. The lack of Jurassic clasts within synrift deposits may be explained by a thicker hangingwall flat that here included thick Valanginian series. The catchments 6 and 7 (Fig. 17) were hanging-wall flat catchments. In summary, the major feature during stage 2 is that the lateral changes of clast composition in the different fan systems are related to the lateral changes of hangingwall flat thickness related of hanging-wall flat stratigraphy. The precise structure of the 'mixed hangingwall-footwall catchments' (4 and 5 in Fig. 17) is highly speculative. In a first hypothesis catchments were small, but deep, and incisions in the Lutetian limestones of the upper hangingwall flat were deep enough to reach the Jurassic of the footwall in the transition zone between the upper fault flat and the major ramp. In a second hypothesis catchments were larger, but shallower, and they extended beyond the hanging-wall flat,
102
A. B E N E D I C T O E T A L .
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS reaching the denudated Jurassic of the footwall flat. In Fig. 17, catchment 4 is drowned using the second hypothesis because the gravitational sliding of Lutetian limestones into the basin probably contributed to the denudation of the footwall (see section in Fig. 17). Catchment 5 is drawn using the first hypothesis because the hangingwall flat should be wide and thick enough to provide all the material for stage III (see next). Stage 3. It is represented only by small fans fed by 100% of Valanginian clasts. However, this stage represents a drastic change in drainage conditions, especially in the middle part of the basin where these deposits (5 and 8 in Figs 7 and 15) fed by hanging-wall catchments (Valanginian) overlie deposits fed by 'footwall-Jurassic and hangingwall-Valanginian catchment' (mixed hanging-wall-footwall catchment 6) of stage 2. This vertical evolution seems not to be coherent with a progressive deepening of the catchments. On the contrary, it argues for a rise of the level of the base of the catchment relative to the top of the hanging-wall rocks. Two explanations are possible. In a first hypothesis, the base level actually rose and the base of the catchment valley cutting through the Jurassic footwall was filled by sediments. In a second hypothesis, the base level remained fixed relative to the footwall but a thicker part of the hangingwall flat entered into the fixed catchment providing more new material to be eroded. The presence of a small percentage (1%) of Valanginian bioclastic limestones ('Calcaires Miroitants') restricted to the breccia deposits of the third stage, strongly supports the second hypothesis. It suggests that a residual cuesta relief of the Upper Valanginian bioclastic limestones were preserved in the hanging-wall flat and entered during stage 3 into the catchment (Fig. 18). At present, these monogenic (Valanginian) breccia deposits, unconformably resting on Valanginian beds of the hanging wall, are at a lower elevation than the Jurassic limestones of the footwall. This situation may be explained by the activity of a through-going normal fault (Withjack et al. 1990) terminating the evolution of the extensional ramp-flat fault system (b in Fig. 10).
103
Conclusions The Matelles basin provides excellent conditions to study the evolution of hanging-wall syncline basins associated with ramp-flat fault systems. In this type of system, as the hanging wall slides along the fault, is folded by both active axial surfaces fixed to the fault curvatures and passive axial surfaces fixed to the sliding hanging wall. If sedimentation completely fills the hanging-wall ramp syncline and there is no erosion, the axial surface which links the growth syncline axis (the axial growth syncline plane) is parallel to the fault ramp. Our study of the interaction between normal faulting, erosion and basin development in the Matelles basin, points out that erosion leads to specific geometrical relationships within the synrift deposits and between the latter and prerift strata, different of those in a system without erosion. As the erosion affects the upper hanging-wall flat, this becomes thinner and thinner. Displacement of a progressively thinner upper hanging-wall flat above the ramp induces the migration of the growth syncline depocentres towards the fault ramp; as a consequence, the axial growth syncline plane dips less than the fault ramp. This effect leads to development of a progressive unconformity within the synrift sediments, with the growth syncline 'climbing up' the ramp-side steeply dipping limb of the hanging-wall ramp syncline. This particular geometrical pattern, with the axis of the younger and younger growth synclines unconformably overlying the monoclinal prerift strata, results from a simultaneous effect of: (1) anticlinal folding of the hanging-wall flat passing the flat-ramp fault bend, (2) erosion of the hanging-wall anticline fold, and (3) unfolding of the hanging-wall flat passing entirely on the ramp while flat synrift strata undergo synclinal folding (ramp syncline). Our study also highlights that the sedimentological analysis of the clast composition within the synrift deposits in ancient hanging-wall ramp syncline basins, allows palinspastic reconstructions of the hanging wall in order to establish the palaeo-drainage system and its evolution during the extension. In the Matelles ramp syncline basin, distribution of coarse breccia and breccia clast composition within the synrift deposits indicates that the ramp syncline basin was
Fig. 16. (a) Palinspastic map of the Matelles area and (b) synthetic cross-section corresponding to stage 1 of extension and sedimentation. The map is drawn by reference to the present cartographic trace of the Matelles fault system. It shows: (1) the probable position of the emergent ramp (footwall cut-off), (2) the hanging-wall cut-off of the Valanginian marly limestones and Lutetian limestones, (3) the inferred exposed lithologies, and (4) the inferred position of the catchments corresponding to the individualized fan systems from the distribution of the exposed dominantly breccia deposits.
104
A. B E N E D I C T O E T A L .
TECTONICS & SEDIMENTATION IN RAMP SYNCLINE BASINS
105
Fig. 18. (a) Synthetic cross-section corresponding to the stage 2 (see Fig. 17a) through the fan system 5. During stage 2, a residual cuesta of Upper Valanginian bioclastic limestone was preserved in the hangingwall flat and entered into the catchment during stage 3 (b). See text for explanation. mainly supplied by material eroded from the upper hanging-wall flat, U p p e r hanging-wall catchments were fixed through time on the faultbend transition between the upper fault flat and
the lower fault ramp. As the fault system evolved, the rocks of the sliding hanging wall passed through these catchments. A steady state may be reached if material of the sliding hanging
Fig. 17. (a) Palinspastic map of the Matelles area and (b) synthetic cross-section corresponding to stage 2 of extension and sedimentation. The map is drawn by reference to the present cartographic trace of the Matelles fault system. It shows: (1) the probable position of the emergent ramp (footwaU cut-off), (2) the hanging-wall cut-off of the Valanginian marly limestones and Lutetian limestones, (3) the inferred exposed lithologies, and (4) the inferred position of the catchments corresponding to the individualized fan systems from the distribution of the outcropping dominantly breccia deposits,
106
A. BENEDICTO ETAL.
wall e n t e r i n g the c a t c h m e n t balances the r e m o v ing m a t e r i a l by erosion. E v o l u t i o n of clast composition within the synrift deposits argue for the differentiation of t h r e e stages during the evolu t i o n of the d r a i n a g e system in the Matelles r a m p syncline basin: (a) during stage 1 the catchm e n t s w e r e not d e e p e n o u g h to r e a c h the u p p e r footwall flat and w e r e upper hanging-wall flat catchments in the w h o l e basin; (b) during stage 2 s o m e of the c a t c h m e n t s w e r e d e e p e n o u g h to r e a c h the u p p e r footwall flat and b e c a m e 'mixed hanging-wall flat-footwall catchments'; (c) d u r i n g stage 3 c a t c h m e n t s b e c a m e u p p e r hanging-wall flat type again b e c a u s e the hanging wall e n t e r i n g the c a t c h m e n t was thicker due to the p r e s e r v a t i o n of a residual cuesta. T h e insight g a i n e d f r o m this s t u d y into t h e g e o m e t r y a n d evolution of extensional hangingwall r a m p syncline structures contributes to the c o m p r e h e n s i o n of n o r m a l fault systems. It highlights structural features to differentiate r a m p flat fault systems from forced fold structures, and suggests that a n u m b e r of examples interp r e t e d as forced fold m o d e l s should be re-examined k e e p i n g in m i n d a ramp-flat fault system model. This research was founded by the European Community DGXII (contract JOULE II - CEC Project n. PL 920287 'Integrated Basin Studies'). R. Collier, P. D. Egerton and J. Hart acquired high-resolution seismic data. J. Hart processed high-resolution seismic data (Fig. 6a) and participated to the interpretation of the section (Fig. 6b). We are grateful to Total Exploration for providing seismic data for this study. Some ideas developed in this manuscript benefited from discussions with M. S6ranne and K. R. McClay. We are grateful to the reviewers M. R. Leeder and R. Vially for their comments which improved the paper. We thank C. Wibberley for correcting English spelling.
References ALLMENDINGER, R. W., GAUGE, T. A., HAUSER, E. C., POTrER, C. J. & OLWER, J. 1987. Tectonic heredity and the layered lower crust in the Basin and Range Province, western United States. In: COWARD, M. P., DEWEY, J. E & HANCOCK, P. L. (eds) Continental Extension Tectonics. Geological Society, London, Special Publications, 28, 223-246. ANDRIEUX, J., MATTAUER, M., TOMASI, P., MARTINEZ, C., REILLE, J. L., MATTE,P., BOUSQUET, J. C., RAOUF, K. A., BEL, F., VERRIER,J., BONNET,M. & SAUVEL, M. 1971. Carte Gdologique de la France au 1/50 000. Feuille de Montpellier. Bureau R6gional Geologique et Miniere, France, Orl6ans. ARTHAUD,E & LAURENT,P. 1995. Contraintes, ddformation et d6placement dans l'avant-pays
Nord-pyrdnden du Languedoc mdditerranden. Geodynamica Acta, 8, 142-157. -& SEGURET,M. 1981. Les structures pyr6n6enes du Languedoc et du Golfe du Lion (sud de la France). Bulletin de la SociOtO g~ologique de France, 7, 1-63. AUZENDE, J. M., BONNIN J. & OLIVET, J. L. 1973. The origin of the Western Mediterranean basin. Journal of the Geological Society, London, 129, 607-620. BALLY,A. W., BERNOUroLl, G. DAVIS,A. & MONTARDET, L. 1981. Listric normal faults. Oceanologica Acta, 4, supplement, 87-101. BEACH, A., BmD, T. &GmBS, A. 1987. Extensional tectonics and crustal structure: deep seismic reflection data from the northern North Sea Viking Graben. In: COWARD, M. P., DEWEY, J. E & HANCOCK, E L. (eds) Continental Extension Tectonics, Geological Society, London, Special Publications, 28, 467-476. BENEDICTO, A. 1996. ModOles tectono-sOdimentaires des bassins en extension et style structural de la marge passive du Golfe du Lion (partie nord du Golfe du Lion, SE France). PhD thesis, Univ. Montpellier II. - - , LABAUME,P., SI~GURET,M. & SI~RANNE,M. 1996. Low-angle crustal ramp and basin geometry in the Gulf of Lion passive margin: the OligoceneAquitanian Vistrenque graben, SE France. Tectonics, 15, 1192-1212. Buu-DUVAL, B., LETOUZEY,J. & MONTARDET,L. 1979. Variety of margins and deep basins in the Mediterranean. In: Geological and Geophysical investigation of continental margins. American Association of Petroleum Geologists Bulletin, 29, 293-317. BRADSHAW, G. A. & ZOBACK, M. D. 1988. Listric normal faulting, stress refraction, and the state of the stress in the Gulf Coast basin. Geology, 16, 271-274. BRUCE, C. H. 1984. Smectite dehydration: its relation to structural development and hydrocarbon accumulation in northern Gulf of Mexico Basin. American Association of Petroleum Geologists Bulletin, 68, 673-683. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provenqal basin (North-western Mediterranean). Marine Geology, 55, 247-269. CROCHET, J. Y. 1984. Gdologie et pal6ontologie de la partie septentrionale du foss6 oligoc6ne des Matelles (H6rault, sud de la France). GOologie de la France, 1, 91-104. DAVISON, I. 1987. Normal fault geometry related to sediment compaction and burial. Journal of Structural Geology, 9, 393-401. DULA, W. E 1991. Geometric models of listric normal faults and rollovers. American Association of Petroleum Geologists Bulletin, 75, 1609-1625. ELIET, R R & GATHORPE, R. L. 1995. Drainage deveiopment ans sediment supply within rifts, examples from the Sperchios basin, Central Greece. Journal of Structural Geology, 152, 883-893. ELLIS, P. G. & MCCLAY,K. R. 1988. Listric extensional
TECTONICS & SEDIMENTATION IN R A M P SYNCLINE BASINS fault systems: results of analogue model experiments. 3asin Research, 1, 55-70. ERSLEV,E . ~. 1991. Trishear fault-propagation folding. Geology, 19, 617-620. FAURE, J. L. 1990. Failles normales, coupes equilibrOes et subsidence dans les bassins en extension. Le bassin Viking (Mer du Nord) et le domaine brianqonnais (Alpes occidentales) au Jurassique. Th6se. Univ. Montpellier II. & CHERMEX-rE,J. C. 1989. Deformation of tilted blocks, consequences on block geometry and extension measurements. Bulletin de la Socidtd gOologique de France, 8, 3,461-476. GAWTHORPE,R. L. & HURST,J. M. 1993. Transfer zones in extensional basins: their structural style and influence on drainage development and stratigraphy. Journal of the Geological Society, London, 150, 1137-1152. GraBS, A. B. 1984. Structural evolution of extensional basin margins. Journal of the Geological Society of London, 141, 609-620. 1987. Development of extensional and mixed mode sedimentary basins. In Continental extension tectonics. In: COWARD,M. P., DEWEY, J. E & HANCOCK, P. L. (eds) Continental Extension Tectonics. Geological Society, London, Special Publications, 28, 19-33. GRAMSAST,L. 1962. Aperqu des charofhytes tertiaires du Languedoc et leur signification stratigraphique. Compte Rendu Sommaire ~ la Societ~ g~ologique de France, 10, 96-105. HAUGUE, T. A. & GRAY G. G. 1996. A critique of techniques for modelling normal-fault and rollover geometries. In: BUCHAMAN,P. G. & NIEUWLAND, D. A. (eds) Modern Developments in Structural Interpretation, Validation and Modelling. Geological Society, London, Special Publications, 99, 89-97. JACKSON,J. & LEEOER, M. 1994. Drainage systems and the development of normal faults: an example from Pleasant Valley, Nevada. Journal of Structural Geology, 16, 1041-1059. KERR, H. G. & WHITE, N. 1992. Laboratory testing of an automatic method for determining normal fault geometry at depth. Journal of Structural Geology, 14, 873-885. LEEDER, M. R. & JACKSON,J. A. 1993. The interaction between normal faulting and drainage in active extensional basins, with examples from the western United States and central Greece. Basin research, 5, 79-102. MATTAUER,M. 1962. Sur l'fige oligocdne de la br6che dite "bartonienne" des Matelles (H6rault). Compte Rendu Sommaire d la Soci~tO gdologique de France, 10, 316-317. MCCLAu K. R. 1990. Extensional fault systems in sedimentary basins: a review of analogue models studies. Marine and Petrolleum Geologist, 7, 206-233. 1996. Recent advances in analogue modelling: uses in section interpretation and validation, In: BUCHANAN, P. G. & NIEUWLAND, D. A. (eds) Modern Developments in Structural Interpretaion, -
-
-
-
-
-
107
Validation and Modelling. Geological Society, London, Special Publications, 99, 201-225. -& EI~LIS,P. G. 1987a. Geometries of extensional fault systems developed in model experiments. Geology, 1, 341-344 & -1987b. Analogue models of extensional fault geometries. In: Coward, M. R, DEWEY, J. E & HANCOCK, R L. (eds) Continental Extension Tectonics. Geological Society, London, Special Publications, 28, 109-125. -& SCOTT, A. D. 1991. Experimental models of hangingwall deformation in ramp-flat listric extensional fault systems. Tectonophysics, 188, 85-96. PATTON,T. L. 1984. Normal-fault and fold development in sedimentary rocks above preexisting basement normal fault. PhD thesis, Texas A. & M. Universityl College Station, Texas. PEACOCK, D. C. P. & SANDERSON,D. J. 1991. Displacements, segment linkage and relay ramps in normal fault zones. Journal of Structural Geology, 13, 102-103. PHILIBERT, S. 1992. Gdomdtrie et cin~matique de l'extension oligocOne en Languedoc. Example du synclinal des Matelles (H~rault). DEA, Univ. Montpellier II. PHILIP, H., BODEUR, Y., MATFEI, J., MATFAUER, M., THEROND, R., PALOC,H. & FERAUD,J. 1979. Carte GOologique de la France au 1/50 000. Feuille de Saint-Martin de Londres. Bureau Regional G6ologique et Miniere, France. REHAULT,J. P., BOILLOT,G. & MAUFFRET,A. 1984. The western Mediterranean basin geological evolution. Marine Geology, 55, 447-477. SCHLICHE, R. W. 1991. Half-graben basin filling models: new constraints on continental extensional basin development. Basin Research, 3, 123-141. St~RANNE,M., BENEDICTO,A.,TRUFFERT, C., PASCAL,G. & LABAUME, P. 1995. Structural style and evolution of the Gulf of Lion Oligo-Miocene rifting: Role of the Pyrenean orogeny. Marine and Petroleum Geology, 12, Special issue IBS. TRAVIS, C. J. & NUNN, J. A. 1994. Stratigraphic architecture of extensional basins: insights from a numerical model of sedimentation in evolving half grabens. Journal of Geophysical Research, 99, 15 653-15 666. VENDEVILLE, B. 1987. Champs de failles et tectonique en extension: modelisation experimentale. Thbse, Universit6 de Rennes, Rennes, France. 1991. Mechanisms generating normal fault curvature: a review illustrated by physical models. In: ROBERTS,A. M., YIELDING, G. & FREEMAN,B. (eds) The geometry of normal faults. Geological Socciety, London, Special Publications, 56, 241-249. WHITE, N. J. 1987. Constraints on the measurement of extension in the brittle upper crust Norsk geologiske Tidsskrifi, 67, 269-279. - - . , JACKSON, J. A. & MCKENZIE, D. P. 1986. The relationship between the geometry of normal faults and that of the sedimentary layers in their -
-
108
A. B E N E D I C T O ETAL.
hanging walls. Journal of Structural Geology, 8, 897-909. WITHJACK, M. O., OLSON, J. & PETERSON, E. 1990. Experimental models of extensional forced folds. American Association of Petroleum Geologists Bulletin, 74, 1038-1054. X1AO,H. & SuPez, J. 1989. Role of compaction in listric
shape of growth normal faults. American Association of Petroleum Geologists Bulletin, 73, 777-786. & 1992. Origin of rollover. American Association of Petroleum Geologists Bulletin, 76, 509-529.
Alpine plate kinematics in the western Mediterranean: a westwarddirected subduction regime followed by slab roll-back and slab detachment H . P. Z E C K
Geological Institute, Copenhagen University, Oster Voldgade 10, 1350K Copenhagen, D e n m a r k (e-mail
[email protected]. ku. dk) Abstract: Integration of a wide range of multidisciplinary data, including seismic tomogra-
phy features, regional tectonic transport directions and ages of subduction/collision, lithospheric extension and slab detachment, suggests a new model for the development of the Alpine belt in the western Mediterranean. The model implies that the backbone of the Alpine Orogeny was formed by a composite SW-NE-striking subduction system, active until some time before 22 Ma. The system which consumed Mesozoic Tethyan lithosphere was dipping westward under the leading edge of Iberia which was drifting eastward with respect to North America under the influence of the opening of the North Atlantic. Allowing for a series of late-stage extensional regimes, with local formation of Neogene oceanic lithosphere, in the Valencia Gulf, Provenqal-Algerian basin and southern Tyrrhenian basin, and inherent slab roll-back, the original collision belt may be reconstructed comprising all present Alpine metamorphic core complexes in the western Mediterranean: Betic-Rif, Kabylies and the Sicily-Apennines(-Corsica) belt. It follows that in the western Mediterranean the northward drift of Africa against Iberia/Europe, although influential, e.g., in creating the E-W grain of the Betic Cordilleras, has not been the controlling factor. In contrast, in the eastern Mediterranean the influence of N-S convergence has been much more pronounced due to the hinged sinistral movement of the African plate. A major part of the Europe-Africa N-S convergence was accommodated in a coherent N-dipping subduction zone running from North Africa through Sicily-Calabria to Crete, becoming increasingly significant to the east.
Tracing on a map the connection between the A l p i n e orogenic complexes in the western M e d i t e r r a n e a n - from the Rif, the Betic Cordilleras, the Kabylies, Sicily, Calabria, Apennines and Corsica to Liguria - does not readily suggest a geodynamic rationale. A recent attempt was made by Wortel & Spakman (1992) who, on the basis of seismic tomography data, suggested a model (Fig. 1) that recognizes the importance of slab detachment for the geodynamic evolution of the area, and which starts with a peripheral subduction system, which through a series of successive eastward-directed slab roll-back phases arrived at its present position beneath Italy. It will be demonstrated below that the large-scale slab roll-back mechanism applied by these authors can indeed explain some of the more enigmatic geological relations which have burdened Mediterranean tectonic modelling for years. However, the model of Wortel & Spakman (1992) is perhaps less successful in accounting for Alpine core complexes in N o r t h Africa, the Rif and notably the Kabylies. Wortel & S p a k m a n introduced a major, 1000-1500 km long, E-W-trending transcurrent structure north of the African plate
(Fig. 1) which appears unsupported by geological evidence. The present paper outlines an alternative solution which has some aspects in c o m m o n with earlier models suggested by Argand (1922) and notably Alvarez (1976). Dewey et al. (1989) presented a tectono-stratigraphic evolutionary model for a major part of the western Mediterranean; unexpectedly the tectonic transport pattern established within restricted areas did not reflect the working hypothesis of an overall N-S convergent stress system between the African and E u r o p e a n plates. To remedy this paradox Dewey et al. suggested that the overall N-S stress system had been diffused in a mosaic of fault-bounded crustal fragments. Notably absent in the area surveyed by Dewey et al. (1989) is the Betic-Rif mountain belt, which for a long time has been a serious stumbling block for reaching an allcomprising Alpine kinematic reconstruction for the western Mediterranean. Most regional tectonic models have interpreted this part of the Alpine orogenic belt as the result of N-S convergence between Africa and Iberia producing H i m a l a y a n - t y p e lithospheric thickening and detachment, with body forces causing a regime of
ZECK,H. P. 1999. Alpine plate kinematics in the western Mediterranean: a westward-directed subduction regime followed by slab roll-back and slab detachment. In: DURAND,B., JOLIVET,L., HORVATH,F. & SI~RANNE, M. (eds) The MediterraneanBasins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 109-120.
110
H.P. ZECK
/ J
......................
/
7 ..... /--.............. / j~ _~'V
]
~
,
<"
i
'
-
\ -8So ----_ i ' ~ ~/''~ // ,f-/ " - - ' ~ _vk~..~_ ,_.~/ II -
tl
~1 I
I
500km
Fig. 1. After Wortel & Spakman (1992). A geodynamic model for the Alpine system in the western Mediterranean based on a subduction system initially developed at the present eastern Iberian coast and subsequently transferred eastward by a series of slab roll-back phases (1 through 4) to end beneath the present Apennines. extensional collapse within the overall compressive system (e.g., Dewey 1988; P l a t t & Vissers 1989; Van der Wal & Vissers 1993; Vissers et al. 1995). The present paper considers an alternative interpretation recently suggested by Zeck (1996a), and uses this model for the Betic-Rif as a starting point to arrive at an orogenic model for the western Mediterranean as a whole.
Betic-Rif Orogeny: sinking slab model The sinking slab model for the Betic-Rif Orogeny (Zeck 1996a) is summarized in Fig. 2, a geological sketch map, and Fig. 3, a general cross-section. The model implies that primary nappe emplacement in the orogen was by successive westward understacking (De Jong 1991; Zeck et al. 1992a, b) in a subduction zone dipping westward under Iberia, which was drifting eastwards under the influence of the opening of the North Atlantic (Fig. 3). Initially the subduction system processed Mesozoic Tethyan oceanic lithosphere. Subduction reached its final stage when the system was fed continental lithosphere from the Tethyan realm (Betic-Ligurian lithosphere). The collisional wedge thus produced comprised the primary metamorphic core complex nappes of the Betic-Rif orogen, consisting of crustal material containing local slices of mantle peridotite. After cessation of subduction activity, the subducted slab steepened, broke off and sank rapidly into the sub-lithospheric mantle. Approximately 400 km of Mesozoic oceanic Tethyan lithosphere has been subducted in the Betic Rif system, and has sunk c. 200 km since slab break-off (Fig. 3) shortly
before 22 Ma. This age indication is based on isotopic and palaeontological dating (Zeck et aL 1989, 1990, 1992a; Moni6 et al. 1994; Andriessen & Zeck 1996), which restrict the very rapid rock uplift and cooling of not less than 500~ Ma -1 (Zeck 1996a), following slab break-off, to the period 22-17 Ma (Latest Aquitanian-Burdigalian; Zeck 1996a; all numerical age calibrations after Harland et al. 1990). The forced cooling is thought to have been associated with very fast rock uplift which allegedly was triggered by the slab break-off (Zeck 1996a), but might have been controlled mainly by tectonic extrusion (Thompson et al. 1997), involving the primary Alpine collisional nappe pile thermally softened in the HT regime above the sinking slab and compressed between the rigid Iberian and Betic-Ligurian lithospheric plates (Fig. 3). Lower crustal tectonic levels would thus be characterized by regional compression and higher levels by large-scale extensional structures comprising the tectonically extruded complexes (cf. Dewey 1988). This late extensional tectonic stage produced the present, secondary, nappe sequence, involving large-scale (150-200 km) lateral displacements; Flinch et al. 1996; Zeck 1997) and represents a tectonometamorphic regime radically different from the earlier nappe stacking stage. Whereas this earlier stage is characterized by HP metamorphism (parageneses including omphacite, glaucophane, Mgcarpholite etc.; Nijhuis 1964; Golf6 et al. 1989; Tubfa & Gil Ibarguchi 1991; Bouybaouene et al. 1995; Azafi6n & Goff6 1997), the later stage shows LP parageneses (e.g., Balanyfi et aL 1993; Tubfa 1994).
ALPINE PLATE KINEMATICS, W M E D I T E R R A N E A N
111
Fig. 2. Geological sketch map of the Betic-Rif (modified after Zeck 1996a) featuring the two nappe sequences of the internal zone, the Alpujfirride and Nevado-Filfibride complexes, and showing the vertically projected outline of the sinking lithospheric slab (cf. Figs 3, 4) as revealed by seismic tomography studies (Blanco & Spakman 1993). Open arrows indicate directions of late-stage extensional tectonic transport.
Comparison of orogenic timing in western Mediterranean core complexes
Fig. 3. Cross-section outlining the sinking slab model (modified after Zeck 1996a). The section runs approximately through the island of Alborfin which is located over the SE boundary of the slab (cf. Figs 2, 4). Dashed line outlines the subducted Tethyan oceanic lithospheric slab at the final stage of subduction.
To support the reconstruction of the Alpine collisional belt, a comparison is made of the timing of the main orogenic stages in the various core complexes in the western Mediterranean (Table 1). Based on the detailed North Atlantic spreading chronology obtained by Srivastava et al. (1990), Zeck (1996a) suggested a minimum age of c. 55 Ma for the start of subduction activity in the Betic-Rif system and c. 25 Ma for slab breakoff. These age estimates build on a simplified plate tectonic model based on the assumptions that North Atlantic sea-floor spreading and subduction east of Iberia acted in concert and that the rate of slab sinking is comparable to the rate of subduction/sea-floor-spreading. The position of the nose of the vertical slab at c. 600 km below the general base of the lithosphere (Fig. 3) then translates into an approximate start of subduction at c. 55 Ma, that is the age of chron 22-23, running at c. 600 km east of the Mid-Atlantic ridge (cf. Srivastava et al. 1990, fig.l). Similarly, the age of slab break-off (cf. Fig. 3) would correspond approximately to the age of chron 6c-7
112
H.P. ZECK
9
< ~o0
o~
r r
r r
r162 c-,l
G"
o0
9~
G)
,.a m
~o~
ALPINE PLATE KINEMATICS, W MEDITERRANEAN (c. 25 Ma), running c. 200 km east of the MidAtlantic ridge. The thus obtained age indication for the start of subduction should be regarded as a minimum value as formation of syncollisional nappe complexes will have absorbed part of the Iberian, eastward drift, while another part may have been accommodated further east in the Alpine tectonic system. Initially subduction involved oceanic lithosphere, whereas formation of collisional nappe complexes took place later. Direct age information on timing of Alpine collisional tectonometamorphism in the Betic-Rif orogen (Table 1) is sparse and not very well defined. De Jong (1991) suggested minimum ages of c. 65 Ma (Rb-Sr, phengite-WR) and c. 80 Ma (4~ tourmaline); however analytical results vary widely and do not define an entirely consistent pattern. Moni6 et al. (1996) claimed minimum ages of 50 Ma based on Sm-Nd and 4~ data. As pointed out above, the subsequent stage of regional extensional tectonics is much younger (>22-17 Ma), and the earliest nappe-sealing sedimentary rocks vary regionally in age from c. 22 to 17 Ma (Table 1). Structural, stratigraphic and chronological work in Alpine Corsica, Apennines and Calabria suggest oldest metamorphic ages of c. 80-60 Ma ( K - A r age of c. 90 Ma, Maluski 1977; Rb-Sr isochron age of 105 _+20 Ma, Cohen et al. 1981; and somewhat younger stratigraphic constraints, Egal 1992; De Roever 1972). The older ages should perhaps now be regarded with caution (H. Maluski 1996 pers. comm.). More recent 4~ work suggests ages ranging from c. 65 to 35 Ma (Moni6 et al. 1996). A recent 35 Ma single-zircon age for the HP metamorphism in the western Alps (Gebauer et al. 1997) may be relevant to note. Late E - W - and S W - N E directed extensional tectonics range from c. 27 Ma to recent (e.g., Jolivet et al. 1991; Kastens et al. 1988), oldest nappe sealing rocks from c. 22 to 10 Ma (Egal 1992; Carmignani et al. 1995). In the Kabylies, early Alpine metamorphic ages of c. 80 Ma were reported for high-grade core complexes (4~ muscovite, biotite and feldspar, Moni6 et al. 1988). Younger ages of c. 25 Ma (4~ and Rb-Sr muscovite and biotite, Moni6 et al. 1988) were obtained from regional mylonite zones with regional southdirected tectonic transport (cf. Mahdjoub & Merle 1990), which continued until Recent time (Mauffret et al. 1987). Ages of oldest nappe sealing sedimentary rocks were given as c. 23-21 Ma (Saadallah & Caby, 1996). The analogous structural, stratigraphic and geochronological development of these Alpine complexes, although c. 1000 km apart, suggests that they are part of the same regional
113
geodynamic development. The data indicate an early stage characterized by subduction activity starting in earliest Tertiary time, perhaps latest Cretaceous, and a later orogenic stage characterized by thin-skinned, extensional tectonics which took place from c. 30 Ma to Recent time. The oldest sedimentary rocks sealing these extensional allochthons have ages between c. 25 and 10 Ma.
Reconstruction of the Alpine collisional belt in the western Mediterranean The original configuration of the Alpine subduction zone system may be reconstructed by considering on the one hand the regional pattern of tectonic transport directions and on the other hand the distribution of late-stage extensional basins in the western Mediterranean region. The southern termination of the Alpine subduction zone in the western Mediterranean (here called the Betic-Ligurian subduction system) is given by the sinking slab under the Betic-Rif orogen (Figs 2, 3 and 4b,c). Its abrupt southwestern termination represents the transcurrent contact towards the African plate. The equally abrupt nature of its northeastern termination suggests that the Betic-Ligurian subduction system was formed not as a continuous, smooth belt, but rather had a composite character, segments with active subduction being separated by transform fault zones. Such development of the subduction system may have been controlled mainly by factors such as the shape of the Iberian continental plate and the variable nature of the subducting/colliding Tethyan realm lithosphere. An outline of its preferred configuration is given in Fig. 4b and c; constraints are discussed below. Two approximately N W - S E transform fault zones are proposed (Fig. 4b): one north of Alicante, the other, following an earlier suggestion by Cohen (1980), south of the pre-drift position of Sardinia. It is further proposed that subsequent Miocene extensional regimes which broke up the subduction system were patterned upon these fundamental discontinuities and may be described in part by two spreading poles (Dewey et al. 1989), one in the Genoa area for the Corsica-Sardinia block, and one southwest of Alicante for the Balearic Islands. The length of the Sardinia-Menorca transform fault zone is suggested to be c. 250 km (Fig. 4b). This estimate is based on reconstruction of the CorsicaSardinia rotation (sinistra150~50 ~ c. 400 km for south Sardinia; c. 23-18 Ma, Dewey et al. 1989; 21-16 Ma, Todesco & Vigliotti 1993) and the much later Tyrrhenian extension. The length of the Alicante transform fault zone is not very well
114
H.P. ZECK
defined; it may range from very small to up to c. 100 km. The Balearic subduction zone segment, located between the two proposed transform
fault zones (Fig. 4b, c), comprised the Kabylian collisional prism. The segment was translated towards the SE from its suggested pre-drift position along the present Iberian coast (Dewey et
ALPINE PLATE KINEMATICS, W MEDITERRANEAN al. 1989; Fig. 4b) through opening of the Gulf of Valencia (c. 23-19 Ma, Dewey et al. 1989, Banda & Santenach 1992; c. (25-)15-10 Ma, Torres et al. 1993). Additional extension south of the Balearic Islands (c. 22-13 Ma, Dewey et al. 1989) would have translated it further southeastward (Fig. 4c). This translation is suggested to have been by slab roll-back which requires old subductable oceanic lithosphere southeast of the Balearic subduction zone segment and concomitant spreading of back-arc character between the Kabylian collision prism and the present Balearic Isles (for details on the slab roll-back mechanism, see Elsasser 1971 and Le Pichon & Angelier 1981). Concurrent N-S AfricanIberian convergence (Dewey et al. 1989; Srivastava et al. 1990) might have been accommodated in part along this retreating slab system and in part along a parallel subduction zone somewhat further south (Fig. 4c; see below). This kinematic evolution resulted in the Kabylian core complex being translated from its collisional position close to the slab trench (Fig. 4b) towards the African plate where it was emplaced onto the African foreland (Fig. 4c), such in agreement with consistent top-to-south, Miocene and younger regional tectonic transport directions observed in the Kabylies (see above). Seismic tomography information on the northern termination of the African plate is not conclusive (cf. Mueller 1989) and therefore it is difficult to decide whether late-stage slab breakoff which would have supported uplift of the Kabylian core complex and extensional emplacement upon the African foreland, similar to the model proposed for the Betic-Rif, has been part of the process. Neither is the crucial cooling information, which was an important argument supporting the Betic-Rif model, available for the Kabylies. There is some tomographic support, though, for a steeply N-dipping slab north of the African plate (Mueller 1989), and this might represent the 20-30 ~ dextrally rotated, peeled back oceanic lithosphere of the Balearic segment (Fig. 4b,c). The model is
115
supported by older Nll0-140~ stretching lineations (c. N85-115~ after drift correction) reported by Moni6 et al. (1988) and Saadallah & Caby (1996) for Kabylian metamorphic core complexes. These directions are identical to syncollisional directions of tectonic transport in core complexes of the Betic Cordilleras (Vauchez & Nicolas 1991; Zeck 1996a) and thus conform to the overall E - W convergent character of the Betic-Ligurian subduction system implied by the regional model here presented (Fig. 4b). The Corsica-Sardinia segment of the Betic-Ligurian subduction system also went through a slab roll-back translation after subduction (Fig. 4b, c). This involved coeval backarc type extension west of the Corsica-Sardinia block and formation of the (in part) oceanic Provencal-Algerian basin (Fig. 4c; cf. ChamotRooke et al. 1996). In analogy with the model outlined above for the Balearic-Kabylies segment, this roll-back operation requires old subductable oceanic lithosphere east of the west-dipping subduction zone indicated in Fig. 4b. Results of earlier work which concluded that Alpine Corsica consists of an ophiolite containing nappe sequence which for its major part was formed in a west-dipping subduction zone, mainly by underthrusting from the east (Egal 1992; Jolivet et al. 1994; Carmignani et al. 1995) are in good agreement with this model. Final allochthon emplacement was by west-directed thin-skinned tectonics (Warburton 1986; Carmignani et al. 1995). Complementary eastward sense of tectonic transport in the northern Apennines and Elba suggests regional rock uplift over a sinking slab in the area between Corsica and Apennines and agrees well with earlier work in this particular region (e.g., Jolivet et al. 1994; Carmignani et al. 1995; Keller & Coward 1996; and references therein) and conforms to the Betic-Rif model outlined above. Seismic tomography information (Wortel & Spakman 1992) indicates that the detached, lithospheric slab at present is located c. 100 km
Fig. 4. Outline of the kinematic evolution of the western Mediterranean (modified after Zeck 1996b). (a) Location of metamorphic core complexes, modified after Coward and Dietrich (1989), showing major geographic entities. (b) Final stage of subduction with Iberian plate colliding with continental Betic-Ligurian lithosphere after Mesozoic oceanic Tethyan lithosphere has been subducted westwards under eastward drifting Iberia. The subduction system did not form a smooth, continuous belt: three segments of active subduction being separated by two transform fault zones. (c) Miocene stage of extensional regimes with slab roll-back and local slab detachment involving opening of the Gulf of Valencia, the Provenqal-Algerian basin and Tyrrhenian basin, but prior to the latest stage extension in the southern Tyrrhenian Sea basin (Vavilov and Marsili oceanic basins in dark shading) which translated the slab further east (hatched double arrows) towards its present position (Fig. 1, stage 4) and assisted in the final emplacement of the Calabria core complex, overriding the E-W-trending subduction system (hinge line in bold) accomodating major N-S Europe-Africa convergence (cf. Figs 5, 6). Black arrows indicate late stage extensional tectonic transport directions.
116
H.P. ZECK
more to the east, directly below t h e Apennine chain (Fig. 1, stage 4), that is below the Evergent cover complexes, suggesting that the detached slab underwent some late stage movement towards the east relative to the overlying lithosphere. This suggestion is supported by a recent fault kinematic study by Keller & Coward (1996) indicating an eastward migration of the centre of extension across the Tyrrhenian Sea. Displacement of the southern part of the Corsica-Sardinia subduction zone segment has been much larger than for its northern part (Fig. 4c) and therefore a scenario similar to that given above for the Balearic-Kabylian segment is suggested with additional extension taking place between the Sardinian block and the Calabrian collisional wedge which was located close to the slab trench (Fig. 4c). This resulted in slab rollback towards its present position under the Apennines and initial emplacement of Neogene, E-vergent allochthons in the Apennine chain.
The situation in the southern part of the Corsica-Sardinia segment is complicated by considerable, very fast latest stage extension (Figs 4c, 5, 6) with neoformation of oceanic crust in the southern part of the Tyrrhenian Sea (c. 250 km long, NW-SE elongated area with Vavilov and Marsili basins, c. 7 Ma to Present, younging southeastwards, Kastens et al. 1988). This latestage, predominantly E - W - and N W - S E directed extension has advanced final emplacement of E-(SE-)vergent allochthons in the southern Apennines and brought the Calabrian core complex into its present position, possibly, as in the Betic Rif, involving extrusion tectonics (Figs 4c, 6). This model for the Calabrian region explains a number of local geological features which otherwise appear enigmatic. Consistent top N E - w a r d tectonic transport directions in collisional tectonic melanges in Calabria and southern Apennines (Monaco &
Fig. 5. Sicily-Calabria area with seismic epicentres of events >50 km depth during the period 1988-1993; data from Selvaggi & Chiarabba (1995). Marsili oceanic crust floored basin in striped signature. In bold: hinge line of subduction system which steeply N-dipping Tethyan oceanic lithosphere is overridden by European (crustal) lithosphere (cf. Figs 4c, 6).
ALPINE PLATE KINEMATICS, W MEDITERRANEAN
NW
SE MARSILI
~ CALABRIA
o
,,
-k
r
117
area, the subduction zone has much lower level of >50 km depth seismicity (Figs 5, 6). The present lithospheric slab configuration in the Calabria region according to current interpretations (e.g., Channel & Mareschal 1989; Wortel & Spakman 1992; Robertson & Grasso 1995; Catalano et al. 1995) would comprise a continuous subduction zone (Fig. 1, stage 4) which runs southwards along the Apennines, curves westward around Calabria and continues north of the African plate. The model suggested here implies an alternative lithospheric slab configuration in which the southern Apennines, Calabria and Sicily are dominated by thrust units derived from the Betic-Ligurian collisional realm and emplaced in a regionally extensional setting controlled by eastward slab roll-back and E - W and N W - S E extension (Fig. 4c). The subduction system running north of the African plate and eastward over Sicily and Calabria (Figs 4c, 5, 6) in this model is a separate entity, continuing eastward to connect with the subduction zone dipping northward under Crete.
r 100
km
Fig. 6. Cross section showing seismicity outlining steeply N-dipping subducted Tethyan oceanic lithospheric slab; for location see Fig. 5; diamond on top frame indicates intersection point with the subduction zone hinge line indicated in Figs 4c and 5.
Tortorici 1995) are at variance with a N-S-convergent subduction regime, which is currently claimed (e.g., Dewey et al. 1989), but are in good agreement with the top-to-the-E tectonic regime of the Betic-Ligurian subduction zone, if the 45 ~ post-collisional sinistral rotation is taken into account (Fig. 4c). A Cretaceous age for H P metamorphism in core complexes in Calabria (De Roever 1972) is difficult to combine with Late Oligocene or even later collision as currently suggested (Monaco & Tortorici 1995), but agrees well with the collision timing in the Betic-Ligurian subduction system (see above; Table 1). A high level o f seismicity in a restricted area N W o f Calabria is explained by N W - S E convergence under the influence of extension centered in Vavilov and Marsili basins. Earthquake hypocentres are concentrated within the denser lithospheric slab. Outside this restricted
Conclusions An outline for the tectonic evolution of the Alpine orogen in the western Mediterranean is presented which has the westward subduction of Tethyan oceanic lithosphere under eastward drifting Iberia as its driving force. The resulting SW-NE-striking Betic-Ligurian subduction zone system had a segmented character: three sectors with subduction activity being separated by two transform fault zones. The Tethyan realm lithosphere which was subducted under, and ultimately collided with, the Iberia plate had a laterally variable character. In the Betic segment, the southern one, the Betic-Rif collisional wedge, formed after Tethyan oceanic lithosphere had been subducted, was not subsequently affected by a slab roll-back development. The subducted, and detached, Betic-Rif lithospheric slab is at present located under the Betic-Rif orogen (Figs 2, 3, 4). The intermediate, BalearicKabylian, segment underwent slab roll-back after collision which indicates that the Kabylian collisional wedge represents a small continental fragment located between Tethyan oceanic lithosphere which was subducted westward before collision and Tethyan oceanic lithosphere which was peeled back southeastward after collision. The northern, Corsica-Sardinia, segment likewise underwent considerable slab roll-back after subduction, concomitant with back-arc type extension both in the Provenqal-Algerian basin west of Sardinia and
118
H.P. ZECK
b e t w e e n eastern Sardinia and the Calabrian collisional wedge which was located close to the slab trench (Figs 4b, c). This indicates that also the Calabrian collisional prism represents a small continental fragment within the Tethyan realm. It thus seems that the rather irregular p a t t e r n of extensional basins and roll-back t r e n c h e s which characterizes the w e s t e r n M e d i t e r r a n e a n reflects the distribution pattern of continental and oceanic lithosphere in the Tethyan realm prior to Alpine tectogenesis (cf. A l v a r e z 1976). The ultimate s o u t h e a s t w a r d e m p l a c e m e n t of the Calabrian core complex took place under the influence of the latest stage, c. 7-0 Ma extension involving generation of oceanic crust in Vavilov and Marsili basins. The tectonic evolution model presented here implies that N-S convergence played a minor role in western M e d i t e r r a n e a n tectonics, whereas in the eastern Mediterranean its influence was much more important due to the sinistral hinged m o v e m e n t of the African plate with respect to Eurasia (Dewey et al. 1989; Mueller 1989; Srivastava et al. 1990). This led to developm e n t of the well defined E - W - t r e n d i n g subduction z o n e system dipping n o r t h w a r d u n d e r Crete. It is suggested here that this subduction zone continues westward over Calabria and Sicily and further westward north of the African plate, loosing importance on the way. The tectonic complexities in the SicilyCalabria region (e.g., Dewey et al. 1989; Channel & Marechal 1989; Robertson & Grasso 1995; Catalano et al. 1995; Monaco & Tortorici 1995) may be explained by its location in a zone where two different tectonic regimes meet, one controlled by the E-W-convergent Betic-Ligurian subduction zone system and its off-spin of E(SE)-vergent allochthons following slab roll-back translations, the other controlled by northward subduction of Tethyan oceanic lithosphere under the influence of northward drift of Africa with respect to Europe. Supported by the Danish Research Council (SNF) and Carlsberg Foundation, and in earlier stages by NATO. I thank H. Maluski (Montpellier) for hospitality, A. Saadallah (Z&S, Stavanger) for introduction in North African geology, L. Jolivet (Paris) and A. Mascle (IFP, Rueil Mahnaison) for organizing the 1996 Mediterranean geology meeting in Cergy, and R. Caby (Montpellier) and J. Verg6s (Barcelona) for inspiring and constructive reviews.
References AGUADO, R., FEINBERG, H., DURAND-DEI,GA, M., MART[N-ALGARRA,A., ESTERAS, M. & DIDON,J. 1990. Nuevos datos sobre la edad de formaciones
miocenas transgresivas sobre las Zonas Internas b6ticas: la formaci6n de San Pedro de Alc~ntera (Prov. de M~ilaga). Revista Sociedad Geolologia Espa~a, 3, 79-85. ALVAREZ,W. 1976. A former continuation of the Alps. Bulletin Geolological Society of America, 87, 891-896. ANDRIESSEN,P. A. M. & ZECK, H. P. 1996. Fission track constraints on timing of Alpine nappe emplacement and rates of cooling and exhumation, Torrox area, Betic Cordilleras, S Spain. Chemical Geology (Isotope Geosciences), 131, 199-206. ARGAND, E. 1922. La tectonique de l'Asie. In: 8th International Geolological Congress, 171-372. AZA/q0N, J. M. & GOFFE, B. 1997. Ferro- and magnesiocarpholite assemblages as record of high-P, Iow-T metamorphism in the Central Alpujarrides, Betic Cordillera (SE Spain). European Journal of Mineralogy, 9, 1035-1051. BALANYA,J. C., AZAIq0N,J. M., SANCHEZ-GOMEZ,M. & GAr~C|A-DUESAS, V. 1993. Pervasive ductile extension, isothermal decompression and thinning of the Jubrique unit in the Paleogene (Alpujarride Complex, W Betics, Spain). Comptes Rendus de l'Acad~mie des Sciences Paris, 316, s6r. II, 1595-1601. BANDA, E. • SANTANACH,P. 1992. The Valencia trough (western Mediterranean): an overview. Tectonophysics, 208, 183-202. BLANCO,M. J. & SPAKMAN,W. 1993. The P-wave velocity structure of the mantle below the Iberian Peninsula: evidence for subducted lithosphere below southern Spain. Tectonophysics, 221, 13-34. BOUYBAOUENE, M. L., GOFFE, B. & MICHARD, A. 1995. HP-LP metamorphism in the Sebtides nappes, northern Rif, Morocco. Geogaceta, 17, 117-119. CARMIGNANI,L., DECANDIA,E A., DISPERA'FI,L., FANTOZZI, P. L., LAZZAROTTO, A., LIOTFA, D. & OGGIANO, G. 1995. Relationships between the Tertiary structural evolution of the SardiniaCorsica-Provenqal Domain and the Northern Apennines. Terra Nova, 7, 128-137. CATALANO,R., INFUSO,S. & SULH, A. 1995. Tectonic history of the submerged Maghrebian Chain from the Southern Tyrrhenian Sea to the Pelagian Foreland. Terra Nova, 7, 179-188. CHAMOT-ROOKE, N., MAILI.ARD,A., GAUI.IER, J. M. & PASCAl., G. 1996. Crustal structure of the LiguroProvenqal basin from gravity modelling: geodynamic implications. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop 11-13.12.1996, Cergy-Pontoise, Paris, 35-36. CHANNELL, J. E. T. & MARESCHAI,,J. C. 1989. Delamination and asymmetric lithospheric thickening in the development of the Tyrrhenian rift. In: COWARD, M. P., PARK, R. G. & DIETRICH,D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 285-302. COHEN,C. R. ! 980. Plate tectonic model for the OligoMiocene evolution of the western Mediterranean. Tectonophysics, 68, 283-311.
A L P I N E PLATE KINEMATICS, W M E D I T E R R A N E A N , SCHWEICKERT,R. A. & ODOM, A. L. 1981. Age of emplacement of the Schistes Lustr6s nappe, Alpine Corsica. Tectonophysics, 73, 267-283. COWARD, M. & DmTmCH, D. 1989. Alpine tectonics an overview. In: COWARD, M. P., PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics, Geological Society, London, Special Publications, 45, 1-29. DE JONG, K. 1991. Tectono-metamorphic studies and radiometric dating in the Betic Cordilleras (SE Spain) - with implications for the dynamics of extension and compression in the western Mediterranean area. Thesis, Amsterdam Christian University. DE ROEVER, E. W. E 1972. Lawsonite-albite-facies metamorphism near Fuscaldo, Calabria (S Italy), its geological significance and petrological aspects. Proefschrift Amsterdam University. DEWEY, J. E 1988. Extensional collapse of orogens. Tectonics, 7, 1123-1139. - - , HELMAN, M. L., TURCO, E., HuTroN, D. H. W. & KNOTT, S. D. 1989. Kinematics of the western Mediterranean. In: COWARD,M. R, PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. EGAL, E. 1992. Structures and tectonic evolution of the external zone of Alpine Corsica. Journal of Structural Geology, 14, 1215-1228. ELSASSER,W. M. 1971. Seafloor spreading as thermal convection. Journal of Geophysical Research, 76, 1101-1112. FLINCH,J. E, BALLY,A. W. & WU, S. 1996. Emplacement of a passive-margin evaporitic allochthon in the Betic Cordillera of Spain. Geology, 24, 67-70. GEBAUER, D., SCHERTL, H. R, BRIX, M. & SCHREYER, W. 1997. 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora Maira massif, western Alps. Lithos, 41, 5-24. HARLAND,W. B., ARMSTRONG,R. L., Cox, A. V., CRAIG, L. E., SMITH,A. G. & SMITH,D.G. 1990. A geologic time scale 1989. Cambridge Univ. Press, Cambridge. JOLIVET, L., DANIEL, J. M. & FOURNIER, M. 1991. Geometry and kinematics of extension in Alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. - - , TRUFFERT,C. & GOFFE, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in arc and back-arc regions. Lithos, 33, 3-30. KASTENS ETAL. 1988. ODP Leg 107 in the Tyrrhenian Sea: insights into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KELLER, J. V. A. & COWARD,M. P. 1996. The structure and evolution of the N Tyrrhenian Sea. Geolological Magazine, 133, 1-16. LE PICHON,X. & ANGELIER, J. 1979. The Hellenic arc and trench system: a key to the evolution of the eastern Mediterranean area. Tectonophysics, 60, 1-42. MAHDJOUB, Y. & MERLE, 0. 1990. Cindmatique des ddformations tertiaires dans le massif de Petite
119
Kabylie (Alg6rie orientale). Bulletin de la Soci~tO GOologique de France, 6, 629-634. MALUSKI, H. 1977. Application de la mOthode 4~ 39Ar aux min~raux des" roches cristallines perturbOes par des ~v~nements thermiques et tectoniques en Corse. Thesis Univ. Montpellier. MAUFFRET, A., EL-ROBRINI, M. & GENNESSEAUX, M. 1987. Indice de la compression r6cente en mer M6diterrande: un basin losangique sur la marge nord-alg6rienne. Bulletin de la SociOtd Gdologique de France, 3, 1195-1206. MONACO, C. & TORTORICI, L. 1995. Tectonic role of ophiolite-bearing terranes in the development of the Southern Apennines orogenic belt. Terra Nova, 7, 153-160. MONIfl, P., JOLIVET, L., BRUNET, C., TORRES-ROLDAN, R. L., CABY, R., GOFFE, B. & DUBOIS, R. 1996. Cooling paths of metamorphic rocks in the western Mediterranean region and tectonic implications. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop, Cergy-Pontoise, Paris, 16-17. - - , MALUSKI,H., SAADALLAH,A. & CABY, R. 1988. New 39Ar-4~ ages of Hercynian and Alpine thermotectonic events in Grande Kabylie (Algeria). Tectonophysics, 152, 53-69. --, TORRES-ROLDAN, R. L. & GARCiA-CASCO, A. 1994. Cooling and exhumation of the Western Betic Cordilleras, 4~ thermochronological constraints on a collapsed terrane. Tectonophysics, 238, 353-379. MUELLER, S. 1989. Deep-reaching geodynamic processes in the Alps. In: COWARD,M. R, PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 303-328. NIJHUIS, M. 1964. Plurifacial Alpine metamorphism in the SE Sierra de los Filabres S, of Lubrfn, SE Spain. Proefschrift, Amsterdam University. PLATT,J. P. & VISSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: a working hypothesis for the Alboran Sea and Gibraltar arc. Geology, 17, 540-543. ROBERTSON,A. H. E & GRASSO,M. 1995. Overview of the Late Tertiary-Recent tectonic and palaeoenvironmental development of the Mediterranean region. Terra Nova, 7, 114-127. SAADALLAH,A. & CABY,R. 1996. Alpine extensional detachment tectonics in the Grande Kabylie metamorphic core complex of the Maghrebides (northern Algeria). Tectonophysics, 267, 257-273. SELVAGGI,G. • CHIARABBA,C. 1995. Seismicity and Pwave velocity image of the southern Tyrrhenian subduction zone. Geophysical Journal International, 121, 818-826. SERRANO, E 1990. Presencia de Serravalliense marino en la cuenca de Nijar (Cordillera B6tica, Espafia). Geogaceta, 7, 95-97. SRIVASTAVA,S. P., ROEST, W. R., KOVACS,G., OAKLEY, G., Lt~VESQUE,S.,VERHOEF,J. & MACNAB,R. 1990. Motion of Iberia since the Late Jurassic: results from detailed aeromagnetic measurements in the Newfoundland Basin. Tectonophysics, 184, 229-260.
120
H.P. Z E C K
THOMPSON, A. B., SCHULMAN, K. & JEZEK, J. 1997. Extrusion tectonics and elevation of lower crustal metamorphic rocks in convergent orogens. Geology, 25, 491-494. TODESCO, M. & VIOLIOTTI,L. 1993. When did Sardinia rotate? Statistical evaluation of the paleomagnetic data. Annali di Geofisica, 36, 119-134. TORRES, J., BOIS, C. & BURRUS,J. 1993. Initiation and evolution of the Valencia Trough (western Mediterranean): constraints from deep seismic profiling and subsidence analysis. Tectonophysics, 228, 57-80. TUBiA, J. M. 1994. The Ronda peridotites (Los Reales nappe): an example of the relationship between lithosphere thickening by oblique tectonics and late extensional deformation within the Betic Cordillera (Spain). Tectonophysics, 238, 381-398. & IBARGUCHI, J. I. 1991. Eclogites of the Oj6n nappe: a record of subduction in the Alpuj~irride complex (Betic Cordilleras, S Spain). Journal of the Geological Society, London, 148, 801-804. VAN DER WAL, D. & VISSERS, R. L. M. 1993. Uplift and emplacement of upper mantle rocks in the western Mediterranean. Geology, 21, 1119-1122. VAUCHEZ, A. & NrCOLAS,A. 1991. Mountain building: strike-parallel motion and mantle anisotropy. Tectonophysics, 185, 183-201. VISSERS, R. L. M., PLATr, J. P. & VAN DER WAL, D. 1995. Late orogenic extension of the Betic Cordillera and the Alboran domain: a lithospheric view. Tectonics, 14, 786-803. WARBURTON, J. 1986. The ophiolite-bearing Schistes Lustr6s nappe in Alpine Corsica: a model for the emplacement of ophiolites that have suffered HP/LT metamorphism. Memoirs of the Geological Society of America, 164, 313-331. WORTEL, M. J. R. d(z SPAKMAN,W. 1992. Structure and -
-
dynamics of subducted lithosphere in the Mediterranean region. Verhandelingen der
Koninklijke Nederlandse Akademie van Wetenschappen, Amsterdam, 95, 325-347. ZECK, H. E 1996a. Betic-Rif orogeny: subduction of Mesozoic Tethys lithosphere under E-ward drifting Iberia, slab detachment shortly before 22 Ma, and subsequent uplift and extensional tectonics. Tectonophysics, 254, 1-16. 1996b. Alpine kinematic evolution in the W Mediterranean - mosaic of local extensional tectonic regimes following upon regional W-ward directed subduction of Betic-Ligurian lithosphere. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop, Cergy-Pontoise, Paris, 34. 1997. Mantle peridotites outlining the Gibraltar Arc - centrifugal extensional allochthons derived from the earlier Alpine, westward subducted nappe pile. Tectonophysics, 281, 195-207. , ALBAT, E, HANSEN, B. T., TORRES-ROLDAN, R. L., GARCIA-CASCO, A. & MART[N-ALGARRA, A. 1989. A 21 _+2 Ma age for the termination of the ductile Alpine deformation in the internal zone of the Betic Cordilleras, S Spain. Tectonophysics, 169, 215-220. , MONI[!, R,VILLA, I. & HANSEN,B.T. 1990. Mantle diapirism in the W-Mediterranean and high rates of regional uplift, denudation and cooling. In: -
-
Proceedings of the Symposium on Diapirism, 2. Tehran, Iran Geological Survey, 403-422. & -1992a. Very high rates of cooling and uplift in the Alpine belt of the Betic Cordilleras, southern Spain. Geology, 20, 79-82. -- & -1992b. Reply to K. de Jong (1992), Comment on "Very high rates of cooling... " Geology, 20, 1053-1054.
The petroleum systems of the Southeast Basin and Gulf of Lion (France) ALAIN MASCLE 1& ROLAND
VIALLY 2
IlFP School, 228 avenue Napol(on Bonaparte, 92 852 Rueil Malmaison cedex, France 2Institut Franfais du P(trole, 1-4 avenue de Bois Pr6au, 92 500 Rueil Malmaison, France
Abstract: The Southeast Basin is the thickest onshore French sedimentary basin where up to 10 km of Mesozoic-Cenozoic sediments can locally be found. This basin is surrounded to the east and to the south by two segments of the Alpine Thrust Belt, the Western Alps and the Pyrenees-Provence respectively, and to the west by recently uplifted elements of the Palaeozoic Basement (Massif Central). The development of the basin was related to several stages of subsidence between late Carboniferous and late Cretaceous times. Partial tectonic inversion took place during two Alpine compressive events in early Tertiary and late Tertiary times. They were separated by an intervening stretching event of Oligocene age which further south led to the opening of the western Mediterranean oceanic basin in Burdigalian times and, as a result, to the formation of the Gulf of Lion passive continental margin. In Neogene times the Palaeozoic basement of the Massif Central was uplifted to approximately 2000 m as the result of an ascending athenospheric plume. A large oil seep near the city of Gabian has been exploited since the beginning of the seventeenth century. Most of the exploration undertaken from 1945 (onshore) and 1965 (offshore) to the present time has, however, been disappointing as no significant oil or gas accumulations have been discovered, despite drilling of about 150 wells. A recent re-assessment of the potential remaining prospectivity of the Southeast Basin and Gulf of Lion basin has been undertaken by IFP. This study has benefited from scientific researches developed within the Integrated Basin Studies programme. This paper focuses on the source rocks and the petroleum system aspects. The review of all potential source rocks indicates that, from a qualitative and quantitative point of view, the best source rocks are located within three specific stratigraphic intervals (Stephanian-Autunian, late Lias and late EoceneOligocene). This analytical work allowed the reconstruction of the history of the different petroleum systems from the Jurassic to the present day. Because of the severe tectonic disturbances that these areas have experienced in Tertiary times, we can conclude that the best potential for economical discoveries are within the Gulf of Lion and some sub-basins of the Southeast Basin where subsidence has been active in Neogene times, or, in other words, where the processes of hydrocarbon generation, expulsion and migration can still be active today.
Following the detailed structural analysis undertaken at different crustal levels in the Southeast Basin and Gulf of Lion (Fig. 1) within the IBS project, a re-assessment of petroleum plays in these two areas is proposed. This should convince operators that opportunities still exist for significant discoveries in these two basins despite the previous disappointing exploration campaigns. This re-assessment is based upon the identification of all potential source rocks that have b e e n deposited at different stages of the basin's development. The reconstruction of their burial and uplift history allows modelling of the timing of hydrocarbon generation and expulsion. Correlations b e t w e e n p r o d u c e d hydrocarbons and the source rocks are based upon geochemical analyses. A more complete definition of petroleum systems requires, however, additional geological data (or h y p o t h e s e s ) including those related to tectonic and thermal
histories, distribution of potential reservoirs and migration paths (Perrodon 1992; Demaison & Huizinga 1991). These data are more difficult to discuss at a regional scale as they are mostly closely related to the local geology. They will thus be shown in different geological settings that illustrate a variety of unexplored potential plays still present in the Southeast Basin and the Gulf of Lion.
Regional Framework The Palaeozoic basement was consolidated in Carboniferous time during the last stages of Variscan thrusting and magmatism (Fig. 2). This basement crops out in the Massif Central (Ledru et al. 1994), the massif des Maures (Crevola & Pupin 1994) and in the axial zone of the Pyrenees (Majeste-Menjoulas & D e b a t 1994). Basin d e v e l o p m e n t started in Stephanian and/or
MASCLE,A. & VIALLY,R. 1999. The petroleum systems of the Southeast Basin and Gulf of Lion (France). In: DURAND,B., JOLIVET,L., HORVATH,F. & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156,121-140.
122
A. M A S C L E & R. V I A L L Y
PETROLEUM SYSTEMS, SE FRANCE
123
9
r
124
A. MASCLE & R. VIALLY
Permian times as a result of the post-collisional collapse of the Variscan thrust belt (Malavielle et al. 1990; Echtler & Malavieille 1990; Costa & Rey 1995). These basins were approximatively 2-4 km thick but of limited lateral extend. However, late Permian-early Triassic erosions probably removed large volumes of sediments thus explaining why their actual distribution and initial thicknesses are poorly known. The basin subsidence and morphology were controlled by the previous Variscan structures. The older ones (Stephanian) contain the inprint of the latest Variscan compressive events (Genna & Debriette
MA -. 0
STRATIGRAPHY
GEODYNAMIC EVENTS
TECTONIC PHASES
OROGENY
Alpine phase
II~
~
L,t*
9
.~
M., :w..,...M~,. .......... p~,...
23-- ~
PALAEO-STRESS SOURCE-ROCK FIELD DISTRIBUTION |
l
i : WEST-EUROPEAN I Olig....ne . RIFTING
~[
1994) as opposed to the distensive stress regime established in Permian times (Mascle 1989; Legrand et al. 1994). The late CarboniferousAutunian deposits are contemporaneous with HT/LP metamorphism, anatexis and granite intrusions (Chenevoy et al. 1995), and to metallogenesis processes well documented in the Massif Central (Bril et al. 1994). The Triassic-Jurassic history of the basin represents the development of a passive margin related to the Tethyan rifting (Trias to Malta) and to the subsequent opening of the 'LiguroPidmontais' ocean in Callovian times. The
Oligocene exten.
I, III
+
Mhh
Paleocene
PYRENEAN OROGENY
65 Late
Pyrenean phase
4
Bay of Bi,'~'ay riftingand ,,pening
Early
~ ~
/\
9
III
9
II
9
II
"~
~
135 --
Maim
[~ ~=~ ~i ~ 205-
~1
Dogger Li'*s
i
TETHYAN AND ATLANTIC RIFT AND OPENING -
+ E - Mid Jurassic rifting /
~
I1
,r
Late
Middle Early
245
295-- ~
9
[~
360
_
,.2
LATE VARISCAN
Autunian
COLLAPSE
Stephanian
Permian exten. I, III III
Saalian phase
~
Westphalian
Na,..aa,,
5": "':;';"+ :5 :.%.+<.'" :V':-4"5"4/.. ; .~~///., / ~ / , . . / . . , . " / " ." / "//.///
~
Dillantiall
""'"""
i ~9
Thu~h,~ia.,
_
Middle Early
/
VARISCAN
" "/'///
. / / / /
/
""
"
""
//
/
" / / / / / / " / / / / / , / / , , C , ' / / " / , , ' / , "
/ / /"/////-., ' . ,
, / ,, ' / /",' ,/ C / // / /,.//" /," "@
~(7////////
"
," / / ,
, , / / / / , // // / /
"/," /'/"/,,/,/,"/ / ' / / .,'."/,,,,- / / . / / / / / / ,
/ /
/ /--///(..
/
Fig. 2. Tectonic chart with the location of the main source rock intervals. I, II and III refer to the source rock types (lacustrine, marine, continental respectively).
~
PETROLEUM SYSTEMS, SE FRANCE Southeast Basin s.s. started developing at that time between two stable subsiding carbonate platform to the west (Ard~che) and east (Verdon-Durance platform; Baudrimont & Dubois 1977; Debrand-Passard 1984). The subsidence of the basin was controlled on both sides by steeply dipping basement faults (Roure et al. 1994; Mascle et al. 1996) resulting in a spectacular increase of thickness of the sedimentary infill, particularly in the Liassic interval (Razin et al. 1996). Metallogenesic processes were still very active at that time as documented by uranium deposits (Lancelot et al. 1995). It is important to note the presence of late Triassic salt deposits responsible of a major d6collement level within the Southeast Basin, but also of diapirism movement that were initiated during Bathonian time (Dardeau et al. 1990). The present Gulf of Lion was at that time the southern continuation of the Southeast Basin (although possibly subsiding less) and of its two lateral carbonate platforms (Vially & Tr6moli~res 1996). Subsidence was still strongly active in early Cretaceous time with pelagic and hemipelagic facies extending far to the west of the previous Jurassic 'Ard~che' carbonate platform. The geodynamic control of such a subsidence, i.e. thermic versus tectonic, is still poorly understood. This led in Barremian times to the largescale development of platform carbonates (Urgonian formation) in the southern part of the Southeast basin possibly extending south to the present Gulf of Lion (Arnaud-Vanneau & Arnaud 1990). A narrow trough with deeper open marine sedimentation (Vocontian Basin; Ferry & Rubino 1989) remains between the Vercors Urgonian platform to the north and the Provenqal Urgonian platform to the South. The Southeast Basin was thus the locus of almost exclusively marine clay-carbonate sedimentation from Rhaetian to Barremian times, with a total present (compacted) thickness locally in excess of 10 km (Fig. 3). This would probably rank the Southeast Basin as one of the thickest basins of this type in the world. The crustal processes at the origin of this subsidence remain completely unknown as the few available geophysical data do not show evidence of any upper crustal stretching in the range of what would be required for such a subsidence. In the present range of conceptual models developed within the IBS project, the best fit with observed data could possibly be an initial lithospheric rheology with a shallow level of necking and prominent athenospheric uplift below the rift basin, giving way to a large downward flexuring of the lithosphere (Cloetingh et al. 1996). This would also provide an explanation for the
125
absence of any large siliciclastic influxes within the basin during the syn and post-rift stages as the model predicts the absence of any rift-shoulder topography. Subsidence significantly slows down in Aptian to Senonian times, with major changes in the area surrounding the basin, as evidenced by erosion down to the Palaeozoic basement to feed the basin with siliciclastics. The southern half of the Southeast basin was uplifted in Albian-Cenomanian times along an E - W trending wide arch known as the 'Durancian Isthmus', possibly as a result of compressive stresses induced by the anticlockwise rotation of the Iberian Peninsula following the opening of the Bay of Biscay in Aptian-Campanian times. Late Cretaceous sediments were deposited to the north (northern part of the previous Southeast Basin) and to the south (present Gulf of Lion) with total thicknesses barely in excess of 1500 m (Debrand-Passard 1984). Sedimentation in Tertiary times was closely related to several major tectonic events. (a) From Senonian to late Eocene times, the Alpine Orogeny results in the formation of the east-west-trending Pyrenees-Provence thrust belts that was linked further east with what will later become the present Western Alps. Sedimentation prevailed at that time in the foredeep, with significant thicknesses of Eocene clastics (several hundred metres) still preserved in front of the eastern Pyrenees (Debrand-Passard 1984), or within the southern French Alps (Annot sandstones; Vially 1994). (b) A major rifting event in Oligocene-Aquitanian times led to the formation of narrow and elongated N E - S W - t r e n d i n g half grabens within the Southeast Basin, Gulf of Lion (Bois 1993; S6ranne et al. 1995) and Sardinia (still attached at that time to southern France, together with Corsica). The more spectacular onshore half graben is the Camargue Basin (Fig. 4) where about 4000 m of synrift Aquitanian-Oligocene sediments have been penetrated by the well Pierrefeu-1 (Valette & Benedicto 1995; Benedicto 1996). The Gulf of Lion passive continental margin was initiated in a similar way at that time, but in this case the rifting event led to oceanic accretion in the Burdigalian to form the present western Mediterranean basin. (c) Tectonism was still very active in Neogene times. Four events are more particularly noteworthy as they are at the origin of the present physiography of southern France: (1)
to the east the final emplacement of the western Alps (Vercors, Baronnies, Digne
126
A. MASCLE & R. VIALLY
9
~,-~
:al
~z
"~
"
o ~z
r,~
PETROLEUM SYSTEMS, SE FRANCE
127
9, -
o
~
0
~.n
0
~ ~'~
c..) .~ ~
~d2 ,.~
. ~ ,-"~
~
9,..-t ~.....~
"~ = < m9 .~
9~ N
~
N
,a ['.)
~-
o
..~,~
.~ 54-~
9
.~,
~
128
(2)
(3)
(4)
A. MASCLE & R. VIALLY and Castelanne arcs) was achieved, without any major foredeep development; to the west the uplift of the Massif Central was associated with active volcanism as a result of a possible local athenospheric plume (Wilson & Downes 1991; Granet et al. 1995); to the southwest the Pyrenean axial zone was uplifted (Calvet 1985), in response to post-collisional isostatic re-equilibrium (Vergds et al. 1995) or to the reactivation of deep crustal ramps related to the present intra-plate N-S compressive stress; to the south the western Mediterranean oceanic basin opened in Burdigalian time resulting in the anticlockwise rotation of the Corsica-Sardinia block, and the subsidence of the Gulf of Lion continental passive margin where more than 3000 m of sediments have been, and are still being, deposited (Vially & Tr6moli6res 1996; Chamot-Rooke et al. this volume).
This recent and present-day tectonic activity, is well reflected in today's onshore physiography of the whole area (Fig. 1), with lowlands (Rhone valley, Tertiary basins) close to either regional highs (Massif Central-Causses, Western Alps, eastern Pyrenees, Maures Massif) or local highs (Pic Saint-Loup, Ventoux-Lure). In addition the offshore physiography is largely controlled by Plio-Quaternary depositional processes, with very large sedimentary inputs from the Rhone River.
History of petroleum exploration: hydrocarbons fields and shows A large oil seep has been exploited since at least 1608 in the vicinity of what will become later the Gabian oil field. Oil was extracted from water flowing from a natural spring at the 'Font de Oly' (which means 'oil spring' in old French). The water also contained significant amounts of CO2 and H2S (Barrab6 & Schneegan 1935; Heritier 1994). Production at that time averaged 3.6 metric tons per year. Subsequent improvement of the production in the middle of the nineteenth century raised this production to 23 metric tons per year up to 1885 when the spring suddenly dried up. Exploration resumed in 1923 and a total production of 23 400 metric tons was achieved from 32 producing wells in 1945 when the field closed. Oil was hosted in early Triassic sandstones. Recent geochemical studies strongly suggest that this oil was expelled from Autunian lacustrine oil shales (Jacquart et al. 1993).
Most of the subsequent exploration undertaken from 1945 (onshore) and 1965 (offshore) to the present time has been disappointing as no significant oil or gas accumulations have ever been discovered and as many wells unexpectedly encountered CO2. Carbon dioxide does indeed commonly occur in the Southeast Basin to the great benefit of the sparkling water manufacturers. A large part of this gas derives from the mantle or lower crust while locally it appears to be mixed with biogenic CO2 according to 13C analysis (Arthaud et al. 1994). Very small oil productions have, however, been achieved at two localities, Saint Jean de Marudjols and Gallician (with total production of about 2000 and 5000 metric tons respectively), supporting the idea that some petroleum systems have locally been active. At St Jean de Maru6jol both source rocks and reservoirs are of Sannoisian age, whereas they are of late Oligocene age at Gallician. Gas (CH 4 and CO2) was also tested at St Jean de Marudjols and in a few other nearby wells. Hydrocarbon shows have been reported from a few tenth of the 150 wells drilled between 1948 and 1995. The more significant consisted of gas that was tested in 1959 at Grdoux-les-Bains from dolomites of Liassic age. No convincing hydrocarbon shows has been reported from the 11 offshore wells drilled from 1969 to 1985 in the Gulf of Lion. This lack of success is interpreted as related to the poor geological knowledge of the area in these early stages of exploration (1945--75). The Southeast Basin actually appeared to contain a much thicker Mesozoic-Tertiary sediment infilling than previously expected, and the potential reservoirs (of Triassic and Liassic age) were far below the drilling-depth capabilities of the rigs that were available in France at that time. When reached, on the edges of the basin for instance, these reservoirs were also found to be of unpredictable quality. The distribution and quality of source rocks were not fully considered and thermal modelling techniques were not available to propose timings of hydrocarbon generation until the middle of the 1980s. The poor quality of the seismic data at that time also resulted in inaccurate location of the first wells with respect to the complex geometry at depth of structural traps. An example of a site drilled in the early 1950s and revisited by modern seismic in the early 1980s is shown on Fig. 5. The quality of seismic acquisition and processing significantly improved from the early 1980s. At that time also, the first regional seismic profiles were shot. They correctly imaged for the first time the whole sedimentary package to depths locally in excess of 10 km (Deville et al.
PETROLEUM SYSTEMS, SE FRANCE SW
Sigoyer l
129
Synthetic structural and lithological log
~~
~
Petroleum systems
~
ff~~g ~ 'Prov. POS.ISpec. Ad.
Q. P.
:
._~'~
[] ....
"r
~
.................... B.- ............................ [ ]
i
o= ii ii
<.--~ ...,ii
',I iI
A.
,r _1 !
[] F.
0
~"'1 I/ i
:'--~ I
!i,/
[] q'-
', ~'~-
9~
i! /
i
II Fig. 5. Simplified section at the site of the Sigoyer 1 well (1959; location on Fig. 1). This section is derived from a seimic line shot in 1981 (IFP, unpublished report). The location of the hole was decided after conventional field mapping showing a large anticline of Dogger limestones within late Jurassic shales (see also Gidon et al. 1991). The well actually went through an unknown shallow d6collement or thrust and bottomed at 1927 m in upper Lias strata. Major structures below are now shown by modern seismics (including the first onshore 3D shot in France), but have still never been tested. A and B could be early Lias and/or Triassic horizons, C and D could be intervals ranging in age from early Lias to late Carboniferous. E: Palaeozoic basement. 1994). M o r e recently, d e e p seismic profiles through the Western Alps and Gulf of Lion finally allowed a better understanding of crustal processes at the start of subsidence of sedimentary basins and subsequent inversions, and to the related t h e r m a l events which partly control the maturation history of source rocks (Vially & Tr6moli~res 1996). Integration of m o d e r n basin modelling techniques, source-rock analyses and recent seismic imaging of the subsurface allow drastic improvem e n t of our k n o w l e d g e of the p e t r o l e u m systems in the Southeast Basin and Gulf of Lion. More specifically, the understanding of both deep geodynamic processes and shallower tectonic styles, as investigated in the course of the IBS project, allows a better assessment of remaining plays in these areas.
Source
rocks
Stephanian A few Stephanian troughs crop out at the edges of the Mesozoic basin (Fig. 1 and Table 1). The
m
.--h --ii
I .,,_ i ;
F.
i
[]
g
Fig. 6. Petroleum systems in the recently subsiding areas of the Southeast Basin and Gulf of Lions. Log: Q., Quaternary; P., Pliocene; M., Miocene; B., Burdigalian; A., Aquitanian; O., Oligocene; Pa., Palaeocene; Cr., Cretaceous; J., Jurassic; D., Dogger; L., Lias; T., Trias; Mz., Mesozoic; P., Permian; S., Stephanian; Pz., late Palaeozoic. Reservoirs: large and small squares for good quality and poor quality reservoirs respectively, and F. for fractured or karstified reservoirs. Petroleum Systems: Prov., proven; Pos., possible; Spec., speculative; Ad., additional; the arrows are the related migration pathways from the source rocks to the reservoirs; CBM: coal-bed methane. largest of t h e m is the Gard Basin (also know as the C6vennes or Al~s Basin) w h e r e coal measures are still being mined in open pits. Coal is generally c o n s i d e r e d to be an i m p o r t a n t source rock (for c o n v e n t i o n a l or coal-bed m e t h a n e exploration), but the generation potential of these rocks is highly speculative as: (1) the
130
A. MASCLE & R. VIALLY Synthetic structural
=,
lithological log
I~ 2
and
Petroleum
-~
~ .~ ~ nee
systems
~rov Pos. Spec Ad.
~.
.--J f"
f"
o o
o o ~ 1 7 6
"
"~',o~176176176176 i []
Q
!
"
Fig. 7. Petroleum systems in the Tertiary thrust belts surrounding the Southeast Basin (see Fig. 6 for the legend). extension of these basins below the Mesozoic strata is unknown (except near the present outcrops) and (2) the present maturation stage is variable, as a result of complex and possibly polyphased tectonic and thermal histories. This complexity is illustrated by the rank of coal in the Gard Basin as expressed by volatile matter isovalues (Courel et al. 1994) or rock-eval pyrolysis (Table 2). Some coals are far within the gas window whereas others are still within the oil window and have thus preserved some hydroc a r b o n - g e n e r a t i o n potential. The timing of hydrocarbon generation in the Gard Basin is
very poorly constrained, as we do not know the initial configuration of the basin and the value of heat flows prevailing at that time (possibly quite high). As a result, the present maturation could have occurred as early as late Palaeozoic times or, alternatively, much more recently, in Cretaceous times, following possible burial below 3000 m of Mesozoic sediments (Courel et aL 1994). R e m n a n t Stephanian coals with high ranks in the central part of the Massif Central, i.e. at a place where the Mesozoic cover is not believed to have ever exceeded a few hundred metres, strongly suggest that most of the maturations occurred in late Palaeozoic times. In which case these coals have potential as effective source rocks only in the central part of the Southeast Basin or below the Tertiary thrust belts, i.e. where their Mesozoic-Tertiary burial has been deep enough to provide the high temperatures required for further hydrocarbon generation. A l o n g the edge of the basin, however, some potential could remain for coalbed methane exploration as the gas content of these coals is very high (collieries of the Gard basin have a bad reputation for firedamp explosions). Autunian
A few Permian basins crop out around the Southeast Basin and Gulf of Lion (Table 3; Chateauneuf & Farjanel 1989; Mascle 1989). They are only partly related to previous Stephanian troughs, suggesting that their d e v e l o p m e n t followed different structural trends. Their outcrops are less tectonically disturbed than those of Stephanian strata. Potential source rocks are restricted to the Autunian interval. The younger Permian sediments were deposited in a hot, dry and oxidizing continental environment, which produced the typical red colour of these strata and which was not favorable to the preservation of the original organic matter. Some coal measures are still present in
Table 1. Main Stephanian basins around the Southeast Basin and Gulf o f Lion
Name Segure Durban Graissessac Neffi6s Decazeville Le Vigan Gard Prades Reyran
Location Mouthoumet Massif Mouthoumet Massif Montagne Noire Montagne Noire Massif Central C6vennes C6vennes C6vennes Maures Massif
References Caz6tien (1982) Caz6tien (1982) Becq-Giraudon & Rodriguez (1986) Desrousseaux (1938) Vetter (1968) Desrousseaux (1938) Wang (1991); Genna & Debriette (1994) Desrousseaux (1938) Basso (1985)
PETROLEUM SYSTEMS, SE FRANCE
131
Table 2. Rock eval pyrolys&: contrasting results from Stephanian coals, Gard Basin TOC
S1
$2
Tmax
HI
OI
80.7
5.9
177.0
444
219
6
76.7
4.1
130.7
464
170
3
68.4
0.1
35.0
511
51
2
66.6
0.4
12.5
564
19
15
Location (S) X: 742.85 Y: 3232.1 (S) X: 741.9 Y: 3228.5 (S) X 739.1 Y: 3213.2 (O) X: 738.95 Y: 3218.6
TOC, total organic carbon; S1, free hydrocarbons (mg HC/g rock); $2, residual petroleum potential (mg HC/g rock); HI, hydrocarbon index; OI, oil index (Espitali6 et al. 1985-86). the Autunian interval, but the more important potential source rocks are oil shales or bituminous limestones with a Type I (lacustrine, algal), or a type III (continental) organic matter, or a mixture of both ( C h a t e a u n e u f & BecqGiraudon 1990). The maturity values of these exposed source rocks are slightly lower than the values measured of Stephanian rocks. They typically fall within the oil window, although some of them occasionally reach the gas window (Table 4). The maturation history of these source rocks is poorly constrained. As for the Stephanian coals, a late stage (late Cretaceous-Tertiary) of significant h y d r o c a r b o n generation in the central part of the Southeast Basin or below the Tertiary thrust belts is considered to be a reasonable assumption. Geochemical data suggests that the oil produced at the Gabian field originated from such Autunian shales through either primary or secondary migration (Jacquart et aL 1993). Triassic-Lias
The Triassic interval is generally characterized by continental deposits. These contain a limited amount of potential source rocks such as very thin and discontinuous layers of coal or bitu-
ruinous shales. Marine sedimentation started in Rhaetian time and prevailed during the Jurassic, resulting in the f o r m a t i o n of widespread organic-rich (type II, marine) intervals. The richest intervals are within the Rhaetian, late Liassic and Malm sequences (except in the V e r d o n - D u r a n c e platform d o m i n a t e d by shallow-marine carbonate sedimentation). The best source rock potential is undoubtedly found in the lower Toarcian interval consisting of several metres thick black shales (known as the 'Schistes Cartons'), with initial TOC between 1 and 10% (Table 5). These shales were deposited during a major flooding event (Razin et aL 1996). They are well known on the western platform (south of Albs) of the Southeast Basin and in the Causses. They are not however observed further east within the several hundred metres thick upper Liassic marly interval of the Southeast Basin. The high thermal evolution of these rocks prevents any detailed analysis of their initial potential of generating h y d r o c a r b o n s to be carried on (but an average initial T O C between 0.5 and 2 is generally considered to be a reasonable assumption). It should be noted that the present high level of maturation of the Liassic outcrop and well samples from the western edge of the basin and from the nearby C6vennes
Table 3. Main Permian basins around the Southeast Basin and Gulf o f Lion
Name Gabian Lodbve Saint Affrique Rodez Largenti~res Var Barrot Argentera Balagne Mulargia Lake
Location
References
Montagne Noire Montagne Noire Causses Causses C6vennes Provence Western Alps Western Alps Corsica Sardinia
Barrab6 & Schneegans (1935) Odin (1986) Rolando (1988) Legrand (1990) Bourges (1987) Rouire & Chiron (1980) Toutin (1980) Vinchon (1984) Romain (1978) Vellutini (1977) Barca et al. (1995)
132
A. MASCLE & R. VIALLY
Table 4. Rock eval analysis of selected samples from four Autunian basins Name
TOC
$2
Tmax
HI
Rodez Lod6ve Gabian Largentibre
6.03 2.43 2.33 1.21
35.65 11.78 2.78 0.3
439 440 455 499
591 502 119 22
window or even sometime within the gas window (Aurel area). Mid-Cretaceous
The increasing order of maturation does not reflect an overall trend between the basins but has been arbitrarily chosen to show different classes of rock eval results. See Table 2 for abbreviations.
Table 5. Rock eval analysis of selected samples from the Toarcian section TOC
$2
Tmax
HI
0.54 0.84 0.69 0.62 0.60 1.99" 3.84*
0.27 0.39 0.30 0.19 0.19 0.94 2.25
466 466 468 468 468 464 466
50 46 43 30 31 47 68
(top)
(bottom)
*: 'Schistes Carton' of the Pic Saint Loup area
(Combe de Morties) See Table 2 for abbreviations.
platform required that 1500-2000 m of Jurassic-Cretaceous sediments have been eroded in Tertiary times following the uplift of the Massif Central. A similar conclusion has been obtained from detailed geochemical (organic and mineral) analysis performed on materials from the Balazuc scientific borehole at the northwestern edge of the basin (Steinberg et al. 1991)). This further implies that the main period of hydrocarbon generation from the Liassic rocks was the Cretaceous i.e. prior to the main tectonic events of the Tertiary.
Malta A n y geologist travelling through the Southeast Basin may have been impressed by the 'bad lands'-type outcrops of dark silty claystones from the late Bathonian-early Oxfordian interval (Artru 1972). These rocks known as the 'Terre Noires' are widespread within the Southeast Basin where their thickness is locally in excess of 2000 m. Their petroleum potential is, however, low with an average initial T O C not exceeding 1%. Most of the samples collected on the field (western Alps) are within the oil
Pelagic facies in mid-Cretaceous times (Barr e m i a n - C e n o m a n i a n ) were restricted to the northern part of the Southeast Basin ('Vocontian' Basin) and to a 'Southern Provence' Basin extending from the eastern Pyrenees to the Marseille area, i.e. north and south respectively of the 'Durancian Isthmus'. This interval includes in the Vocontian basin and southern Provence several horizons a few metres thick displaying a black shale lithology with T O C between 4 and 8% (Brdhdret 1994; Oberti 1992). The samples collected on the field (Baronnies, Digne, southern Provence) are immature or at the beginning of the oil window. They are not considered as of economic interest in our area. Time-equivalent continental lignites of Cenomanian age have been mined in the past at the vicinity of the Alhs Tertiary basin (Desrousseaux 1938). They are very thin and of local interest only. Several hundred metres thick late A p t i a n - A l b i a n shallow-marine black shales have been deposited in the eastern Pyrenees in an E-W-elongated trough, possibly related to the sinistral displacement of the Iberian plate at that time (Arango 1989). They probably had an average initial T O C of about 1%, with higher values near the base as expressed in the present residual T O C (Table 6). They could have generated some hydrocarbons before the Pyrenean inversion with possible migrations and trappings in the northern early Tertiary foredeep. L a t e Cretaceous Shallow marine condition prevailed in late Cretaceous time, with as a result the deposition of laterally discontinuous coal seams and bituminous shales. Most of them are immature and Table 6. Rock eval analysis of selected samples from the upper Aptian-Albian section of Quillan (eastern Pyrenees) TOC
$2
Tma•
HI
0.38 0.74 0.90 0.54 0.76 0.97 1.22
0.27 0.35 0.41 0.26 0.18 0.01 0.08
441 443 446 448 493 519 532
71 47 45 48 23 1 6
See Table 2 for abbreviations.
(top)
(bottom)
PETROLEUM SYSTEMS, SE FRANCE Table 7. Rock eval analysis of two Senonian potential rich source rocks Location
TOC
$2
Tmax
HI
Vagnas (northern edge of Al6s Basin) Gardanne
9.14
66.33
438
726
439
354
75.9
268.8
See Table 2 for abbreviations.
only of local interest as their thickness and lateral extent are limited. The rock-eval analysis of two samples are shown on Table 7. The first one (Vagnas) is an example of a very rich bituminous shale of Santonian age mined in the past for the production of synthetic oil. The second one is a sample of coal of late Campanian age, still being mined at depths in excess of 1000 m in the area of Aix-en-Provence ( G a r d a n n e coalpit; Arcamone et al. 1980). Eocene-Oligocene
Most of the Palaeogene basins contain potential source rocks, either lignites or bituminous shales and limestones (Table 8). All these basins are of limited extent and thus have a different lithological content. The five main onshore basins are from west to east. 9 The Minervois basin in front of the eastern Pyrenees and Corbi6res thrust belt contains lignites of L u t e t i a n age that have been occasionally mined in the past (Desrousseaux 1938). They are split into several beds with a cumulative thickness not exceeding a few metres. 9 T h e N a r b o n n e - H 6 r a u l t basin is poorly exposed on the field but a few lignites (two seams at Armissan south of Narbonne) and bituminous limestones and shales of late
133
Oligocene age have been described locally. These shales are frequently associated with sulphur accumulations (Malvesi mine near Narbonne for instance; Estival & Schneegans 1935; R o y 1946). Bituminous coatings found on Oligocene normal fault scarps could have originated from these rocks, but their present severe meteoritic weathering does not allow any detailed geochemical analysis. 9 The A16s Basin includes the abandoned St Jean De Maru6jols oil field. Geochemichal correlations have shown that the hydrocarbons have been sourced from just mature Sannoisian (latest Eocene) bituminous limestones (Mascle et al. 1996). These rocks are approximatively 10-20 m thick and contain heavy oil coatings associated with a fracture network. They are still mined at depths of about 400 m to supply pavement brick constituants. Other potential source rocks are lignites of slightly younger Sannoisian age that were mined in the past along the eastern edge of the basin. Individual coal seams were between 0.5 and 2 m thick but only two or three, probably discontinuous, levels were present at a same locality (Desrousseaux 1938). 9 The Camargue Basin (Fig. 3) includes the abandoned Gallician oil field. A 3180 m thick Oligocene section has been drilled at the Pierrefeu-1 well, with bituminous limestones near the top and coal beds near the bottom. The oil produced or tested in different wells show differences in composition possibly linked to these two different types of kerogen: sulphurrich crudes from Gallician and Ste C6cile wells have geochemical characteristics suggesting an origin from the confined lagunal-lacustrine bituminous limestones, while paraffinic crudes from Pierrefeu show characteristics suggesting an origin from the continental kerogen within the coals (Blanc & Connan 1993). 9 The Manosque Basin contains relatively thick lignites and bituminous shales/limestones of
Table 8. Rock eval analysis of selected Eocene and Oligocene potential source rocks Location
Nature
TOC
Minervois (La Caunette) Al6s (Barjac) A16s (Mons) Camargue (well) Camargue (well) Manosque (Piferat) Manosque (Jas)
Lignite Lignite Bituminous limestone Bituminous limestone Coal Lignite Bituminous limestone
57.3 55.2 7.22 2.7 14 51.7 5.5
See Table 2 for abbreviations.
$2
Tmax
HI
120.5 107.7 56.8 6.2 3.3 28.6 44.4
409 391 427 446 546 431 420
211 197 786 277 23 55 804
134
A. MASCLE & R. VIALLY
late E o c e n e - O l i g o c e n e age (Desrousseaux 1938), with organic matter of continental and lacustrine origin respectively (Biondi et al. 1992). Both of them were mined in the past. They show steep bedding dips as a result of the partial inversion of the basin in Neogene times and related salt diapirism (Fig. 2). All the samples collected on the field are immature, but, as in the A16s Basin, bitumen coatings can be observed in fractures or cavities.
related to previous tectonic event could locally have been preserved. The petroleum systems that are, or could be present, in the Southeast Basin and Gulf of Lion are thus related to basins or sub-basins that have been continuously subsiding, buried below or involved within thrust belts, uplifted or have remain stable during this period of time.
With the exception of the Camargue Basin, all the source-rock samples from the onshore Palaeogene basins are immature. These onshore basins have actually been slightly inverted, or have followed the uplift of the Massif Central in Neogene times, with, as a result, the erosion of the youngest strata and a limited burial of the potential source rock intervals. They are of limited interest unless tectonic overthrusting allowed additional burial, as it has been interpreted in front of the Corbi6res thrust belt (Mascle et al. 1996). Furthermore the Camargue Basin has followed the evolution of the Gulf of Lion, with continuous subsidence in Neogene times. As a result, the different source-rocks intervals are within the oil or gas windows. Oligocene basins similar to the Camargue Basin are present in the Gulf of Lion and are concealed below 3 or 4 km of Neogene sediments (Fig. 2). Their lithological content is, however, unknown as the 11 wells already drilled in this area were spudded on regional highs at the edge of the Oligocene troughs (Lamiraux & Mascle 1995).
Such basins are present in the Gulf of Lion and in the related Camargue and Roussillon onshore basins to the northeast and southwest respectively. Continuous subsidence in Neogene times led to the deposition of 1000 to 4000 m of sediments. A spectacular drop in the sea level in Messinian time was followed by the development of a spectacular canyons network on the upper part of the margin, and to salt deposition in the deep western M e d i t e r r a n e a n Basin. Proven petroleum systems relate to the onshore Gallician oil field where Oligocene sandstones and limestones have been charged with hydrocarbons sourced from Oligocene rocks. Similar plays could hopefully be present in the Guff of Lion as a more important burial of these rocks in Neogene times could have led to the generation of larger volumes of hydrocarbons (Vially & Tr6moli6res 1996). These hydrocarbons could also possibly have been trapped in Neogene sandstones, although the distribution of these rocks is poorly constrained. It should be emphasized that the presence of such potential source rocks is only an hypothesis as none of the eleven drilled wells have penetrated the Oligocene basins. These wells were located on regional structural highs at the edge of these basins to explore possible karstified limestones of Mesozoic age, as similar plays led to significant discoveries in the near Gulf of Valence in Spain (Clavel & Berfistegui 1991). Much more speculative plays could include gas generated from deep Mesozoic or late Palaeozoic rocks, or biogenic gas within the recent fine-grained turbidites of the Rhone delta.
Petroleum systems If the distribution of all potential source rocks is obviously the first step in the evaluation of a tectonically complex and little explored area, the nature and distribution of the youngest tectonic events is the next parameter to take into consideration. These events are likely to control most of the structural traps, and to either stop or induce the generation of hydrocarbons according to the uplift and burial history of the sourcerock intervals. They will also control the distribution of migration pathways (either secondary from the source rocks, or tertiary from previous accumulations), but this aspect is still too poorly documented to allow further discussion. The three Tertiary tectonic events that have been recorded in the Southeast Basin and Gulf of Lion led to rapid and contrasted evolution of these two areas. As far as petroleum exploration is concerned, however, we will consider that the more important is the last one in Neogene times, although earlier accumulations
Subsiding areas (Fig. 4)
Thrust belts (Fig. 5) Neogene thrust belts are located in the western Alps and include, within the limits of Fig. 1, the Diois-Baronnies, Digne and Castelanne Arcs. They extend further north within the Vercors and Chartreuse thrust belts where 2D 'Thrustpack' modelling has shown the potential interest of Liassic source rocks (Mascle et al. 1996). The western Alps can be considered as unexplored as very little seismic has been shot and as only
PETROLEUM SYSTEMS, SE FRANCE three wells (Aurel 1, Sigoyer 1 and 2) have been drilled in these areas. These wells were furthermore quite unconclusive as they did not reach their initial early Mesozoic objectives due to the unexpected thickness of the Jurassic strata (Fig. 5). No seaps have ever been reported from these thrust belts (but oil and gas seaps are known in the Vercor-Chartreuse areas). Petroleum systems are thus quite speculative. The more realistic one could correspond to hydrocarbons generated from Liassic source rocks during the Neogene tectonic burial, and trapped within late Triassic or Jurassic fractured carbonates on top deep anticlines or duplexes. The late Triassic salt, which usually hosts the main d6collement, should probably prevent any migration to older reservoirs. Still more speculative plays could be gas generated from late Palaeozoic source rocks (if any) and trapped within Permian, early Triassic (or even late Triassic-Jurassic limestones as the d6collement may locally deepen to deep stratigraphic level as evidenced by the 'D6me du Barrot' Permian outcrops within the Digne-Castellane arc). A late generation of gas could actually have been induced by the Neogene tectonic burial, but, alternatively, early (pre-Neogene) generations of hydrocarbons could have been trapped and preserved in reservoirs below the late Triassic salt acting as a quite efficient seal in the outer part of the belt during alpine deformations.
Uplifted areas (Fig. 6) These include all the western edge of the Southeast Basin and the western onshore part of the Gulf of Lion. These uplifts followed the regional uplift of the Massif Central. This area is superimposed to the initial Triassic-Jurassic margin of the Southeast Basin. Many wells have been drilled without significant results, with the exception of the two tiny Gabian and St Jean de Maru6jol oil fields. These fields, however, indicate that two petroleum systems have been operating, the first one between the Permian source rocks and the early Triassic reservoir, the second one entirely within the late EoceneOligocene interval. Gas tested in several wells (including both CH4 and CO2 with a quite different ratio from well to well, or even inside a single well) give indication of additional possible petroleum systems. Gas could have been generated from known deep late Palaeozoic or Liassic source rocks and could have migrated to any Mesozoic or Tertiary reservoirs. More speculative is the presence of deep late Palaeozoic source rocks within the hanging-wall side of the main Mesozoic margin fault. The significant
135
erosion and the weak compressive tectonic deformations that these areas have experienced in Neogene times is likely to have prevented any significant hydrocarbon generation at that time, and to have, furthermore, destroyed any previous accumulation. Coal-bed methane could be another source of interest on the footwall of the Mesozoic fault as coal seams could be reached less than 2000 m below the Mesozoic cover. Two wells drilled a few years ago to test such a play have been inconclusive.
Stable area (Fig. 7) Although most of the Southeast Basin was submitted to compressive stresses in Neogene times, some areas remained little affected by tectonic deformation. This is particularly the case of the Provence area, between the southern deformation front of the Baronnies to the north, and the northern front of the Provence thrust belt to the south. Existing Eocene or Oligocene faults were reactivated in the Neogene (Luberon-Manosque, Alpilles, Ventoux-Lure, for instance), resulting in deformations restricted to their vicinity. Little or no generation of hydrocarbons can be expected at that time. It is however possible that the gentle tectonic deformation did not affect earlier hydrocarbon accumulations. Less than ten wells have been drilled in this area, and the occurrence of gas in some of them, especially in the Gr6ouxles-Bains wells, suggests that petroleum systems were active in the past. The most likely hypothesis is a Cretaceous and/or early Tertiary generation of gas from Liassic source rocks in the deep basin combined with trapping in fractured Triassic to middle Jurassic limestones and dolomites in the less subsiding eastern and southern platforms. More speculative plays could be associated with the generation of gas from late Palaeozoic rocks, but the distribution and infill nature of these basins is poorly constrained. Santonian coal seams could be of local interest for coal-bed methane exploration if they are shallower than 1500 m. Their low level of maturation could have possibly prevented the generation-adsorption of large volumes of gas.
Conclusions The position of the Southeast Basin and Gulf of Lion relative to the Alpine thrust belt controlled their present geological complexity and contrasted physiography. The superimposition of Tertiary basins that developed in a back-arc setting in the Oligocene and ultimately led to oceanic accretion in Burdigalian times, in
136
A. MASCLE & R. VIALLY
addition to the poorly understood infracrustal thermal events at the start of the Massif Central uplift and Neogene volcanism, are supplementary complications to a classic fold and thrust belt. These complications are increased by the variety of tectonic styles in both compressional and extensional regimes, with both thin and thick skinned deformation (Roure et al. 1994; Roure & Colletta 1996). The large volume of late Triassic salt acts as a major d6collement surface, but is also at the origin of diapirism in Mesozoic-Tertiary times, the interference of which with the propagation of thrusts is still poorly understood (Dardeau et al. 1990). The propagation of the Tertiary thrust belt has also been strongly controlled by Mesozoic structures, and more particularly by the development of an extremely thick (10 km) Triassic-Early Cretaceous shale-carbonate basin, following crustal processes possibly controlled in turn by some late Palaeozoic-Variscan structures. R e c e n t studies conducted within the IBS project (this volume), and the results from the scientific drilling program conducted by the B R G M in the Balazuc area (Bonijoly 1996) have brought new insights into the tectonic and sedimentary
Synthetic structural and lithological log
~ Petroleum ~_~ ~ systems ~ ~ ~ Prov. Pos. spec Ad.
Synthetic structural aria lithological log
~ ~ ~> ~c ~
Petroleum systems ,roy F .
Ad.
n-
I
o
.
.
.
.
9~176176176176
r
r
r
M r
"'* L'.*.'. X
r
*§247247
r
r
+
r
. . . .
9
*,*,',"
+
'.'.,.'.'.'. ,
9
:
9
L.
I
!
§
9 Pz
Fig. 9. Petroleum systems in the parts of the Southast Basins that have remained stable, or have been only moderatly deformed in Tertiary times (see Fig. 6 for the legend).
er
E
~
~.
"6 .=_ ~f
-
Mz
=.
+--:,t +-
7,,
D? p.-s.
Fig. 8. Petroleum systems in the parts of the Southeast Basin that have been uplifted in Tertiary times (see Fig. 6 for the legend).
processes that were active in these areas. What is the impact of this complex geological history on hydrocarbon exploration potential? The first consequence is obviously that this whole area is now divided up in many sub-basins, which were formed during distinct phases of tectonic and subsidence between the late Palaeozoic and the Neogene. The identification of the source-rock interval specific to each of the sub-basins is of critical importance. It allows the unravelling of different timings of hydrocarbon generation and expulsion. Short-lived petroleum systems can thus have been active locally and at different periods of time, but not at a regional scale (Fig. 8). Pre-Tertiary petroleum systems are poorly understood due to the lack of thermal and burial constraints, and to the strong overprint of the Tertiary tectonic events. The Tertiary to Recent thermal and burial history is better constrained thus allowing assessment of the most recent petroleum systems that are still active today. In this respect the Gulf of Lion can certainly be considered as the more attractive area for further exploration, the most critical uncertainty being the presence and volumes of source rocks
P E T R O L E U M SYSTEMS, SE F R A N C E <---
Late Generation
I TRUSTING TBUSllNGI I SUBSIDENCE I FOLDING
SUBSIDENCE
Subsiding area . . . .
137
I TRUSTING ITRUSllNOI FOLDNG I SUBS DENCE FOLDING EROSION [< "IEROSION I I I I I I I
I
Thrust belt and uplift area Stable area
I ' I'
I
i I
t I l
I I ~.1 I --'~'41 i'~ " i
I I--'2~-"I
!
'- -- ~ ~
I LIASSIC SOURCE-ROCK
]
,~-
E v
F--r----
_1 IX
I
,
I I
I I
II;". el/!X
I
I
i
i----
--
rl
T--- - -
I
J~ll
~l--l--
t
',
Early Generation Palaeogene I i
150
"1'--
I
i
m
200
I
:'""'t,, 1
A
10o
I
i
i
Neogene i
5O PYRENEAN OROGENY
ALPINE OROGENY
PYRENEAN OROGENY
ALPINE OROGENY
Fig. 10. Maturation history of Liassic and Palaeogene source rocks from different parts of the Southeast Basin and Gulf of Lion.
within the (undrilled) Oligocene sequences (Vially & TrOmoliOres 1996). Mention should also be made of some unexplored segments of the Alpine thrust belt (Chartreuse Massif for instance; Mascle et al. 1996) where Mesozoic source rocks have recently been matured, as a result of late Miocene-Pliocene tectonic burials. The search for earlier (pre-Tertiary) accumulations is considered to be more risky. This could be attempted where the Tertiary tectonic events have been less destructive, such as the Provence area, or below the main dOcollement, above which only thin-skinned deformation has occurred, possibly in basins of unknown age below the CorbiOres thrust belt (Mascle et aL 1996). The Tertiary uplift of the Western margin of the Southeast Basin, together with the Massif Central, have prevented any generation of hydrocarbons in Tertiary times (with the exception of small rift basins such as the Albs Basin where Oligocene source rocks have locally reached the oil window threshold). This uplift and the synchronous compressive-extensive tectonic events probably led to the loss of any previous accumulation.
no. PL 920287). We thank J. Borgomano from Norske Shell for his extensive annotations and for the resulting improvement of our manuscript.
This work has been partly supported by the European Community DG XII (contract Joule II-CEC Project
BARCA, S., CARMIGNANI, L., ELTRUDIS, A.
References ARANGO, J. C. 1988. Sddimentologie et stratigraphie sdquentielle de l'Albian dans le bassin de Quillan et le synclinorium d'Axat (zone nord-pyrenednne, Aude). Document du BRGM, 169. BRGM, Orl6ans. ARCAMONE, J., BIDEAUD, M. ET AL. 1980. Le gisement de charbon du bassin de l'Arc, Provence Occidentale. M6moire du BRGM, 122. ARNAUD-VANNEAU,A. & ARNAUD,D. 1990. Hauterivien to lower Aptian carbonate shelf sedimentation and sequence stratigraphy in the Jura and northern Subalpine chains (southeastern France and Swiss Jura). In: TUCKER, M. E. ET AL. (eds) Special Publications o f the International Association o f Sedimentologists, 9. Blackwell Scientific Publications, Oxford, 203-233. ARTHAUD, E, DAZY, J. & GRILLOT, J. P. 1994. Distribution of deep carbon dioxide in relation to the structure and tectonic evolution of south-east France. Geodinamica Acta, 7, 86-102. ARTRU, P. 1972. Les Terres Noires du bassin rhodanien. Th6se de l'Universit6 CI. Bernard, Lyon.
&
FRANCESCHELLI,M. 1995. Origin and evolution of
138
A. M A S C L E & R. VIALLY
the Permian-Carboniferous basin of Mulargia lake (South-Central Sardinia, Italy) related to the late Hercynian extensional tectonics. Comptes Rendus de l'Academie des Sciences Paris, 321, s6rie IIa, 171-178. BARRABE, M. L. & SCHNEEGANS,M. D. 1935. Rapport d'ensemble sur le gisement p6trolifbre de Gabian (H6rault). Annales de l'Office National des Combustibles Liquides, 10, 595~563, 819-887. BASSO,A. M. 1985. Le Carbonifdre de Basse-Provence. Evolution s~dimentaire et structurale. Thbse de l'Universit6 de Provence, Aix-Marseille. BAUDRIMONT, A. E & DUBOIS, P. 1977. Un bassin m6sog6en du domaine p6re-alpin: le sud-est de la France. Bulletin des Centres Recherche et Exploration Elf-Aquitaine, 1, 261-308. BECQ-GIRAUDON, J. E & RODRIGUEZ, G. 1986. Maturation de la mati6re organique dans le bassin de Graissessac (H6rault): liaison entre structure et m6tamorphisme des charbons; signification r6gionale. G~ologie de la France, 3, 339-344. BENEDICTO,A. 1996. ModOles tectono-s~dimentaires de bassins en extension et style structural de la marge passive du Golfe du Lion (partie Nord), sud-est de la France. Thbse de doctorat europ6en pr6sent6e l'universit6 Montpellier II Science et Techniques du Languedoc. BIONTI, E, LERAT, O. & PHILLIPS, J. 1992. SynthOse structurale, sOdimentologique et gOochimique du bassin docOne-oligocOne de ManosqueForcalquier (Alpes de Haute-Provence). Rapport IFP inddit, 40 151. BLANC, P. & CONNAN,J. 1993. Crude oil in reservoirs: the factors influencing their composition. In: BORDENAVE, M. L. (ed.) Applied Petroleum Geochemistry. Editions Technip, Paris, 149-174. BoIs, CH. 1993. Initiation and evolution of the OligoMiocene rift basins of southwestern Europe: contribution of deep seismic profiling. Tectonophysics, 226, 227-252. BONIJOLu D. (ed.) 1996. Mesozoic evolution of the western margin of the French Southeast Basin (G. P. F. Programme). Marine and Petroleum Geology, 13, 605-735. BOURGES, R. 1987. SOdimentation alluviale et tectonique extensive dans le Permien du ddtroit de Rodez (Aveyron, France). Th6se de l'Universit6 de Toulouse. BRIL, H., MARIGNAC, M., CATHELINEAU,M., TOLLON, E, CUNEY,M. & BOIRON, M. C. 1994. Metallogenesis of the French Massif Central: time-space relationships between ore deposition and tectono-magmatic events. In: KEPPIE, J. D. (ed.) Pre-Mesozoic Geology in France. SpringerVerlag, 379-402. CALVET,M. 1985. N6otectonique et mise en place des reliefs dans l'Est des Pyren6es; l'exemple du horst des Alb6res. Revue de GOologie Dynamique et de GOographie Physique, 26, 119-130. CAZI~TIEN,R. 1982. Le StOphanien des CorbiOres orientales. Th6se de l'Universit6 de Toulouse. CHAMOT-ROOKE, M., GAULIER,J.-M. & JESTIN,E 1999. Constants on Moho depth and crustal thickness in the Liguro-Provenqal Basin from a 3D gravity invasion: geodynamic implication. This volume.
CHATEAUNEUF,J. J. ~: BECQ-GIRAUDON,J. E 1990. Les combustibles solides du Permien en France. In: Potentiel dconomique des bassins permiens franqais. Chronique de la recherche mini6re. BRGM, 499, 87-102. -& FARJANEL, G. 1989. SynthOse g~ologique des bassins permiens franfais. M6moire du B R G M 128. CHENEVOY, M., GAY. M. & DUTHOU. J.-L. 1995. Age Rb-Sr autunien des migmatites ~ cordi6rite nordc6venoles (Massif Central franqais). Comptes Rendus de l'Academi des Sciences, Paris, 320, sdrie IIa, 491-497. CLAVEL, E. & BERASTEGUI, X. 1991. Petroleum geology of the Golfe of Valencia. In: SPENCER,m. M. (ed.) Generation, Accumulation and Production of Europe's hydrocarbons. Oxford Science Publications, 355-368. CLOETINGH. S., VANWEES, P. A., VAN DER BECK, P. A. & SPADINI,G. 1995. Role of pre-rift theology in kinematics of extensional basin formation: constraints from thermomechanical models of Mediterranean and intracratonic basins. Marine and Petroleum Geology, 12, 793-808. COSTA, S. & RE',', E 1995. Lower crustal rejuvenation and growth during post-thickeneing collapse: insights from a crustal cross-section through a Variscan metamorphic core complex. Geology, 3, 905-908. COUREL. L., LIu. X. B. & WANG, H. 1994. Polyphase coalification in French intermontane coal basins of the late Carboniferous. In: MASCLE, A. (ed.) Hydrocarbon and Petroleum Geology of France. Springer-Verlag, 399--407. CREVOLA,G. & PUPIN,J.-R 1994. Crystalline Provence: structure and Variscan evolution. In: Keppie, J. D. (ed.) Pre-Mesozoic Geology in France. SpringerVerlag, 426-441. DARDEAU, G., FORTWENGLER, D., DE GRACIANSKY,P. C., JACQUIN, T., MARCHAND, D. & MARTINOD, J. 1990. Halocin6se et jeux de blocs dans les Baronnies: diapirs de Propiac, Montaulieu, Condorcet (d6partement de la Dr6me, France). Bulletin du Centres de Recherches Exploration Production Elf-Aquitaine, 14, 111-159. DEBRAND-PASSARD,S. (ed.) 1984. SynthOse gOologique du Sud-Est de la France. M6moires du B R G M 125 and 126. DEMAISON, G. & HUIZINGA, B. J. 1991. Genetic classification of Petroleum systems. American Association of Petroleum Geologists, Bulletin, 75, 1626-1643. DESROUSSEAUX,J. 1938. Bassins houillers et lignifOres de France. M6moire annexe de la statistique de I'industrie min6rale. Paris, Imprimerie Nationale. DEVILLE, E., MASCLE,A., LAMIRAUX,C. & LE BRAS,A. 1994. Tectonic styles, reevaluation of plays in southeastern France. Oil and Gas Journal, Oct. 31, 53-58. ECHTLER,M. & MALAVIEILLE,J. 1990. Extensional tectonics, basement uplift and Stephano-Permian collapse basin in a late Variscan metamorphic core complex (Montagne Noire, Southern Massif Central). Tectonophysics, 177, 125-138. ESPITALI~,J., DEROO, G. & MARQUIS,E 1985-1986. La
PETROLEUM SYSTEMS, SE FRANCE Pyrolise Rock-Eval et ses applications. Revue de l'Institut Franqais du P~trole, 40, 563-579, 755-784, 41, 73-89. ESTIVAL,J. & SCHNEEGANS,D. 1935. Sur les gisements de soufre natif des bassins oligoc~nes du midi de la France. In: CongrOs international des mines, de la m~tallurgie et de la g~ologie appliqu~e, section G~ologie AppliquOe, 1, 341-349. FERRY, S. & RUBINO, J. L. 1989. Mesozoic eustacy record on western Tethyan margins. Deuxibme Congres Franqais de S6dimentologie; postmeeting field trip in the Vocontian Trough, Association des Sddimentologistes Franqais. GENNA,A. & DEBRIETrE, R J. 1994. Structures en fleur dans le bassin houiller d'Ales; implications structurales. Comptes Rendus de l'Academie dies Sciences, Paris, 3, s6rie II, 977-984. GIDON, M., MONJUVENT,G., FLANDRIN,J., MOULLADE, M., DUROZOY, G. & DAMIANI,L. 1991. LaragneMonteglin. Carte g6ologique de la France 1:50 000. BRGM, Orldans, France, notice explicative. GRANET, M., STOLL, G., DOREL, J., ACHAUER, U., POUPINET, G. & FUCHS, K. 1995. Massif Central (France): new constraints on the geodynamical evolution from teleseismic tomography. Geophysical Journal International, 121, 33-48. H~.RITmR, E 1994. A history of petroleum exploration in France. In: MASCLE,A. (ed.) Hydrocarbon and Petroleum Geology in France. Springer-Verlag, 29-45. HmN, A., PHILIP, M., ROCHE, A. & WEBER, C. 1980. Images g6ophysiques de la France. In: 26th International Geological Congress, C7 section, GOologie de la France, 25-50. JACQUART, G., DEVILLE, E., MASCLE, A. • RICO, C. 1993. Languedoc-Roussillon. Nonexclusive regional report, Institut Franqais du P6trole. LAMIRAUX, C. & MASCLE,A. 1995. Operators renewing exploration in offshore basins of France. Oil and Gas Journal, July 3, 70-74. LANCELOT,J., BRIQUEU,L., RESPAUT,J. P. & CLAUER,N. 1995. G6ochimie isotopique des systemes U-Pb/Pb-Pb et 6volution polyphas6e des gites d'uranium du Lod6vois et du sud du Massif Central. Chronique de la Recherche Mini&e, 521, 3-18. LEDRU, P., AUTRAN,A. ~: SANTALIER,O. 1994. Lithostratigraphy of Variscan Terranes in the French Massif Central: a basis for paleogeographical reconstructions. In: KEPnE, J. D. (ed.) PreMesozoic Geology in France. Springer-Verlag, 276-288. LEGRAND, X. 1990. Effets de la tectonique extensive en milieu continental." le bassin permien de Saint Affrique. These de l'Universit6 de Toulouse III. , SOULA, J. C. & ROLANDO,J. E 1994. The SaintAffrique Permian Basin (southern France): an example of a roll-over controlled alluvial sedimentation during regional extensional tectonics. Geodinamica Acta, 7, 103-120. MAJESTt~-MENJOULAS, C. & DEBAT, P. 1994. Pyrenees. In: KEPPIE, J. D. (ed.) Pre-Mesozoic Geology in France. Springer-Verlag, 442-457. MALAVIEILLE,J., GUIHOT, P., COSTA, S., LARDEAUX,J.
139
M. & GARDIEN,V. 1990. Collapse of a thickened Variscan crust in the French Massif Central: Mont Pilat extensional shear zone and Saint-Etienne upper Carboniferous basin. Tectonophysics, 177, 139-149. MASCLE,A. 1989. G6ologie p6troliere des bassins permiens fran~ais. In: Potentiel ~conomique des bassins permiens franfais. Chronique de la Recherche Miniere, 499, 69-86. - - , VIALLY,R., DEVILLE,E., BHu-DUVAL, B. & RoY, J. E 1996. The petroleum evaluation of a tectonically complex area: the western margin of the South-East Basin (France). Marine and Petroleum Geology, 13,941-961. OBERTI, G. 1992. Bassin de l'Arc-Gardanne: apport des donners de terrain et de subsurface. IFP-ENSPM report, unpublished. ODIN, B. 1986. Les formations permiennes, Autunien sup&ieur d Thuringien du bassin de Lod~ve (H~rault, France). These de l'Universit6 de Provence, Aix-Marseille. PERRODON, A. 1992. Petroleums systems: models and applications. Journal of Petroleum Geology, 15, 319-326. RAZ1N,P., BONIJOLY,D., LE STRAT,E, COUREL,L., POLI, E., DROMART, G. & ELMI, S. 1996. Stratigraphic record of the structural evolution of the western extensional margin of the Subalpine Basin during the Triassic and the Jurassic, Ard6che, France. In: Mesozoic evolution of the western margin of the French South-East Basin (GPF Programme). Marine and Petroleum Geology, 13, 000-000. ROLANDO,J. P. 1988. Sddimentologie et stratigraphie du bassin permien de Saint Affrique. These de l'Universit6 de Toulouse. ROMAIN,J. 1978. Etude pdtrographique et structurale de la bordure sud-occidentale du massif de l'Argentera, de St Martin de V~subie d la cime du diable. These de la facult6 des sciences et techniques de Lille. ROUIRE, J. & CmRON, J. C. 1980. Valence. Carte g6ologique de la France h 1/250 000. BRGM, Orl6ans, notice explicative. ROURE, E & COLLETTA,B. 1996. Cenozoic inversion structures in the foreland of the Pyrenees and Alps. In: ZIEGLER, P. A. & HORVATH, E (eds) Structure and Prospects of Alpine Basins and Foreland. M6moire du Museum National d'Histoire Naturelle, Paris, 170, 173-209. - - . , BRVN, J. R, COLLETTA,B. & VIALLY,R. 1994. Multiphase extensional structures, fault reactivation, and petroleum plays in the alpine foreland basin of southeastern France. In: MASCLE, A. (ed.) Petroleum Geology o f France. SpringerVerlag, 245-268. Roy, R. 1946. Etude des gisements franqais de Soufre. Houille, Minerais, P6trole, 2, 6676. SERANNE,M., BENEDICTO,A., TRUFFERT,C., PASCAL,G. & LABAUME, P. 1995. Structural style and evolution of the Gulf of Lions Oligo-Miocene rifting: role of the Pyrenean Orogeny. Marine and Petroleum Geology, 12, 809-820. STEINBERG,M., GIOT, D., DEGOUY,M., ELMI, S., FRITZ, B., MILLON,R., PERRIN,J., ROURE, E & SUREAU,J. E 1991. Inter-actions fluides-roches sur une
140
A. MASCLE & R. VIALLY
paldomarge distensive. R6sultats prdldminaires du forage Balazuc-1 (Ard6che). Comptes Rendus de l'Academie des Sciences, Paris, 312, s6rie II, 875-882. TOVTIN,N. 1980. Le Permien continental de la Provence orientale (France). Th6se d'dtat de l'Universit6 de Nice. VALETrE, M. & BENEDICTO,A. 1995. Chevauchements gravitaires halotectoniques dans le bassin distensif de Camargue (Marge du Golfe du Lion, SE de la France). Bulletin de la SociOtO GOologique del France, 166, 137-147. VELLUTINI, E 1977. Le magmatisme permien de la Corse du NW. Thbse de doctorat de l'Universit6 de Marseille. VERGt~S, J., MILLAN, H., ROCA, E., Mu•oz, J. A., MARZO, M., GIRl'S, J., DEN BEZENER, T., ZOETEMEIJER, R. & CLOETINGH, S. 1995. Eastern Pyrenees and related foreland basins: pre-, synand post-collisional crustal-scale cross-sections. Marine and Petroleum Geology, 12, 903-916. VETTER,P. 1968. Gdologie et paldontologie des bassins houillers de Decazeville, de Figeac et du ddtroit de Rodez. Th6se d'Etat de l'Universit6 de Toulouse.
VIALLY,R. 1994. The southern French Alps Paleogene Basin: subsidence modelling and geodynamic implications. In: MASCLE, A. (ed.) Hydrocarbon and Petroleum Geology of France. SpringerVerlag, 281-294. & TngMOLIERES, E (1996) Geodynamics of the Gulf of Lions: implication for petroleum exploration. In: ZIEGLER, E A. & HoavAxI-I, E (eds) Structure and Prospects of Alpine Basins and Forelands. Mdmoire du Museum National d'Histoire Naturelle Paris, 170, 129-158. VINCHON, C. 1984. Sddimento-genOse du Permien du Dome du Barrot (Alpes Maritimes, France). Document du BRGM 70. WANG, H. 1991. Dynamique sOdimentaire, structuration et houillification dans le bassin houiller stOphanien des COvennes. Th6se de l'Universit6 de Dijon. WILSON,M. & DOWNES,H. 1991. Tertiary-Quaternary extension-related alkaline magmatism in Western and Central Europe. Journal of Petrology, 811-849.
-
-
Tertiary-Quaternary magmatism within the Mediterranean and surrounding regions MARJORIE
WILSON 1 & GIANLUCA
B I A N C H I N I 1,2
1School o f Earth Sciences, Leeds University, Leeds LS2 9JT, UK 2Istituto Di Mineralogia, Corso Ercole I D'este 32, 44100 Ferrara, Italy Abstract: Tertiary-Quaternary magmatism within the Mediterranean and surrounding regions, including the European foreland of the Alps, the North African margin and the Eastern Mediterranean, occurs in three distinct associations: 9 anorogenic, extension-related intra-plate magmas, typically Na-rich alkali basalts, basanites and their differentiates, but locally including subalkaline basalts and potassic partial melts (e.g. leucitites) of the mantle lithosphere; 9 orogenic, subduction-related/post-collisionalmagma series related to plate convergence; typically with potassium-rich geochemical characteristics which may be related to subduction of continental crustal materials; 9 subalkaline basalts, similar to mid-ocean ridge basalts, formed at localized oceanic spreading centres. The magmatism is spatially and temporally associated with the Late Cretaceous-Cenozoic convergence of Africa-Arabia with Eurasia which resulted in the progressive closure (by subduction) of oceanic basins in the Mediterranean domain and ultimately the collision of the Alpine orogen with the southern passive continental margin of Europe. Break-off of subducted lithospheric slabs may have provided an important trigger for magmatism in several localities. The timing and geographical distribution of magmatism of orogenic and anorogenic affinity is summarized. Detailed discussion of the major and trace element and Sr-Nd-Pb isotope characteristics of the magmatism in the Central Mediterranean region provides important insights into the petrogenesis of the magmas and the nature of the main mantle source components.
T e r t i a r y - Q u a t e r n a r y magmatism within the Mediterranean and surrounding regions (Fig. 1), including the European foreland of the Alps, the East Iberian and North African margins and the Eastern Mediterranean, occurs in three distinct associations: extension-related intra-plate magmatism, typically alkali basalts, basanites and their differentiates, but locally including subalkaline (tholeiitic) basalts and rare potassic magma types (leucitites and leucite nephelinites); 9 subduction/post-collisional magmatism related to plate convergence characterized by a spectrum of calcalkaline, High-K calcalkaline and potassic magma series (shoshonites and lamproites), including relatively primitive mafic magmas and their differentiates; * localized oceanic spreading centres, erupting subalkaline basalts with affinities to midocean ridge basalts (MORB).
9
Magmatism is closely associated with the Late C r e t a c e o u s - C e n o z o i c convergence of A f r i c a - A r a b i a with Eurasia, involving the deformation of intervening microplates (e.g. the
Italo-Dinarides block), the gradual closure of oceanic basins in the Mediterranean domain and ultimately the collision of the evolving Alpine orogen with the southern passive continental margin of Europe (Ziegler et al. 1995). The stress field affecting the A l p i n e - M e d i t e r r a n e a n region changed repeatedly during this period, related to changes in the convergence direction. There was a gradual shift of compressional tectonic activity away from the foreland of the Carpathians and Eastern Alps to the foreland of the Central and Western Alps, related partly to dextral translation during the late E o c e n e Pliocene. There appears to have been an intermittent build-up of horizontal compressive stresses in the E u r o p e a n foreland, transmitted from the Alpine and Pyrenean collision fronts. Stresses related to the collision of Iberia with Europe, the latest phases of which occurred in the Oligocene-earliest Miocene, probably interfered with stresses transmitted from the Alpine collision zone, at least during the main phases of the Pyrenean orogeny. Phases of compressional deformation occurred during the late Palaeocene, late
WILSON,M. & BIANCHINI,G. 1999. Tertiary-Quaternary magmatism within the Mediterranean and surrounding regions. In: DURAND,B., JOLIVET,L., HORVATH,F. & SgRANNE,M. (eds) The Mediterranean Basins: TertiaryExtension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 141-168.
142
M. WILSON & G. BIANCHINI
Fig. 1. Map of the Mediterranean and surrounding regions showing the main thrust belts and Neogene extensional basins. After Lonergan & White (1997).
Eocene-early Oligocene, late Oligocene-early Miocene and Pliocene (Ziegler et al. 1995). The Eocene and younger compressional deformation of the northwestern Alpine foreland is broadly synchronous with the evolution of the Cenozoic rift system of eastern and central Europe (ECRIS; Fig. 2), which initiated in midEocene-early Oligocene times (Ziegler 1992). The geodynamic setting of the western Mediterranean during the Miocene is complex because of the mutual interference of the Alpine-Betic (Cretaceous-Miocene) and Apennine-Maghrebides (late Oligocene-Pleistocene) collision systems. Continental collision between Africa and Europe, related to the Alpine-Betic system, occurred in the early Miocene and since the Tortonian (8-9 Ma) dextral oblique shortening has been occurring between North Africa and Iberia (Lonergan & White 1997). Doglioni et al. (1997) consider that the major basins, the Alboran Sea, Valencia Trough and Liguro-Provenqal Basin, developed as a coherent system of back-arc basins related to eastward roll-back of a westward directed
Apennines-Maghrebides subduction zone (Fig. 1). These late Oligocene-mid-Miocene basins developed both within the Betic Cordillera (Alboran Sea) and in its foreland (Valencia and Provenqal troughs), cross-cutting the Betic orogenic front. The Alboran Sea opened mainly during the early Miocene (22-10 Ma), with the zone of extension migrating progressively eastwards (Comas et al. 1992; Docherty & Banda 1995). Extension in the Liguro-Provenqal Basin probably started in the late Eocene-Oligocene (Faccenna et aL 1997) with the main syn-rift phase in Oligocene-Aquitanian (32-23 Ma). From the Late Aquitanian, sea-floor spreading resulted in a 25-30 ~ counterclockwise rotation of the Sardinia-Corsica block at a rate of 4-5 cm a -l. During the early-mid-Miocene the zone of extension migrated eastwards from the LiguroProvenqal Basin to the Tyrrhenian Sea. The Western Mediterranean is characterized by large variations in crustal and lithospheric thickness (Gueguen et al. 1997). The lithosphere has been thinned to less than 60 km beneath the major basins (e.g. Valencia Trough - 50-60 km; East Alboran Sea - 40 km; Tyrrhenian Sea -
TERTIARY-QUATERNARY MAGMATISM
143
Fig. 2. The distribution of Tertiary-Quaternary magmatism in western and central Europe. Data sources are given in the text.
20-25 km) while it remains 65-80 km thick below the Corsica-Sardinia block and the Balearic Promontary, which may be regarded as continental boudins separating zones of intensely stretched lithosphere. The Tertiary-Quaternary evolution of the Eastern Mediterranean has been controlled by the collision of the African and Arabian plates with the Eurasian plate along the Hellenic arc and, further to the east, along the Bitlis-Zagros suture zone. Northward subduction of a remnant of the Tethys ocean has generated a broad zone of extension and associated magmatism in the Aegean which may be related to slab roll-back to the south. In order to understand the petrogenesis of the magmas within this wider geodynamic context it is necessary to establish the ages and geochemical characteristics of the main magmatic episodes and their relationship to the local tectonic setting. In subsequent sections magmatism spatially and temporally associated with plate convergence is referred to as being of orogenic affinity, whilst that associated with extensional tectonics is termed anoroge,:ic. The igneous
rocks of each association can be distinguished on the basis of their major and trace element and S r - N d - P b isotope geochemistry; this is discussed in detail in subsequent sections in the context of the Central Mediterranean domain. Emphasis is placed on the geochemistry of the most primitive mafic igneous rocks as this provides the best indicator of the geodynamic setting of the magmatism. In general, the geochemical and isotopic characteristics of more differentiated (i.e. more silica-rich) magmatic rocks are strongly influenced by high-level crustal contamination of their primitive, mantlederived, mafic parent magmas.
European and East Iberian margins Extension-related, predominantly alkaline, magmatic activity within the European foreland of the Alpine orogen (Fig. 2) commonly occurs as small volume monogenetic centres, scattered necks and plugs and fissure-controlled plateau basalts (Wilson & D o w n e s 1991). Rarer central volcanic complexes (e.g. Cantal and Mont Dore in the French Massif Central) include more
144
M. WILSON & G. BIANCHINI
differentiated magmas which can be related to processes of magmatic differentiation in subvolcanic magma chambers (e.g. Wilson et al. 1995). Although magmatism initiated locally in the latest Cretaceous-Palaeocene, the major phase of activity occurs in the Neogene (10-5 Ma) with a subsidiary peak in the Pliocene (4-2 Ma). Magmatism locally continues to a few thousand years Be. Many of the volcanic fields are located on domally uplifted basement massifs (e.g. Massif Central) which appear to be dynamically supported by hot upwelling asthenospheric mantle diapirs (Granet et al. 1995). The magmatism is broadly synchronous with the evolution of the ECRIS rift system (Wilson & Downes 1992; Ziegler 1992). A detailed compilation of the timing of magmatic activity in the French Massif Central (Fig. 3) suggests that the main volcanic phases may be associated with periods of compressional stress relaxation in the Alpine foreland, although the correlation is by no means perfect. Extension-related alkaline magmatism also occurred at scattered localities throughout the eastern part of the Iberian peninsula in
Neogene-Quaternary times (Figs 2 & 4). The earliest reported igneous activity in the central and western parts of the Internal Zones of the Betics is a tholeiitic basaltic dyke swarm with K - A r ages of c. 22 Ma (Torres-Roldfin et al. 1986). Along the southeastern margin of Iberia magmatism of orogenic affinity, including ultrapotassic lamproites, occurred from 17 to 6 Ma (Venturelli et al. 1984a; Nixon et al. 1984; Hern~indez et al. 1987). Breccia pipes, plugs, sills, flows and tephra of lamproite, inferred to be partial melts of subduction-modified mantle lithosphere, were emplaced from 8 to 6 Ma at the SE edge of the Betic and Sub-Betic orogenic belts over an area of 15 000 km 2 (Bergman 1987; Wilson 1989). During the late Miocene-Quaternary extension-related (anorogenic) alkaline magmatism occurred in the Calatrava province (L6pez-Ruiz et al. 1993; Cebria & L6pez-Ruiz 1995), near Olot just south of the Pyrenees, to the NW of Cartagena (Tallante) within the Betics and at Cofrentes and Columbretes Island (Fig. 2). The Calatrava province lies within the Iberian Hercynian Massif close to the external sector of the
Fig. 3. Timing of Tertiary-Quaternary anorogenic magmatism in the Massif Central, France, in relation to the major phases of Alpine compression (Ziegler et al. 1995). Data sources: Downes (1987); Mergoil & Boivin (1993); Patterson (1996).
TERTIARY-QUATERNARY MAGMATISM
145
Fig. 4. Neogene volcanic activity in the Western Mediterranean. After Lonergan & White (1997).
Alpine Betic Cordillera. Volcanism initiated in late Miocene times (7.7-6.4 Ma) with the extrusion of highly potassic olivine leucitites, inferred to be partial melts of the lithospheric mantle, followed, after a hiatus of about 1.7 Ma, by a spectrum of alkaline mafic magmas, including melilitites, nephelinites and alkali olivine basalts.
North African margin-Aiboran Sea Tertiary-Quaternary magmatic activity occurs along the whole western Mediterranean margin of North Africa (Fig. 1) from Morocco to Tunisia. This may be divided into two stages: Late Cretaceous-mid-Eocene (c. 42 Ma) and early Miocene-Recent (Wilson & Guiraud 1998). There is little recorded magmatic activity of Late Cretaceous-mid-Eocene age in the Alpine domain of North Africa apart from some basic alkaline sills and dykes (aiounites) in northern Morocco dated at 57-37 Ma (Michard 1976). In contrast, along the eastern Tunisian and north western Libyan Mediterranean margin abundant Aptian to Palaeocene volcanics have been penetrated by oil wells (Ellouz 1984; Hammuda
et al. 1992). This anorogenic magmatic province
is elongated NW-SE and appears to be associated with active Early Cretaceous and/or Late Cretaceous-Palaeocene rifting on the Pelagian Shelf (Guiraud & Maurin 1991; Guiraud et aL 1992; Van der Meer & Cloetingh 1993). In contrast, Neogene-Recent magmatic activity was widespread along the North African Mediterranean margin from Morocco to Tunisia. The bulk of the magmatism appears to postdate a major phase of Aquitainian-Burdigalien (c. 20 Ma) compression induced by the Alpine collision and has many similarities to that of the Betic Cordillera of southern Spain (Hern~indez et al. 1987; Lonergan & White 1997; Fig. 4). The earliest magmatic rocks have calcalkaline affinities becoming progressively more alkaline with time. Calcalkaline magmatism appears to have initiated in northeastern Algeria during the early Miocene and subsequently spread eastwards along the Mediterranean margin. Given the complex geodynamic setting of the Western Mediterranean, and the limited amount of geochemical and geochronological data on the magmatic rocks in both the Betic Rif and Maghrebide belts, it is difficult to constrain the
146
M. WILSON & G. BIANCHINI
precise tectonic setting in which the magmas were generated. Some authors consider that the Miocene calcalkaline volcanic episode postdated active subduction (e.g. Hern~indez & Lepvrier 1979), whilst others link it directly to subduction (Tricart et al. 1994; Lonergan & White 1997) or to slab detachment in the Tertiary (Zeck et al. 1992; Moni6 et al. 1992). A l b o r a n Sea
The Alboran Sea appears to be underlain by thinned (c. 15 km) continental crust including metamorphic rocks similar to those of the adjacent Internal Zones of the Betic Rif orogen (Fig. 4; Platt etaL 1996). Along with the Betic Rif Internal Zones, it underwent extension until the late Tortonian (c. 7 Ma) concurrent with northwest-vergent thrusting in the External Zone of the Betics and SSW-vergent thrusting in the External Rif (Lonergan & White 1997). Most models proposed to explain this apparent paradox of coeval shortening and extension involve extensional collapse of previously thickened continental lithosphere (Dewey 1988; Platt & Vissers 1989; Doblas & Oyarzun 1989; Vissers et al. 1995). More recently, however, Lonergan & White (1997) have suggested that the Neogene structure and evolution of the Betic Rif orogen can better be explained by westward roll-back, between 23 and 10 Ma, of a short, eastdipping subduction zone segment. The existence of an extensive Tertiary, calcalkaline magmatic province has been inferred within the Alboran Sea (Fig. 4) based upon available magnetic data and samples obtained by dredging (Hern~indez et al. 1987). Alboran Island is a calcalkaline volcanic edifice, dated at 18-7 Ma (Aparico et al. 1991). The volcanism within the Alboran Sea, north Africa and southern Spain is generally younger than the main phase of deformation of the Internal Zones and is both coeval with, and younger than, extension (Lonergan & White 1997). Morocco
Calcalkaline volcanics, mainly rhyolites, were erupted between 16 and 10 Ma along the Mediterranean coast of Morocco followed by trachyandesites between 8 and 4 Ma, and finally by alkali basaltic lavas during the Plio-Pleistocene (Bellon 1981; Lonergan & White 1997; Fig. 4). In the Anti Atlas mountains two large Neogene stratovolcanoes, Jbel Saghro and Siroua, have been recognized. K-Ar dating of samples from Jbel Saghro (nephelinites, tephrites, phonolites) has yielded ages between
10 and 2.8 Ma (Berrahma et al. 1993). Quaternary (1.5-0.5 Ma) alkali basalts and basanites outcrop in the Tabular Middle Atlas, in the High Moulouya Rise within the folded Middle Atlas and also further to the west in the Morocco Meseta (Morel & Cabanis 1993; Gomez et aL 1996; Wilson & Guiraud 1998). Widespread magmatic activity also occurred in the Rekkame Province (Rachdi et al. 1997) where many Palaeogene, Neogene and Quaternary (based on K - A t ages and stratigraphic constraints) volcanic centres (including diatremes) erupted mafic alkaline lavas (mainly basanites, nephelinires). Algeria
Along the NW part of the Algerian coast, in the Sahel d'Oran-Mt Sirda area, calcalkaline to shoshonitic andesites and dacites were erupted between 11.7 and 7.2 Ma, giving way to more alkaline (anorogenic) magmas in the Tafna valley during the Pliocene (c. 4 Ma; LouniHacini et aL 1995). To the east of Algiers, Miocene (16-12 Ma) magmatic activity is represented by basaltic and andesitic lava flows and intrusions (Dellys, Cap Djinet), a granodioritic plug (Thenia), and dacitic to rhyolitic lavas and pyroclastics (Belanteur et al. 1995). These ages are consistent with those reported by Moni6 et al. (1992) for the granitic massif of Edough (15-17 Ma). Cohen (1980) described the occurrence of Aquitainian (19.1_+1) andesites and andesitic tufts intercalated with sedimentary rocks in Grande Kabylie and slightly younger (Burdigalian) submarine andesitic volcanics in Petite Kabylie. Tun&ia
In Tunisia there are two main volcanic areas, Les Nefza and Les Mogods (Bellon 1981). At Les Nefza there are two distinct phases of activity: The first (12.9-8.2 Ma) is characterised by the eruption of potassic acid magmas of probable orogenic affinity, whilst the second phase (8.4-6.6 Ma) is dominated by the eruption of alkali basalts. In Les Mogods (7.4-5.2 Ma) only basic alkaline lavas were erupted. Libya
Widespread magmatic activity occurred in N Tripolitania between the Lower Eocene and the Pliocene. Piccoli (1970) describes the occurrence of fissure-fed lava fields and several central volcanoes, characterized by the eruption of alkaline to transitional basalts and their
TERTIARY-QUATERNARY MAGMATISM
147
differentiates. Available K - A r ages fall in a wide range from 52 to 3.5 Ma, some of which are almost certainly in error. Cahen et al. (1984) have reported ages of 30-12 Ma for basalts from the Gharian area of NW Libya.
It includes localized oceanic spreading centres in the Balearic Basin and Tyrrhenian Sea and a number of distinct orogenic (subduction-related and post-collisional) and anorogenic provinces, considered in detail in the following sections.
Central Mediterranean domain
Provence
Magmatism within the Central Mediterranean domain is mostly of Miocene to Quaternary age (Figs 5 & 6), although in the northern part of the region (Provence-Sardinia, Insubric-Periadriatic Line, Veneto district, Slovenia-Croatia) much of the activity is of Eocene-Oligocene age.
Tertiary volcanic and subvolcanic rocks occur along the French coast from Drammont (west of Cannes) to Monaco. Oligo-Miocene (34-20 Ma; Bellon 1981) basalts, andesites, dacites and microdiorites have been reported from the Est~rel region, along with andesitic clasts in
Fig. 5. The distribution of anorogenic and orogenic volcanism in the Central Mediterranean region. Data sources are given in the text. Abbreviations: AD, Adamello; AE, Aeolian Archipelago; CA, Campania; CO, Corsica; E, Ernici; ET, Mount Etna; IB, Iblean area; L, Linosa; LA, Latium; MA, Magnaghi seamount; MB, Marsili Basin; MS, Marsili seamount; P, Pantelleria; PN, Pietre Nere; PR, Provence; R, Roccamortfina; SA, Sardinia; SC, Sicily Channel; SL, Slovenia; TU, Tuscany; V, Vavilov seamount; VB, Vavilov Basin; VE, Veneto district; V1, Vulsini; Vu, Vulture.
148
M. WILSON & G. BIANCHINI
conglomerates near Nice. Magmatic activity is contemporary with the Oligocene-Miocene volcanism of Sardinia, suggesting that this volcanic district, before the opening of the Balearic basin, would have been part of a wider magmatic province. Beccaluva et al. (1994) consider that the magmatic rocks have an island-arc tholeiite/calcalkaline affinity based on their major and trace element geochemistry.
Sardinia-Corsica
Cenozoic magmatism in Sardinia and Corsica appears to have occurred in two distinct cycles, based upon stratigraphic constraints and K - A r age determinations (Macciotta et al. 1978; Beccaluva et al. 1987). The older Oligocene-Miocene cycle (32-13 Ma; climax of the activity between 21 and 18) on Sardinia has
Fig. 6. (a) Timing of orogenic and anorogenic magmatism, rifting and sea-floor spreading in the Central Mediterranean region. (b) Timing of orogenic and anorogenic magmatism in North Italy and surrounding areas. Data sources are given in the text.
TERTIARY-QUATERNARY MAGMATISM
orogenic characteristics (Savelli et al. 1979; Dostal et al. 1982a; Beccaluva et al. 1994; Galassi 1995a, b; Morra et al. 1997) and may be related to the subduction of oceanic lithosphere towards the NNW under the Sardinia-Corsica microcontinent. Magmatic activity commenced with the eruption of subduction-related basaltic andesites with an island arc tholeiitic/calcalkaline affinity, similar to those in Provence. This was followed at about 26 Ma by the eruption of a complex sequence of alternating basaltic andesites and rhyodacites (including ignimbrites) with a marked calcalkaline affinity; magmatic rocks of shoshonitic affinity have also been recorded but these are subordinate in volume. A thick sequence of dacitic pyroclastic rocks in the Valencia Trough, between the Spanish coast and the Balearic Islands, dated at 20.5 + 1.5 Ma (Weze11976), ignimbrites in southern Corsica (18.9-19.3 Ma; Ottaviani-Spella et al. 1996) and lamproites in north Corsica at Sisco (14 Ma; Peccerillo et al. 1987) may also be related to this phase of activity. A later, extension-related, anorogenic,
149
volcanic cycle (5-0.1 Ma) occurred in Sardinia, characterized by the eruption of both subalkaline and alkaline lavas (including primitive and more differentiated magma types; Beccaluva et al. 1977, 1987; Rutter 1987). Whilst there is no clear temporal trend, the eruption of subalkaline lavas appears to have occurred preferentially during a short period with a climax at about 3.5-3 Ma (Montanini & Villa 1993), followed by more widespread eruption of alkaline magmas. Interpretation of the palaeotectonic setting of the Plio-Pleistocene subalkaline basic-acid magmatism of Mt Arci (Cioni et al. 1982; Dostal et al. 1982b) is somewhat ambiguous. The lavas appear to be transitional in chemistry between those of the earlier Oligo-Miocene subductionrelated cycle and the younger alkaline magmas. This may partially reflect an inherited 'subduction-related' signature in their mantle source. Liguro-Provenqal-Balearic
Basin
Oceanic crust occurs in a narrow zone, no more than 100 km wide, between Corsica and
150
M. WILSON & G. BIANCHINI
Provence, Sardinia and the Balearic Islands and north of the Algerian coastline (Fig. 5). Seafloor spreading is most likely to be related to back-arc extension above a westward dipping subduction zone which generated the OligoMiocene subduction-related magmatism of Sardinia, Corsica and Provence. Palaeomagnetic studies of Tertiary volcanic rocks cropping out on Sardinia indicate an anticlockwise rotation of the island between 21.7 and 16.5 Ma (late Aquitanian-late Burdigalian), constraining the age of the major spreading phase (Vigliotti & Langenheim 1995). The Burdigalian opening of this basin is also supported by the age of the sedimentary sequences preserved around its margins and by available heat flow data (Rehault et aL 1984). A sample of a potassic igneous rock (tristanite), of intermediate composition, dredged from the sea floor between Provence and Corsica has a K - A r age of 18 _+0.5 Ma (Rehault et al. 1984). Samples from submarine volcanic edifices to the NW and SW of Corsica have given ages of 17.5-16 Ma (Genesseaux & Rehault 1975; Deverch~re et al. 1996). Tyrrhenian Sea
The Tyrrhenian Sea can be subdivided geologically, at the 41~ parallel, into two different sectors (Fig. 5). The tectonomagmatic evolution of the northern sector reflects the geological history of the adjacent Tuscan area to the east, with E - W extension associated with widespread acid magmatism, commencing at c. 7-9 Ma (Spadini et al. 1995). The southern sector was affected by more recent E - W / N W - S E extensional tectonics associated with widespread basic magmatism and the formation of true oceanic crust in two different sub-basins (sites 650, 651, 655 of O D P Leg 107 and DSDP site 373, Leg 42; Beccaluva et aL 1990). Sea-floor spreading commenced in the Vavilov Basin around 4-3.5 Ma and in the Marsili Basin around 1.9-1.7 Ma, based on palaeontological constraints from associated sediments, magnetostratigraphy and limited A r - A r dating (Savelli 1988; Feraud 1990; Kastens et al. 1990). Igneous rocks from sites 650 (Marsili Basin) and 651 (Vavilov Basin) have a subduction-related affinity (calcalkaline to high-K calcalkaline) while those from sites 655 and 373 (Vavilov Basin) and site 654 (Sardinian c o a s t ) h a v e anorogenic characteristics (transitional MORB). Two large seamounts in the Vavilov Basin (Magnaghi and Vavilov) have anorogenic geochemical characteristics similar to those of oceanic island basalts (OIB), whilst the Marsili seamount in the
Marsili Basin has a lower part characterized by OIB-iike magmas and an upper part with a marked calcalkaline signature (Serri 1990; Beccaluva et al. 1990; Savelli & Gasparotto 1994). The heterogeneous nature of the mantle source of the seamount magmas may reflect fluid fluxing of the shallow mantle above a westward dipping subduction zone.
Sicily area. Several phases of MesozoicCenozoic volcanic activity have been recognised in the Iblean area of southern Sicily (Fig. 5). These include outcrops of Cretaceous, Miocene and Plio-Pleistocene volcanics and sub-surface Triassic-Jurassic magmatic rocks, known only from drill cores. Cretaceous alkaline volcanics are exposed discontinuously in the eastern part of the area (Capo Passero, Siracusa and Augusta), while Miocene and Plio-Pleistocene volcanics occur in the northern part of the Iblean platform, toward the Apennine-Maghrebian compressional front (Beccaluva et al. 1998). The Miocene (mainly Tortonian) magmatic phase is predominantly alkaline in composition, characterized by a low melt production rate. After a period of low-level activity from about 6.5 to 4 Ma ago (according to biostratigraphical data) a new cycle, with an higher melt production rate, initiated in the early-mid-Pliocene, lasting until the early Pleistocene. This cycle is characterized by an abrupt compositional change to magmas of predominantly tholeiitic affinity. The melt production rate gradually decreased after the climax of volcanic activity and magmas of various geochemical affinities were erupted simultaneously. During the early Pleistocene sporadic eruptions of highly undersaturated alkaline lavas occurred in the northernmost part of the plateau, towards Mount Etna. It is difficult, however, to study the migration of volcanism from the Iblean area to the Etna area because the transition zone is obscured by recent sediments. Most of the Tertiary-Quaternary lavas from the Iblean area are relatively primitive basalts. The alkaline mafic magmas frequently contain peridotite mantle xenoliths, indicating that the magma rose rapidly to the surface from their mantle source region with little opportunity for differentiation or crustal contamination. Iblean
M o u n t Etna. Mount Etna is the largest active volcano in Europe (3300 m high) with an estimated volume of 500-600 km 3. It is polygenetic, having several distinct stages to its evolution. The oldest volcanic products (c. 600 000 years
TERTIARY-QUATERNARY MAGMATISM BP; Romano 1983) are of tholeiitic affinity and outcrop sporadically at considerable distances from the present volcanic focus. Alkaline mafic magmatism commenced around 220 000 years BP (Condomines et al. 1982; Gillot et al. 1994) and for a short period tholeiitic and alkaline magmas erupted simultaneously. More recent magmatism has been entirely alkaline and the erupted lavas have become progressively more differentiated with time, consistent with the development of a high-level magma chamber beneath the volcano (Clocchiatti et al. 1988). Sicily Channel
This area is characterized by a shallow submarine platform (average water depth c. 350 m) transected by three deep tectonic graben (water depth > 1000 m). Most of our knowledge of the volcanism in this sector is based on the subaerial record of activity on the islands of Pantelleria and Linosa (Fig. 5). Stratigraphic evidence, supported by K - A t dating, indicates that volcanic activity commenced in the late Miocene (c. 10 Ma) and has continued to the present day. The last observed manifestation of activity was in 1891 when a small volcanic island emerged and was subsequently eroded. The most primitive volcanic rocks on Pantelleria are transitional-mildly alkaline basalts and hawaiites. More silica-rich magma types (trachytes and peralkaline rhyolites) are widespread and are usually interpreted as the differentiation products of parental basic magmas. Linosa is predominantly composed of alkali-basalts with subordinate hawaiites. Beccaluva et al. (1981) and Calanchi et al. (1989) studied samples of dredged submarine volcanic rocks from the area, noting the occurrence of a wider range of basic magma compositions from tholeiites to nepheline basanites. There are many similarities with the volcanism of the Iblean area of Sicily to the NE, although the latter is characterised by predominantly mafic magma compositions. Aeolian Archipelago
The Aeolian Archipelago consists of seven volcanic islands plus a number of seamounts (located in the western and northeastern parts of the archipelago) which define a horse-shoe shaped structure. A wide spectrum of volcanic rocks of orogenic affinity (island arc tholeiite, calcalkaline, high-K calcalkaline and shoshonitic) have been recorded (Beccaluva et al. 1985; Ellam et al. 1989; Francalanci et aL 1993; Galassi 1995a, b). In contrast with the volcanism of central Italy, that of the Aeolian Archipelago
151
is predominantly calcalkaline to high-K calcalkaline, ranging in composition from basalt to rhyolite. In several volcanoes a transition from calcalkaline to highly potassic shoshonitic and leucite tephrite magmas has been observed. Such potassium-rich alkaline magma compositions are, however, generally subordinate in volume. Volcanism is clearly related to subduction of a slab of Ionian Sea oceanic crust towards the NW along the Calabrian arc (Spakman 1990; Giardini & Velona 1991; Francalanci & Manetti 1994). Magmatic activity initiated in the Pleistocene and becomes progressively younger moving counterclockwise along the archipelago towards the northeast (Beccaluva et al. 1985). The islands of Stromboli and Vulcano are volcanically active. The oldest calcalkaline magmatism, based on a dredge sample from the westernmost seamount, commenced at c. 1.3 Ma whilst the oldest recorded shoshonitic activity occurred at 0.85-0.64 Ma. Central-south Italy
Tertiary-Quaternary potassic magmatism occurred in central-south Italy in a postcollisional extensional tectonic setting. The magmatism of this area (Fig. 5) has traditionally been divided into two provinces: the Tuscan magmatic province in the north and the Roman magmatic province (including Campania and Latium) further south. More recently, some authors (e.g. Serri 1990; Beccaluva et al. 1991) have proposed a further subdivision of the Roman Province into a N W Campania-Latium sub-province and a Central Campania subprovince, based on significant differences in the nature of the erupted magmas. Many authors (e.g. Peccerillo et al. 1987; Conticelli & Peccerillo 1992; Serri et al. 1993) have suggested that the onset of magmatism in the area occurred at c. 14 Ma, based upon a small outcrop of ultrapotassic igneous rock at Sisco in North Corsica. This remains a subject for debate, however, since the Sisco magmatism is both temporally and spatially separated from the magmatism of central-south Italy. As reported by Serri et al. (1993) the first recorded occurrence of magmatic activity in the area (c. 7 Ma) is in t h e Tuscan archipelago in northern part of the Tyrrhenian Sea and along the Tuscan coast (Capraia, Elba, Giglio and Montecristo islands, Tolfa and S.Vincenzo, with activity between 7 and 3.5 Ma,), with subsequent migration of the volcanism towards the east and southeast (Roccastrada, Mt Amiata, Radicofani, Mt Cimini, where activity is much younger, ranging from 2.3 to 0.18 Ma).
152
M. WILSON & G. BIANCHINI
In the Roman Province the oldest magmatism occurs in the island of Ponza (4.4-3.9 Ma), whilst on the adjacent mainland the oldest volcanic products (c. 2 Ma) have been located in boreholes in the westernmost part of the Volturno plain (Beccaluva et al. 1991). In the Latium-Roccamonfina area volcanic activity ended between 0.05 and 0.1 Ma, but continues to the present day further south in Campania around the Gulf of Naples (Ischia, the Phlegrean Fields and Mt Vesuvius). The Pleistocene (0.8-0.58 Ma) volcanic activity of Mount Vulture represents the southeasternmost extent of the province. This is the only volcano located in the eastern side of the Apenninic orogenic belt (De Fino et al. 1986) and, consistent with its unique geodynamic setting, the volcanic products have distinctive geochemical characteristics. Within the Roman Province, most of the volcanic rocks belong to potassic and highly potassic alkaline and calcalkaline magma series (Appleton 1972; Serri et aL 1993). Among the most potassium-rich rocks are exotic lamproitic and kamafugitic magma types, similar to those which occur in the Ugandan segment of the East African rift (Peccerillo et al. 1987; Conticelli & Peccerillo 1992). In addition, rare carbonatite magmas, closely associated with rocks of kamafugitic affinity, have been recorded in the easternmost part of the area (Stoppa & Cundari 1996). Subalkaline intermediate-acid volcanic and plutonic rocks, mainly of high-K calcalkaline affinity, occur in Tuscany and in the Tuscan archipelago. The more acid magmas are usually interpreted as hybrids between crustal and mantle derived melts, although a pure crustal, anatectic origin has been proposed for rhyolites from S. Vincenzo, Roccastrada, and Cerveteri (Pinarelli 1987; Ferrara et al. 1989) and for some acid plutonic rocks exposed in the Vercelli seamount and on the islands of Montecristo, Elba and Giglio. The geodynamic setting of the volcanism of south-central Italy has been hotly debated in the past. Its association with extensional tectonic structures (grabens) has led some authors to assume that the magmas are of anorogenic affinity. The recognition of typical calcalkaline magmas associated with the more prevalent Krich alkaline magmas (Beccaluva et al. 1991), however, clearly demonstrates an orogenic affinity, related to subduction. Serri et al. (1993) have recognized a temporal evolution within the province from regional compression to extension (rifting) and the formation of sedimentary basins, followed by uplift and the onset of vol-
canism. Eastward migration of the focus of the extensional tectonic activity and of the location of the magmatism, while compression was still occurring in more external portions of the orogenic belt, suggests that the geodynamic setting of the magmatism is most probably postcollisional related to subducted slab roll-back, possibly associated with tearing of the slab and the formation of a slab window. N o r t h I t a l y a n d s u r r o u n d i n g areas: I n s u b r i c - P e r i a d r i a t i c line a n d the V e n e t o v o l c a n i c district
There is no direct evidence for Eoalpine (prelate Eocene) orogenic magmatic activity associated with Neotethys subduction apart from the presence of andesitic clasts in the Tavayanne formation of the Haute Savoie, France (Delaloye & Sawatzky 1975). Magmatism within the Alpine orogenic belt occurred between 42 and 25 Ma, with a climax between 33 and 29 Ma, postdating the main compressional phases. This included the emplacement of granitoid intrusions (e.g. Adamello) as well as of extensive basic-acid dyke swarms cutting Austroalpine, Southalpine and, rarely, Penninic units. Most of the dykes are located along the Insubric-Periadriatic tectonic lineament (Fig. 5). Geochemical studies of the basic dykes indicate a spectrum of magma compositions ranging from calcalkaline and high-K calcalkaline to shoshonitic and ultrapotassic types (Beccaluva et al. 1979; 1983; Venturelli et al. 1984b; von Blanckenburg & Davies 1995). The most primitive magma compositions appear to be mantle-derived melts, minimally affected by crustal contamination or high-level differentiation processes. Their geochemical characteristics indicate a subductionrelated affinity (von Blanckenburg & Davies 1995) and it has been suggested that they were derived from a lithospheric mantle source intensely metasomatized by subduction zone fluids/melts. Partial melting of the lithospheric mantle may have been triggered by slab breakoff. It is possible that the widespread occurrence of shoshonitic/ultrapotassic magmas in this region could reflect the subduction of continental crustal material. Beccaluva et al. (1983) recognized a spatial zonation in the chemical composition of the dykes with an increase in the K20 content from the SE (where the magmas are typically of calcalkaline affinity) towards the NW (shoshonitic/ultrapotassic magma types). On this basis they proposed the existence of an Oligocene subduction zone dipping towards the NW, in the opposite sense to the generally
TERTIARY-QUATERNARY MAGMATISM accepted southward directed Eoalpine subduction of the European plate. If correct, the OligoMiocene magmatism of this sector could represent the westward continuation of the Sardinian Oligo-Miocene orogenic (subductionrelated) magmatic arc. Late Eocene-early Miocene magmatism of calcalkaline affinity also occurs in Croatia and Slovenia in the easternmost part of the Periadriatic zone close to the southwestern part of the Pannonian Basin (Pamic 1993; Altherr et al. 1995). Miocene-Pliocene basalts, andesites and pyroclastics have been reported from the Adriatic coast of Croatia (Marjanac pers. comm. 1996), although these have not been characterized geochemically and therefore their tectonic affinity is unknown. Localized occurrences of Palaeocene-Oligocene extension-related (anorogenic) alkali basalts, basanites and subordinate transitional basalts occur in the Veneto (Lessini Mountains, Southern Trentino area, Euganei Hills, Berici Hills, Asiago Plateau and Marostica Hills) region of northern Italy (Siena & Coltorti 1989; De Vecchi & Sedea 1995; Milani 1996). More differentiated magmas occur only in the Euganei Hills. The age of the magmatism is based primarily upon stratigraphical constraints and may extend into the Miocene.
Aegean area Fytikas et al. (1984) recognized two main phases of volcanic activity in the Aegean area (Fig. 7).
153
The first developed in the northernmost sector from Oligocene to mid-Miocene times, mainly consisting of intermediate magmas of calcalkaline to shoshonitic affinity. There appears to have been a continuous migration of the focus of volcanic activity towards the south, accompanied by a variation in the chemical composition; the younger volcanic products becoming more K-rich. An exception to this trend, however, are the islands of Skiros and Evia (15-13 Ma) where more normal calcalkaline magmas were erupted (Pe-Piper & Piper 1994). After a hiatus in activity during the midMiocene to Pliocene a second volcanic cycle initiated in the early Pliocene, along a restricted zone in the southern part of the Aegean erupting a typical calcalkaline association with both basic and more evolved magmas (Briqueu et al. 1986; Pe-Piper & Piper 1989; Robert et al. 1992; Pe-Piper 1994). This magmatic arc is considered to be the expression of active subduction towards the north of the oceanic crust of the Herodotus abyssal plain (a remnant of the Mesozoic Tethys; Robertson & Grasso 1995).
Macedonia Late Miocene to early Pleistocene volcanism occurred in Macedonia characterized by high-K calcalkaline to shoshonitic products, including both basic and more evolved magmas (Kolios et al. 1980; Karamata et al. 1994).
Oligocene Miocene Pliocene Quaternary Fig. 7. Distribution of TertiaryQuaternary magmatism of different ages in the Aegean. Data sources are given in the text. Abbreviations: AL, Alexandroupolis; AT, Athens; AF, Afyon; BO, Bodrum; CH, Chios; CR, Crete; DO, Doirani; ED, Edessa; EV, Evia; EZ, Ezine; IP, Isparta; IS, Istanbul; IZ, Izmir; KO, Kos; KU, Knla; LE, Lesbos; LI, Limnos; PA, Patmos; SA, Samothrathrakj; SM, Samos; ST, Santorini; TH, Thessaloniki; VO, Volos.
154
M. WILSON & G. BIANCHINI
Western Turkey and the Levant The Neotectonic evolution of Turkey (Fig. 7) has been dominated by the collision of the African and Arabian plates with the Eurasian plate along the Hellenic arc to the west and the Bitlis-Zagros suture to the east (Seng6r & Yilmaz 1981; Seng6r et at. 1985; Dewey et al. 1986; Yilmaz 1993; Oral et al. 1995). This resulted in extensive mafic to felsic volcanism in eastern, central and western Anatolia, as well as along the North Anatolian and East Anatolian strike-slip fault zones (Yilmaz 1990; Pearce et al. 1990; Notsu et al. 1995; Wilson et al. 1997). The Western Anatolian volcanic province (Fig. 7) is located at the eastern end of the Aegean volcanic arc, which results from the northward subduction of the African plate beneath the Aegean. Calcalkaline volcanic activity commenced in the late Oligocene-early Miocene followed by alkali basaltic volcanism from late Miocene to Recent times. This change in the style of the volcanism has been attributed by some authors to a change in the regional stress field from N-S compression to N-S extension (Yilmaz 1990; Gt~leq 1991). Seyitoglu & Scott (1992), however, consider that the transition to N-S extensional tectonics actually commenced much earlier in the latest Oligocene-early Miocene. Seyitoglu et al. (1997) have demonstrated that within the youngest volcanic sequence there is a change from potassic magmatism in the Miocene to more sodic (anorogenic) alkaline magmatism in the Quaternary. The geochemical characteristics of the potassic magmas are inferred to reflect the presence of an inherited subduction-modified component in their mantle source. Widespread, extension-related alkali basaltic volcanism of Late Miocene-Pliocene to Recent age also occurs throughout the Middle East in Israel, Syria and eastward within the Zagros collision zone (Giannerini et al. 1988; Garfunkel 1989; Sawaf et al. 1993; Seng6r et al. 1993; Alavi 1994).
Geodynamic setting of magmatism in the central Mediterranean region It is generally accepted that the main orogenic phase of the Alpine chain was associated with southward subduction of the European lithosphere beneath the African plate. As noted previously, however, there is little evidence for Eoalpine subduction-related magmatic activity. Eocene-Oligocene igneous activity associated with the Insubric-Peradriatic lineament has been interpreted by some authors to reflect the
post-collisional detachment of the subducted slab (e.g. von Blanckenburg & Davies 1995). If correct, the geochemical characteristics of the magmas suggest that there may be a correlation between the generation of highly potassic magmas and slab detachment which deserves further investigation. It seems reasonable to assume that after the main Eoalpine compressional phase several oceanic strands (former parts of the Mesozoic Tethys) remained in the Mediterranean area which were subsequently consumed by subduction. Focussing on the circum-Tyrrhenian region (Fig. 5), it is possible to recognize two distinct subduction-related magmatic phases (Beccaluva et aL 1987, 1994; Galassi 1995b) which can be attributed to the subduction of these remnants of Tethyan oceanic lithosphere. Oligo-Miocene
cycle
Two distinct subduction systems seem to have been operative at this time. In the western Mediterranean a subduction system directed towards the northwest (Fig. 8) is required to explain the oldest orogenic volcanism in Provence (34-20 Ma), Sardinia (32-13 Ma), North Africa (Kabylies; c. 18-20 Ma) and southern Spain (c. 18 Ma). Calcalkaline igneous rocks dated at c. 20 Ma have also been recorded from the Alboran Sea and Valencia Trough (DSDP holes 122, 123; K - A r ages confirmed by fission track dating). The easternmost propagation of this subduction system could be represented by the Oligocene magmatism along the Periadriatic-Insubric lineament in the Alpine domain, providing an alternative hypothesis to the slabbreakoff model of von Blanckenburg & Davies (1995). During this phase the Sardinia-Corsica microcontinent and the Kabilies blocks would have been joined to the European continent as the Liguro-Provenqal-Balearic Basin had not yet opened. A second Oligo-Miocene subduction system must have dipped towards the north in the eastern part of the Mediterranean in the Aegean area (Fytikas et al. 1994), progressively consuming the northernmost oceanic strand present in this sector. The southernmost oceanic strand present in the eastern Mediterranean area was not involved during this stage (Robertson & Grasso 1995). Lonergan & White (1997) have recently suggested that the collision of the Kabylies block with the North African margin by about 18 Ma effectively divided the previously continuous Oligo-Miocene subduction zone into two segments (Fig. 8). They attribute the Miocene calcalkaline magmatism of the Western
TERTIARY-QUATERNARY MAGMATISM
155
Fig. 8. Tectonic reconstructions illustrating the Neogene evolution of the Western Mediterranean at c. 30 and 18 Ma. Modified after Lonergan & White (1997). By 18 Ma Sardinia and Corsica had rotated counterclockwise and the Balearic Islands clockwise as a consequence of the opening of the the Liguro-Provenqal and Valencia basins. The Kabylies block had collided with the North African margin splitting the formerly continuous subduction system into two branches.
Mediterranean to a short, arcuate, eastward dipping subduction zone. This view is somewhat controversial as most authors consider that the subduction zone polarity was towards the north at this stage (e.g. Doglioni et aL 1997).
N e o g e n e - Q u a t e r n a r y cycle (15-0 Ma) By 16-13 Ma orogenic magmatic activity along the two Oligo-Miocene subduction systems seems to have ceased. It is possible that in some sectors (e.g. North A p p e n n i n e area) no more oceanic crust was left to 'feed' the arc magmatism, with the result that these subduction zones became partially locked by continental collision (e.g. Mantovani et al. 1997). To accommodate the continuing convergence between Europe and Africa subduction is inferred to h a v e migrated, by a process of slab roll-back, towards the southeast in the Western Mediterranean and to the south in the Eastern Mediterranean. In these areas subductable oceanic lithosphere was still present (Ionian Sea lithosphere in the west and the southernmost oceanic strand of
Tethys in the Eastern Mediterranean). Beccaluva et al. (1987) proposed that there was a single subduction system in the Central Mediterranean which migrated progressively, if discontinuously, from its Oligocene configuration to the present one. In the Central Mediterranean region a new orogenic magmatic cycle began in the uppermost Tortonian (c. 7 Ma). The products of this cycle are widespread in Central Italy (Tuscany, Latium, C a m p a n i a and Umbria) and in the Aeolian archipelago. Active subduction is currently only occurring along the Calabrian arc, where subduction of Ionian Sea oceanic lithosphere is associated with the Pleistocene-Recent orogenic volcanism of the Aeolian Archipelago, and further east in the south Aegean, where subduction of the oceanic lithosphere of the Herodotus abyssal plain has been associated with orogenic volcanism from Pliocene to historical times. The relative positions of the subduction systems of this younger orogenic phase are clearly revealed by seismic tomography (e.g. Spakman et aL 1993).
156
M. WILSON & G. BIANCHINI
Relation between subduction and extensional and cornpressional tectonics A combination of compressional tectonics and simultaneous extension (in the more internal areas) seems to be a common feature of the subduction systems of both the Oligo-Miocene and Neogene-Quaternary orogenic phases within the Mediterranean region. For example, extension occurs in the internal zone of the South Aegean arc while subduction and a compressive regime (associated with the formation of the socalled Mediterranean ridge) is active in the external part of the arc; extension occurs in the internal zone of the Apenninic belt while the more external zones further east are under compression; Oligo-Miocene extension occurred in South France and Sardinia simultaneous with subduction. SengOr (1993) noted that such behaviour is not unique to the Mediterranean area, reporting similar cases of coeval and codirectional shortening and extension in active orogenic zones (e.g. Western Pacific island arcs). Lonergan & White (1997) argue that this is an inevitable consequence of slab roll-back.
Sea-floor spreading Sea-floor spreading in the Balearic Basin (c. 20-15 Ma) and the Tyrrhenian Sea (<4 Ma) has traditionally been attributed to back-arc extension above a subduction system which was continually retreating towards the southeast (Beccaluva et aL 1987,1994; Serri 1990; Doglioni et al. 1997; Lonergan & White 1997). There appears, however, to be a problem with this simple model in that the formation of oceanic crust in the Tyrrhenian Sea appears to precede the onset of magmatic activity in the Aeolian arc by at least three million years. Francalanci & Manetti (1994) explain this discrepancy by recognising the existence of an older submarine volcanic arc of Pliocene age in the Tyrrhenian Sea between the Vavilov and Marsili basins, represented by the calcalkaline submarine volcanoes Glauco and Aceste Anchise, by other minor volcanoes at the western boundary of the Marsili basin and by igneous rocks of orogenic affinity drilled at ODP leg 107 sites 650 and 651. The easternmost continuation of this arc may be represented by the calcalkaline Pliocene volcanism of the Pontine Islands near the south Latium coast. Accepting the presence of this submerged arc, the magmatic evolution of the area along a NW-SE profile from Provence to the Aeolian arc can be considered schematically as follows:
(1)
(2)
(3)
Oligo-Miocene calcalkaline magmatism in Sardinia-Provence (34-13 Ma) and formation of the Balearic basin (20-15 Ma) between Provence and Sardinia; migration of the active arc to the position of the inferred Pliocene submarine arc and formation of the Vavilov Basin (oceanic spreading 4-3.5 Ma); opening of the Marsili Basin (1.9-1.6 Ma) and the onset of magmatism in the Aeolian Archipelago (1.3 Ma).
It seems reasonable to assume that when no more oceanic lithosphere was present to 'fuel' the Oligo-Miocene arc, the resulting increase in the forces resisting subduction, primarily due to continental collision between the SardiniaCorsica block and the western margin of Apulia, caused a reorganization of the subduction system with its displacement to the southeast where oceanic crust was still available for subduction. According to this model it is the distribution of remnants of Mesozoic Tethyan oceanic crust which has been the most important parameter in controlling the distribution of orogenic volcanism in the Mediterranean domain, in combination with the slab roll-back process. The geochemical characteristics of the orogenic magmas provide important constraints on the nature of the subducted lithosphere. This will be considered in further detail in a subsequent section. Serri et al. (1993) have suggested that the petrogenesis of the anomalously potassium-rich magmas of the Roman Province implies the subduction of continental crustal materials. The same hypothesis could be extrapolated to the other orogenic magmatic districts in which potassium-rich magmas predominate. Conversely, Francalanci & Manetti (1994), in their study of the Aeolian arc magmatism, have demonstrated the oceanic nature of the lithosphere subducted beneath the Calabrian arc. Subduction of oceanic lithosphere is also implicated in the petrogenesis of the Oligo-Miocene magmatism of Sardinia-Provence.
Anorogenic magmatism From the late Eocene (c. 40 Ma) to the present day, anorogenic magmatic activity has occurred throughout the European and African margins of Tethys and within the Mediterranean domain. Major phases of volcanism occur in the early Miocene and in the late Miocene-Pliocene, though activity was not widespread until the late Miocene. This major volcanic flare up may reflect a fundamental reorganization of the convection system within the upper mantle
TERTIARY-QUATERNARY MAGMATISM intimately associated with the Alpine collision. The volcanic fields are generally concentrated in Pan-African/Hercynian mobile belt zones which have experienced tectonothermal events within the past 650 Ma; these generally have higher heat flow and thinner lithosphere than the adjacent cratons. Pliocene-Recent (extensionrelated) anorogenic volcanic activity often occurs in those sectors which previously experienced Oligo-Miocene subduction-related magmatism (e.g. Sardinia, Morocco, NW Algeria, W Turkey and the Pannonian Basin). On the basis of a detailed seismic tomographic study of the French Massif Central, Granet et al. (1995) suggested that those major TertiaryQuaternary volcanic fields associated with 300-500 km diameter basement uplifts may be underlain by active diapiric mantle upwellings. The diapirs were inferred to represent 'hot fingers' of mantle material rising from a much more extensive sheet of anomalous mantle material, thought to reflect a large-scale mantle plume head which had spread out at depths of 400-600 km within the upper mantle beneath the entire Mediterranean domain. The geochemical characteristics of the most primitive anorogenic basaltic magmas erupted throughout the region are similar to those of oceanic island basalts (OIB) related to the activity of mantle plumes. The Plio-Pleistocene extension-related lavas of Sardinia seem to be an exception and it is possible that the mantle source of these magmas is still contaminated by the orogenic imprint of the earlier OligoMiocene subduction. The localized oceanic spreading centres preferentially sample a depleted mantle source component, similar to the source of MORB, although in the Tyrrhenian Sea this is clearly modified by a subductionrelated fluid flux.
Geochemical characteristics of the magmatism in the central Mediterranean region In producing this regional synthesis of the geotectonic setting of magmatism in the Mediterranean region, magmas of orogenic and anorogenic affinity have been distinguished using a combination of major and trace element and S r - N d - P b isotope geochemical characteristics. It is clearly impossible to discuss all of the available data in detail and therefore we have chosen to summarize the key features, focussing on the Central Mediterranean region (Fig. 5).
157
As stated previously, igneous rocks of orogenic affinity encompass a broad spectrum of calcalkaline to ultrapotassic alkaline magma series, including relatively primitive basalts and their differentiates. The anorogenic suites include both alkaline and sub-alkaline magma series, again ranging from basalts to more silicarich magma types. In general, only the most primitive mafic magmas (basalts sensu lato) can provide information about the nature of their mantle source and consequently we have focussed on the geochemical characteristics of such rock types in the following sections. To determine the tectonic affinity of a particular suite, we stress the importance of obtaining a comprehensive set of major and trace element and Sr-Nd-Pb isotope data.
M a j o r elements Orogenic and anorogenic magma series can be clearly distinguished by plotting KzO/Na20 (weight ratio) versus weight % SiO 2 (Fig. 9). Magmas of orogenic affinity typically have a KzO/Na20 ratio >1, whereas the anorogenic magmas have a KzO/Na20 ratio <1. The only exceptions to this general trend are the early eruptives in the Oligo-Miocene lava sequence on Sardinia, which have a low-K transitional island arc tholeiitic to calcalkaline affinity, island-arc tholeiites from the Aeolian Archipelago and some low-K basalts from the Campanian province (Roccamonfina).
Trace elements Primitive mantle-normalized trace element variation diagrams (colloquially referred to as 'spiderdiagrams') are particularly useful for distinguishing orogenic from anorogenic magmas. In Figs 10 and 11 we have plotted representative patterns for the most primitive mafic magma compositions reported in the literature. Basalts with an orogenic affinity have a distinct negative Nb trough, whereas those of anorogenic affinity have a peak at Nb and a negative trough at K. The negative potassium anomaly may reflect the presence of a residual K-bearing phase in their mantle source, such as amphibole or phlogopite (e.g. Wilson & D o w n e s 1991). It should be noted that the trace element patterns of more differentiated magmas are often more complicated to interpret because of the combined effects of crustal assimilation and fractional crystallisation and therefore, for tectonic discrimination purposes, only mafic magmas with >6-7 wt% MgO should be considered.
M. W I L S O N & G. B I A N C H I N I
158 8
,
i
9
i
zx
i
9
orogenic
t'q
9
A
t"q
anorogenic
t'q
r
A~~
A zxzx~
oooo ,
,
~,~rW'*
40
I>
,
I
50
0
70
60
80
40
50
SiO 2 9
Aeolian Archipelago
[]
Ernici (Latium)
O
Roccamonfina
A
Vulsini(Latium)
@
Sardinia 32-13 Ma
6O
70
SiO 2 I~
Pantelleria
@
Etna
@
Iblean tholeiites
V
Iblean transitional
[]
Iblean alkaline
4,
Iblean ankaramites
Sardinia Plio-Pleistocene subalkaline series
t>
Sardinia Plio-Pleistocene alkali basalts
Fig. 9. V a r i a t i o n of K 2 0 / N a 2 0 (weight r a t i o ) versus wt % SiO2 for T e r t i a r y - Q u a t e r n a r y volcanic suites f r o m the C e n t r a l M e d i t e r r a n e a n region. D a t a sources: see c a p t i o n to Fi~. 12.
1000
i
i
I
I
I
I
I
I
I
I
I
I
I
I
000
I
i
i
i
i
i
l
i
i
i
i
i
i
t
i
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Ernici
~ 100
lO0
~ lO
lO
?.
] m
1
i
i
t
RbBaTh
i
i
i
i
i
i
t
K N b L a C e Sr N d P
i000 ...............
i
t
t
[
I
t
R b B a T h K N b L a C e Sr N d P Z r S m E u T i Y b
Z r S m E u Ti Y b
OROGENIC BASIC MAGMAS Vulsini
100
10001................ A e o l i a n
~
100
" , " ~
10
lfl I I I I i I f I R b B a T h K N b L a C e Sr N d P Z r S m E u Ti Y b
10
Islands
,~
..... " O . . . O
\'b
""
"...... ":'-"
1 . . . . . i i , i ~ , , i i i I RbBaTh K NbLaCe SrNd P ZrSmEu TiYb
Fig. 10. P r i m i t i v e m a n t l e - n o r m a l i s e d t r a c e e l e m e n t v a r i a t i o n d i a g r a m s for p r i m i t i v e o r o g e n i c basic volcanic r o c k s f r o m the C e n t r a l M e d i t e r r a n e a n region. D a t a sources: see c a p t i o n to Fig. 12.
TERTIARY-QUATERNARY MAGMATISM 1000
1000 . . . . . . . . . . . . . . . 9-~
Iblean lOO
Plateau
I
.,..~
~ 100
subalkaline
lo
lO E
~ 9 a.
1
RbBaTh KNbLaCeSrNd P ZrSmEuTiYb
1000
I
!{' IbiS'anPlateau~ alkaline I
I
I
I
I
I
I
I
I
I
1
I
@
100
~100
r
I
I
I
I
10
~
........9 ........ P a n t e l l e r i a
1RbBaTh K NbLaCe SrNd P ZrSmEuTi Yb I
I
I
I
I
I
I
I
I
I
I
I
I
9
I
St. Helena
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
---o .... S a r d i n i a . . . . . . 9 ........
9 ..,---~O9
Z.
I
I
ANOROGENICBASIC MAGMAS ~1000
o
-~
(
RbBaTh K NbLaCe SrNd P ZrSmEuTiYb
---~--- E t n a
.,..,
159
"-%
. . . . . . .....
..
10
.,.~ .,--4
1
.
.
.
.
7"7
RbBaTh KNbLaCe SrNd P ZrSmEu TiYb
Fig. 11. Primitive mantle-normalized trace element variation diagrams for anorogenic basic volcanic rocks from the Central Mediterranean region. Data sources: see caption to Fig. 12.
Even for the more primitive magmas, interpretation of the trace element signature can be ambiguous. For example, negative Nb anomalies can indicate either shallow-level crustal contamination or introduction of subducted sediment into the mantle source of the magmas. Plots of Th/Yb versus Ta/Yb (Fig. 12) and Th/Zr versus Nb/Zr (Fig. 13) for primitive mafic magma compositions, clearly discriminate basalts of orogenic from those of anorogenic affinity. The mafic samples (tephrites-phonotephrites) from Mt Vulture plot in an intermediate position between the orogenic and anorogenic fields in both diagrams, consistent with the unique tectonic setting of this volcano.
S r - N d - P b isotopes Diagrams showing the variation of 143Nd/144Nd versus STSr/S6Sr (Fig. 14) and 2~176 versus 2~176 (Fig. 15) clearly distinguish the anorogenic from the subduction-related magmatic suites. These isotope diagrams include all the available data in the literature for both primitive mafic and more differentiated magmatic rocks. Data sources are given in the figure captions. Acid volcanics from the Tuscan province are most likely to have highly radiogenic Sr
isotope compositions as a consequence of high degrees of crustal contamination. The radiogenic Sr isotope compositions of the orogenic volcanic suites from the Aeolian Archipelago, Sardinia, Provence, Campania and Latium most probably reflects a combination of mantle source enrichment by aqueous fluids or silicate melts released from the subducted oceanic crust (which m a y inherit a continental crustal isotopic signature as a consequence of subduction of continentally derived sediments), combined with shallow-level crustal contamination. Anorogenic alkaline magmas from Sicily (Etna and Iblean area) and the Sicily Channel (Pantelleria) have Nd-Sr isotope compositions identical to those of primitive mafic alkaline magmas from central and western Europe (e.g Massif Central, Wilson & Downes 1991; Wilson et al. 1995), inferred to originate by partial melting of a distinct mantle source component within the European upper mantle, the European Asthenospheric Reservoir or E A R , which may be plume-related. The subalkaline (tholeiitic) basalts from the Iblean area of Sicily have N d - S r - P b isotope compositions transitional between those of the E A R and the depleted mantle source of M O R B (DM). The MORBlike basalts from the Tyrrhenian Sea have higher
160
M. WILSON & G. BIANCHINI
Fig. 12. Variation of ThfYb versus Ta/Yb for primitive mafic igneous rocks from the Central Mediterranean region. Data sources: Iblean area, Sicily, Bianchini (1995); Pantelleria, Esperan~a & Crisci (1995); Sardinia, Plio-Pleistocene anorogenic lavas, Rutter (1987; alkaline) and Dostal et al. (1982b; subalkaline); Etna, 1991-1993 eruption Armienti et al. (1994); Treuil & Joron (1994); Mount Vulture, De Fino et al. (1996); Campania, Beccaluva et al. (1991); Aeolian Archipelago, Ellam et al. (1989); Galassi (1995a, b); Vulsini (Latium), Coltorti et al. (1991); Rogers et al. (1985); Ernici (Latium), Civetta et al. (1981): South Tuscany-North Latium high-Mg# samples, Conticelli & Peccerillo (1992).
Fig. 13. Variation of Th/Zr versus Nb/Zr for primitive mafic igneous rocks from the Central Mediterranean region. Data sources: see caption to Fig. 12.
TERTIARY-QUATERNARY MAGMATISM
161
Fig. 14. Variation of 143Nd/144Ndversus 8VSr/86Sr(age corrected) for orogenic and anorogenic magma series from the Central Mediterranean region. Data sources: Iblean area, Beccaluva et al. (1998); Tonarini et al. (1996); Bianchini (1995). Pantelleria, Esperan~a & Crisci (1995). Etna, D'Orazio (1994). Sardinia, Rutter (1987; alkaline); Cioni et al. (1982; subalkaline). Aeolian Archipelago, Ellam et aL (1989); Francalanci et aL (1993); Galassi (1995a, b). Sardinia, Oligocene-Miocene orogenic lavas, Galassi (1995a, b). Provence, Galassi (1995a, b). Mount Vulture, Vollmer, (1976); Hawkesworth & Vollmer (1979). Campania, Vollmer (1976); Hawkesworth & Vollmer (1979); Civetta et al. (1991a, b); Galassi (1995a, b). Latium, Vollmer (1976); Hawkesworth & Vollmer (1979); Rogers et al. (1985). Tuscany, Vollmer (1976); Hawkesworth & Vollmer (1979); Peccerillo et al. (1987).
87Sr/86Sr and 2~176 and lower 143Nd/144Nd ratios than typical depleted mantle, consistent with the involvement of a subduction-related fluid flux in their petrogenesis. The subalkaline Plio-Pleistocene basalts of Mt Arci in Sardinia have Sr isotope compositions intermediate between those of the Plio-Pleistocene alkali basalts and the subduction-related OligoMiocene andesites, consistent with the involvement of a subduction-modified mantle source component in their petrogenesis.
Discussion: Magma source regions and mantle dynamics The generation of Tertiary-Quaternary basaltic magmas within the Mediterranean domain was most probably triggered by adiabatic decompression partial melting of the asthenopheric upper mantle, the solidus of which was locally lowered by the infiltration of slab-derived fluids
above adjacent subduction zones. Throughout the region, both orogenic and anorogenic magmas appear to share a common asthenospheric mantle source component (EAR) which could be plume-related. Locally (e.g. central Spain, French Massif Central), asthenospherederived anorogenic basaltic magmas appear to mix with potassium-rich partial melts of enriched domains within the mantle lithosphere, which modifies their trace element and Sr-Nd-Pb isotope signature (Wilson &Downes 1991; Cebrigl & L6pez-Ruiz 1995). The widespread pollution of the shallow upper mantle by a geochemically distinct mantle plume component (EAR) need not necessarily be a Tertiary phenomenon. Available data (Patterson 1996) suggest that it may have occurred during the Mesozoic. The convective instabilities (mantle diapirs) which appear to trigger the magmatism in many areas have been imaged by seismic tomography (e.g. Granet e t al. 1995). Their scale-length
162
M. WILSON & G. BIANCHINI
Fig. 15. Variation of 2~176
versus 2~176 for orogenic and anorogenic magma series from the Central Mediterranean region. Data sources: see captions to Figs 12 & 14.
suggest that they most probably originate from t h e r m a l b o u n d a r y layers within the upper mantle (400 or 650 km discontinuities). Convective de-stabilization of the upper mantle may have been initiated by the Alpine collision and the consequent global reorganization of plate motions. The distinctive geochemical and isotopic characteristics of the orogenic magmas can be related to the 'pollution' of the shallow asthenospheric mantle by subduction zone fluids/melts, which in some regions (e.g. T u s c a n y - C a m p a n i a - L a t i u m ) carry a particularly strong crustal signature. This may locally reflect subduction of continental lithosphere during the collision of continental micro-plates. The distribution of highly potassic magmas within the Mediterranean region may provide an important indicator of those locations at which collision of continental micro-plates triggered slab break-off. We would like to thank L. Beccaluva for his encouragement and support for this project. Constructive
comments from B. Bonin, L. Jolivet and an anonymous reviewer helped to clarify our ideas.
References ALAVI,M. 1994. Tectonics of the Zagros orogenic belt of Iran: new data and interpretations. Tectonophysics, 229, 211-238. ALTHERR,R., LUGOViC,B., MEYER, H.-P. & MAYER,V. 1995. Early Miocene post-collisional calc-alkaline magmatism along the easternmost segment of the Periadriatic fault system (Slovenia and Croatia). Mineralogy and Petrology, 54, 225-247. APARICO,A., MITJAVILA,J. M., ARANA,g. • gILA, I. M. 1991. La edad del volcanism de las islas Columbrete Grande Alboran (Mediterraneo Occidentale). Boletim Geologica y Minero, 102, 562-570. APPLETON, J. D. 1972. Petrogenesis of potassium rich lavas from the Roccamonfina volcano, Roman Region, Italy. Journal of Petrology, 13, 425-456. ARMIENTI, P., CLOCCHIATFI,R., D'ORAZIO, M., INNOCENTI, F., PETRINI, R., POMPILIO,M., TONARINI,S. & Vn~LAm, L. 1994. The long standing 1991-1993 Mount Etna eruption: petrography and geochemistry of lavas. Acta Vulcanologica, 4, 15-28.
TERTIARY-QUATERNARY MAGMATISM BECCALUVA,L., BIGIOGGERO,B., CHIESA,S., COLOMBO, A., FANTI,G., GATTO, G. O., GREGNANIN,A., MONTRASIO,A., PICCmlLLO, E. M. & TUNESI,A. 1983. Post collisional orogenic dyke magmatism in the Alps. Memorie della Societd Geologica Italiana, 26, 341-359. , BONATTI, E., DuPuY, C., FERRARA, G., INNOCENTI, E, LUCCHINI,E, MACERA, P., PETRINI, R., Rossl, R L., SERRI, G., SEYLER, M. & SIENA, E 1990. Geochemistry and mineralogy of volcanic rocks from ODP sites 650, 651,655, and 654 in the Tyrrhenian sea. Proceedings of the ODP: Scientific Results, 107, 49-74. , BROTZU, P., MACCIOTTA, G., MORBIDELLI, L., SERRI, G. & TRAVERSA, G. 1987. Cainozoic tectonomagmatic evolution and inferred mantle in the Sardo-Tyrrhenian area. The lithosphere in Italy. Advances in Earth Science Research. Accadamia Nazionale Lincei, 80, 229-248. , COLANTONI,P., DI GIROLAMO, P. • SAVELLI.C. 1981. Upper-Miocene submarine volcanism in the Strait of Sicily (Banco senza nome). Bulletin Volcanologique, 44, 573-581. --, COLTORTI, M., GALASSI, B., MACCIOTTA, G. & SIENA, E 1994. The Cainozoic calcalkaline magmatism of the western Mediterranean and its geodynamic significance. Bollettino Geofisica Teoretica ed Applicata, 36, 294-308. - - , DERIU, M., MACCIOTrA, G., SAVELLI. C., VENTURELLI G. 1977. Geochronology and magmatic character of the Pliocene-Pleistocene volcanism in Sardinia (Italy). Bulletin Volcanologique, 40-3, 1-16. - - , DI GIROLAlVIO,P. & SERRI, G. 1991. Petrogenesis and tectonic settings of the Roman volcanic province. Lithos, 26, 191-221. --, GABBIANELLI,G., LUCCHINI, E, ROSSI, P. L. & SAVELL1,C. 1985. Petrology and K/Ar ages of volcanics dredged from the Eolian seamounts: implication for geodynamic evolution of the southern Tyrrhenian basin. Earth and Planetary Science Letters, 74, 187-208. - - , GATTO,G.D., GREGNANIN,A., PICCIRILLO,E.M. & SCOLAm,A. 1979. Geochemistry and petrology of dyke magmatism in the Alto Adige (Eastern Alps) and its geodynamic implications. Neues Jahrbuch fiir Geologic und Palaiiontology Monatshefte, 6, 321-339. --, SIENA, E, COLTORTI, M., Dt GRANDE, A., LO GIUDICE, A., MACCIOTrA, G., TASSINARI, R. & VACCARO, C. 1998. Nephelinitic to tholeiitic magma generation in a transtensional tectonic setting: an integrated petrogenetic model for the Iblean volcanism, Sicily. Journal of Petrology, 39, 1547-1577. BELANTEUR, 0, BELLON, H., MAURY, R. C., OUABADI, A., COUTELLE,A., SEMROUD,B., MEGARTSI,M. & FOURCADE, S. 1995. Le magmatisme miocbne de l'Est Alg6rois gdologie, g6ochimie et g6ochronologic 40K-40Ar. Comptes Rendus de l'Acaddmie des Sciences, Paris, 321, serie II a, 489-496. BELLON, H. 1981. Chronologic radiom6trique (K-Ar) des manifestations magmatiques autour de la
163
M6diterran6e occidentale entre 33-1 MA. F.C. In: WEZEL, E C. (ed.) Sedimentary basin of Mediterranean margins, CNR Italian Project of Oceanography, Tectonoprint Bologna, 341-360. BERGMAN, S. 1987. Lamproites and other potassiumrich igneous rocks: a review of their occurrence, mineralogy and geochemistry. In: FITTON,J. G. & UPTON, B. G. J. (eds) Alkaline Igneous Rocks. Geological Society, London, Special Publications, 30, 103-190. BERRAHMA, M., DELALOYE, M., FAUREMURET,A. & RACHDI, H. 1993. Preliminary and geochronological data on alkaline volcanism of Jbel-Saghro, Anti-Atlas (Morocco). Journal of African Earth Sciences, 17, 333-341. B1ANCHINI, G. 1995. Petrogenesi delle vulcaniti sottomarine dell'area Iblea. PhD thesis, Univ. Ferrara, Italy. BRIQUEU, L., JAVOY,M. LANCELOT,J. R. & TATSUMOTO, M. 1986. Isotope geochemistry, of recent magmatism in the Aegean arc: Sr, Nd, Hf, and O isotopic ratios in the lavas of Milos and Santorini - geodynamic implications. Earth and Planetary Science Letters, 80, 41-54. CAHEN, L., SNELLING,N. J., DELHAL, J. & VAIL, J. R. 1984. The Geochronology and Evolution of Africa. Clarendon Press. CALANCHI,N., COLANTONI,P., RossI, P .L., SAITTA,M. & SERRI, G. 1989. The Strait of Sicily continental rift system: physiography and petrochemistry of the submarine volcanic centres. Marine Geology, 87, 55-83. CEBmA, J. M. & LOeEz-Ruiz, J. 1995. Alkali basalts and leucitites in an extensional intracontinental plate setting: The Late Cenozoic Calatrava Volcanic Province (Central Spain). Lithos, 35, 27-46. C1ONI, R., CLOCCHIATTI, R., DI PAOLA, G. M., SANTACROCE,R. & TONARINI,M. 1982. Miocene calcalkaline heritage in the Pliocene volcanism of Monte Arci (Sardinia Italy). Journal of Volcanology and Geothermal Research, 14, 133-167. CIVETTA,L., CARLUCCIO,E., INNOCENTI,E, SBRANA,A. & TADDEUCCI, G. 1991a. Magma chamber evolution under the Phlegrean Fields during the last 10 Ka: trace element and isotope data. European Journal of Mineralogy, 3, 415-428. --, GALATI, R. d~c SANTACROCE,R. 1991b. Magma 9 mixing and convective compositional layering within the Vesuvius magma chamber. Bulletin Volcanologique, 53, 287-300. --, INNOCENTI, E, MANETrI, P., PECCERILLO,A. & POLI, G. 1981. Geochemical characteristics of potassic volcanics from Mts. Ernici (Southern Latium-Italy). Contribributions to Mineralogy and Petrology, 78, 37-47. CLOCCHIATI'I, R., JORON, J. ~; TREUIL, M. 1988. The role of selective alkali contamination in the evolution of recent historic lavas of Mount Etna. Journal of Volcanology and Geothermal Research, 34, 241-249. CONEY, C. R. 1980. Plate tectonic model for the OligoMiocene evolution of the Western Mediterranean. Tectonophysics, 68, 283-311. COLTORTI,M., DI BATTISTINI,G., NAPPI, G., RENZULLI,
164
M. WILSON & G. BIANCHINI
A. & ZEDA, O. 1991. Structural setting and magmatic evolution of Montefiascone volcanic complex, Vulsini district, Central Italy. Journal of Volcanology of Geothermal Research, 46, 99-124. COMAS, M. C., GARC[A-DUElqAS,V. JURADO,M. J. 1992. Neogene tectonic evolution of the Alboran Sea from MCS data. Geomarine Letters, 12, 157-164. CONDOMINES, M., TANGUY, J. C., KIEFFER, G. & ALLEGRE, C .J. 1982. Magmatic evolution of a volcano studied by 23~ disequilibrium and trace elements systematics: the Etna case. Geochimica et Cosmochimica Acta, 46,1397-1416. CONTICELLI, S. & PECCERILLO,A. 1992. Petrology and geochemistry of potassic and ultrapotassic volcanism in central Italy: petrogenesis and inferences on the evolution of the mantle sources. Lithos, 28, 221-240. D'ORAZIO, M. 1994. Natura ed evoluzione delle vulcaniti dell'Etna e loro relazioni con il magmatismo Ibleo. Doctoral thesis, Universith di Pisa, Italy. DE FINO, M., LA VOLPE, L., PECCERILLO,A., PICCARRETTA, G. &POLI, G. 1986. Petrogenesis of Monte Vulture volcano (Italy): inferences from mineral chemistry, major and trace element data. Contributions to Mineralogy and Petrology, 92, 135-145. DE VECCHI, G. & SEDEA, R. 1995. The Paleogene basalts of the Veneto Region. Memorie della Societd Geologica Italiana, 47, 253-274. DELALOYE, M. & SAWATZKY, G. 1975. G6ochronom6trie des 616ments volcaniques du flysch Helv6tique du synclinal de Th6ne (Haute Savoie, France). Archives du SociOt~ GenOve, 28, 95-99. DEVERCHERE, J., BELLON, H., BESLIER, M-O. ET AL. 1996. Geodynamics of the Ligurian Basins and margins. In: Workshop Abstracts: The Mediterranean Basins: Tertiary extension within the Alpine Orogen. Cergy-Pontoise, Paris, France, December 1996. DEWEY, J. E 1988. Extensional collapse of orogens. Tectonics, 7, 1123-1139. - - , HEMPTON, M. R., KIDD, W. S. E, SAROGLU, E & SENGOR, A. M. C. 1986. Shortenening of continental lithosphere: the neotectonics of Eastern Anatolia-a young collision zone. In: COWARD,M. E & RIES, A. C. (eds) Collision Tectonics. Geological Society, London, Special Publications, 19, 3-36. DOBLAS, M. & OYARZUN, R. 1989. Neogene extensional collapse in the Western Mediterranean (Betic-Rif orogenic belt): Implications for the generation of the Gibraltar arc and magmatic activity. Geology, 17, 430-433. DOCHERTY, C. & BANDA, E. 1995. Evidence for the eastward migration of the Alboran Sea based on regional subsidence analysis: a case for basin formation by delamination of the subcrustal lithosphere? Tectonics, 14, 804-818. DOGLIONI, C., GUEGUEN, E, SABAT,F & FERNANDEZ, M. 1997. The Western Mediterranean extensional basins and the Alpine orogen. Terra Nova, 9, 109-112. DOSTAL, J., COULON, C. & DUPUY, C, 1982a. Cainozoic andesitic rocks of Sardinia (Italy). In: THORPE R. S. (ed.) Andesites, Wiley and Sons Ltd, 353-370.
, DuPuY, C., & VENTURELLI,G. 1982b. Geochemistry of volcanic rocks from the Monte Arci (West Sardinia, Italy). Chemical Geology, 35, 247-264. DOWNES, H. 1987. Tertiary and Quaternary volcanism in the Massif Central, France. In: F~TTON,J. G. & UPTON, B. G. J. (eds) Alkaline Igneous Rocks. Geological Society, London, Special Publications, 30, 517-530. ELLAM, R. M., HAWKESWORTH, C. J, MENZIES, M. A. & ROGERS N. W. 1989. The volcanism of Southern Italy: Role of subduction and the relationship between potassic and sodic alkaline magmatism. Journal of Geophysical Research, 9 4 , 4589-4601. ELLOUZ, N. 1984. Etude de la subsidence de la Tunisie atlasique, orientale et de lamer POlagienne. Th6se 3e cycle, Paris 6. ESPERANfA, S. & CRISCI, G. M. 1995. The Island of Pantelleria: a case for development of DMMHIMU isotopic composition in a long lived extensional setting. Earth and Planetary Science Letters, 136, 167-182. FACCENNA,C., MATTEI, M., FUNICIELLO,R. & JOLIVET, L. 1997. Styles of back-arc extension in the Central Mediterranean. Terra Nova, 9, 126-130. FERRARA, G., PETRINI, R., SERRI, G. & TONARINI, S. 1989. Petrology and isotope-geochemistry of San Vincenzo rhyolites (Tuscany, Italy). Bulletin Volcanologique, 51, 379-388. FERAUD G. 1990. 39Ar/4~ analysis on basaltic lavas series of Vavilov basin, Tyrrhenian sea (Ocean Drilling Program, Leg 107, Holes 655B and 651A). In: Proceedings of the O D P Sci. Results, 107, 93-97. FRANCALANCI, L. & MANETTI, P. 1994. Geodynamic models of the southern Tyrrhenian region: constraints from the petrology and geochemistry of the Aeolian volcanic rocks. Bollenino Geofisica Teoretica Applicata, 36, 283-292. - - , T A Y L O R , S. R., MCCULLOCH,M.T. & WOODHEAD, J. D. 1993. Geochemical and isotopic variations in the calc-alkaline rocks of Aeolian arc, southern Tyrrhenian Sea, Italy: constraints on magma genesis. Contributions to Mineralogy and Petrology, 113, 300-313. FYTIKAS, M., INNOCENTI, E, MANETTI, P., MAZZUOLI, R., PECCEmLLO,A. & VmLAm, L. 1984. Tertiary to Quaternary evolution of volcanism in the Aegean region. In: RO~ERTSON, A. H. E & D~XON, J. G. (eds) Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 687-699. GALASSI, B. 1995a. Evoluzione magmatologica delle associazioni orogeniche cenozoiche del Mediterraneo Centro-Occidentale. Phd thesis, Universitia di Ferrara, Italy. - 1995b. Magmatological evolution of the Cainozoic orogenic associations of the Central-Western Mediterranean. Plinius - European Journal of Mineralogy Italian Supplement, 13, 86-91. GARFUNKEL, Z. 1989. Tectonic setting of Phanerozoic magmatism in Israel. Israeli Journal of Earth Sciences, 38, 51-74.
TERTIARY-QUATERNARY MAGMATISM GENESSEAUX,M. & REHAULT,J. P. 1975. La marge continentale corse. Bulletin de la Soci~tO GOologique de France, 17, 505-518. GIANNERINI, G., CAMPREDON,R. ~:; FERAUD, G. 1988. Gen~se des rifts intracontinentaux par propagation polygonale de transformantes continentales: exemple du triplet golfe d'Aden, mer Rouge, Afrique de l'Est. Comptes Rendus de l'Acaddmie des Sciences, Paris, 306, Sdrie II, 1507-1514. GIANNETTI, B. & ELLAM, R. 1994. The primitive lavas of Roccamonfina volcano, Roman region, Italy: new constraints on melting processes and source mineralogy. Contributions to Mineralogy and Petrology, 116, 21-31. GIARDINI, D & VELONA,M. 1991. The deep seismicity of the Tyrrhenian Sea. Terra Nova, 3, 57-64. GILLOT, R Y., KIEFFER, G. & ROMANO, R. 1994. The evolution of Mount Etna in the light of potassiumargon dating. Acta Volcanologica, 5, 81-87. GOMEZ, E, BARAZANGI, M. & BENSAID, M. 1996. Active tectonism in the intracontinental Middle Atlas Mountains of Morocco: synchronous crustal shortening and extension. Journal of the Geological Society, London, 153, 389-402. GRANET,M.,WILSON,M. & ACHAUER,U. 1995. Imaging a mantle plume beneath the French Massif Central. Earth and Planetary Science Letters, 136, 281-296 GUEGUEN, E., DOGLIONI, C. & FERNANDEZ,M. 1997. Lithospheric boudinage in the Western Mediterranean back-arc basin. Terra Nova, 9, 184-187. GUIRAUD, R. & MAURIN, J. C. 1991. Le rifting en Afrique au Cr6tac6 inf6rieur: synthbse structurale, mise en 6vidence de deux 6tapes dans la genese des bassins, relations avec les ouvertures oc6aniques p6ri-africaines. Bulletin de la SociOt~ G~ologique de France, 162, 811-823. GUIRAUD, R., BINKS,R. M., FAIRHEAD,J. D. ~%WILSON, M. 1992. Chronology and geodynamic setting of Cretaceous-Cenozoic rifting in West and Central Africa. Tectonophysics, 213, 227-234. G12LE~, N. 1991. Crust-Mantle interaction in Western Turkey: Implications from Sr and Nd geochemistry of Tertiary and Quaternary volcanics. Geological Magazine, 128, 417-435. HAMMUDA, O. S., VAN HINTE, J. E. & NEDERBRAGT,S. 1992. Geohistory Analysis Mapping in Central and Southern Tarabulus Basin, Northwestern Offshore of Libya. In: SALEM,M. J. (ed.) The geology of Libya. Elsevier, Amsterdam, 1657-1680. HAWKESWORTH,C. J. & VOLLMER,R. 1979. Crustal contamination versus enriched mantle: 143Nd/144Nd and 87Sr/a6Sr evidence from the Italian volcanics. Contributions to Mineralogy and Petrology, 69, 151-165. HERNANDEZ, J., LAROUZIERE DE, E D., BOLZE, J. BORDEr, R 1987. Le magmatisme n6ogbne b6ticorifain et le couloir de d6crochement transAlboran. Bulletin de la SociOtO GOologique de France, III, 2, 257-267. - - & LEPVRIER,C. 1979. Le volcanisme calco-alcalin miocene de la r6gion d'Alger (Alg6rie): p6trologie et signification g6odynamique. Bulletin de la Socidt6 GOologique de France, XXI, 73-86.
165
KARAMATA, S., PAVICEVIC, K., KORIKOVSKIJ, S. K., BORONIHIN,V. A. & AMTHAUER, G. 1994. Petrology and mineralogy of Neogene volcanic rocks from the Allchar area, the FY Republic of Macedonia. Neues Jahrbuch fiir Mineralogie Abhandlung, 167, 317-328. KASTENS, K., MASCLE, J., ET AL. 1990. The geological evolution of the Tyrrhenian Sea: an introduction to the scientific results of ODP Leg 107. In: KASTENS,K., MASCLE,J. ET AL. Proceedings of the ODP, Scientific Results, 107, 3-26. KOLIOS,N., INNOCENTI,E, MANETTI,P., PECCERILLO,A. & GIULIANI, O. 1980. The Pliocene volcanism of the Voras Mts (Central Macedonia, Greece). Bulletin Volcanologique, 43, 553-568. LONERGAN, L. & WHITE, N. 1997. Origin of the BeticRif mountain belt. Tectonics, 16, 504-522. LOUNI-HACINI, A., BELLON, H., MAURY, R.C., MEGARTSI,M., COULON,C., SEMROUD,B., GOTTEN, J. & COUTELLE,A. 1995. Datation 40K-40Ar de la transition du volcanisme calco-alcalin au volcanisme alcalin en Oranie au Miocbne supdrieur. Comptes Rendus de l'AcadOmie des Sciences, Paris, 321, serie II a, 975-982. LOPEZ-RuIz, J., CEBRIA,J. M., DOBLAS, M., OYARZUN, R., HoYOS, M. & MARTIN,C. 1993. The late Cenozoic alkaline volcanism of the Central Iberian Peninsula (Calatrava Volcanic Province, Spain): Intra-plate volcanism related to extensional tectonics. Journal of the Geological Society, London, 150, 915-922. MACCIOTTA, G., VENTURELLI, G. d~; BECCALUVA, L. 1978. Geochemistry of mafic Cainozoic volcanic rocks from Sardinia (Western Mediterranean). Bulletin Volcanologique, 41, 56-78. MANTOVANI,E., ALBARELLO,D., TAMBURELLI,C., BABBUCCI, D. & VITI, M. 1997. Plate convergence, crustal delamination, extrusion tectonics and minimization of shortening work as a main controlling factor of the recent Mediterranean deformation pattern. In: MANTOVANI,E. (ed.) Geodynamics of the Mediterranean Region and its implications. Annali di Geofisica, 40, 611-643. MERGOIL, J. & BOIVIN,P. 1993. Le Velay. Son volcanisme et les formations associ~es. Notice de la carte 1/100 000. Geologie de la France. Orl6ans, BRGM. 3. MICHARD, A. 1976. ElOments de GOologie Marocaine. Notes et Mdmoires du Service G~ologique Maroc, 252. MILANI, X. 1996. Petrogenesis and metallogenic aspects of Tertiary volcanics of the Euganean Hills magmatic complex (Venetian Province). Plinius, European Journal of Mineralogy Italian Supplement, 15, 129-132. MONIE, E, MONTIGNY, R. & MALUSKI, H. 1992. Age burdigalien de la tectonique ductile extensive dans le massif de l'Edough (Kabylie, Alg6rie). Donn6es radiom6triques 39Ar-4~ Bulletin de la Socidt~ G~ologique de France, 163, 571-584. MONTANINI, A. & VILLA, I. M. 1993. 39Ar-4~ chronostratigraphy of Monte Arci volcanic complex (Western Sardinia, Italy). Acta Vulcanologica, 3, 229-233.
166
M. WILSON & G. B I A N C H I N I
MOl~Er~, J.-M. & CABANIS,B. 1993. Mise en 6vidence d'une association magmatique dans le volcanisme plio-quaternaire du Moyen-Atlas marocain. Comptes Rendus de l'Acad~mie des Sciences, Paris, 316, S6rie II, 537-362. MORRA, V., SECCHI, F .A.G., MELLUSO, L. & FRANCIOSl, L. 1997. High-Mg subduction related Tertiary basalts in Sardinia, Italy. Lithos, 40, 69-91. NIXON, P. E., THIRLWALL,M. F., BUCKLEY,E & DAVIES, C. J. 1984. Spanish and Western Australian lamproites: aspects of whole rock geochemistry. In: KORNPROBST,J. (ed.) Kimberlites I: Kimberlites and Related Rocks. Elsevier, Amsterdam, 285-296. NOTSU, K., FUJITANI,T., UI, T., MATSUDA,J. & ERCAN, T. 1995. Geochemical features of collision-related volcanic rocks in central and eastern Anatolia, Turkey. Journal of Volcanology and Geothermal Research, 64, 171-192. ORAL, M. B., RELINGER, R. E.,TOKSOZ, M. N., KING, R. W., BARKA, A. A., KINIK, I. & LENK, 0. 1995. Global positioning system offers evidence of plate motions in eastern Mediterranean. EOS Transactions of the AGU, 76, 1-3. OTTAVIANI-SPELLA,M.-M, GIRAD, M. & CHEILLETZ,A. 1996. Les ignimbrites burdigaliennes du Sud de la Corse. Comptes Rendus de l'AcadOmie des Sciences, Paris', 323, serie II a, 771-778. PAM1C, J. 1993. Eoalpine to Neoalpine magmatic and metamorphic processes in the northwestern Vardar Zone, the Easternmost Periadriatic zone and the southwestern Pannonian Basin. Tectonophysics, 226, 503-518. PATTERSON, R. 1996. Geochemical characteristics and geodynamic setting of Cenozoic mafic magmatism in western and central Europe. PhD thesis, University of Leeds, UK. PEARCE, J. A., BENDER, J. E, DELONG, S. E., KIDD,W. S. E, Low, R J., GONER, Y., SAROGLU,E, YILMAZ,Y., MOOI~BATH,S. & MYTCHELL,J. G. 1990. Genesis of collision volcanism in Eastern Anatolia, Turkey. Journal of Volcanology and Geothermal Research, 44, 189-229. PECCERILLO, A., CONTICELLI, S. & MANETTI, E 1987. Petrological characteristics and the genesis of the recent magmatism of South Tuscany and North Latium. Period. Mineral., 56, 167-183. PE-PIPER, G. 1994. Lead isotopic composition of Neogene volcanic rocks from the Aegean extensional area. Chemical Geology, 118, 27-41. & PWER, D. J. V. 1989. Spatial and temporal variation in Late Cenozoic back-arc volcanic rocks, Aegean Sea region. Tectonophysics, 169, 113-134. & -1994. Miocene magnesian andesites and dacites, Evia, Greece: adakites associated with subducting slab detachment and extension. Lithos, 31, 125-140. PICCOLI, G. 1970. Outlines of volcanism in northern Tripolitania (Libya). Bollettino della Societd Geologica Italiana, 89, 449-461. PINARELLI,L. 1987. Considerazioni geochimiche-petroIogiche sul vulcanismo acido della Provincia Magmatica Toscana. Univ. Firenze, Italy. PLATT, J. P. & V1SSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: A -
-
-
-
working hypothesis for the Alboran Sea and the Gibraltar Arc. Geology, 17, 540-543. - - , SOTO, J. I., COMAS, M. C. & LEG 161 SHIPBOARD SCIENTISTS 1996. Decompression and high temperature, low-pressure metamorphism in the exhumed floor of an extensional basin, Alboran Sea, western Mediterranean. Geology, 24, 447-45O. RACHDI, H., BERRAHMA, M., DELALOYE, M., FAUREMURET,A. & DAHMANI M. 1997. Tertiary volcanism in Rekkame (Morocco) - Petrology, geochemistry and geochronology. Journal of African Earth Sciences, 24, 259-269. REHAULT,J. P., BOILLOT,G. & MAUFFRET,A. 1984. The Western Mediterranean basin geological evolution. Marine Geology, 55,447-477. ROBERT, U., FODEN, J. & VARNE, R. 1992. The Dodecanese Province, SE Aegean: a model for tectonic control on potassic magmatism. Lithos, 28, 241-260. ROBERTSON,A. H. E & GRASSO,M. 1995. Overview of the Late Tertiary-Recent tectonic and paleoenviromental development of the Mediterranean region. Terra Nova, 7, 114-127. ROGERS, N. W., HAWKESWORTH,C. J,, PARKER, R. J. & MARSH, J. S. 1985. The Geochemistry of potassic lavas from Vulsini, Central Italy and implications for mantle enrichment processes beneath the Roman region. Contributions to Mineralogy and Petrology, 90, 224-257. ROMANO,R. 1983. Succession of the volcanic activity in the Etnean area. Memorie della Societd Geolologica Italiana, 23, 27-28. RtJrrER, M. J. 1987. Evidence for crustal assimilation by turbulently convecting, mafic alkaline magmas: geochemistry of mantle xenoliths-bearing lavas from Northern Sardinia. Journal of Volcanology and Geothermal Research, 32, 343-354. SAVELI,I, C. 1988. Late Oligocene to Recent episodes of magmatism in and around the Tyrrhenian Sea: implications for the process of opening in a young inter-arc basin of intra-arc-orogenic (Mediterranean) type. Tectonophysics, 146, 163-181. - - - , BECCALUVA, L., DERIU, M., MACCIOTTA, G. & MACCIONI, L. 1979. K/Ar geochronology and evolution of the Tertiary "calc-alkalic" volcanism of Sardinia (Italy). Journal of Volcanology and Geothermal Research, 5, 257-269. dh; GASPAROTTO,G. 1994. Calc-alkaline magmatism and rifting of the deep water volcano of Marsili (Aeolian back-arc, Tyrrhenian sea). Marine Geology, 119, 137-157. SAWAF,T., AL-SAAD, D., GEBRAN, A., BARAZANGI,M., BEST, J. A. & CHAIMOV,T. A. 1993. Stratigraphy and structure of eastern Syria across the Euphrates depression. Tectonophysics, 220, 267-281. SENGOR, A. M. C. 1993. Some current problems on the Tectonic evolution of the Mediterranean during the Cainozoic. In: B o s c m E., MANTOVANIE. 8,: MORELLI A. (eds) Recent volution and seismicity of the Mediterranean Region. Nato ASI Series C, D Reidel, Dordrecht-Boston, 402, 403-422. & YILMAZ, Y. (1981) Tethyan evolution of -
-
T E R T I A R Y - Q U A T E R N A R Y MAGMATISM Turkey: a plate tectonic approach. Tectono-
physics, 75, 181-241. , CIN, A., ROWLEY, D. B. & NIE, S. Y. 1993. Spacetime patterns of magmatism along the Tethysides: a preliminary study. Journal of Geology, 101, 51-84. , GOROR, N. & SAROGLU,E 1985. Strike-slip faulting and related basin formation in zones of tectonic escape: Turkey as a case study. In: B1DDLE, K. T. & CRISTIE, B. N. (eds) Strike-slip defor-
mation,
basin formation and sedimentation.
Society of Economic Paleontologists and Mineralogists. Special Publications, 37, 227-64. SERRI, G. 1990. Neogene-Quaternary magmatism of the Tyrrhenian region: characterization of the magma sources and geodynamic implications. Memorie della Societti Geologica Italiana, 41, 219-242. - - , INNOCENTI,E & MANETTI,P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism of central Italy. Tectonophysics, 223, 117-147. SEYITOGLU,G. & SCOTT,B. C. 1992. Late Cenozoic volcanic evolution of the northeastern Aegean region. Journal of Volcanology and Geothermal Research, 54, 157-176. - - , ANDERSON,D., NOWELL,G. & SCOTT,B. 1997. The evolution from Miocene Potassic to Quaternary sodic magmatism in western Turkey: implication for enrichment processes in the lithospheric mantle. Journal of Volcanology and Geothermal Research, 79, 127-148. SIENA, E ~ COLTORTI,M. 1989. Lithospheric mantle evolution: Evidences from ultramafic xenoliths in the Lessinian volcanics (northern Italy). Chemical Geology, 77, 47-364. SPADINI, G., BERTOTTI, G. & CLOETINGH, S. 1995. Tectono-stratigraphic modelling of the Sardinian margin of the Tyrrhenian Sea. Tectonophysics, 252, 269-284. SPAKMAN, W. 1990. Images of the upper mantle of central Europe and the Mediterranean. Terra Nova, 2, 542-553. , VAN DER LEE, S. AND VAN DER HILST,R.D. (1993) Travel-time tomography of the EuropeanMediterranean mantle down to 1400 km. Physics of the Earth and Planetary Interiors, 79, 3-74. STOPPA,E & CUNDAR[,A. 1995. A new Italian carbonatite occurrence at Cupaelto (Rieti) and its genetic significance. Contributions to Mineralogy and Petrology, 122, 275-288. TONARINI, S., D'ORAZIO, M., ARMIENTI,P., INNOCENTI, E, SCRIBANO V. 1996. Geochemical features of eastern Sicily lithosphere as probed by Hyblean xenoliths and lavas. European Journal of Mineralogy, 8, 1153-1173. TORRES-ROLDAN, R. L., POLI, G. & PECCERILLO, A. 1986. An early Miocene arc-tholeiitic magmatic dyke event from the Alboran Sea: Evidence for pre-collisional subduction and back arc crustal extension in the westernmost Mediterranean. Geologisches Rundschau, 71, 219-234.
167
TREUIL, M. & JORON, J.-L. 1994. Etude g6ochimique des 616ments en traces dans les laves 6mises au cours de l'6ruption 1991-1993 de l'Etna. Mise en evidence des contributions de la source, de la fusion partielle, de la differenciation et des modalitds de tranfert des magmas. Acta Volcanologica, 4, 29-45. TRICART, P., TORELLI, L., ARGNANI, A., REKHISS, E & ZITELLINI,N. 1994. Extensional collapse related to compressional uplift in the Alpine Chain off northern Tunisia (Central Mediterranean). Tectonophysics, 238, 317-329. VAN DER MEER, E & CLOETINGH, S. 1993. Intraplate stresses and the subsidence history of the Sirte basin (Libya). Tectonophysics, 226, 37-58. VENTURELLI, G., CA PEDRI, S., DI BATTISTINI,G., CRAWFORD, A., KOGARKO, L. N. & CELESTINI,S. 1984a. The ultrapotassic rocks from southeastern Spain. Lithos, 17, 37-54. - - , THORPE, R. S., DAL PIAZ, G. V., DEL MORO, A. & POTTS, E J. 1984b. Petrogenesis of calc-alkaline, shoshonitic and associated ultrapotassic Oligocene volcanic rocks from the Northwestern Alps, Italy. Contributions to Mineralogy and Petrology, 86, 209-220. VIGLIOTTI, L.& LANGENHEIM, V. E. 1995. When did Sardinia stop rotating? New paleomagnetic results. Terra Nova, 7, 424-435. VISSERS,R. L. M., PLATT,J. P. & VANDER WAAL, D. 1995. Late orogenic extension of the Betic Cordillera and Alboran Domain: A lithospheric view. Tectonics, 14, 786-803. VOLLMER, R. 1976. Rb-Sr and U - T h - P b systematics of alkaline rocks: the alkaline rocks from Italy. Geochimica et Cosmochimica Acta, 40, 283-295. VON BLANCKENBURG, E & DAVIES, J. H. 1995. Slab breakoff: a model for syncollisional magmatism and tectonics in the Alps. Tectonics, 14, 120-131. WEZEL, E C. 1976. Widespread manifestations of Oligocene-Lower Miocene volcanism around the Western Mediterranean. In: BIJU-DUvAL, X. & MONTADERT,X. (eds) International symposium on
the structural history of the Mediterranean basin. Split, Yugoslavia. Editions Technip, Paris, 287-302. WILSON, M. 1989. Igneous Petrogenesis. International Thompson, London. & DOWNES, H. 1991. Tertiary-Quaternary extension related alkaline magmatism in Western and Central Europe. Journal of Petrology, 32, 811-849. & -1992. Mafic alkaline magmatism in the European Cenozoic rift system. Tectonophysics, 173-182. & GUIRAUD, R. 1998. Late Permian to Recent magmatic activity on the African-Arabian margin of Tethys. In: MACGREGOR. D. CLARK-LOWES,D. D. & MOODY, R. (eds) Petroleum Geology of North Africa. Geological Society, London, Special Publications, 132, 231-263. - - - , DOWNES, H. & CEBRIA, J.-M. 1995. Contrasting fractionation trends in coexisting continental alkaline magma series; Cantal, Massif Central, France. Journal of Petrology, 36, 1729-53.
-
-
-
-
-
-
2 0 8 ,
168
M. WILSON & G. BIANCHINI
-,TANKU'I,A. &GOLE~ N. 1997. Tertiary volcanism of the Galatia Province, Central Anatolia, Turkey. Lithos, 42, 105-121. YILMAZ,Y. 1990. Comparison of young volcanic associations of western and eastern Anatolia formed under a compressional regime: a review.
1992. Very high rates of cooling and uplift in the Alpine belt of the Betic Cordilleras, southern Spain. Geology, 20, 79-82. ZmGLER, R A. 1992. European Cenozoic rift system. In: ZIEGLER, P. A. (ed.) Geodynamics of Rifting,
Volume I. Case History Studies on Rifts: Europe and Asia. Tectonophysics, 208, 91-111.
Journal of Volcanology and Geothermal Research, 44, 69-87. 1993. New evidence and model for the evolution of the southeast Anatolian orogen. Geological Society of America Bulletin, 195, 251-271. ZECK, H. R, MONIE, P., VILLA, I. M. & HANSEN, B. T.
-
-
--,
CLOETINGH, S. & VAN WEES, J.-D. 1995. Dynamics of intra-plate compressional deformation: the Alpine foreland and other examples. Tectonophysics, 252, 7-59.
Crustal structure of the North Tyrrhenian Sea: first result of the multichannel seismic LISA cruise A. M A U F F R E T
& I. C O N T R U C C I
D @ a r t e m e n t de Gdotectonique, Case 129, Universitd Pierre et M a r i e Curie, 4 Place Jussieu, 75252, Paris C e d e x 05, France
Abstract: During a deep-penetration multichannel seismic cruise, the North Tyrrhenian basin was investigated. This basin has a shallow Moho along the Italian coast. Two kinds of dipping reflections were identified in the crust. The westward-dipping reflections are interpreted as Apennines or/and Alpine thrusts whereas the eastward-dipping reflections may be the seismic expression of extensional detachments. The deep part of the 10 km thick Corsica Basin is well imaged on the continental shelf. This basin was formed before or during the opening of the Balearic-Ligurian basin and is related to the collapse of the Alpine Corsica belt. We emphasize the active role of plutonic bodies that rise from a 3 km depth during the mid-Pliocene. The thinning of the North Tyrrhenian crust could be explained by a delamination of the Adriatic lithosphere that retreats towards the east.
The formation of extensional basins in convergent setting is a major problem of Earth sciences (Faccena et al. 1996). In the past years formation of marginal basins behind subduction zones was the best explanation for extensional tectonics. Indentation theory was an alternate hypothesis to explain extension in the collided plate. In recent years several studies highlighted the collapse of former orogens to explain the extension. The Mediterranean Sea is an ideal place to study the formation of extensional basins that were fully developed in the Western Mediterranean basins (Balearic and Ligurian Seas) with early Miocene emplacement of oceanic crust. The North Tyrrhenian Basin is underlain by continental basement and its evolution is very recent (late Miocene to Pliocene). Consequently in this region we can study the young extensional processes that are inactive in the others basins of the Western Mediterranean Sea.
Tectonic framework The African plate has converged towards the E u r o p e a n plate since the Late Cretaceous (Olivet 1987). The Adriatic Promontory (Fig. 1) represents a n o r t h e r n prolongation of the African plate (Alvarez 1991). This promontory collided with E u r o p e during the Cretaceous-Eocene with the closure of the LiguroPiemont Ocean (Stampfli & Marchant 1997) and formation of the Alps. During this event the Corsican Alps were overthrust onto western Hercynian Corsica (Mattauer et al. 1981). Recent dates, obtained (Brunet et al. 1997) by the 39Ar/4~ method, suggest a Late Cretaceous
(65 Ma) to late Eocene (37 Ma) age for the emplacement of Corsican nappes. Soon after this compressional episode a large extension occurred in the European plate with formation of Oligocene grabens (Rhine, Limagnes and Bresse Basins; Fig.l). The subsequent formation of oceanic crust in the Balearic and Ligurian Seas and the coeval rotation of Corsica-Sardinia block occurred (Cravatte et al. 1974; Montigny et al. 1981; Rehault et al. 1984) from the mid-Aquitanian to Burdigalian (22-19 Ma). During the late Oligocene (33-29 Ma and 25 Ma; Brunet et al. 1997) Alpine Corsica underwent an extensional event (Jolivet et al. 1990, 1991, 1994; Daniel & J o l i v e t 1996) while compression migrated towards the east, in the inner Apennines Belt (27 Ma; Carmigniani & Kligfield 1990), and then in the present-day front of deformation along the Adriatic coast of Italy (Fig. 1). During the same time span extension also migrated from the North Tyrrhenian Sea to the central part of Apennines Belt (Elter et al. 1975; D'Offizi et al. 1994).
Geological and geophysical setting of the North Tyrrhenian Basin This region is limited (Figs 2 and 3) by Corsica to the west and the Tuscany coast of Italy to the east, the north Ligurian Sea to the north and the South Tyrrhenian sea to the south. Four physiographic provinces can be distinguished (Fig. 2): the Corsica Basin with moderate depth and smooth topography, a central ridge where Capraia, Elba, Pianosa, and M o n t e Cristo
MAUFFRET,A. dCz;CONTRUCCI,I. 1999. Crustal structure of the North Tyrrhenian Sea: first result of the muitichannel seismic LISA cruise. In: DURAND,B., JOUVE1,L., HORVATH,F. & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 169-193.
170
A. MAUFFRET & I. CONTRUCCI European Plate Iberian Plate African Plate
Molassic Basin
Adriatic Plate Austro-Alpine appes
Po Basin
i " I
"-,..J ~
v--,c_,L.gJ ~-,
Apennines Brianconnais and Alpine Corsica
I Neogene basin Oceanic crust
v
0 I
500 km I
I
I
I
i
Fig. 1. Tectonic framework of the North Tyrrhenian Basin. From Stampfli & Marchant (1997) modified.
Islands are located, the Tuscany shelf and the north-south-trending Etruschi and Cialdi seamounts that formed the transition with the deeper South Tyrrhenian Basin. The Corsica basin has an onshore extension (Aleria Basin) flanked by Alpine Corsica (Fig. 3). The central high is divided into two ridges: Pianosa sedimentary ridge and Capraia-Monte Cristo Ridge constituted of recent granitic intrusions (Elba, Monte Cristo) and volcanic extrusion (Capraia). Similarly, the Giglio Island intrusion is a prominent feature of the Tuscany Shelf that is characterized by a horst and graben structure (Bartole et al. 1991; Bartole 1995). Monte Etruschi and Cialdi are also recent extensional features (Zitellini et al. 1986). The main offshore structures of the North Tyrrhenian Sea have been investigated by an industrial multichannel seismic survey described by Bartole et al. (1991) and Bartole (1995). Two exploratory wells located on the Capraia ridge (Martina and Mimosa, Fig. 3) help to calibrate
the seismic sections. However the results of coring and dredging surveys (Aleria 1980) show that this ridge has a composite and irregular basement (Upper Cretaceous and Eocene; Fig. 3). Moreover, Triassic rocks crop out in Africa islet (Figs 2 and 3). Therefore, the results of the exploratory wells projected on a distant seismic line (Fig. 4a) may not be accurate. The line drawing L 122 (Fig. 4a; Bartole et al. 1991) illustrates the main feature of the North Tyrrhenian Sea. The Ligurian nappes are characterized by eastern vergent thrusts. Martina well bottomed into Palaeocene-middle Eocene layers of Alpine affinity that are overlain unconformably by Oligocene to early Miocene pelites. This layer, 1 km thick in the Marina and Mimosa wells, is not deformed by compression, but affected by the Tortonian to early Pliocene extension. However, the seismic sequence attributed to the Oligocene (Bartole et al. 1991) may be younger if the correlation between the well and the seismic profile is wrong. The Ligurian nappes have been
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
171
Fig. 2. Track map of the LISA cruise. Bathymetric contour: 0.2 km interval. The position of the seismic profiles shown are indicated by the number of the figures. emplaced in an accretionary prism (Principe & Treeves 1984) that has been overprinted by the late Oligocene collision between Corsica and Adria continental crust (Keller & Pialli 1990;
D'Offizi et al. 1994; Keller & Coward, 1996). In this context the Oligocene-early Miocene unit is interpreted as piggy back deposits if the correlation between the line 122 and the Marina well
172
A. MAUFFRET & I. CONTRUCC1
Fig. 3. Structural map of the North Tyrrhenian Basin. Extension of the synrifts sediments are from Bartole et (1991); Bartole (1995). Results of sampling are from Aleria (1980) and Bartole et al. (1991).
al.
Fig. 4. (a) Line drawing of the industrial seismic L 122 redrawn from Bartole et al. (1991). (b) Transect crossing the North Tyrrhenian Basin constrained by refraction (Hirn & Sapin 1976; Letz et al. 1977a,b; Egger et al. 1988; Egger 1992) and gravity (Carrozzo & Nicolich 1977) data. The depth of the Corsica Basin is only 5 km deep in the northern part of the Corsica Basin; it is much deeper in the southern part. Note the thin crust beneath the Tuscany Margin. The refractor indicating a deep European Moho (Letz et al. 1977a,b) could be a multiple (Ponziani et al. 1995).
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
173
174
A. MAUFFRET & I. CONTRUCCI
is correct (Bartole et al. 1991). In an alternate hypothesis this unit may represent the base of the sedimentary fill deposited in the Corsica fore-arc basin (Rehault 1981). The acoustic basement off Elba Island is supposed to be formed by the Tuscan metamorphic rocks that outcrop in the island (Fig. 4a; Bartole et al. 1991; Keller & Coward 1996), but this basement may be also representative of the 6.2-5.1 Ma old (Saup6 et al. 1982)granitic intrusion of Porto Azzuro located on the eastern coast of Elba Island. The refraction ( H i r n & Sapin 1976; Letz et al. 1977a,b; Egger et al. 1988; Egger 1992) and gravity studies (Carrozzo & Nicolich 1977) show the transition from the 32 km thick crust of Corsica (Fig. 4b) to the 22 km thick thinned crust of Tuscany. The base of the Corsica basin is 5.2
km deep (Carrozzo & Nicolich 1977) that corresponds to 3.6 s two-way travel time (TWTT). However, this transect crosses the northern part of the Corsica basin whereas the deepest part of the basin is located southwards (see later). In the former refraction experiments a 50-70 km deep 'European' Moho was supposed beneath the 'Italian' Moho (Letz et al. 1977a,b) but a recent study (Ponziani et al. 1995) show that the supposed refracted deep arrival was in fact a multiple artifact. The penetration of the industrial multichannel seismic (MCS) lines in the north Tyrrhenian Sea are no greater than 4 s TTWT (Fig. 4a). On the other hand the refraction results (Fig. 4b) penetrated deeply, but do not show the fine structure. The Lisa MCS lines fill the gap between the two types of seismic data.
Fig. 5. Eastern part of the LISA 7 seismic profile showing dipping reflections with Apennine vergence. A 7 s TWTT deep reflector is correlated with the Moho identified on the refraction line shown in Fig. 4b. The horizontal reflection at 7.5 s is an artifact generated by the migration processing. Position of the seismic profile is indicated in Fig. 2.
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
LISA Cruise A multichannel seismic cruise (LISA, Fig. 2) was carried out in 1995 on the RV Nadir in the Western Mediterranean Sea. We used a 2.4 km streamer, 96 channels, towed 20 m below sea level. The same depth was adopted for the immersion of 10 GI guns with a total volume of 1140 cubic inches. These guns were tuned in single bubble mode (Avedik et aL 1993). This special mode and the deep immersion of streamer and guns generate low frequencies and allow deep penetration of the crust, although the volume of guns is relatively small. We presented the results of on board processing: velocity analysis, normal move out correction and stack. LISA seismic data need a deconvolution and filtering in the F-K domain. Multiple and reverberation is an acute problem, but we were lucky that the primary reflections are not parallel to the sea floor and can be distinguished, particularly in the Corsica Basin, through the multiples.
Main results of the LISA cruise Structures with Apenninic
vergence
T h e east vergent A p e n n i n e s thrusts are identified by p r o m i n e n t reflections that dip towards the west (Bartole et al. 1991; B a r t o l e 1995).
175
T h e s e reflections w e r e not visible b e l o w 4 s T W T T in the industrial MCS lines (Fig. 4a). T h e Lisa 7 profile (Fig. 5) shows w e s t w a r d - d i p p i n g reflectors as d e e p as 7 s TWTT. A t this d e p t h an horizontal reflector can be c o r r e l a t e d to the M o h o (a d e p t h of 22 km, Fig. 4b, c o r r e s p o n d s to 7 s T W T T ) . T h e s e d i m e n t a r y c o v e r of t h e Capraia Basin is separated into two parts (Fig. 5) by a p r o m i n e n t unconformity, probably Messinian in age. T h e strong reflection, which is linked to this u n c o n f o r m i t y , is characteristic of the Messinian level s h o w e d in the Corsica basin (Fig. 6), although Bartole et al. (1991) attributed a m i d - P l i o c e n e age to this unconformity. T h e u p p e r part of the s e d i m e n t a r y fill is horizontal w h e r e a s the lower part is affected by extensional m o v e m e n t s . T h e base of the s e d i m e n t a r y fill could be O l i g o c e n e (Bartole et al. 1991). We n o t e d that the Oligocene, if t h e assumption of Bartole et al. (1991) is correct, is not involved in the thrusts, which can be d a t e d as late E o c e n e . In this case the f o r m a t i o n of thrusts is coeval to the A l p i n e Corsica c o m p r e s s i o n with already an A p e n n i n i c vergence. S o m e n o r m a l faults s e e m to b e b r a n c h e d a n d r e a c t i v a t e d t h e f o r m e r thrusts (200-250 shot poins, Fig. 5).
Fig. 6. LISA 7 seismic profile. The northern Corsica Basin is relatively shallow (3.5 s TWq-T). The Messinian reflector is prominent and cut by a normal fault. Position of the seismic profile is indicated in Fig. 2.
176
A. MAUFFRET & I. CONTRUCCI
Fig. 7. Western part of LISA 8 seismic profile. The northern Corsica Basin is relatively shallow (3 s TWTT). The Messinian reflector is upturned on the western flank of the Pianosa Ridge. Burdigalian sediments have been sampled by coring (Viaris de Lesegno 1978) on the Pianosa Ridge near the seismic profile. 3 km of uplift are calculated if the erosion on the continental shelf of Elba Island is taken into account. The chaotic configuration of the seismic reflectors below the Messinian unconformity indicates that erosion on Elba Island was already active during the late Messinian. Position of the seismic profile is indicated in Fig. 2.
Corsica Basin (Figs 6 to 11) In the onshore extension of the Corsica Basin (Aleria Basin) the marine lower Pliocene overlies the Messinian erosional unconformity. Beneath this unconformity lies the upper Tortonian which rest u n c o n f o r m a b l y upon the L a n g h i a n (middle Miocene). The Miocene layers has a 20 ~ dip towards the east (OrzagSperber & Pilot 1976). In the Corsica basin the P l i o c e n e - Q u a t e r n a r y sedimentary unit is limited at its base by a prominent reflector that is Messinian in age
(Viaris de Lesegno 1978; Viaris de Lesegno et al. 1978). The lower part of the Pliocene-Quaternary unit forms a wedge that thins towards the east (Figs 6 and 7). This sedimentary wedge shape suggests that the main source of sediments is Corsica (Viaris de Lesegno, 1978; Viaris de Lesegno et al. 1978), although we cannot exclude a recent tilting of Corsica in the n o r t h e r n m o s t seismic profile (Fig. 6). The Messinian reflector is affected by normal faulting and uplifted near Corsica in the northern part of the Corsica Basin (Fig. 6). The normal fault is also evident in the early Pliocene. The
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
177
Fig. 8. Western part of LiSA 10 seismic profile. The Corsica Basin is 5-6 s TWTT thick (horizontal arrows show the real reflectors, dipping arrows show the multiples) Pianosa Ridge is an uplifted part of the Corsica Basin. The Messinian reflector and a Miocene horizon are 1.1 km and 2.4 km uplifted respectively. Position of the seismic profile is indicated in Fig. 2.
Messinian reflector is flat and deep (more than 2 s TWTT) in the middle part of the basin (Figs 7 and 8). It is shallower and cut by several canyons in the southern part of the study area (Figs 10 and 11). The acoustic basement is moderately deep in the northern part of the Corsica Basin (3.5 and 3 s T W T T Figs 6 and 7 respectively). This depth corresponds to the 5.2 km depth (3.6 s TWTT) observed on the refraction and gravity results (Fig. 4b; Carrozzo & Nicolich 1977). This basement is very deep in the middle part of the Corsica Basin (about 5.5 s TWTT; Fig. 8; Finetti & Morelli 1973). For the first time the deepest part of the Corsica Basin is imaged on the Corsica continental shelf (Fig. 9). The basal unit (between 5 and 6 s TWTT) is tilted and the middle unit (between 3.5 and 5 s
TWTT, Fig. 9) shows a fan-shaped configuration characteristic of a synrift sequence. This huge graben is limited by the Solenzara listric fault (Fig. 3) and by other normal faults evident in the easternmost Alpine Corsica (Daniel et al. 1996). South of this fault the acoustic basement is shallow (3-2 s TWTT, Figs 10 and 11). The onshore extension (Aleria Basin) of the Corsica Basin is 4 km thick (Bayer et al. 1976) and the acoustic basement is more than 10 km deep in the offshore part of this basin. The thickness of the Corsica Basin is shallower (5.2 km) in the transect presented in Fig. 4b, which crosses the northern part of the basin. The Bouguer gravity map on land (Bayer et al. 1976) and the free air gravity map at sea (Sandwell et al. 1995) show (Fig. 12) the shape of the Corsica basin and in
178
A. MAUFFRET & I. CONTRUCCI
Fig. 9. LISA 11 seismic profile. This profile shows the listric Solenzara fault (Fig. 3) that limits the Corsica Basin. The acoustic basement is 6 s TWTT deep. The synrift formation has a fan shaped configuration. Position of the seismic profile is indicated in Fig. 2.
particular its northern and southern termination (+50 milligal contour). The southern positive gravity anomaly is related to Hercynian Corsica whereas the northern one is linked with Alpine Corsica.
Capraia-Monte Cristo and Pianosa ridges (Figs 6, 7, 8, 10, 13 and 14) The e m p l a c e m e n t of magmatic bodies of Capraia, M o n t e Capanne (Elba), Giglio and Monte Cristo have been dated at 6.9-3.5 Ma,
C R U S T A L S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N S E A
179
Fig. 10, Western part of LISA 12 seismic profile. The Corsica Basin has a shallow basement. Note the Monte Cristo Ridge that is correlated with a prominent magnetic anomaly. Position of the seismic profile is indicated in Fig. 2.
Fig. 11. Western part of L I S A 13 seismic profile. This profile shows the eastern extension of the Hercynian Corsica (see Fig. 12). Position of the seismic profile is indicated in Fig. 2.
180
A. MAUFFRET & I. CONTRUCCI
., /0
/~
~'.50../
25 /
f~O'~ l 'z
~
25__---~
"
i
]
,
,
-50
25
.,
-- 7---.~.:.v.~.~ .
41"N
1 C~
rN
Fig. 12. Gravity map of Corsica (Bayer et al. 1976) and North Tyrrhenian Basin (Sandwell et al. 1995). Bouguer anomalies on land and free air anomalies at sea. Alpine Corsica is characterized by positive anomalies and Aleria Basin (on shore extension of the Corsica Basin, see Fig. 3) by negative anomalies. The Corsica Basin is correlated to negative gravity anomalies (0 to -25 mGal) whereas the western extension of the Hercynian Corsica shows a positive anomaly (50 mGal).
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA 5 Ma and 7 Ma respectively (Serri et al. 1993 for a review). This ridge is delineated by a prominent magnetic anomaly (up to 500 nT near Monte Cristo Island). A relationship between magmatism and magnetism is likely, although the magnetic positive anomaly has been attributed to an ophiolitic body (Cassano 1991). However, Monte Capanne is not related to a magnetic anomaly that can be explained by the low magnetic susceptibility of the granite that forms this intrusion (Boullin et al. 1993). In Pianosa Island (Colantoni & Borsetti 1974) the middle Pliocene-Quaternary series overlies directly the lower Miocene (Langhian, 15 Ma). On the continental shelf of this island Burdigalian marls have been sampled by coring (Viaris de Lesegno 1978). The Miocene layers of the island and on the Pianosa continental shelf dip 20 ~ towards the west. In the exploratory well Martina (Fig. 3), located south of Pianosa Island the upper Pliocene overlies directly 1 km thick Oligocene series unaffected by compressional tectonics, if the distant correlation (Fig. 4a) between the seismic profile L 122 and the Martina well is correct. In the Martina well this thick Oligocene sequence rests on Palaeocene-middle Eocene turbiditic layer and ophiolitic breccias that is deformed by east vergent thrusts (Bartole et aL 1991). Triassic rocks crop out in the Africa Islet (Figs 1 and 2). Eocene, Upper Cretaceous Palaeocene (Helmintoides flysch) rocks were sampled (Aleria 1980) in the southern part of Pianosa Ridge (near Monte Cristo Island, Fig. 3). From these results it is evident that the Corsica Basin is underlain by Alpine Corsica formations (Helmintoides flysch, Shistes Lustr6s and ophiolites) and that the Miocene and Oligocene sedimentary layers of the Pianosa Ridge are equivalent, but uplifted, to the deep part of the Corsica Basin. In the Corsica Basin the Messinian reflector is flat but it shows a i km tilt (Figs 6, 7 and 8) along the western flank of Pianosa Ridge. The Lower Pliocene is restricted to the basin (sedimentary wedge, Figs 6 and 7); therefore, we conclude that the main episode of uplift occurred during the early Pliocene. However, Messinian uplift is also recorded by the accumulation of erosional detritus below the Messinian unconformity (Fig. 7). The Pianosa continental shelf is an erosional surface (Fig. 8) and 1 km of uplift is a minimum value; therefore, 3 km can be estimated if we take into account this erosion (Fig. 7). A greater value (10 km) may be proposed if we assume that the Oligocene layers drilled on the Pianosa Ridge Martina and Mimosa wells (Fig. 3) were initially located in the deepest part of the Corsica Basin. However, we do not know the
181
initial topography of the Pianosa Ridge area that was probably the hanging wall of the Corsica Basin and consequently shallower than the bottom of the basin. Moreover the basement that underlies the basin must be backstripped (work in progress) to obtain a correct evaluation. On the eastern flank of the Monte Cristo Ridge tilting is also evident (Figs 13 and 14). The mid-Pliocene tilting can be evaluated to 0.9 km (Fig. 13). A 1 km thick Oligocene layer was drilled in Mimosa well (Bartole et al. 1991), but folded Alpine units crop out at the western end of the seismic profile presented in Fig. 14 although this profile is very close to Mimosa well. This seismic profile illustrates the difficulty in tying the results of the exploratory wells to seismic stratigraphy. In this profile the minimum tilting is evaluated to 1 km. We note that the normal faults are west facing (Figs 13 and 14) and these faults may be antithetic relative to a deep detachment. This detachment is not observed because prominent multiples obscure the seismic profiles. 1 to 3 km of mid-Pliocene uplift are recorded by the tilting of the sedimentary layers and acoustic basement on the two flanks of the Elba-Monte Cristo Ridge. Monte Capanne (Elba) was emplaced 6-7 Ma ago (late Tortonian) at a depth of 2.5 km, then a very rapid 2 Ma old uplift is recorded by fission tracks in apatites (Bouillin et al. 1994). Monte Capanne is 1 km high and Elba Island underwent a 3.5 km total uplift during the mid-Pliocene. These results are in complete agreement with our observations. Eurite (tourmaline-rich aplite) pebbles has been found in the early Pliocene (Tongiorgi & Tongiorgi 1964) and latest Messinian (Marinelli et al. 1993) basins of Tuscany. These pebbles originated from the Monte Capanne intrusion of Elba Island, which was an extension of the Tuscany continental landmass at this time, Marinelli et al. (1993) concluded that he Monte Capanne was 2.5 km high during the Messinian. The presence of erosional products beneath the Pliocene-Miocene unconformity (Fig. 7) suggests, indeed, that the Monte Capanne was already high during the latest Messinian. The tectonic denudation of the Monte Capanne is commonly attributed to low-angle simple shear fault dipping to the east (Keller & Pialli 1990; Keller & Coward 1996; Daniel & Jolivet 1995) associated with the Tyrrhenian Sea extension. The emplacement of the plutonic body is passive in this hypothesis. However, we suggest, in agreement with Zitellini et al. (1986), that the mid-Pliocene tilting on the two flanks of the Monte Capanne-Monte Cristo Ridge is caused by the rise of the intrusion. Therefore, we favour
182
A. MAUFFRET
& I. C O N T R U C C I
. ,.,,,, .,,,,a
9
~z
9
9
o ~3
+,,.,
o3
r
r~
.~
,,,~
C R U S T A L S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA
183
Fig. 14. Eastern part of LISA 10 seismic profile. We observe again a prominent tilting along the eastern flank of the Monte Cristo Ridge. Alpine Formation outcrops on the ridge (see Fig. 2) but the Mimosa well sampled 1 km thick Oligocene series. These Oligocene series may be included in the compressional deformed layers or undeformed (Fig. 4a) but absent in the area crossed by the seismic profile. This profile illustrates the difficulty to tie the results of the wells and the seismic stratigraphy. Position of the seismic profile is indicated in Fig. 2.
184
A. MAUFFRET & I. CONTRUCCI
an active role for the magmatic body. Ductile extensional shear zone localized at the pluton boundary accompanied the exhumation at depth of the Elba intrusives (Daniel & Jolivet 1995). This history does not contradict the fact that the pluton in its last stages of ascent, already in its solid-state, rises up by isostatic forces and the brittle extension of its cover. A part of the uplift may be due to elastic rebound and extensional unloading of the footwall during the tectonic denudation. Offshore Elba several eastwarddipping reflectors are observed in the upper part of the crust (3 s TWTT, Fig. 15) but also in the lower part (6-7 s TWTT, Fig. 15). A well-layered lower crust shows a prominent eastward dip whereas the reflectors corresponding to the Moho (7 s TWTT, Fig. 15) are horizontal. The presence of the magmatic body may favour the localization of detachment faults at crustal scale, probably in a ductile regime. Tuscany M a r g i n a n d extensional tectonics (Figs 14, 15 a n d 16) We crossed two main basins: one between Monte Cristo and Giglio islands that we name Monte Glio Basin (Fig. 14) and the Punta Ala Basin (Figs 15 and 16). A shallow unconformity is related to the end of the mid-Pliocene extensional event (Bartole et al. 1991). A Tortonian extension is sealed by the Messinian reflector. The Monte Glio Basin has a relatively simple geometry (Fig. 14), but the mid-Pliocene uplift of the Monte Cristo Ridge tilted the basin (see above). The Punta Ala Basin is a graben where the fan-shaped synrift layers thin alternatively through time towards the east and west (Fig. 15). This basin is underlain by a shallow horizontal detachment (3 and 2 s TWTT Figs 15 and and 16 respectively) also found by Bartole (1995). S o u t h e r n p a r t o f the study area (Fig. 17) Two highs and adjacent basins that trends north-south form the main structures of this area. A mid-Pliocene rifting event (Zitellini et al. 1986) was defined in this area. Calcshists and ophiolites dredged on the two flanks of the Monte Cialdi (Aleria 1980). These samples demonstrated the extension towards the south of the Alpine formations. We found in the western part of this area (Fig. 3) several reflectors dipping towards the west (Fig. 11). These dipping reflections, also noted by Zitellini et al. (1986), could be related to Alpine thrusts with eastwards vergence.
Discussion The tectonic evolution of the North Tyrrhenian Sea is not fully understood in terms of time and space. The collision between Europe and Adria occurred during the late Cretaceous-Eocene. In Corsica the last compressional event is 37 Ma old (late Eocene) (Brunet et al. 1997). During or before the early Oligocene the oceanic Liguride Domain was closed and subducted beneath Corsica (Principe & Treeves 1984; Abbate et aL 1994), whereas the Alpine Corsica overthrust the Hercynian crust. Seismic profile (Fig. 5) north of Elba Island shows that late Eocene thrusts, but with an eastwards vergence, cut the entire crust. The Ligurides Units of the Northern Apennines can be considered as an accretionary prism at a crustal scale (Principe & Treeves 1984; Abbate et al. 1994) related to westwards subduction. The Adria continental crust is involved in the collisional prism (Figs 18 and 19). In other studies (Alvarez 1991; Keller & Coward 1996), the Liguride oceanic crust is not completely consumed after the Eocene epoch. A space problem is related, indeed, to the subsequent opening of the Balearic-Ligurian Basin (Carmignani et al. 1994, 1995). In this basin the rifting starts during the Oligocene (30 Ma) and the onset of oceanic crust is clearly dated (Gorini 1994; Gorini et al. 1993, 1994; Mauffret et al. 1995) by the middle Aquitanian (22 Ma) break-up unconformity. The drifting of the Corsica-Sardinia block occurred (Montigny et al. 1981; Rehault et al. 1984) from the mid-Aquitanian to the Burdigalian (22-19 Ma). The first compression in the modern Apennines is 27 Ma old (Carmigniani & Kligfield 1990). This compression cannot be related to the drifting in the Western Mediterranean Sea (Keller & Coward 1996), but is probably controlled by the subduction of the Adria continental lithosphere. After the emplacement of Alpine nappes in Corsica, Eastern Corsica underwent an extensional event (33-29 Ma and 25 Ma; Brunet et al. 1997) along east-dipping detachments faults (Jolivet etal. 1990, 1991,1994; Daniel etal. 1996). The formation of the Corsica Basin is related to this extensional event. The 5 km thick synrifl formation (Fig. 9) is at least Burdigalian (21-16 Ma) old but the sampling by industrial wells of a thick Oligocene layer beneath Pianosa Ridge suggests that the basal infilling of the Corsica Basin could be Oligocene. Therefore, this basin may have been formed at the same time as the grabens of the European margin before the rotation of the Corsica-Sardinia block. The Adriatic Promontory may be an indenter that collided with the European plate during the Eocene
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
185
Fig. 15. Eastern part of LISA 8 seismic profile. Note several detachments and the dipping of the layered lower crust. A reflection between 7 and 8 s TWTT corresponds to the 22 km deep Moho of the Tuscany margin (Letz et al. 1977a, b; Fig. 4b). The Punta Ala Basin has a complex geometry. Position of the seismic profile is indicated in Fig. 2. (Stampfli & Marchant 1997; Fig. 18a). During the late Oligocene-early Miocene Sardinia and may be Corsica u n d e r w e n t a transpressive
motion with a left-lateral reactivation of the N E - S W H e r c y n i a n faults (Carmignani e t al. 1994, 1995). The Corsica Basin, that shows a
186
A. M A U F F R E T & I. C O N T R U C C I
Fig. 16. LISA 9 seismic profile. Note the shallow detachment beneath the Formiche di Giglio Basin. Observe a deep reflection (white arrow) corresponding to the shallow Moho beneath the Tuscany margin. Position of the seismic profile is indicated in Fig. 2.
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
187
Fig. 17. Eastern part of LISA 12 seismic profile. Monte Etruschi and Cialdi are two horst formed during the middle Pliocene extension (Zitellini et al. 1986). Rocks of Alpine affinity have been dredged (arrows) along the Monte Cialdi scarps (Aleria 1980).
rhomboedric shape (Fig. 3), could have been controlled by these strike-slip faults or preferably by the conjugate N W - S E right-lateral faults (Fig. 18b) before and/or during the rotation of the Corsica-Sardinia block. During the formation of the deep Corsica Basin by extension, the crust was stretched but we do not know if the thinning is local, beneath the basin, or regional because the North Tyrrhenian Basin has subsequently been affected by recent rifting episodes. During and after its formation the Corsica Basin was not affected by the Apennines compression; consequently, at that time, the backstop of the Apennines nappes is located eastwards of this basin probably along the Capraia-Monte-Cristo Ridge. During Tortonian time (10-6 Ma) the whole Tyrrhenian Sea underwent a rifting event (Kastens et al. 1988; Sartori 1990). In the South Tyrrhenian Sea this extensional tectonics preceded the emplacement of oceanic crust whereas
the North Tyrrhenian Sea has yet a continental crust but intruded by plutonic and volcanic bodies. These intrusions were located in the deep and intermediate levels of the crust. The present geometry of the North TyrrhenJan Basin results from a middle Pliocene extension. This extension was coeval to a 3 km ascent of the plutonic bodies that upturned the sedimentary cover of the Corsica Basin (Fig. 19). 0.6 to 1 km Pliocene uplift is also observed on land above the geothermal fields (Mongelli et al. 1991; Marinelli et al. 1993). A regional high heat flow (100 W m -2) indicates that the lithosphere is thin in the study area (Mongelli & Zitto 1991). Very high heat flow anomalies are superimposed on the regional field. These anomalies are related to geothermal fields like Larderello. However, the extensional unloading of the western Elba footwall may also concur to the tectonic uplift (Fig. 19e, right side). The extensional tectonics of the North Tyrrhenian is
188
A. M A U F F R E T & I. C O N T R U C C I
A. Late Eocene
B. Late Oligocene
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA characterized by small basins that trend n o r t h - s o u t h separated by transfer faults (Bartole et al. 1991; Bartole 1995). The emplacement of Porto Azzurro granite (eastern Elba) and Larderello intrusions (Camelli et al. 1994) could be related to the Piombino-Faenza transfer fault (Fig. 20) that intersects the N-S to NNW-SSE grabens. A magnetic susceptibility study of Monte Capanne (Boullin et al. 1993) suggests that this pluton was emplaced in a pullapart environment along E - W to N E - S W leftlateral strike-slip faults although an E - W extensional regime is proposed by Daniel & Jolivet (1995). Monte Cristo Island can also be related to the Grosseto-Pienza transfer fault. We observe that Monte Capanne and Monte Cristo correspond to an orientation change of the Capraia-Monte Cristo Ridge and the trend of the C a p r a i a - M o n t e Cristo Ridge and the trend of the ridge south of Monte Cristo Island is compatible with a left-lateral strike-slip fault (Fig. 20) with a small component of extension. These transfer faults and the change of trend of the extensional basins from N W - S E in the north to N-S in the south suggest a pole of rotation located in the north. Although no tectonic rotation since the Messinian is recorded on land by palaeomagnetic studies (Mattei et al. 1996) a Tortonian rotation is probable in the North Tyrrhenian Basin. Several studies (Channel & Mareschal 1989; Serri et al. 1993) suggest that the thinning of the North Tyrrhenian lithosphere is related to a d e l a m i n a t i o n of the A d r i a lithosphere that retreats towards the east. High heat flow, the active role of the plutons and space problems are in favour of this hypothesis but we will wait the results of the land recording of the shots fired during the LISA cruise to propose an interpretation of the deep structure.
Summary and conclusions The LISA seismic cruise investigated the upper sedimentary layers and the deep crust. The results of this cruise are complementary to the industrial seismic lines, where the resolution is
189
good but the penetration too shallow, and the refraction data, which show the deep structure but fail to give a good resolution. The Tertiary history of the North Tyrrhenian Basin began with the Alpine collision in Corsica. Probably the Liguride oceanic crust has been consumed at this time but several authors disagree with this complete subduction of the oceanic crust. After the formation of Alpine nappes in Corsica, this mountain collapsed along an east-dipping extensional detachment. The Corsica Basin is related to this extension, which is probably coeval with the rifting of the Western Mediterranean Basin. The rhombohedric shape of the 10 km deep Corsica Basin suggests a strike-slip component related to late Oligocene-early Miocene transpressional tectonics in the Corsica-Sardinia block. The Corsica basin is preserved from the A p e n n i n e compression that began during the Oligocene, and is characterized in the North Tyrrhenian Basin by crustal thrusts. We have no evidence of tectonic activity in the N o r t h T y r r h e n i a n Basin from the early to midMiocene. A n intense normal faulting event is recorded from the Tortonian to the midPliocene in the grabens of the Tuscany margin and in the southern part of the study area. The late Miocene rifting is contemporaneous with the emplacement of plutonic bodies at intermediate and deep levels in the crust. During the mid-Pliocene extensional event these plutonic bodies rose to the surface and induced a rapid uplift of the sedimentary layers. During this event deep detachments were formed and the lithosphere was t h i n n e d by an hypothetical delamination process. We thank the officers and crew of RV N a d i r for assistance in the LISA project. The authors are grateful to the scientists particularly C. Truffert, and J. Begot who processed on board the LISA seismic data. This study benefited from the Diplome d'Etudes Approfondies presented by S. Poignard. ELF-SNEA(P), CFPTOTAL and IFP kindly provided a seismic line and allow us to publish this seismic profile. We thank C. Faccena and J. M. Daniel for their constructive reviews and helpful suggestions. This work is supported by the IDYL-INSU program. URA 1759 Contribution.
Fig. 18. (a) Late Eocene reconstruction. Position of Corsica-Sardinia block from (Olivet 1987; Gueguen et aL 1993; Mauffret et al. 1995). General tectonic framework from Stampfli & Marchant (1996). A. C., Alpine Corsica. (b) Oligocene-early Miocene reconstruction. Position of Corsica-Sardinia block from (Olivet 1987; Gueguen et aL 1993; Mauffret et aL 1995). General tectonic framework from Stampfli & Marchant (1996). The Corsica Basin, adjacent to the Alpine Corsica, may have been initiated before the rotation of Corsica by normal faulting along the former thrusts of Alpine Corsica (Jolivet et al. 1990, 1991, 1994; Daniel et al. 1996). The Corsica pull-apart basin may be related to the northwards motion of the Sardinia-Corsica block (Carmignani et al. 1994, 1995).
190
A. M A U F F R E T & I. C O N T R U C C I
Fig. 19. (a to c) Sketch of the North Tyrrhenian tectonic evolution. The delamination model is inspired from Serri et al. (1993). (d) Uplifting of the sedimentary layers fill of the Corsica Basin by the ascent of an intrusive body. (e) Left: formation of the Corsica Basin during the Oligocene-Early Miocene. (e) Right: the rise of the intrusive body is coeval to an extensional unloading of the Elba footwall. The horst and graben structures of the Tuscany margin form the hanging wall.
CRUSTAL S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA
191
Llrderelio
Porto Azzurro ~.~
4"
~,
......
~ ....
,
~
~.__:-~.::::::
"-'-
i-:-U.-.'::'-.'-'.-----_ ...4 ............ ~,d,c,~o
',
~
"
~ 'i
. . . . . . . . . .
r
___3
..........
t
sf
~--~2~ ,!
/
Fig. 20. Transfer faults in the North Tyrrhenian Basin (from Bartole 1995). Pull-apart basin interpretation of the Monte Capanne emplacement is from Bouillin et aL (1993). Larderello geothermal field and Porto Azzurro granite are in strike along the Piombino-Faenza transfer fault. Monte Cristo Island is located on the southwestern extension of the Grosseto-Pienza transfer fault. The ridge south of Monte Cristo Island may correspond to left-lateral strike-slip fault with a component of extension.
References
ABBATE,E., BORTOLOTTI,V., PASSERINI,P., PRINCIPI,a. & TREVES, B. 1994. Oceanisation processes and sedimentary evolution of the Northern Apennine ophiolitic suite: a discussion. Memorie della Societd Geologica Italiana, 48, 117-136. ALERIA, A. 1980. Alpine metamorphic rocks and Late Cretaceous rocks and Late Cretaceous-Eocene flysch sediments in the Northern Tyrrhenian Sea: tectonic implications. Centro Grafico, Universith di Parma. ALVAREZ,W. 1991. Tectonic evolution of the CorsicaApennines-Alps studied by the method of Successive Approximations. Tectonics, 10, 936-947. AVEDIK, E, RENARD, g., ALLENOU, J. P. • MORVAN,B. 1993. "Single Bubble" air-gun array for deep exploration. Geophysics, 58, 366-382.
BARTOLE, R. 1995. The North Tyrrhenian-Northern Apennines post-collisional system:constraints for a geodynamic model. Terra Nova, 7, 7-30. - - . , TORELLI,L., MATTEI, G., PEIS, D. & BRANCOLINI, G. 1991. Assetto stratigrafico-structurale del Tirreno settentrionale: stato dell'arte. In: Studi Geologici Camerti, Volume speciale 1991/1, AGIP and CNR, 115-140. BAYER, M., BAYER, R. & LESQUER,A. 1976. Quelques remarques sur la structure g6ologique de la Corse d'apr~s la gravim6trie et le magn6tisme. Bulletin de la SociOt~ GOologique de France, 18,1189-1194. BOULLIN, J. E, BOUCHEZ, J. L., LESHNASSE, E & PgCHER, A. 1993. Granite emplacement in an extensional setting: an AMS study of the magmatic structures of Monte Capanne (Elba, Italy). Earth and Planetary Science Letters, 118, 263-279.
192
A. M A U F F R E T & I. C O N T R U C C I
, POUPEAU, G. & SABIL, N. 1994. Etude thermochronologique de la ddnudation du pluton du Monte Capanne (ile d'Elbe, Italie). Bulletin de la SociOtd GOologique de France, 165, 19-25. BRUNET, C., MONIt~,P. 8`: JOLIVET,L. 1997. Geodynamic evolution of Alpine Corsica based on new Ar/Ar data. Terra Nova, 9, 493. CAMELI, G. M., DINI, I. 8`: LIOTTA, D. 1994. Upper crustal structure of the Larderello geothermal field as a feature of post-collisional extensional tectonics (Southern Tuscany, Italy). Tectonophysics, 413-423. CARMIGNIANI,L . 8`: KLIGFIELD,R. 1990. Crustal extension in the Northern Apennines: the transition from compression to extension in the Alpi Apuanes core complex. Tectonics, 9, 1275-1303. - - , BARCA,S., D1SPERATI,L., FANTOZZI,P., FUNEDDA, A., OGGIANO, G. 8`: PASCI, S. 1994. Tertiary compression and extension in the Sardinian basement. Bollettino di Geofisica Teorica ed Applicata, 36, 45-62. - - , DECANDIA, E A., DISPERATI,L., FANTOZZI,P. L., LAZZAROTTO,A., LIOTTA,D. 8`: OGGIANO, G. 1995. Relationships between the Tertiary structural evolution of the Sardinia-Corsica-Provenqal domain and the Northern Apennines. Terra Nova, 7, 128-137. CARROZZO, M. T. 8`: NICOLICH, R. 1977. Quantitative interpretation of gravity and magnetic data. Bollettino di Geofisica Teorica e Applicata, 19, 236-248. CASSANO, E. 1991. Dati magnetici lungo il profilo CROP 03. In: Studi Geologici Carnerti, Volume speciale 1991/1. AGIP and CNR 4%53. CHANNEL, J. E. Y. 8`: MARESCHAL,J. C. 1989. Delamination and asymmetric lithospheric thickening in the development of the Tyrrhenian rift. In: COWARD, M. E, DmTRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 285-302. COLANTONI, R & BORSETTI,A. M. 1974. Geologia e stratigrafia dell'isola Pianosa (Archipelago toscano-Mar tireno). Giornale di Geologia, 29, 287-302. CRAVATI'E, J., DUFAURE, P., PRIM, M. & ROUAIX, S. 1974. Les sondages du Golfe du Lion: stratigraphie et sddimentologie. Notes et Memoire de la CFP, 2, 209-274. DANIEL, J. M. 8,: JOLIVET, L. 1995. Detachment fault and pluton emplacement: Elba Island (TyrrhenJan Sea). Bulletin de la SociOtO G~ologique de France, 166, 341-354. - - , GOFFE, B. & POINSSOq; C. 1996. Crustalscale strain partitioning: footwall deformation blow the Alpine Oligo-Miocene detachment of Corsica. Journal of Structural Geology, 18, 41-59. D'OFFIZI, S., MINELLI, G. & PIALLI, G. 1994. Foredeeps and thrust systems in the Northern Apennines. Bollettino di Geofisica Teorica ed Applicata, 36 91-102. EGGER, A.P. 1992. Lithospheric structure along a transect from the northern Apennines to Tunisia derived from seismic refraction data. Th6se, Universit6 de Lausanne.
--,
DEMARTIN, M., ANSORGE, J., BANDA, E. & MAIESTRELLO,M. 1988. The gross structure of the crust under Corsica and Sardinia. Tectonophysics, 150, 363-389. ELSTER, P., GIGLIA, G., TONGIORGI, M. & TREVISAN,L. 1975. Tensional and compressional areas in the recent (Tortonian to present)evolution of the Northern Apennines. Bollettino Geofisica Teorica ed Applicata, 17, 3-19. FACCENA, C., DAVY, P., BRUN, J. R, FUNICELLO, R., GIARDIM, D., MATTEI, M. & NALPAS,T. 1996. The Dynamics of back-arc extension: an experimental approach to the opening of the Tyrrhenian Sea. Geophysical Journal International, 126, 781-795. FINETrI, L. & MORELLI, C. 1973. Geophysical exploration of the Mediterranean Sea. Bollettino Geofisica Teorica ed Applicata, 15, 263-341. GORINI, C. 1994. GOodynamique d'une marge passive: le Golfe du Lion (MOditerranOe occidentale). Thase de l'Univ. Toulouse, 256 p. --, LE MARREC, A. & MAUFFRET, A. 1993. Contribution to the structural and sedimentary history of the Gulf of Lions, (Western Mediterranean), from the ECORS profiles, industrial seisnaic profiles and well data. Bulletin de la SociOtO GOologique de France, 164, 353-363. , MAUFFRET,A., GUENNOC, E & LE MARREC, A. 1994. Structure of the Gulf of Lions (Northwestern Mediterranean sea): a review. In: MASCLE,A. (ed.) Hydrocarbon and Petroleum Geology of France. European Association of Petroleum Geologists, Special Publications, 4, 223-243. GUEGUEN, E., OLIVET,J. L. & REHAULT,J. R 1993. Kinematics model in the Western Mediterranean: new constraints. Terra Abstracts', 5, 167. HmN, A. & SAPIN,M. 1976. La crofite terrestre sous la Corse: donn6es sismiques. Bulletin de la SociOtO G(ologique de France, 18, 1195-1199. JOLIVET, L., DANIEL, J. M. & FOURNIER, M. 1991. Geometry and kinematics of extension in Alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. - - , TRUFFERT,C. & GOFFI~, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in back-arc regions. Lithos, 33, 30. - - , DUBOlS, R., FOURN1ER,M., GOFFI~, B., MICHARD, A. & JOURDAN, C. 1990. Ductile extension in Alpine Corsica. Geology, 18, 1007-1010. KASTENS, K., MASCLE,J. t~r AL. 1988. ODP Leg 107 in the Tyrrhenian Sea: insights into passive margin and black-arc evolution. Geological Society of America Bulletin, 100, 1140-1156. KELLER, J. V. A. & COWARD, M. P. 1996. The structure and evolution of the Northern Tyrrhenian Sea. Geological Magazine, 133, 1-16. - 8`: PIALLI,G. 1990. Tectonics of the island of Elba: a reappraisal. Bolletino della Societd Geologica Italiana, 109, 413-425. LETZ, H., REICHERT,C. & WIGGER, P. 1977a. Results of two refraction lines in the Northern Apennines. Bolletino di Geofisica Teorica e Applicata, 19, 225-232. - - , - - , - - & GIESE, P. 1977b. Seismic refraction
CRUSTAL S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA measurements in the Ligurian Sea and in Northern Apennines. In: CLOSS, H. & ROEDER, D. (eds) Alps, Apennines and Hellenides. Stuttgart, 413-430. MARINELLI, G., BARBERI,E & CIONI, R. 1993. Sollevamento neogenice e intrusion acide della Toscana e del Lazio Settentrionale. Memorie della Societd Geologica Italiana, 49, 279-288. MATTAUER, M., FAURE, M. & MALAVIELLE,J. 1981. Transverse lineation and large scale structure related to alpine obduction in Corsica. Journal of Structural Geology, 3, 401-409. MATTEI, M., KISSEL, C. & FUNICELLO,R. 1996. No tectonic rotation of the Tuscan Tyrrhenian margin (Italy) since late Messinian. Journal of Geophysical Research, 101, 2835-2845. MAUFFRET,A., PASCAL,G., MAILLARD,A. & GORINI, C. 1995. Structure of the deep Northwestern Mediterranean basin. Journal of Petroleum Geology, 12, 645-666. MONGELLI, E & ZITO, G. 1991. Flusso di calore helle regiona toscana. In: Studi Geologici Camerti, Volume speciale 1991/1. AGIP and CNR, 91-98. , PUXEDDU, M., SQUARCI, P., TAFFI, L. & ZITO, G. 1991. I1 flusso di calore e l'anomalia geotermica dell'area Tosco-Laziale: implicazioni profonde. In: Studi Geologici Camerti, Volume speciale 1991/1. A G I P and CNR, 399-402. MONTIGNY, R., EDEL, J. B. & THUIZAT,R. 1981. OligoMiocene rotation of Sardinia: K - A r ages and palaeomagnetism data of tertiary volcanics. Earth and Planetary Science Letters, 54, 261-271. OLIVET,J. L. 1987. L'origine du bassin nord-occidental de la M6diterran6e du point de vue de la cin6matique des plaques. In: BURRUS,J. & OLIVET, J. L. (eds) Profils ECORS. Golfe du Lion: rapport d'implantation. IFP, Paris, 35 941-1, 10-49. OZRAG-SPERBER,E & PILOT,M. D. 1976. Grands traits du N6ogbne en Corse. Bulletin de la Socidt~ G~ologique de France, 18, 1183-1187. PONZIANI, E, DE FRANCO,R., MINELLI, G., BIELLA, G., FEDERICO, C. &; PIALLI, G. 1995. Crustal shortening and duplication of the Moho in the northern Apennines: a view from seismic refraction data. Tectonophysics, 252, 391-418. PRINCIPE, G. & TREEVES, B. 1984. I1 systema CorsoApennino come prisma d'accrezione. Riflessi sul problema generale del limite Alpi Apennino, I1
193
systema Corso-Apennino come prism d'accrezione. Riflessi sul problema generale del limite Alpi Apennino, 28, 549-576. REHAULT, J. P. 1981. Evolution tectonique et sddimentaire du bassin Ligure (MOditerran~e occidentale). Th6se d'Etat de l'Univ. P. et M. Curie, Paris. - - , BOILLOT, G. & MAUFFRET,A. 1984. The western Mediterranean Basin, geological evolution. Marine Geology, 55, 447-477. SANDWELL,D. Y., YALE, M. M. & SMITH,W. H. E 1995. Gravity anomaly profiles from ERS-1, Topex and Geosat Altimetry. LOS, Transactions of the AGU, 76, 89. SARTORI,R. 1990. The main results of O D P Leg 107 in the frame of Neogene to recent geology of Perithyrrhenian areas: In: KASTENS, K. MASCLE, J. (eds) Proceeding of Ocean Drilling Program, Scientific Results, 107, 715-730. SAUPt~,E, MARIGNAC, C., MOINE, B., SONET, J. • ZIMMERMANN, J. L. 1982. Datation par les m6thodes K/Ar et Rb/Sr de quelques roches de la partie orientale de l'~le d'Elbe province de Livourne, Italie. Bulletin de Min~ralogie, 105, 236-245. SERRI, G., INNOCENTI, E & MANETTI, P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the NeogeneQuaternary magmatism of central Italy. Tectonophysics, 223, 117-147. STAMPFLI, G. M. & MARCHANT, E H. 1997. Geodynamic evolution of the Tethyan margins of the western Alps, In: PFIFFNER, O. A., ET AL. (eds) Deep structure of Switzerland. Results from NPF 20, in press. TONGIORGI, E. & TONGIORGI, M. 1964. Age of the Miocene-Pliocene limit in Italy. Nature, 201, 365-367. VIARIS DE LESEGNO, L. 1978. Etude structurale de la met Tyrrh~nienne septentrionale. Th6se 3 ~ cycle, Paris, 6, 1-170. - - , GENNESSEAUX,M. & REHAULT,J. P. 1978. La tectonique n6og6ne et les s6ries s6dimentaires dans le bassin nord-tyrrh6nien. Bulletin de la Soci~t~ G~ologique de France, 20, 29-42. ZITELLINI,N., TRINCARDI,E, MARANI,M. & FABBRI,A. 1986. Neogene tectonics of the Northern Tyrrhenian Sea. Giornale di Geologia, 48, 25-40.
IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration FRANK
HORVATH
1 & GABOR
TARI 2
1Geophysical Department, EOtvOs University, H-1083, Budapest, Ludovika tOr 2. 2AMOCO Production Company, 501 WestLake Park Bld, Houston, Texas, USA
Abstract: The IBS Pannonian Basin project presents a good example of fruitful joint activity between Hungarian and other European scientists, and beneficial co-operation of academia and the petroleum industry. This allowed us to achieve significant progress in the understanding of the structure, tectonic evolution, stratigraphic features and hydrocarbon generation in the Pannonian Basin. The Pannonian region has been an integral part of the Alpine belt, and it reveals the complexity of orogenic evolution. Continental to oceanic rifting, followed by convergence, subduction and continental collision shaped the Palaeozoic-Mesozoic substrata of the region. Subsequently, two periods of basin formation (Late Cretaceous and Palaeogene) occurred, most probably in compressional setting. From the earliest Miocene large scale lateral displacement and block rotation took place in the internal domain of the orogen, together with the formation of the Pannonian Basin. This has been characterized by lithospheric extension, however, interrupted by compressional events. The modern Pannonian Basin is in an initial phase of positive structural inversion. All of these tectonic events had significant impacts on the formation and the economic value of the various petroleum systems in the area. Located completely within the E u r o p e a n Alpine belt, the Pannonian Basin has been traditionally an area of intensive geological research, and a classical test site of models to explain areas of subsidence within active orogens. As in many other regions of the world, a significant part of the geological knowledge has come from data acquired during h y d r o c a r b o n exploration. Deliberate search for hydrocarbons has been going on for more than 80 years in this area. The Pannonian basin is now a mature exploration area, and known hydrocarbon reserves together with production in H u n g a r y has undergone a slow but steady decrease during the past decade (Fig. 1). One way to keep exploration prosperous and effective is to make further progress in understanding the structure and evolution of the basin and its orogenic substrata. The aim of this introductory paper is to review the main scientific results achieved during the IBS project and, in the light of this progress, redefine the petroleum systems of the basin. First, a summary is presented about the general geological setting and tectonostratigraphic units. This is followed by a listing of some of the classical geological problems in the Pannonian Basin, which have been addressed and partially solved during the past three years. More detailed reports about recent results in different subjects can be found in subsequent papers in this volume. In the final part of the paper three petroleum systems are outlined. First, potential traps in the
Neogene basin fill, which can be charged by Miocene source rocks are reviewed. Then, reservoirs in the underlying Palaeogene rocks which can contain h y d r o c a r b o n s derived from an Oligocene source rock are reported. Last, but not least, prospect possibilities are discussed on the basis of the distribution of potential traps and source rocks in the structurally complex Mesozoic/Palaeozoic substrata of the basin.
Geological setting and main tectonostratigraphic units The P a n n o n i a n Basin is located in eastern central Europe, and situated inside the European Alpine belt (Fig. 2). At the western margin of the Pannonian basin, the Eastern Alps apparently bifurcate and continue to the SE (Dinarides) and the N E (Carpathians). While the Dinarides constitute a remarkably linear mountain belt, the Western, Eastern and Southern C a r p a t h i a n s form an almost complete loop before continuing into the Balkans. Thus, the Alpine chain is about 300, 1000 and 400 km wide respectively in the Alps, Dinarides-Pannonian b a s i n - C a r p a t h i a n s , and in the H e l l e n i d e s Balkans transects. The pronounced widening in the Pannonian sector is mainly the result of the Neogene extensional basin formation in this area. The surrounding Eastern Alps, Carpathians and Dinarides indeed project below the
HORVATH,F. 8r TARI,G. 1999. IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration. In: DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 195-213.
196
F. HORVATH & G. TARI
PANNONIAN BASIN OVERVIEW
197
Fig. 2. Main geological and geographic units of the Pannonian region and surrounding Alpine-Carpathian-Dinaric mountains. Pannonian Basin, which is superimposed on these A l p i n e fold-thrust belts. While these mountains are characterized by relatively high average elevation (Carpathians: 1500 m; Dinarides: 1000 m; Eastern and Southern Alps: 2000 m), the Pannonian Basin is a lowland with an average height of 150 m above sea level. In this lowland, which is about 400 km wide from N to S and 800 km long from E to W, isolated mountains (or inselberg) emerge from the plain with elevations up to 1000 m. These ranges subdivide the P a n n o n i a n Basin into a number of subbasins. The Vienna Basin is not part of the Pannonian Basin strictly speaking. It is located between the Eastern Alps and the West Carpathians (Fig. 2). The Danube Basin is bounded by the Eastern
Alps and the West Carpathians to the west and north, respectively, while it is bordered by the Transdanubian Range to the south. The Hungarian (southern) part of this basin is called the Little Hungarian Plain. Transdanubia is that part of Hungary which is located to the south and west of the river Danube. It is interesting to know that two thousand years ago this western part of the basin was a R o m a n province for about four centuries, and called Pannonia. Historically, this R o m a n name was extended to apply to the whole basin. The southern edge of Transdanubia is given by the Drava Trough, which is an elongate and curvilinear basin, just like the Sava trough further to the South. The Zala Basin constitutes the southwestern corner of Transdanubia and passes to the Styrian and
Fig. 1. Diagrams showing the history of the cumulative amount of hydrocarbon reserves (a), annual production (b), and a few data to illustrate that Hungary is a mature exploration area (K6kai 1994; Hungarian Geological Survey 1998).
198
E HORVATH & G. TARI
Mura Basins in Austria and Slovenia, respectively. The Great Hungarian Plain is the largest sub-basin of the Pannonian Basin and occupies the eastern portion of the area. The Transylvanian Basin located between the Apuseni Mts and the Eastern and Southern Carpathians (Fig. 2) does not belong in a strict sense to the Pannonian Basin. The separation of the Vienna and Transylvanian basins from the Pannonian Basin can be considered as mainly geographical. There is, however, a more profound geological reason for it: the Vienna and Transylvanian Basins are different in tectonic origin from the Pannonian Basin. The Pannonian Basin is the result of whole-scale lithospheric extension (Tari et al. this volume). In contrast, the Vienna Basin formed by thin-skinned extension, while the Transylvanian Basin is not of extensional origin (Royden 1988). Accordingly, when the name Pannonian Basin is used, this excludes the Vienna and Transylvanian Basins. However, the term intra-Carpathian basins or area refers to all the basins and ranges inside the Carpathian loop. The present-day assembly of the superimposed basins and the underlying orogenic structures suggest a complex origin of the intraCarpathian area. Four major periods of tectonic evolution can be distinguished. (i) Late Permian-Early Cretaceous. Two distinct episodes of continental break-up occurred during this period in the future Alpine-Mediterranean domain of Pangaea. While the Triassic rifting aborted in most of the area, the Jurassic extension led to the formation of the Tethys ocean, flanked by two rifted continental margins: the African-Adriatic (on the south) and the European (on the north). After the rifting, the continental margins and the intervening ocean were controlled by thermal subsidence until the end of Early Cretaceous. (ii) Late Cretaceous-Palaeocene. It was a period of first major compressions in the Alpine system, when many of the oceanic troughs and passive margins disappeared, due to convergence of the continental margins and subduction of the Tethys ocean. During this period, three subperiods of compressional events can be recognized: the Austrian phase (Aptian-Albian), the PreGosau phase (Cenomanian-Turonian) and the
Laramian phase (Maastrichtian-Danian). These events played a decisive role in shaping the structure of the pre-Tertiary strata in the intraCarpathian area. In addition, it was an important period of basin formation, which took place in the Senonian, after the second and before the third compressional event. Accordingly, the Senonian basin fill always represents a seal on the thrusts and folds developed during the Austrian and pre-Gosau phases. (iii) Eocene-Early Miocene. This period represents the second major interval of collision and compression in the Alps, which mostly affected the more external parts. In the internal part, a set of basins developed. The two most remarkable basins are the Palaeogene 'epicontinental' basin and the Szolnok-Maramures 'flysch' basin (Fig. 3). At the end of this period (latest Oligocene-Early Miocene) large-scale lateral displacement (continental extrusion) and/or rotation of internal blocks occurred, which disintegrated the former Alpine fold-thrust belt, and also strongly dismembered the Palaeogene basins. Disintegration includes juxtaposition in the intra-Carpathian area of two Alpine terranes of different early Mesozoic palaeographic position (Yilmaz et al. 1996): the North Pannonian terrane (African-Adriatic continental margin) and the South Pannonian terrane (European continental margin). The boundary of the two juxtaposed terranes is actually a wide zone of intensive early Miocene deformation, called the mid-Hungarian shear zone (Fig. 3). (iv) Mid-Miocene-Recent. Continuing convergence between Europe and Africa has formed further fold-thrust belts in the outermost domains. In the rearranged internal domain the Mid-Miocene was the period when widespread continental rifting initiated the formation of the intra-Carpathian basins. It was followed by the postrift thermal subsidence; however, compressional events causing local basin inversion and fault reactivation also occurred. These broad periods are close to those defined by Trtimpy (1973) in the classic Swiss sector of the Alps. Although there are slight differences in the timing, one can refer to his terms for the major stages outlined above. These are the Early Alpine, the Eoalpine, the Mesoalpine and the Neoalpine stages, respectively. In the intra-Carpathian area, the main Alpine
Fig. 3. Distribution of Palaeogene basins in the intra-Carpathian area. Keys: C, Central Carpathian flysch basin; Sz, Szolnok flysch Basin; Tc, Maramures flysch Basin; K, Krappfeld Basin; S, Slovenian Basin; H, Hungarian Basin; Ts, Transylvanian Basin.
PANNONIAN BASIN OVERVIEW
199
200
F. HORVATH & G. TARI
evolutionary stages led to the formation of four major tectonostratigraphic units. A simplified chronostratigraphic diagram of the area illustrates these major units (Fig. 4). The lower tectonostratigraphic unit is a composite of two different terranes made up of Mesozoic and Palaeozoic successions, which developed during the Early Alpine stage. These are the North Pannonian terrane (often called Alcapa block) and the South Pannonian terrane (often called Tisza-Dacia block). Structurally, this is the most complex unit, as it was affected also by all later phases of deformations. The overlying tectonostratigraphic unit is composed of the Senonian basins, which developed during the Eoalpine stage. They are now preserved in the Little and Great Hungarian Plains and in Transylvania. In these parts of the intra-Carpathian area they represent the immediate post-tectonic cover of the nappes developed in the lower tectonostratigraphic unit. The next tectonostratigraphic unit above comprises the Palaeogene epicontinental basin and the flysch basin formed during the Mesoalpine stage (Fig. 3). The uppermost tectonostratigraphic unit
"
=--1 "~;:;in-
comprises the Neogene Pannonian and Vienna basins which opened and filled during the Neoalpine stage. It is interesting to note that the Transylvanian Basin developed during both the Neoalpine and Mesoalpine stages.
Major problems of the Alpine evolution of the Pannonian basin and progress during the IBS project The Miocene to Quaternary basin fill of the uppermost tectonostratigraphic unit covers almost the entire area and locally is very thick (8 km). Neogene volcanics crop out at the basin margins, but they are also abundant in the subsurface. The middle and lower units are mostly unexposed. There are only isolated blocks where they crop out, particularly in the Transdanubian and North Hungarian Ranges, and Transylvania (Fig. 3). This clearly shows the importance of subsurface geology. Many geological problems in the Pannonian Basin, debated for almost a century, can be resolved only by systematic evaluation of surface and subsurface geological information, combined with acquisition of new data applying modern
P a n n o n i a
z-1
b a s i n
~
"~Transylvanlar~
J
basin
z
~ ~z _
w z
~
r
Trans.
.
~
m ..a
,<
.g
ul >z
North Pannonian
(Alcapa
)!
,
'
South Pannonian
(Tisz
")tlll
~ ~ <
-ff
Fig. 4. Generalized chronostratigraphic diagram showing the main tectonostratigraphic units in the intraCarpathian area. In the case of the Senonian basins LHP, GHP and Trans refer to the Little and Great Hungarian Plains, and Transylvania, resp.
PANNONIAN BASIN OVERVIEW technologies and concepts. This has been the main research philosophy behind our IBS project. A few 'classic' problems are lined up below, which have been addressed and progress made during the project. We proceed in a sequence from the lower toward the upper tectonostratigraphic units.
Structural style o f the Early Alpine unit The existence of Cretaceous Alpine nappes in the lowermost tectonostratigraphic unit of the Pannonian Basin has been debated since the beginning of the century. Although in many boreholes repeated sections were found, they were interpreted as local disturbances along major wrench fault zones. The large-scale allochthony of the Transdanubian Range was recently suggested (Horv~ith 1993; Tari 1994) in contradiction to the more traditional view, that claims negligible Alpine deformation in this major tectonic unit. A closely related development is the documentation of the much-debated continuation of the magnificent nappe pile of the Eastern Alps underneath the Danube Basin towards the Transdanubian Range and the Western Carpathians (Tari 1995, 1996; Mattick et al. 1996). Nappes of Late Cretaceous in age have been proven also below the Great Hungarian Plain (Grow et al. 1994). Furthermore, interpretation of the deep seismic profile PGT-4 and oil company lines suggest NW-vergent nappes comprising Palaeozoic crystalline basement rocks and/or their Mesozoic sedimentary cover (Tari et al. this volume). These nappes are actually the subsurface continuation of the nappe system cropping out in the Apuseni Mts, Romania (Fig. 2).
Mechanism o f Palaeogene basin formation The origin of the different Palaeogene basins in the intra-Carpathian area (Fig. 3) is poorly understood. These basins are traditionally classified either as 'flysch' or 'epicontinental', without referring to a geodynamic context. Recently a pull-apart origin was proposed by analogy to the overlying, much better understood Pannonian Basin (Royden & B~ildi 1988). Such a backward extrapolation of a transtensional model does not seem tenable for a number of reasons. Alternative geodynamic scenario suggests flexural origin in a retroarc position due to Mesoalpine backthrusting of the most internal West Carpathian nappes (Tari et al. 1993). This problem is further discussed and supporting evidence presented in Fodor et al. (this volume).
201
Kinematics o f extrusion and juxtaposition o f Alpine terranes For the last decade, the location and stratigraphic features of the Palaeogene basins (Fig. 3) have been considered as manifestation of major lateral displacements in the area that has significantly reshaped the palaeogeography of the Alps-Carpathians-Pannonian basin system after the Palaeogene (Csontos et al. 1992). Very different models were proposed involving the exact timing, magnitude and geometry of this displacement. There is, however, a general consensus that the main lateral displacement is related to the extrusion of the A l p i n e Carpathian-North Pannonian terrane (Alcapa) from the main Alpine collision zone towards the East (Ratschbacher et al. 1991). The restoration of the eastwardly extruding Alcapa terrane to their original place within the Alpine edifice was also a subject of major controversy (Balla 1988). Reliable and accurate structural markers have been used for better restoration and results show that escape of about 400 km occurred quite rapidly in the latest Oligocene-Early Miocene time (Tari et al. 1995). At about the same time, the South Pannonian (Tisza-Dacia) terrane went through a completely different kinematic history, characterized by some 90~ clockwise rotation (Patrascu et al. 1994). Juxtaposition of the two different terranes is discussed by Gy6rfi & Csontos (this volume). Furthermore, they offer new field observations from outside of the Pannonian Basin, in northern Transylvania, where the contact zone of the two different terranes is exposed. Similarly, at the southeastern rim of the Pannonian Basin in Slovenia, the contact zone is also uncovered, and Fodor et al. (this volume) present stress field determinations from new field observations in this area.
Modes and timing o f extension in the Pannonian Basin The understanding of the Neogene evolution of the Pannonian Basin largely improved during the last decade, but some fundamental questions remained unanswered. The Neogene Pannonian Basin was superimposed on an earlier Alpine compressional realm (Figs 2 and 4). Thus, the compressionally pre-conditioned structure of the basement supposedly influenced the magnitude and geometry of the subsequent continental extension by reactivation of regional decollement levels (Tari et al. 1992; Horv~th 1993). Extensive use of regional reflection seismic data has revealed that, at shallow depth,
202
F. HORVATH & G. TARI
high-angle normal faults cross-cut these d6collement levels. However, due to the listric geometry of normal faults they flatten with depth and, typically, sole out in compressional d6collement levels (Taft & Horv~th 1995; Tari 1996). This geometry, where extension was of significant magnitude, led to rapid unroofing of deep-lying rock masses, and eventually the exposure of metamorphic core complexes. One core complex has been documented at the western margin of the Little Hungarian Plain and another in the Great Hungarian Plain by seismic and gravity interpretation, and fissiontrack studies (Szafi~n & Tari 1995; Tari et al. this volume). In addition, timing and other modes of continental extension (e.g. wide-rift), have become better understood (Tari et al. this volume). The exact timing of the main phase of extension is an interesting problem. Generally speaking, the onset of rifting is easy to define because the oldest basin fill (Eggenburgian) rests unconformably on pre-rift strata. The first rhyolite tuff horizon is interbedded in this lower part of the basin fill. The age of this widespread horizon is radiometrically well constrained (19.6 + 1.4 Ma), and it is convenient to define 20 Ma as the beginning of the main rifting event in the Pannonian Basin. However, in the main depocentres (e.g. the deep basins of the Great Hungarian Plain, Fig. 2) the tuff horizon is missing and the occasionally thick (up to 2 km) basal conglomerates and other continental beds can not be dated accurately. In other words, taking 20 Ma for initiation of rifting is just an assumption in this part of the basin system. Termination of the main phase of extension was originally postulated by Sclater et al. (1980) and Royden (1988) as the end of Sarmatian (Fig. 5), because a regional unconformity can be found below the Pannonian beds that always seals the rifted strata. It has been realized, however, that this is an unconformity that is due to uplift and erosion caused by an early postrift event of basin inversion (Tari 1994; Horv~th 1995). Recent analyses of seismic sections and core samples in areas where this event was weak or absent has led to the conclusion that rifting actually terminated sometime during the Late Badenian (Tari 1994). Again for convenience, we adopt 14 Ma, which is approximately the date
of the Upper Rhyolite Tuff horizon (Fig. 5). In summary, 20-14 Ma is the best estimate for the time period of main rifting in the Pannonian Basin. This is, however, just a good general time bracket, and time shift between the opening of different sub-basins can not be excluded. Fodor et al. (this volume) make attempts to study this problem in the context of their palaeostress data.
C o m p r e s s i o n a l e v e n t s d u r i n g the p o s t r i f t phase
The postrift period of evolution in the Pannonian Basin was thought to be tectonically quiet and controlled by thermal cooling of the lithosphere (Sclater et al. 1980). A most important result of the IBS project has been the demonstration that this was not the case (Figs 4 and 5). Detailed seismic interpretation, including new data from the Lake Balaton (Sacchi et al. this volume) and Hungarian rivers (T6th & Horv~th 1997) were combined with palaeo- and recent stress determinations, and seismotectonic studies (Fodor et al. and Gerner et al. this volume). We arrived at the conclusion that during the postrift period two events of compression could be recognized. Both phases are associated with fault reactivation and structural inversions on local and regional scale (Horv~th 1995). The earlier compressional event took place soon after the termination of the synrift phase (c. 11-8 Ma). The later event started sometime during the late Pliocene and has continued until recent time (c. 3-0 Ma) (Fig. 5). Broad upwarping of the basement from below the Neogene succession are responsible for the characteristic 'inselberg' pattern of present-day outcrops in the Transdanubian Range (Fig. 6). Synchronously with this late-stage uplift, rapid subsidence has taken place in large parts of the Great Hungarian Plain during the Quaternary. Horv~th & Cloetingh (1996) have summarized data showing the pattern and amplitude of the Quaternary differential movements. Their 2D model calculations show that a recent increase in magnitude of horizontal compressional intraplate stress can explain fairly well the observed pattern of Quaternary subsidence and uplift.
Fig. 5. Generalized evolutionary diagram of the Pannonian Basin showing the different time scales, volcanic activity, stratigraphy and main tectonic phases (modified after Tari 1994 and Horv~th 1995). The legend defines the main depositional environments and facies, and lines up the names of the most important lithostratigraphic formations of the Pannonian Basin.
P A N N O N I A N BASIN O V E R V I E W
203
204
F. HORVATH & G. TARI
Fig. 6. Block diagram of the Transdanubian part of the Pannonian Basin to illustrate the structural and stratigraphic conditions, and their relation to surface morphology. The diagram suggests that outcrop and elevation of bedrocks to the North (left) of the Lake Balaton (Transdanubian Range, see in Fig. 2) is the consequence of recent uplift and erosional truncation of the former Neogene sedimentary cover. Keys: 1, Mesozoic-Paleozoic bedrocks; 2, mid-Miocene syn-rift strata; 3, late Miocene to Pliocene postrift strata.
Controls on the stratigraphy o f the N e o g e n e basin fill Establishing a reliable chronostratigraphic framework for the sedimentary fill of the Pannonian Basin, and understanding the variation of depositional environments in space and time (Sacchi et al. this volume) have been longlasting problems. Current debate circles around the applicability of sequence stratigraphic principles to the Pannonian basin. It is generally accepted that marine conditions prevailed practically throughout the synrift phase. However, since the early postrift phase (Sarmatian, Fig. 5a-b) the basin became an isolated brackish-water lake, which has been progressively filled up and lacustrine sedimentary rocks were replaced heterochronously by terrestrial deposits. Vakarcs et al. (1994) put forward an interpretation where all the Middle Miocene-Pliocene third-order sequence boundaries of Haq et al. (1987) were identified in the basin fill. Pogficsfis et al. (1994) used magnetostratigraphic age data to argue that during the Late Miocene to earliest Pliocene the lake level in the Pannonian Basin followed in-phase the global sea level. In contrast, Mattick et al. (1994) argued that observed
sequences are caused by delta lobe switching in a progressively shoaling basin, which was rapidly decreasing in areal extent. Juh~sz et al. (this volume) and Sacchi et al. (this volume) also favour the view that sequences in the postrift fill of the isolated Pannonian lake are not controlled by global eustatic sea-level oscillations. As is shown in Fig. 5 we share this view and, in fact, we see very pronounced tectonic control on postrift stratigraphy. Furthermore, we are realizing that the late-stage terrestrial depositional environments have been particularly sensitive for climatic fluctuations driven by Milankovitchtype cyclicity.
Petroleum systems in the Pannonian Basin The term 'petroleum system' was introduced by Magoon (1988) as a classification scheme for research, exploration and resource assessment. In our view, petroleum system consists of a set of inter-related geological, geochemical, tectonic and hydraulic conditions which is to be met in a particular basin in order to allow formation, migration and preservation of hydrocarbons. System analysis consists, accordingly, of an evaluation of source-rock formation and
PANNONIAN BASIN OVERVIEW maturation, primary and secondary migration, trap formation and infilling, retaining capacity of seals and mapping of the possible petroleum accumulations. Hydrocarbon fields in and around Hungary are shown in Fig. 7. In order to understand their origin we attempt to define petroleum systems in the Pannonian basin with respect to the most recent outcomes of our evaluation. These systems will be reviewed in a sequence from the top of the basin toward the bottom. In other words, the upper system is associated with the Neogene tectonostratigraphic unit, the middle system is related to the Palaeogene tectonostratigraphic unit, and the lower system comprises the Senonian and the Mesozoic-Palaeozoic tectonostratigraphic units. These systems are, however, not always isolated from each other, therefore their communications will be also mentioned. The petroleum systems to be defined can be seen in Figs 8 to 10. It is to be noted, that Figs 9 and 10 are not intended to illustrate a single petroleum system. Instead, they represent idealized cross-sections, showing characteristic structural features and important other elements of petroleum systems available along the sections. The position of the sections shown in Fig. 7 is just for orientation and not location. Neogene tectonostratigraphic unit
Most of the known hydrocarbon pools in the Pannonian Basin are located in the Neogene basin fill. Figure 8 presents an idealized profile in the Great Hungarian Plain showing the distribution of the mature source rocks and potential reservoirs. Vitrinite reflectance (R0) data and maturation history modelling show that, in general, the oilgeneration window (0.6 % < R0 < 1.3 %) is located in the depth range 2.4-4.3 km (Horv~th et al. 1988; Szalay & Koncz 1991). Badenian, Sarmatian and locally Pannonian strata are situated in this range, and their pelagic marls (Formations 10, 13, 16, and 19 in Fig. 5) are characterized by total organic carbon content of 1-2 wt % in average, but locally as high as 5 wt % (Clayton et al. 1994, Clayton & Koncz 1994a). Rock-eval pyrolysis studies indicate that they contain type III and also type II-III kerogens, hence, generation of both oil and gas have occurred (Szalay & Koncz 1991). Migration pathways are usually given by the unconformities at the bottom and top of the synrift strata (3 and 4 in Fig. 8). Structural closures form traps in different structural settings, most obviously above basement highs. This is due to differential compaction which led to drape folds above these highs (Fig. 8).
205
Basement highs are composed of Palaeozoic and Mesozoic rocks, and their occurrence can be explained by two contrasting mechanisms. These highs often represent the tips of crustal blocks, which were rigid enough to avoid significant Miocene extensional deformation and isostatic subsidence relative to the surrounding domains. Alternatively, these highs formed at places of very high extension, where the footwall block of low-angle normal faults experienced large uplift due to unloading (metamorphic core-complex, Tari et al. this volume). The largest fields in compactional anticlines (Play A in Fig. 8) can be found in the southeastern part of the Great Hungarian Plain, above basement highs at the western and eastern flank of the Mak6 trough (Fig. 7). The Algy6 field has been the largest finding in Hungary in 1965, with an initial recoverable oil and gas reserve of about 500 million barrels and 100 billion Sm 3, respectively (K6kai & Pogficsfis 1991). Production derives from the 2500-1600 m depth interval and the individual oil, gas and mixed pools can be found in the following stratigraphic units: fractured basement and overlying basal conglomerate, prodelta turbidites, delta-slope sandstone bodies, and delta-plain channel fill and point-bar deposits. The most favourable reservoirs are the delta-plain sandstones with porosities up to 29%, and permeabilities up to 235 mD (T6rtel Formation, number 23 in Fig. 5). Seals to these reservoirs are given by pelitic beds associated with the prodelta, delta slope and plain environments (Szalay & Koncz 1993). It is to be emphasized that the anticline play has been an obvious concept for hydrocarbon exploration in Hungary, and most seismically detected structures have been tested by drilling. It applies, however, only for the apex of the structures, and significant satellite fields can be present in the vicinity of known fields. It is also possible that hydrocarbon pools can be found apart from compactional anticlines, in stratigraphic traps. Above the deep basins isolated sandstone bodies of gravity flow origin (i.e. basin-floor fans, turbidites and slumps, Phillips et al. 1994) occur, and they can trap hydrocarbons migrating upward from deeper source rocks (Play B in Fig. 8). There is a further important type of positive structures in the Neogene strata of the PannonJan Basin, which was generated during phases of basin inversion. Examples for the early inversion are given by the Kiskunhalas sub-basin (Horv~th 1995) and the Palaeogene flysch trough (L6rincz & Szab6 1993). In the Kiskunhalas area (Fig. 7) the Karpatian to Sarmatian synrift strata were affected by inversion and the
206
F. H O R V A T H & G. TARI
,.e--t
0
o
0
z c~
o . ,...~ 0
el?
0
;~
~.~
.
PANNONIAN BASIN OVERVIEW resulting positive forms have been draped by younger sedimentary beds (Play D in Fig. 8). A few good pools have been found here in Karpatian conglomerates and Badenian sandstones (Formations 8 and 14 in Fig. 5) Further to the N E in the same sub-basin oil accumulations occur in fractured crystalline rocks and Mesozoic carbonates. A set of fields has been found in positive structures above the Szolnok Palaeogene flysch belt (Figs 3 and 7), with the dominance of gases and only occasional occurrence of h y d r o c a r b o n liquids. Gas content is highly variable both vertically and laterally. Hydrocarbons alternate and mix with carbondioxide, nitrogen and inert gases. Carbon isotope (Szalay & Koncz 1991) and rare gas (Ballentine et al. 1991) data suggest deeper crustal origin for the carbon dioxide and nitrogen, and thermogenic, and also bacterial origin for the h y d r o c a r b o n gases. U n d e r favourable conditions large amount of hydrocarbon gases have been accumulated, like in the Hajddszoboszl6 field, which contained 30 x 10 9 Sm 3 recoverable gas reserves. This is a set of fields in Neogene strata, but located above and most probably fed by Palaeogene source rocks. They represents an example of possible c o m m u n i c a t i o n between p e t r o l e u m systems
207
associated with different tectonostratigraphic units (Fig. 4). The Late Pliocene-Quaternary evolution of the Pannonian Basin is characterized by a tectonic reactivation. This includes inversion of the former normal faults and uplift in the western and eastern flanks of the Pannonian Basin, and continuing subsidence and renewal of strike-slip faulting in its central part (Horvfith & Cloetingh 1996). This leads to two types of structural features which can offer good targets for hydrocarbon exploration. Most characteristically in the Zala Basin (Fig. 7) graben inversion has been going on by reactivation of the former main normal fault as a reverse fault (Play A in Fig. 10). Large dome forms and all the top-sealed reservoir rocks in Neogene strata can offer good traps for hydrocarbons migrating from the deeper Miocene source rocks (Formation 10 in Fig. 5) towards the apex of the structure. In the Budafa field most of the oil pools can be found between 800 and 1300 m depth in Pannonian prodelta turbidites and delta-plain mouth-bar sandstones (Formations 20 and 23 in Fig. 5). Initial in-place reserves amounted to 105 million barrel oil and 5 X 109 Sm 3 of gas (K6kai & Pog~icsfis 1991). A n o t h e r type of structure is associated with
Fig. 8. Idealized cross-section to show the petroleum system of the Neogene basin flU at the Great Hungarian Plain. Very approximate location of the section can be seen in Fig. 7. The Neogene tectonostratigraphic unit is bounded by unconformity 4 and the surface. Older tectonostratigraphic units are not distinguished in this figure and shown collectively by the dark stippled pattern.
208
F. HORVATH & G. TARI
Fig. 9. Idealized cross-section to show the petroleum system of the Palaeogene basin fill in North Hungary (left), and of the Mesozoic-Palaeozoic substrata below the Great Hungarian Plain (right). Very approximate location of the section can be seen in Fig. 7. On the left side of the figure the Neogene and Palaeogene tectonostratigraphic units are shown between the surface and unconformity 3", and unconformities 3* and 4, resp. Pz designates the Mesozoic and Palaeozoic rocks of the lowermost tectonostratigraphic unit. On the right side of the figure the Neogene tectonostratigraphic unit is bounded by the surface and unconformity 4. Below that, the Senonian tectonostratigraphic unit (S) can be seen with a limited areal extent. The lowermost unit is composed of Mesozoic (Mz) and Palaeozoic (Crl) rocks which acquired their nappe structure during the Austrian and pre-Gosau phases of the Eoalpine stage. Cr 2 designates the autochthonous basement.
young wrench fault zones. These faults often produce positive flower structures and en echelon folds in Pannonian to Quaternary strata (Play C in Fig. 8). Again, turbidites and deltaplain sandstone reservoirs can trap hydrocarbons of deeper origin (Szalay & Koncz 1993). These two types of play offer only little prospects for future exploration. It is because the inverted basins are known and drilled. In the wrench fault zones the exploration risk is low, but only small fields can be expected.
Paleogene tectonostratigraphic unit In addition to the Palaeogene flysch trough, the Palaeogene epicontinental basin represents an other petroleum system in the second from top tectonostratigraphic unit (Fig. 4). The Lower Oligocene Tard Clay is a good source rock in the Palaeogene basin of northeastern Hungary with a typical total organic carbon content of 2 4 wt %, and type II-III kerogens (Milota et al.
1995). If it reached a depth of more than 2500 m, maturity became adequate to start hydrocarbon generation. The structural style of the Paleogene basin is characterized by high-angle normal faults. Strike-slips and reverse faults are predicted by the interpretation of Tari et al. (1993). The play concept in Fig. 9 depicts a typical structural setting and shows trapping situations which are known in the area. In the Demj6n and Mez6keresztes fields (Fig. 7) oil and some gas have been trapped in Oligocene turbiditic sandstones. More towards the southwest small pools have been found in Sarmatian limestones (Formation 17 in Fig. 5) and Lower Miocene sandstones at the top of uplifted footwall blocks. The Oligocene turbidites are not good reservoir rocks because of their low permeability. Average value is about 2 mD, and even the extreme values do not exceed 100 mD. In addition, the originally small sandstone bodies are usually dismembered and hydrodynamically
PANNONIAN BASIN OVERVIEW
209
Fig. 10. Idealized cross-section to show the petroleum system of the Neogene basin fill in the Zala Basin (left), and of the Mesozoic substrata below the Little Hungarian Plain (right). Very approximate location of the section can be seen in Fig. 7. On the left side of the figure the Neogene tectonostratigraphic unit is bounded by the surface and unconformity 4. This is underlain by the Mesozoic-Palaeozoic tectonostratigraphic unit. On the right side of the figure the Neogene tectonostratigraphic unit is bounded by the surface and unconformity 3*. The Senonian tectonostratigraphic unit (K2) is between unconformities 3* and 4. This is underlain by Mesozoic (mostly Triassic, Tr) and Paleozoic rocks, which acquired their nappe structure during the Austrian and pre-Gosau phases of the Eoalpine stage. Cr designates the autochthonous basement.
isolated by faults. This can be illustrated by the fact that in the Demj6n-East region 461 wells were drilled in an area of 7 km 2 to arrive at a total oil production of 7 million barrels. Better reservoir conditions may be available in the deeper part of the Paleogene basin towards Budapest. If it is not the case, then h y d r o c a r b o n prospects in this tectonostratigraphic unit are very limited.
Senonian and Mesozoic to Palaeozoic tectonostratigraphic units These two tectonostratigraphic units generally constitute an interconnected system, therefore it is reasonable to discuss them together. Due to major differences in Mesozoic stratigraphy, however, we will review separately the petroleum system in the North Pannonian and South Pannonian terranes. The p e t r o l e u m system of the N o r t h Pannonian terrane (Fig. 10, right side) has been
constructed on the basis of data from the Nagylengyel field (Fig. 7), and new knowledge of the deep structure below the Little Hungarian Plain (Clayton & Koncz 1994b; Tari 1994). The field is situated at the central part of the Zala basin in the subsided flank of the Transdanubian Central Range. This is the largest known oil field of the Pannonian basin in Mesozoic reservoir, and the second largest of all fields in Hungary. Original in-place oil reserves amounted to 300 million barrels, and its 40% to 60% has been recoverable (K6kai & Pog~ics~is 1991). As a matter of fact, the total production in the period 1951-1976 already reached 120 million barrels (K6r6ssy 1988). The main reservoir is given by reef limestone in Senonian beds, and the Upper Triassic Hauptdolomit is also often strongly fractured and saturated by oil. Thickness of pools varies between 7-90 m in the Cretaceous and 20-160 m in the Triassic reservoirs. The carbonate reservoirs are highly heterogeneous, small intragranular voids are associated with fracture permeability and
210
F. HORVATH & G. TARI
often karstic holes and even caves are present. This is the explanation for very high yield (up to 4500 barrels a day) of several individual wells. The source rock was penetrated by several wells. It is the Upper Triassic KOssen Marl overlying the Hauptdolomit and interfingering with the Dachstein Limestone. The K6ssen Marl is a good type II source rock with remarkable thickness (up to 600 m) and areal distribution. Total organic carbon content varies between 3 and 6 wt %, with extreme values as high as 19 wt % (Clayton & Koncz 1994b). Seismic coverage of the area is quite poor, and only a couple of interpretable seismic sections is accessible for the Nagylengyel field. Dank (1988) suggested that pools were associated with tilted blocks separated by normal faults of early Miocene age. K6kai & Pog~icsfis (1991) claim that the recent structure of the Nagylengyel field was generated by synrift extensional block faulting of a large older dome. Our model (Fig. 10, right side) suggests that the Nagylengyel structure is a result of complex tectonic evolution, which started with the formation of nappes during the Austrian orogenic phase. The best detachment horizons were the shale layers within the Triassic series, like the K6ssen Marl. This resulted in superposition of Mesozoic sequences and possible repetition of the sourcerock horizons, which may occasionally be located along the thrust contact. The early midMiocene saw a transtensional period of evolution. Nappes were dissected by listric normal faults, and they usually sole out at the former compressional detachment level. Rapid subsidence and sedimentation was interrupted by a compressional event, which resulted in reverse faulting along the former normal faults, and thrusting of the faulted blocks onto Eocene and Lower Badenian strata. It is tempting to speculate that the known and largely exhausted traps in the Nagylengyel region represent only the upper floor of pools, and further reserves can be present in deeper levels below and/or around the known field. Furthermore, the play concept may work elsewhere at the flank of the Transdanubian Central Range where the KOssen Marl can be present. The petroleum system characterizing the South Pannonian terrane relies on two recently realized recognitions. On one hand, new field observations and reinterpretation of well-logs have shown that Toarcian black shale in the Mecsek unit of the South Pannonian terrane is an excellent Type II source rock with total organic carbon content between 4 and 8 wt %. On the other hand, recent drillhole data and seismic interpretation (Pap et al. 1992, Grow et
al. 1994) confirmed earlier suggestions (Horwith
& Rumpler 1984) about the presence of Mesozoic and metamorphic crystalline nappes in the substratum of the Great Hungarian Plain (Tari et al. this volume). The novelty of these recognitions implies that the petroleum system in Fig. 9 is rather speculative, and only a few deliberate explorations have been conducted to test the concept. There are, however, promising findings. These include more than 20 smaller fields that have been found so far in fractured metamorphic basement rocks and the overlying basal conglomerates. These basement rocks have been drilled always at places where they constitute basement highs (Play D in Fig. 9). The traditional explanation for the formation of these pools, i.e. lateral migration of hydrocarbons out of mature Neogene source rocks in the nearby deep basins, may not be always the case. K6kai & Pogficsfis (1991) showed two examples where the crystalline basement high represents only a relatively thin allochthonous flake above Mesozoic rocks. During the preSenonian orogeny NW-vergent nappes formed, and the individual thrust sheets are composed of Triassic to Lower Cretaceous rocks, or older metamorphitic rocks (like granite, gneiss and schists) with, or more frequently, without the Mesozoic cover (Fig. 9). Nappes were reactivated during the midMiocene period of extension, and normal or oblique slip occurred along their contact characterized by ramp-flat geometry. Locally, flats accommodate significant amount of extension, and it resulted in rising and subcropping of highly metamorphic footwall blocks (Tari et al. this volume). As a consequence, elevated blocks can exhibit completely overmature rocks and less mature Mesozoic rocks, in turn, are located at their flanks in deeper position. Interestingly enough it appears that Quaternary tectonic processes have had an impact on the fluid pressure in the fractured metamorphic reservoirs. If one plots the fields on a map showing areas of Quaternary uplift and subsidence, it turns out that most of the geopressured pools are situated in areas of uplift. This suggests that overpressure is mainly controlled by recent uplift and good seal, which hamper fast decay of the excess pressure (Horvfith 1995). In summary, the main point of the play concept is to consider those basement nappes either at highs or flanks, where hydrocarbon generation out of the Toarcian shales could have occurred as late as Neogene. Then, their internal structure is to be deciphered in order to predict traps in deeper position and not
PANNONIAN BASIN OVERVIEW necessarily at the apex of the structure. High exploration risk can only be reduced by using sophisticated seismic technology and intuitive interpretation.
211
throughout the project. The second author (G.T.) acknowledges the support of his company to contribute to the project. Technical preparation of this paper and other Hungarian papers in this volume were partially supported by the Hungarian National Science Foundation (OTKA T 019393 ).
Conclusions The IBS project has b e e n a good example of beneficial co-operation b e t w e e n Hungarian and foreign scientists, and academia and the hydrocarbon industry. R e c e n t progress in understanding the structural e v o l u t i o n of the Pannonian Basin can have a major impact on the future hydrocarbon exploration strategy. Most important points which have e m e r g e d are as follows. 9
9
9
9
Documentation of Alpine nappes all over the area in the substrata of the Pannonian Basin (i.e. Early A l p i n e tectonostratigraphic unit), and the presence of Mesozoic source rocks calls for a deliberate search for traps in these deeper structural levels. This would require new seismic data acquisition and sophisticated interpretation. New k n o w l e d g e raises doubts that the Palaeogene basin is a promising exploration target. A d d i t i o n a l work should be performed, however, to clarify the mechanism of its formation. If it is indeed compressional rather than extensional in origin then m o r e complex m a t u r a t i o n history and migration pathways can be expected. The recognition of complex Neoalpine tectonic evolution of the Pannonian Basin, particularly the events of compression during the postrift phase associated with vertical m o v e m e n t anomalies and fault reactivation have an important application for hydrocarbon exploration. These latestage events have remarkably affected the m a t u r a t i o n history of source rocks, the pathways of secondary migration, trap formation and fluid pressure conditions. The lesson learned in the Pannonian Basin has a message for other intramontain basins in the world. Namely, characterization of their evolution by a phase of rifting followed by a thermal subsidence can be a gross oversimplification. M o r e sophisticated m o d e l s of basin evolution are required and the success of hydrocarbon exploration strongly d e p e n d s on the reliability of these models.
The authors are grateful to the European Community and the management of the IBS Project for funding of the project and the stimulating atmosphere
References BALLA, Z. 1988. Clockwise paleomagnetic rotations in the Alps in the light of the structural pattern of the Transdanubian Range (Hungary). Tectonophysics, 145, 277-292. BALLENTINE, C. J., O'NIoNS, R. K., OXBURGH, E. R., HoavXrH, E & DEAK, J. 1991. Rare gas constraints on hydrocarbon accumulation crustal degassing and groundwater flow in the Pannonian basin. Earth and Planetary Science Letters, 105, 229-246. CSONTOS, L., NAGYMAROSY, A., KOVA~, M. & HoRVATH,E 1992. Tertiary evolution of the intraCarpathian area: a model. Tectonophysics, 208, 221-241. CLAYTON,J. L. & KONCZ,I. 1994a. TGtkomlGs-Szolnok petroleum system of southeastern Hungary. In: MAGOON,L. B. & Dow, W. G. (eds) The petroleum system - f r o m source to trap. Amererican Association of Petroleum Geologists Memoirs, 60, 587-598. -& -1994b. Petroleum geology of the Zala basin, Hungary. American Association of Petroleum Geologists Bulletin, 78, 1-22. , - - , KING, J. D. & TATAR,E. 1994. Organic geochemistry of Crude oils and source rocks, BGkGs basin. In: TELEKI,P. G., MATTICK,R. E. & KOKAI, J. (eds) Basin Analysis' in Petroleum Exploration. Kluwer Academic Publishers, Dordrecht, 161-185. DANK, V. 1988. Petroleum geology of the Pannonian basin, Hungary, an overview. In: ROYDEN,L. & HoRVATH,E (eds) The Pannonian basin a study in basin evolution. American Association of Petroleum Geologists Memoirs, 45, 31%331. FODOR, L., CSONTOS, L., BADA, G., GYGRFI, I. & BENKOVICS,L. 1999. Tertiary tectonic evolution of the Pannonian Basin system and neighbouring orogens: a new synthesis of paleostress data. This volume. GERNER, P., BADA, O., DOVI~NYI, P., MOLLER, B., ONCESCU, M. C., CLOET~NGH,S. & HORVATH, E 1999. Recent tectonic stress and crustal deformation in and around the Pannonian Basin: data and models. This volume. GROW,J. A., MATFICK,R. E., Bt~RCZI-MAKK,A., PIER(5, CS., HAJDI), D., POGACSAS, GY., VARNAI, P. & VARGA, E. 1994. Structure of the Bdkds basin inferred from seismic reflection, well and gravity data. In: TELEKI, EG., MATTICK,R.E. & KOKAI,J. (eds) Basin Analysis in Petroleum Exploration. Kluwer Academic Publishers, Dordrecht, 1-38. GYGRFI, I., CSONTOS, L. & NAGYMAROSY,A. 1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian basins. This volume.
212
F. HORV/kTH & G. T A R I
HAQ, B. U., HARDENBOL,J. d~rVAIL, P. R. 1987. Chronology of fluctuating sea levels since the Triassic. Science, 235, 1156-1167. HORVATH, E 1993. Towards a mechanical model for the formation of the Pannonian basin. Tectonophysics, 226, 333-357. 1995. Phases of compression during the evolution of the Pannonian basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology. 12, 837-844. -& CLOETINGH, S. 1996. Stress-induced late-stage subsidence anomalies in the Pannonian basin. Tectonophysics, 266, 287-300. -& RUMPLER, J. 1984. The Pannonian basement, extension and subsidence of an Alpine orogene. Acta Geologica Hungarica, 27, 229-235. , DOVENYI, P., SZALAY,A. & ROYDEN, L.H. 1988. Subsidence, thermal and maturation history of the Great Hungarian Plain. In: ROYDEN, L. & HOnVATH, E (eds) The Pannonian basin a study in basin evolution. American Association of Petroleum Gelogists, Memoirs, 45, 355-375. HUNGARIAN GEOLOGR2AL SURVEY. 1998. Mineral Resources o f Hungary. Open file report, Budapest. K6ROSSY, L. 1988. Hydrocarbon geology of the Zala basin in Hungary. General Geology Review, 23, 3-162. KOKAL J. 1994: Exploration history and future possibilities in Hungary. In: POPESCU, B. M. (ed.) Hydrocarbons o f eastern Central Europe. Springer Verlag, Berlin, 147-173. --& POGACSAS, GY. 1991. Hydrocarbon plays in Mesozoic nappes, Tertiary wrench basins and interior sags in the Pannonian basin. First Break, 9, 315-334. L6R1NCZ, K. & SZABO, E 1993. Seismic analysis of multiphase tectonism in the central part of the Pannonian basin in Hungary. In: SPENCER,A. M. (ed.) Generation, Accumulation and Production o f Europe ~ Hydrocarbons, EAPG Special Publications 3, Springer, Berlin, 311-323. MAGOON, L. B. 1988. The petroleum system, a classification scheme for research, exploration and resource assessment. In: MA(;OON, L. B. (ed.) Petroleum systems o f the United States. USGS Bulletin, 1870, 2-15. MAITICK, R. E., RUMPLER, J., UJFALUSY,A., SZANYI, B. & NAGY, I. 1994. Sequence stratigraphy of the Bdkds basin. In: TELEKI, P G., MATTI('K, R. E. & KOKAI, J. (eds) Basin Analysis in Petroleum Exploration. Kluwer Academic Publishers, Dordrecht, 39455. , TELEKI, P. G., PHILLIPS,L., CLAYTON,J. L., DAVID, G., POGACSAS,GY., BARDOCZ,B. & SIMON,E. 1996. Structure, stratigraphy, and petroleum geology of the Little Plain Basin, Northwestern Hungary. American Association o f Petroleum Geologists Bulletin, 80, 1780-1800. MILOTA,K., KOV/,CS,A. & GALICZ, Z. 1995. Petroleum potential of the North Hungarian Oligocene sediments. Petroleum Geoscience, 1, 81-87. PAP, S., NOREG,V. & PAP HASZNOS, I. 1992. Exploration of the D6vav~nya south basement structure for -
-
-
hydrocarbons a case history. Geophysical Transactions, 37, 211-228. PATRASCU, ST., PANAIOTU, C., SECLAMAN, M. PANAIOTU,C. E. 1994. Timing of rotational motion of Apuseni Mts. (Romania): paleomagnetic data from Tertiary magmatic rocks. Tectonophysics, 233, 163-176. PHILLIPS, R. L., REVESZ, I. & BI~RCZI.,I. 1994. Lower Pannonian deltaic-lacustrine processes and sedimentation B6k6s basin. In: TELEKI,P.G., MATTICK, R.E. & KOKAI, J. (eds) Basin Analysis in Petroleum Exploration. Kluwer Academic Publishers, Dordrecht, 67-82. POGACSAS, GY., MATTICK, R. E., ELSTON, D. P., HAMOR, T., JAMBOR, A., LAKATOS,L., LANTOS,M., SIMON, E., VAKARCS, G., VARKONYI, L. & VARNAI, E 1994. Correlation of seismo - and magnetostratigraphy in southeastern Hungary. In: TELEKI, P G., MATTICK, R. E. &; KOKAI, J. (eds) Basin Analysis in Petroleum Exploration. Kluwer Academic Publishers, Dordrecht, 143-160. RATSCHBACHER, L., FRISCH, W., LINZER, H. G. & MERLE, D. 1991. Lateral extension in the Eastern Alps, part 2; structural analysis. Tectonics, 10, 257-271. ROYDEN, L. H. 1988. Late Cenozoic tectonics of the Pannonian basin system. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian basin, a study in basin evolution. American Association of Petroleum Geologists, Memoirs, 45, 27-48. -& BALm, T. 1988. Early Cenozoic tectonics and paleogeography of the Pannonian and surrounding regions. In: Royden, L. H. & Horv~ith, F. The Pannonian basin, a study in basin evolution. American Association of Petroleum Geologists, Memoirs, 45, 1-16. SAC('HI, M., HORVATH,E & MAGYARI,0. 1999. Role of unconformity-bounded units in stratigraphy of continental record: a case study from the Late Miocene of western Pannonian Basin, Hungary. This volume. SCALTER, J. G., ROYOEN, L. H., HORVATH, E, BURGHHEL, B. C., SEMKEN, S. & STE~mNA, L. 1980. The formation of the intra-Carpathian basins as determined from subsidence data. Earth and Planetary Science Letters, 51, 139-162. SZAFIAN,P. & TARI, G. 1995. Preliminary results on the gravity modelling of a crustal transect in the Alpine-Pannonian junction, hi: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of a n Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill o f the Western Pannonian Basin. A A P G Internat. Conference and Exhibition, Nice, France, Guidebook to fieidtrip 6, Hungary, 107-118. SZALAY, A. & KON('Z, I. 1991. Genetic relations of hydrocarbons in the Hungarian part of the Pannonian basin. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production o f Europe's Hydrocarbons, E A P G Special Publications 1, Oxford University Press, 317-322. & -1993. Migration and accumulation of oil and natural gas generated from Neogene source rocks in the Hungarian part of the Pannonian -
PANNONIAN BASIN OVERVIEW basin. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons IlL E A P G Special Publication 3. Springer, Berlin, 303-309. TARI, G. 1994. Alpine tectonics of the Pannonian basin. PhD thesis, Rice univ., Houston, Texas. 1995. Eoalpine (Cretaceous) tectonics in the Alpine-Pannonian transition zone. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of an Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G Internat. Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6, Hungary, 133-155. 1996. Neoalpine tectonics of the Danube basin (NW Pannonian basin, Hungary). In: ZmGLER, RA. & HORVATH,F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoir, 2. Editions du Musdum National d'Histoire Naturelle, Paris, 439-454. HORVATH, E 1995. Middle Miocene extensional collapse in the Alpine-Pannonian transition zone. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of an Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G Internat. Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6, Hungary, 75-105. , BALDI, T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian Basin: a geodynamic model. Tectonophysics, 226, 433~455. , DOVI~NYI,P., DUN KL,I., HORVATH,F., LENKEY,L., SZAWAN, R & TOTn, T. 1999. Lithospheric structure of the Pannonian basin derived from seismic, gravity and geothermal data. This' volume.
-
-
-
-
-
-
&
213
, HORVATH,E & RUMPLER,J. 1992. Styles of extension in the Pannonian basin. Tectonophysics, 208, 203-219. , - - & WEIR, G. 1995. Palinspastic reconstruction of the Alpine/Carpathian/Pannonian system. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of an Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G Internat. Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6, Hungary, 119-131. TOTH, T. & HORVATH,E 1997. High resolution seismic profiling for neotectonic investigations. In: MAROSI, S. & MESK0, A. (eds) Seismic risk of the Paks Nuclear Power Plant of Hungary. Publishing House of the Hungarian Academy of Sciences, Budapest, 123-152. TRUMPY,R. 1973. The timing of orogenic events in the Central Alps. In: DE JONG, K.A. & SCHOLTEN,R. (eds) Gravity and Tectonics. Wiley and Sons, 229-251. VAKARCS, G., VAIL, V. R., TARI, G., POGACSAS, GY., MATTICK, R. E. & SZAB0, A. 1994. Third-order Miocene-Pliocene depositional sequences in the prograding delta complex of the Pannonian basin. Tectonophysics, 240, 81-106. Y1LMAZP. O., NORTON,I. 0., LEA RY,D. & CHUCHLA,R. J. 1996. Tectonic evolution and paleogeography of Europe. In: ZIEGLER, R A. & HORVATH, F. (eds) Structure and Prospects" of Alpine Basins and Forelands. Peri-Tethys Memoir, 2, Editions du Mus6um National d'Histoire Naturelle, Paris, 47-60.
Lithospheric structure of the Pannonian basin derived from seismic, gravity and geothermal data G . ~ B O R T A R I 1, P I ~ T E R D O V I ~ N Y I 2, I S T V , ~ N D U N K L 3, F R A N K LASZL0
HORVATH
2,
L E N K E Y 2, M I H A I S T E F A N E S C U 4, P I ~ T E R S Z A F I A N 2 & TAMAS TOTH 2
1BP Amoco, 501 WestLake Park Boulevard, Houston, T X 77079, USA 2E6tvOs University, Geophysical Department, Ludovika t~r 2, 1083 Budapest, Hungary 3Laboratory for Geochemical Research, BudaOrsi ~tt 45, Budapest, Hungary 4 A M O C O Romania, Suite 201, Sevastopo113-17, 1-Bucharest, Romania Abstract: The structure of the Pannonian basin is the result of distinct modes of Mid-Late Miocene extension exerting a profound effect on the lithospheric configuration, which continues even today. As the first manifestation of extensional collapse, large magnitude, metamorphic core complex style extension took place at the beginning of the Mid-Miocene in certain parts of the basin. Extrapolation of the present-day high heat flow in the basin, corrected for the blanketing effect of the basin fill, indicates a hot and thin lithosphere at the onset of extension. This initial condition, combined with the relatively thick crust inherited from earlier Alpine compressional episodes, appears to be responsible for the core complex type extension at the beginning of the syn-rift period. This type of extension is well documented in the northwestern Pannonian basin. Newly obtained deep reflection seismic and fission-track data integrated with well data from the southeastern part of the basin suggests that it developed in a similar fashion. Shortly after the initial period, the style of syn-rift extension changed to a wide-rift style, covering an area of much larger geographic extent. The associated normal faults revealed by industry reflection seismic data tend to dominate within the upper crust, obscuring preexisting structures. However, several deep seismic profiles, constrained by gravity and geothermal modeling, image the entire lithosphere beneath the basin. It is the Mid-Miocene synrift extension which is still reflected in the structure of the Pannonian lithosphere, on the scale of the whole basin system. The gradually diminishing extension during the Late Miocene/Pliocene could not advance to the localization of extension into narrow rift zones in the Pannonian region, except some deep subbasins such as the Mak6/B6k6s and Danube basins. These basins are underlain coincidently by anomalously thin crust (22-25 km) and lithosphere (45-60 km). Significant departures (up to 130 mW m 2) from the average present-day surface heat flow (c. 90 mW m 2) and intensive Pliocene alkaline magmatism are also regarded as evidence for the initiation of two newly defined narrow rift zones (Tisza and Duna) in the PannonJan basin system. However, both of these narrow rifts failed since the final docking of the Eastern Carpathians onto the European foreland excluded any further extension of the back-arc region.
Since the last major overview on the evolution of the Pannonian basin was published (Royden & Horvfith 1988) a number of papers have been published which focus on various new aspects of continental extension in this major Mediterranean back-arc basin (e.g. Tari et al. 1992, 1993; Horvfith 1993, 1995). This paper summarizes the results of an integrated approach to the lithospheric-scale evolution of the intra-Carpathian area. The overview is primarily based on geophysical data (reflection seismic, gravity and geothermal data) complemented with geological data (stratigraphy/structure based on wells, fission track data and magmatic petrology). The Pannonian 'wide' or 'diffuse' rift system
sensu Buck (1991) and Ruppel (1995), respectively, was superimposed on several metamorphic core complexes during the Mid-Miocene (Tari & Bally 1990). The earlier extreme crustal extension is not recognizable by traditional subsidence analysis and thermal data (e.g. Sclater et al. 1980; R o y d e n & DOv6nyi 1988; Demetrescu & Polonic 1989; Lankreijer et al. 1995). It is the style of faulting observed on seismic reflection profiles, and the characteristically retrograde metamorphism of the basement (Szalay 1983), which suggest large-magnitude extension forming metamorphic core complexes (sensu Davis & Lister 1988) in the SE Pannonian basin (Tari et al. 1992). Recent fission-track dating of
TAR[, G., DOVI~NYI,P., DUNKL,I., HORVATH,E, LENKEY,L., STEFANESCU,M., SZAFIAN,E & TOTH,T. 1999. Lithospheric structure of the Pannonian basin derived from seismic, gravity and geothermal data. In: DUaAND, B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 215-250.
216
G. TARI E T A L .
amphibolite to greenschist facies basement rocks in this critical region (see below) verifies this preliminary interpretation. The extreme extension of the upper crust revealed by industry seismic reflection data can also be observed on deep crustal-, or sometimes even lithospheric-scale seismic reflection profiles in the Hungarian part of the basin system (e.g. Posgay et al. 1995, 1996; Hajnal et al. 1996). The integration of this particular data set with gravity modeling results (e.g. Szafi~in & Tari 1995; Szafifin et al. 1997) provided additional constraints on our models. Similar refinement of existing geothermal models was achieved by taking into account the blanketing effect of the thick sedimentary cover (see below). This paper presents original data and interpretations focusing on the SE Pannonian basin after a brief geological introduction (Figs 1 and 2). This area is presented at a sub-regional scale to illustrate typical data constraints on the structure of the entire lithosphere. However, the following description and interpretation of gravity and geothermal conditions cover the entire Pannonian basin system. Several specific
problems on the scale of the entire basin are also discussed, including the influence of prerift rheology on the mechanism of extension (e.g. Cloetingh et al. 1995) and the overall style of lithospheric extension (e.g. Buck et al. 1988). In addition, special attention is given to the temporal/spatial transitions between distinct extensional modes as inferred for highly extended continental areas (Hopper & Buck 1996).
Definition of Neogene structural periods in the Pannonian basin In this work, the traditionally undivided MidMiocene synrift period in the Pannonian basin (e.g. Tari et al. 1992) is separated into an Early-Mid-Badenian 'wide-rift' style and a Karpatian 'metamorphic core complex' style extensional period (Fig. 3), following the terminology of Buck (1991, 1993). As to the synrift/postrift boundary, Tari (1994) placed it between the Upper and Lower Badenian (14.8 Ma) (Fig. 3). Obviously, this particular unconformity is not isochronous across the entire basin, but the 14.8 Ma boundary
Fig. 1. Simplified geological and index map of the Carpatho-Pannonian region. Thermal modeling transects AA' and BB' are shown in Figs 18 and 19, respectively. Transect C is shown in Fig. 24.
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
217
Fig. 2. Regional geologic subcrop map of the SE Pannonian basin, modified from Csontos et al. (1992). Note that all the Eoalpine (Cretaceous) tectonic subunits of the South Pannonian unit are mostly covered by thick Neogene basin fill. Exposures of the we-Tertiary basement are limited to the Apuseni Mts. of Romania, Mecsek/Villfiny Mts of Hungary and Papuk Mts of Croatia (cf. Fig. 1). Locations of a detailed map (Fig. 4) and structure transects A and B (Figs 10 and 11) are also shown.
seems to be a good first approximation. Our Paratethyan correlation scheme is based on the work of Vakarcs et al. (1995, 1998).
Lithospheric structure of the SE Pannonian basin The SE Pannonian basin covers the junction between Hungary, Romania and Serbia (Fig. 2). This paper presents a short review of the regional geology of the area, followed by a description of recently acquired deep reflection seismic and fission track data. Based on the combination of these data sets, two new crustal/lithospheric-scale structure transects are presented. Regional geology
The extensional collapse of the P a n n o n i a n region involved two microplates which were
juxtaposed during the Early Miocene (Tari et al. 1993). The Southern P a n n o n i a n block is subdivided into roughly E N E - W S W - t r e n d i n g units (Fig. 4). This subdivision is based on the distinct Mesozoic facies differences within the pre-Tertiary basement, which suggests a continental margin to the north and an oceanic basin to the south (e.g. B l e a h u et al. 1994). At present, these units are organized in a northvergent Eoalpine (i.e. Cretaceous) folded belt, cropping out in the Apuseni Mts of R o m a n i a (e.g. Rozlozsnik 1936; Ianovici et al. 1976; Bleahu et al. 1981) and the Mecsek and Villgmy Mts of Hungary (e.g. N6medi-Varga 1983). The m a j o r E o a l p i n e t e c t o n o s t r a t i g r a p h i c units include from N to S (or from b o t t o m to top) the M e c s e k - S z o l n o k , the Villfiny-Bihor, the B 6 k 6 s - L o w e r Codru, the U p p e r C o d r u Biharia and the Mures units (Fig. 4). The lowermost M e c s e k - S z o l n o k unit outcrops only in the Mecsek Mrs of Southern H u n g a r y (Fig. 2). Besides its characteristic
218
G. TARI E T A L . Eoalpine fold-thrust belt (e.g. Teleki et al. 1994). The ongoing tectonics in the region, however, are best understood in terms of neotectonic inversion, which is largely responsible for the uplift of the Apuseni Mts during Quaternary times (Horvfith et al. 1995).
carl P-_
__.
GLOBAL
m
STAGES
Ih
STAGES
~
PANNONIAN U.I ..I a Q
m
12.7
SERRAVALLIAN
SARMATIAN
(~
m
R e f l e c t i o n s e i s m i c data
14.8
-15
u.I
LANGHIAN
z I.IJ
16.4
184
KARPATIAN
u>
17.3
OTTNANGIAN I~
O
BURDIGALIAN
-20 O
18.7
EGGENBURGIAN r
UJ 20.5
20.5
..! AQUITANIAN
--
cn cn
23.8
EGERIAN nILl
-25 _
-
~ ~
"
O 0
-30
~
28.5
28.5
RUPE/IAN
KISCELLIAN
.J
O
-
~
-
O
#.
,-I
--
33.7
33.7
Fig. 3. Simplified chronostratigraphic and tectonic framework of the Neogene and Quaternary in the Pannonian basin. The Mediterranean/Paratethyan (global/regional) correlation framework is based on the compilation of Vakarcs et al. (1995, in press).
Mesozoic facies, its most peculiar succession is the Senonian-Palaeogene Szolnok flysch (e.g. Nagymarosy & B~ildi-Beke 1993). The Vilhiny-Bihor unit crops out in the Vilhiny Mts of South Hungary and in the Apuseni Mts of West Romania. Besides the Papuk Mts of Croatia (Fig. 2), the Codru-Biharia units are mostly exposed in the Apuseni Mts. The physical connection between these regions in the subsurface of the Great Hungarian Plain is constrained by numerous wells (e.g. B6rcziMakk 1986). The present-day structure of the SE Pannonian basin is clearly dominated by Miocene extensional features superimposed on the Cretaceous
Two perpendicularly oriented, deep reflection seismic profiles provide the best constraints on the lithospheric-scale structure of SE Hungary (Fig. 4). The first profile concerns the NW-SEtrending Pannonian GeoTraverse (PGT-1) line (Posgay et al. 1990, 1995; Posgay & Szentgy6rgyi 1991; Nagymarosy & B~ildi-Beke 1993). Here we present a reinterpretation of the PGT-1 section in a line drawing manner (Fig. 5). The Mid-Hungarian Line which delimits the South Pannonian unit to the north (sensu Csontos et al. 1992) is identified as a zone of strong reflectivity between 0 and 10 km, through the entire crust. Along this poorly defined zone large-scale strike-slip movements occurred during the Early Miocene (Tari et al. 1993). This significant wrench zone and the associated volcanics (K6r6ssy 1992), however, do not have a clear seismic expression. Similarly, the lateral extent of the Palaeogene Szolnok flysch basin to the south is better constrained by the available well control than by seismic data. Farther to the south, S-dipping low-angle normal faults appear from about 100 km as the regional line enters the deeper part of the Great Hungarian Plain suggesting considerable NNW-SSE extension along the section. The alternation of unmetamorphosed Permomesozoic units with Hercynian crystalline rocks in the basement (Fig. 4b) has little to do with the normal faults or with the regional-scale wrench faults. Rather, this particular map-view pattern is interpreted as the result of Cretaceous thrust faulting. Based on industry boreholes, Pap (1990) documented several examples of Eoalpine thrust sequences in the pre-Tertiary basement of SE Hungary. Some consistently dipping reflector packages seem to define nappe units separated by major overthrust planes, generally verging to the north (Fig. 5). A major synclinal feature dominates the upper/middle crust between about 70 and 110 km. The lowermost part of the crust tends to be very reflective defining a sharp reflection Moho between 26 and 20 km. The surface of the Moho displays small undulations, except at the southern end of the section where it significantly upwarps beneath the B6k6s basin. The lithosphere/asthenosphere boundary
L I T H O S P H E R I C S T R U C T U R E O F P A N N O N I A N BASIN
219
(a)
(b) Fig. 4. (a) D e p t h of pre-Tertiary basement in the SE Pannonian Basin after Kil6nyi et al. (1991) and Dicea et al. (1984). For location see Fig. 2. (b) Subcrop of pre-Tertiary basement in the SE Pannonian Basin after Fiil6p et al. (1987). For location see Fig. 2. The detailed fission track results are listed in Table 1, see also Figs 8 and 9, for zircon and apatite age spectra, respectively.
220
G. TARI E T A L .
Fig. 5. Line drawing reinterpretation of the PGT-1 deep reflection seismic line of Posgay and SzentgyOrgyi (1991), Posgay et al. (1995), simplified from Tari (1994). For location see Fig. 2.
may also have a subdued seismic expression (Posgay et al. 1995), which suggests considerable local thinning of the regionally already thin lithospheric mantle (Fig. 5). The migrated version of the E N E - W S W trending PGT-4 reflection seismic section and its line drawing interpretation are shown in Fig. 6. Note that an additional 25 km was added to this already published profile (Hajnal et al. 1996; Posgay et al. 1996) at its northeastern end (Fig. 4). The details of the seismic acquisition and processing will be published elsewhere (T6th, in preparation). The most prominent feature imaged on this line is the Battonya-Pusztaf61dv~ir basement high separating the H 6 d - M a k 6 basin from the B6k6s basin (Fig. 4a). These deep (>6 km) basins are bound by low-angle normal faults ( G r o w et al. 1989, 1994). Note that in our interpretation the B6k6s basin is controlled by a down-to-the-WSW low-angle normal fault as opposed to other interpretations (e.g. Hajnal et al. 1996; Posgay et al. 1996). This new interpretation is based on a set of regional seismic profiles documenting the regional switch of polarity of normal faulting underneath the B6k6s basin (Tari 1994). Furthermore, the new addition to
the PGT-4 line at its northeastern end also clearly shows the polarity change between the Battonya-Pusztaf01dv~ir and the Sarkad highs (Fig. 6). The detachment of the major normal faults occurs on top of a distinctly reflective lower crust, which is best developed underneath the AlgyO high. The Moho is fairly flat along this section, in constrast to the perpendicularly oriented PGT-I line (cf. Fig. 5). The Pannonian Geotraverse deep reflection seismic profiles clearly reveal the presence of a thin, attenuated crust and lithosphere in the Pannonian basin (Fig. 7). Note that the average crustal and lithospheric thickness values for a normal lithosphere are 33 and 120 km, respectively. Along the PGT lines (Figs 5 and 6) these values are generally around 25 and 60 km, but under the B6k6s basin the depth to the Moho is slightly less than 22 km, whereas the thickness of the lithosphere is approximately 40 kin. Fission-track
data
As part of the IBS project, several core-samples from the Hungarian side of the SE Pannonian basin were analysed for fission track age data
Fig. 7. Lithospheric thickness in the Pannonian-Carpathian region based on seismologic and magnetotelluric data, after Horv~ith (1993).
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
Fig. 6. PGT-4 deep reflection seismic line of Posgay et
221
al. (1996), Hajnal et al. (1996) with its recently acquired and processed extension to the NE (T6th in prep.). The fission track ages of core-samples obtained from nearby wells are projected into the line drawing reinterpretation below (cf. Fig. 4a). For location see Fig. 2.
222
G. TARI E T A L . .=_
~
b
~
+1
,,-,,.i +1+1+1
+1+1+1+1
~
V ~ V
V ~ V V
V V V V
V V V
Im..-
V .~
o,_1 cJ'~
o
~
._=~ d
b
. . . .
d~
,,,-:.
< .,,-, ~
9 9
..~b
o,.-~
-....
H
X
~
~
+~
"-...." r r
,,,_,
=~..~
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
223
(Table 1). For a brief description of the experimental procedure see Appendix A. Several samples in this apatite/zircon fission track study were obtained from the crystalline basement and others were derived from the U p p e r M i o c e n e to Pliocene post-rift sedim e n t a r y succession. W h e r e a s the b a s e m e n t samples significantly constrain the upper crustal structure depicted in transect B (see Fig. 11), the sediment samples give an indirect insight into the regional evolution of the area. The fission track ages of the sediment samples are only interpreted in the context of their age spectras (Figs 8 and 9). Jurassic and Cretaceous zircon fission-track ages were o b t a i n e d from several b a s e m e n t samples (Table 1). The Jurassic age of Permian rhyolite (Bat-18/2 core, 165 + 18 Ma) can be
Fig. 8. Zircon fission-track age spectra from core samples. For location see Fig. 4b, measurement parameters and results are listed in Table 1. Peaks of the age-spectra were calculated following the method of Hurford et al. (1984).
Fig. 9. Apatite fission-track age spectra from core samples. For location see Fig. 4b, measurement parameters and results are listed in Table 1. Peaks of the age-spectra were calculated following the method of Hurford et al. (1984).
224
G. TARI E T A L .
explained in terms of a Jurassic overprint event (probably rifting). Alternatively, it reflects younging due to the burial beneath a Mesozoic cover sequence, overthickened during Alpine thrusting (cf. Fig. 8). A gneiss sample from the PusztafOldv~r area (Pf-89/3 core) was dated as Upper Cretaceous (80.3 _+ 10 Ma). This age is quite common in the Carpathian/Pannonian region, and similar 4~ ages were obtained at the western and eastern margins of the Pannonian basin by Dallmeyer et al. (1992, 1994). This age group is best understood as the result of cooling after the Late Cretaceous thrusting. It is of primary importance that one gneiss sample (A-626/4a) from the basement of the Algy/~ high (Fig. 4) provided Miocene zircon age (17.4 + 0.8 Ma). This age coincides with the beginning of Karpatian sedimentation (B6rczi et al. 1988) and thus also coincides with the onset of extension (Fig. 3). It is remarkable that at the western margin of the Pannonian basin, from the lower plate of the Rechnitz metamorphic core complex (Fig. 2) identical zircon cooling ages (average: 17.3 Ma) were obtained by Dunkl (1992); Dunkl & Dem6ny (1997). This paper also attributes the Miocene age to the rapid uplift of middle crustal rocks in the lower plate of the Algy6 detachment system. Detrital zircon ages of the Neogene sediments reflect on the distribution of cooling ages of the eroding areas at the period of intensive denudation during Neogene. The results are presented in the form of column diagrams and age-spectra (Figs 8 and 9) following the method of Hurford et al. (1984). The sample from a Miocene conglomerate (Szeged-24/1) contains a zircon population of dominantly Triassic age. The colour and the morphology of the crystals are identical. They probably originated from a volcanic or subvolcanic formation. A sandstone sample also (Bat13/2) contains also zircons with rather old fission track ages. This zircon population originated probably from a Permian rhyolite without any indication of Alpine cooling ages. Note that sandstone samples (Bat-80/1 and A-25/1) contain some grains with Mid-Miocene synrift cooling ages (i.e. 15-20 Ma). Apparently, during the deposition of these beds the lower plate of metamorphic core complex(es) has been already exposed at the surface. The apatite fission-track ages of several granite and rhyolite samples (Bat-15/8, Bat-5/5
and Bat-18/2) range between 64 and 74 Ma. The recent temperatures of these samples (72-76~ and the short heating time presumably did not cause significant rejuvenation. Thus, the results can be considered as cooling ages occurring soon after the Cretaceous nappe stacking period. The sample of a micaschist (Pf-88/9) contains very limited number of reliable grains. So the figure of around 42 Ma is regarded as only a week indication of Palaeogene or earlier cooling events (prior the recent thermal overprint). Taking into account the present-day temperature range of the Battonya sediment samples (i.e. <77~ the fission-track age of the detrital apatite grains has probably not been rejuvenated, so the single grain ages can be evaluated as provenance indicators (Fig. 7). The sandstone samples (Bat-19/1 and Bat-13/2) provided apatite grains older than Miocene. Based on the age, these sediments could be derived from the Mesozoic cover of the Battonya basement high itself (Fig. 4). However, a sand sample (Bat80/1) in the same area, but some 300 m higher up in the Late Miocene/Pliocene section provided a very different apatite fission-track age spectrum (Fig. 9). This particular sample is dominated by Mid-Miocene ages, which is similar to its zircon fission track distribution (Fig. 8). The pronounced shift towards younger ages marks the onset of erosion of metamorphic core complex(es) in the provenance area to the NE. The time of subaerial exposure can be dated based on regional seismic reflection data, since a 5.5 Ma old horizon separates the two sampling levels in the Battonya area (Vakarcs et al. 1995). In two samples (A-25/1 and A-312/1) the relatively high present-day temperature (i.e. >96~ produced significant rejuvenation (Fig. 9). However, the fact that the dominant part of the apatite fission-track ages is older then 20 Ma indicates that primary age could be as old as Palaeogene/Cretaceous. Regional
structure transects
Transect A (Fig. 10) is a dip-oriented section with respect to the pre-extensional basement structure in the SE Pannonian basin (Fig. 4b). The perpendicularly oriented transect B (Fig. 11), however, is a dip-section across the major extensional subbasins of the area (Fig. 4a). For both transects, the depth of the pre-Tertiary basement was adopted from Kil6nyi et al. (1991)
Fig. 10. Regional transect A across the SE Pannonian Basin (modified from Tari 1994). For location see Fig. 4a. Compare with Fig. 4b, the tectonic subdivision of the basement shown in the transect is somewhat different from that of Ftil6p et al. (1987). This is due to the recognition of the Battonya anticlinorium (Tari et al. in prep.).
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
225
226
G. T A R I E T A L .
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN and Dicea et al. (1984), in Hungary and in Romania, respectively (see Fig. 4a). The actual basement-type in Hungarian territory was constrained by the map of Ftil6p et al. (1987), whereas the tectonic subdivision of the exposed pre-Tertiary units of the Apuseni Mts is based on the work of Bleahu et al. (1981). The northern Hungarian part of transect A closely follows a composite seismic reflection profile compiled by Tari (1994). The above described PGT-1 deep seismic reflection section (Fig. 5) is located some 25 km to the W of the transect (Fig. 4) but it was projected to the transect to constrain the structure of the lithosphere and in particular, the geometry of the middle/lower crust. The southern continuation of the transect to Romania was simplified from a regional section by Stefanescu (1988). At the northern end of transect A the poorly defined Mid-Hungarian fault zone (cf. Fig. 2) separates the South and North Pannonian units. To the south, the transect emphasizes the superposition of the Neogene extensional basin on a Cretaceous fold-thrust belt. The north-vergent thrusting involved the crystalline basement. Otherwise the interpreted geometry is quite similar to that observed in thin-skinned thrustfold belts (cf. Hatcher & Hooper 1992). The position and dip of the major ddcollement is somewhat speculative, given the seismic control (Fig. 5). In our interpretation, a prominent antiformal stack (Battonya anticlinorium) is responsible for the regional change of dip within the basement in the middle part of the transect. The position of the Moho and the thickness of the lithosphere in the Romanian part of the transect is poorly constrained. At the southern end, the South Transylvanian fault appears to be a major strike-slip fault separating the South Pannonian unit in the north from the Dacides in the south, in the sense of Csontos et al. (1992). The downward continuation of this fault zone to the middle crust and its potential decoupling within the lower crust (?), similarly to the Mid-Hungarian fault, remains to be seen. The eastern Hungarian part of transect B (Fig. 11) more or less coincides with a composite seismic reflection profile compiled by Tari (1994). Since for the most part the transect is parallel with and adjacent to the above described PGT-4 deep reflection seismic section, the middle/lower crustal and the deeper lithospheric geometry is assumed to be identical. At the western, Serbian end of the transect, the basement was reached by several wells in
227
relatively shallow depth of 1000-1500 m. The Subotica basement high is separated from the Szeged/Algy6 high by the Dorozsma basin (see Mattick et al. 1988, fig. 18) underlain by a number of uniformly tilted blocks (see Rumpler & Horvfith 1988, fig. 8). All of these blocks are tilted to the WSW, suggesting a major detachment fault dipping to the ENE. Based on the low-angle normal fault contact separating high-grade metamorphics from unmetamorphosed Mesozoic carbonates; the presence of mylonites; the characteristically retrograde metamorphism within the lower plate (e.g. Ftil6p 1994), and fission-track dating, the Algy/3 high is interpreted as a metamorphic core complex (Fig. 12). This is in sharp contrast to the Pusztaf61dvfir basement high, which is in an upper plate position (cf. Hajnal et al. 1996). The polarity of the extension changes beneath the B6k6s basin. All the WSW-dipping normal faults are decoupled on the Algy6 detachment fault which seems to coincide with the top of the ductile lower crust. Major normal faults in Romania, bounding the Zarand, Beius and VadBorod basins, also dip towards the WSW, but they are shown to be steeper. Transect B displays an overall similarity to the simple-shear model of Wernicke (1985, fig. 12). Adopting his terminology, the Subotica high is interpreted as an upper crustal breakaway in proximal position; the numerous highly tilted blocks around Szeged are extensional allochthons; the Algy/5 high is a metamorphic core complex; and the Codru-Moma, Paduera Craiului and Plopis Mts are large fault-block ranges. Another particular feature constrained by magnetotellurics (Posgay et al. 1996), is the updoming lithosphere/asthenosphere boundary in a position consistent with Wernicke's (1985) model. In map view, the pronounced halfwindow of the Bihor autochthon in the Apuseni Mts (Fig. 2) may correspond to the topographic culmination on the upper or distal plate. Posgay et al. (1996) and Hajnal et al. (1996) interpret the Pusztaf61dvfir high as a block bounded by a low-angle normal fault on its northeastern side. Taking into account the postulations of Jackson & White (1989) on the maximum size of rotated normal fault-blocks, the approximately 65 km wide high could not rotate coherently. Instead, this basement high represents a keystone structure where the polarity of extension changes (Fig. 11). This also explains the symmetric, rounded top of the basement high.
Fig. 11. Regional transect B across the SE Pannonian Basin. For location see Fig. 4. Similarly to Fig. 10, the subdivision of the basement shown in the transect is somewhat different from that of FiilOp et al. (1987).
228
G. TARI E T A L .
Fig. 12. Basement map of the AlgyO high (Kovfics & Kurucz 1984). See Fig. 4a for location. The Pre-Tertiary subcrop was constrained by >95 hydrocarbon exploration wells shown as black dots. Contours in metres show the depth of the basement below the surface. Note the location of the Algy/3-626well which yielded the 17.4 Ma zircon fission track age (Table 1). The Algyti basement high is interpreted as a subsurface Middle Miocene metamorphic core complex based on the presence of a major low-angle normal fault system, the characteristic retrograde metamorphism and mylonitization in its lower plate and fission track age data. Transport direction of the upper plate is to the NE-ENE.
The schematic block diagram shown in Fig. 13 illustrates the complexity of extension in 3D. The H 6 d - M a k 6 and Bdkds basins and the Pusztaf61dvfir high b e t w e e n t h e m record W S W - E N E - o r i e n t e d extension; whereas a set of perpendicularly oriented, but far less important, normal faults is responsible for the extension of the area in a N N W - S S E direction. This second set of faults may have a significant strikeslip c o m p o n e n t since they are located in a transfer zone, separating an extremely extended area from its neighborhood. A n o t h e r important element in our interpretation is the relative importance of extensionally
reactivated thrust planes. Contrary to the expectations (e.g. Grow et al. 1989) Miocene lowangle fault planes typically do not seem to reactivate Cretaceous overthrust planes, at least not in the upper crust of the SE Pannonian basin.
Gravity constraints on the lithospheric structure of the Pannonian basin The regional B o u g u e r anomaly map of the Carpathian-Pannonian area is shown in Fig. 14. The map was constructed using data from Scheffer (1960), Bureau Gravim6trique International
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
229
Fig. 13. Block diagram of the Hungarian part of the SE Pannonian basin showing the surface of the preTertiary basement (see Fig. 4). This figure illustrates the interplay between the pre-existing Cretaceous thrusts and superimposed Miocene low-angle normal faults. (1963, 1966, 1971), Vasiliu et al. (1973), Bilibajkic et al. (1979), Labas et al. (1981), Labas (1987), Mocanu & Radulescu (1994) and Ravnik et al. (1995). The normal values of gravity were computed according to the International Gravity Formula. The anomaly values are in mgal (1 mgal = 10-5 ms-Z). These maps are of different scale and, therefore, of different accuracy. Most of the maps are at least 1:1 million in scale, except the Bouguer anomaly map of Europe and Africa (Bureau Gravim6trique International 1971), which was published at a 1:10 million scale. However, we used this map only for a small part of the Apennine peninsula. Two more factors significantly affect the accuracy of the regional Bouguer anomaly map: (1) recalculations of different data sets for the purpose of using a unique International Gravity Formula and (2) graphic smoothing along the borders of the countries. In spite of the above mentioned artifacts, the resulting map is very suitable for a regional study. Although the map is not complete, it clearly shows the main characteristics of the region. A prominent feature of the gravity field in the Pannonian-Carpathian-Dinaride area is the pronounced negative anomaly belt which can be traced from the Eastern Alps along the whole
Carpathian arcand the Dinarides. However, the anomalies of the Eastern Alps, the Dinarides and the Carpathians are of different character. The negative Bouguer anomaly belts in the Eastern Alps and in the Dinarides are similar in that the minimum axes of the anomalies correlate fairly well with the highest topographic elevations (Karner & Watts 1983; Tomek 1988) and the thickest crust (Horvfith 1993). Within the map frame the Bouguer anomaly reaches its minimum value of less than -160 mgal in the Tauern window of the Eastern Alps (farther to the west actually it drops down to -190 mgal), while in the funnel-shaped negative anomaly belt of the Dinarides the minimum is -95 mgal. The Carpathian gravity low is narrower than its Alpine counterpart and it has a smaller amplitude. Its minimum axis appears to ignore the main tectonic boundaries and the topographic crest. The belt of the Carpathian negative anomaly apparently consists of four different segments (Tomek 1988). The first segment stretches along the Vienna basin, with an amplitude of-50 mgal. The next, quite linear segment of the anomaly begins north of the Vienna basin and continues ENE for approximately 350 kin. Its minimum value is some -60 mgal. The third segment runs SE for more than 500 km along the
230
G. TARI E T A L .
Eastern Carpathians and its minimum (some -100 mgal) is located in the Ukrainian Carpathians at the boundary of the Silesian-Moldavian flysch and the molasse belts (Unrug 1982). The fourth segment of the negative anomaly belt is associated with the Southern Carpathians. The axis of the minimum is completely in the foredeep and reaches some -110 mgal. Seismic data (Tomek et al. 1989) and the results of several gravity modeling studies (Bielik et al. 1990, 1991; Bielik 1991; Lillie & Bielik 1992; Vyskocil et al. 1992; Skorvanek & Biela 1993; Bez~k et al. 1995; Szafi~in et al. 1997) confirm that the Western Carpathians have no real crustal root. This might be the result of different processes: (1) after the termination of Early to Mid-Miocene continental collision approximately 15 Ma ago, the subducted lithospheric slab detached and sank into the deeper asthenosphere or (2) this phase of the orogeny was characterized mostly by transcurrent motion of nappes along the strike of the orogen without major convergence of the subcrustal lithospheres. Most probably both processes were active during the orogeny in the form of strain partitioning. The positive velocity anomaly zone under the Western Carpathians shown by the travel-time tomography (Spakman et al. 1993) can be considered the remnant of the detached slab. One may assume that the detachment process did not terminate under the Western Carpathians but migrated laterally along the strike of the descending slab, recently reaching the Vrancea region in the southernmost part of the Eastern Carpathians. This detachment process causes important changes in the state of stress along the plate boundaries, which has implications for the tectonic transport of nappes, depocentre shifts and volcanic activity (Wortel & Spakman 1992). Modelling of recent and palaeostress fields (Gerner et al. this volume) may provide further constraints on this process. Different geological profiles of the Southern Carpathians (Motas 1983; Stefanescu 1984; Andreescu et al. 1989; Demetrescu & Veliciu 1991) indicates that the foreland is subducted under the thrust sheets and that thrust faulting affected even the Upper Miocene and Pliocene sequences of the foredeep basin. Szafi~n et al. (1996) shows that the observed gravity anomaly may be reproduced reasonably well by assuming a deep crustal root. The modeled shape of the crust implies that due to the right-lateral transpression along the plate boundary, the (lower) crust of the Dacide terrane strongly deformed and indented the crust of the Moesian platform. Another important feature of the region is the
mild (slightly positive to slightly negative) Bouguer anomaly field in the territory of the Pannonian basin. Previous analyses of the Bouguer anomalies in the Carpathian-Pannonian area (Scheffer 1960; Stegena 1964; Renner & Stegena 1965; Ibrmajer 1981; Mesk6 1983, 1988; Bielik 1988) provide information on the crustal structure of the region. These authors point out that the most prominent features, reflected in the regional gravity field, are the shape of the Moho discontinuity, the topography of the basement and the thickness of the sedimentary basin fill. The gravitational effects of the Neogene basin fill and the Moho topography are the subject of detailed investigations (Stegena 1964; Renner & Stegena 1965; Mesk6 1983, 1988; Bielik 1988). Mesk6 (1983) computes a regional Bouguer gravity map of Hungary by filtering out the short wavelength components in order to determine clearly the variation with the depth of the Moho discontinuity. However, the picture obtained was somewhat obscure. To achieve a less distorted view of the deeper density inhomogeneities, Mesk6 (1988) calculates the gravity effect of the uppermost 5 km thick horizontal slab and removes it from the regional Bouguer anomaly map. He terms the result as the reduced regional Bouguer anomaly field of Hungary and concludes that the reduced field depicts quite well the undulation of the Moho. Bielik (1988) studies the gravity anomaly pattern of the Pannonian basin and constructs a stripped gravity map of the region. He evaluates the gravity effect of the basin fill by three-dimensional model calculations and subtracts it from the original Bouguer anomaly field. The stripped map exhibits significantly increased positive anomalies with maximum values in the regions of the deepest subbasins of the Pannonian basin. Bielik (1991) proposes an intrusion of upper mantle material into the lower crust under the deep troughs. This idea was originally suggested by Pospfsil (1980), who carries out a two-dimensional gravity model calculation across the Transcarpathian depression. This modeling shows that the long wavelength (regional) component of the Bouguer anomaly field was caused by a diapiric intrusion of upper mantle material into the thin crust along the axis of the depression. On the basis of his stripped gravity map, Bielik (1991) generalizes this explanation to all of the deep troughs of the Pannonian basin. Nemesi & Stomfai (1992) provide a combined analysis of the gravity and magnetic anomalies in the B6k6s basin, which is the most prominent example of the unexpected pattern of the
13"
14"
24* 15"
16"
17"
18"
19"
20*
21"
22°
25*
26*
27*
280
230 F~UGUEI." R ANOMALY MAP OF THE PANNONIAN BASIN AND ~;ADJACENT AREAS t ~ (mgal) 40 20 0
-20 -40 -60 -80 -100 -120 -140 -160 -180
44"t
0km
100km
.......200km
300km -
400km
Fig. 14. RegionalBouguer anomaly map of the Carpathian-Pannonianregion compiled by P6ter Szafi~,n.
500kin
13"
51
o
o
14
o
o
0
15
o
16
17
-
o
19 °
18 °
20"
21
o
o
22
23
~ 24
25 °
HEAT FLOW MAP OF T H E P A N N O N I A N BASIN A N D ADJACENT AREAS
\
(mWlm~
5o
110 '....
100
,i
49"
90
.....,
80 °
70
48*
~---.,
60 50 40
47*
30 20
46*
45 °
IP"
o
44 I
0km
100km
200km
300km
400kin
500km
Fig. 16. Heat flow m a p o f the P a n n o n i a n - C a r p a t h i a n region. Data in H u n g a r y are based o n the c o m p i l a t i o n s m a d e b y DOv6nyi (1994) and outside H u n g a r y o n the ' G e o t h e r m a l Atlas o f E u r o p e ' (Hurtig e t al. 1992). Stars: heat flow d e t e r m i n a t i o n s , Dots: heat flow estimations. Sections A A ' a n d BB' show locations o f thermal models.
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN Bouguer anomaly field in the region of deep subbasins. According to their model calculations, these anomalies are due to upper mantle material which upwarps to the depth of 15 km beneath the basin. Kov~icsv61gyi (1994) and Posgay et al. (1995) suggest that an uplifted lower crust and upper mantle is responsible for the gravity anomaly under the B6k6s basin. Posgay et al. (1995) assume the intrusion of magmatic rocks into the crust during the basin formation. Grow et al. (1994) put forward two explanations for the origin of the positive gravity anomaly. A high density body representing either a fragment of oceanic crust (ophiolites) sandwiched between pieces of continental crust during Mesozoic collisions, or mafic rocks intruded into the crust in Miocene time during extension. Bielik et al. (1996) and Szafi~n et al. (1997) also suggest the presence of a high density body in the crust under the B6k6s basin since, according to their results, Moho updoming alone can not reproduce the observed positive anomaly. Concerning fragments of oceanic crust under the basin, the stripped gravity map of the Pannonian basin (Bielik 1988) exhibits high positive anomalies in all the deep sub-basins. Therefore, the positive anomaly is related to the extensional origin of the basins. Indeed, the strongly
231
reflecting horizons in the lower crust along the Pannonian Geotraverse (Posgay et al. 1990) may be explained in terms of intrusions of mantlederived mafic melts (Warner 1990). Moho updoming and significant crustal extension accompanied by intrusions of mantle material into the crust during the Neogene is the most plausible mechanism. According to Jordan (1978), 97% of the energy in the Bouguer anomaly field is caused by Moho relief, which in turn is in a close connection with the isostatic compensation processes. This means that the long-wavelength Bouguer anomalies are good indicators for the topography of the Moho, while the short wavelength anomalies represent density inhomogeneities in the shallower parts of the crust. A prominent example is a sedimentary basin, where the large negative anomalies usually correspond to the deep parts of the basin. This is not the case, however, in all the deep sub-basins of the Pannonian basin. In the SE part of Hungary the B6k6s basin is characterized by a large positive Bouguer anomaly, contrary to the expectations (Fig. 15). The area of the Mak6 trough is on the positive side of the Bouguer anomalies, although the depth of this basin is more than 7 km. This behaviour may only be explained with a combined detailed seismic and 3D gravity interpretation. The complicated
Fig. 15. Anomalous gravitational signatures in the SE Pannonian basin. A part of the regional Bouguer map (Fig. 14) was superimposed on the pre-Tertiary basement depth shown in Fig. 4a. Light grey and dark grey shading show the outline of deep subbasins (i.e. >4 km and >6 km depth). Note the coincidence of a large positive Bouguer anomaly with the deep B6k6s basin, contrary to the expectations. See text for detailed explanation.
232
G. TARI E T A L .
nappe structure of the pre-Tertiary basement may partly be responsible for the observed discrepancies (see Szafifin et al. in press), but a complete explanation is most probably hidden in the dynamic processes of the basin's formation.
Lithospheric structure of the Pannonian basin based on geothermal data The Pannonian basin is characterized by high heat flow (Fig. 16). For a description of heat flow determinations the reader is referred to Appendix B. The average near-surface heat flow is about 90 mW m -2, in contrast with a characteristic value of about 60 mW m -~ in the Carpathians and 40 mW m -2 on the Ukrainian shield. The high heat flow in the Eastern Alps is based on a few measurements explained by rapid uplift and erosion (Oxburgh & England 1980). Towards the Dinarides the heat flow is decreasing rapidly, while the Outer Dinarides represent an extremely cold region. The Adriatic Sea is characterized by low, although variable heat flow values. From the Inner Dinarides towards the Pannonian basin, particularly along the Sava and Drava troughs, geothermal highs occur. A large positive heat flow anomaly can be found at the southern part of the Pannonian basin around Belgrade, which continues to the SE along the Vardar zone. The exact areal extent of this anomaly is doubtful because of sporadic data. The Inner Carpathian calcalkaline volcanic regions are also characterized by high heat flow. Simple calculations show that the extra heat cannot be derived from the cooling of magmatic bodies intruded below the subsurface during Miocene and early Pliocene (Horv~th et al. 1986). Such local effect may be expected only in the Eastern Carpathians, where the volcanic activity occurred from late Pliocene to early Pleistocene (Szab6 et al. 1992). The high heat flow is probably the result of elevated temperatures in the deeper crust. The heat flow shown in Fig. 16 also reflects the disturbances caused by groundwater circulation. The outcropping Mesozoic carbonate complex in the Transdanubian Central Range is fractured and karstified and the infiltrating cold meteoric water reduces the heat flow to 40-50 mW m -2. Heat balance calculations (D6v6nyi et al. 1983) show that the background heat flux beneath the Transdanubian Central Range without water circulation is 90 mW m -2. Similar convective systems are active in NE Hungary in karstic areas (Bakk, Aggtelek-Gemer mountains) and in the Outer Dinarides.
The large-scale groundwater flow system, acting in the porous Neogene and Quaternary sediments driven by the difference of hydraulic heads between the recharge and discharge areas, does not alter the regional heat flow significantly. It is because large part of the circulation system is shallow, contained mostly in permeable Quaternary sediments (Erd61yi 1985). The groundwater flow system in the deeper and hotter sediments has an insufficiently low flux, especially vertical flux, to significantly modify the heat flow. The amplitude of the disturbance on the regional heat flow caused by the largescale groundwater system is estimated to be _+10% of the measured heat flow. The rapid Neogene and Quaternary sedimentation decreased the surface heat flow, because sediments have not attained thermal equilibrium yet, and are still colder in the deep subbasins than would be in equilibrium state. The effect of sedimentation on the temperature and heat flow was calculated by solving the heat diffusion equation in vertical direction using finite-difference method (Appendix C). The compaction of sediments was taken into account. The sedimentation history was simplified in the calculations. The Pannonian basin was filled up by a large delta system prograded from NW, N and NE directions into the basin (Vakarcs et al. 1995). We assumed four depositional units: (1) synrift sediments, (2) Lower Pannonian, (3) Upper Pannonian and (4) Quaternary sediments. The Lower Pannonian sediments roughly correspond to prodelta and delta slope sediments and the Upper Pannonian sediments to delta plain sediments. The ages of the Lower and Upper Pannonian units were obtained from the timing of the delta shelf edge progradation given by Vakarcs et al. (1995). All sediments deposited before delta sediments were regarded as synrift sediments. The NW part of the Pannonian basin (Little Hungarian Plain-Danube basin) has been filled up already 10-12 Ma, while the SE part (Mak6 trough-B6k6s basin) only 6 Ma. As a result, in the NW part of the basin the present day heat flow is reduced only by 10%, but in the SE part of the basin the present day heat flow is about 30% less than it would be without sedimentation. The error of the correction, due to the different possibilities for thermal parameters, is 10% of the corrected heat flow (Appendix B), which is within the error of heat flow determinations and estimations. After correction for the Neogene and Quaternary sedimentation the average heat flow in the Pannonian basin increases to 100 mW m -2 and the heat flow
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
233
Fig. 17. Heat flow map of the Pannonian-Carpathian region corrected for Neogene and Quaternary sedimentation, Lenkey (in prep). For detailed explanation see text and also Appendix C.
pattern also changes (Fig. 17). Before the correction the deep subbasins were characterized by local heat flow minimum, after correction their heat flow reaches or becomes higher than the average heat flow value of the Pannonian basin. The Pannonian basin is characterized by thin lithosphere and crust (Fig. 7). In contrast, the mountain arc around the basin system has a deep root with thick crust and lithosphere. Comparing these features to the heat flow distribution the overall correlation is good. It was Stegena et al. (1975), who first suggested that the thin crust and high heat flow are related to the formation of the Pannonian basin. Royden et al. (1983b) show that the heat flow and the subsidence in the Pannonian basin may be explained by nonuniform thinning of the lithosphere. To model the high heat flow and high postrift subsidence rate they assume that the mantle part of the lithosphere was almost completely thinned. We used the non-uniform extensional model of Royden & Keen (1980) to derive crustal and mantle thinning factors along two regional sections (Figs 18 and 19), which are prolongations of sections A and B in Fig. 2. The sections were divided into series of boxes of 5 km width
representing lithospheric columns. According to the nonuniform extensional model, the crustal and mantle thinning factors determine the subsidence and heat flow any time after rifting (Appendix D). The model was used in a simplified form and the thinning factors for each box were calculated from the present day mean basement depth (total subsidence until now) and present day mean heat flow of the box, assuming instantaneous rifting 17 Ma. Parameters of the model are given in Table 2. Before calculating the thinning factors, the heat flow was corrected for sedimentation and was filtered by a low-pass filter to remove heat flow anomalies, which had wavelength less than 100 km. The average lithospheric thickness in the Pannonian basin is 60 km (Fig. 7), therefore we assumed that heat flow variations caused by changes in the lithospheric thickness had the same or even larger wavelength. Therefore we r e m o v e d the smaller wavelength anomalies from the heat flow before modelling (Figs 18a and 19a). We assumed that the error of the corrected and filtered heat flow was _+15%, which primarily derives from the error of heat flow determinations and estimations. Minimum and maximum thinning factors were calculated from
234
G. TARI E T A L .
Fig. 18. Thermal model along section AA' (for location see Fig. 15) based on crustal and mantle thinning factors. (a) Heat flow, crosses: heat flow taken from the heat flow map (Fig. 15), circles: heat flow corrected for Neogene and Quaternary sedimentation, taken from map in Fig. 17, solid line: low-pass filtered corrected heat flow, used as input to calculate thinning factors. The error of the corrected and filtered heat flow, which primarily derives from the uncertainty of heat flow determinations and heat flow estimations, is estimated to be +15% and shown by hatching. (b) Thinning factors derived from the present day corrected and filtered heat flow and basement depth using the nonuniform extensional model of Royden & Keen (1980). The error of the thinning factors due to the error of the heat flow is shown by hatching. (c) Present day temperature distribution in the lithosphere calculated using the nonuniform extensional model with the medium thinning factors, shown by solid lines in (b). Thick line in (c) denotes the bottom of the lithosphere taken from Fig. 7. Note the strong correlation amongst the elevated heat flow, high mantle thinning factors and observed thin lithosphere.
the lower and upper error limits of heat flow, respectively (Figs 18b and 19b). The mantle thinning factor strongly depends on the present day heat flow and relatively small change in the heat flow (+15%) causes large variation in the mantle thinning factor (2-8). In spite of the large error, we obtained that similarly to other subsidence and heat flow modelling studies (Royden et al. 1983b; R o y d e n & D6v6nyi 1988; Lankreijer et al. 1995; Sachsenhofer et al. 1997), the mantle thinning factors
are 1.5-6 times higher than crustal thinning factors. The present day temperature distribution in the lithosphere was calculated using the extensional model with thinning factors corresponding to m e d i u m heat flow. The bottom of the lithosphere, taken from Fig. 7, is also shown in Figs 18c and 19c by thick solid line. Comparing the surface heat flow, mantle thinning factors and temperature distribution in the lithosphere to the thickness of the lithosphere, it is evident
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
235
Table 2. Summary o f parameters used in the thermal model Parameter Thermal conductivity of upper crust Thermal conductivity of lower crust Thermal conductivity of mantle Heat generation in the upper crust Heat generation in the lower crust Heat generation in the mantle Thickness of the upper crust before rifting Thickness of the lower crust before rifting Thickness of the lithosphere before rifting Temperature of the asthenosphere Volumetric heat capacity Volumetric heat expansion coefficient Duration of rifting
Value 3 W in 1 ~ 2.3 W m a ~ 4 W m -1 ~ 2 • 10 .6 W in -3 0.5 • 10 -6 W m -3 0.01 • 10 .6 W m -3 15 km 20 km 120 km 1333~ 3.85 • 106 J m -3 ~ 1 3.1 • 10 -5 ~ 17.5 - 13.5 Ma
Fig, 19, Thermal model along section BB'. For detailed explanation see Fig. 18. (a) Observed and corrected heat flow. (b) Thinning factors derived from heat flow and basement depth using nonuniform extensional model. (e) Modelled present-day temperature in the lithosphere calculated from the extensional model using the thinning factors shown in (b). Note the strong correlation amongst the elevated heat flow, high mantle thinning factors and observed thin lithosphere.
236
G. TARI E T A L .
that the high heat flow in the Pannonian basin is caused by the thinning of the lithosphere. The area of thin lithosphere very well correlates with the increased temperature in the lithosphere and elevated surface heat flow. According to the definition of the mantle thinning factor (Appendix D), the knowledge of the present day thickness of the lithosphere allows to give an independent estimation for the mantle thinning factor:
[3-
a l i - Yci
al - Yc
where all and Yci are the initial thicknesses of the lithosphere and crust, respectively al and Yc are the present day thicknesses of the lithosphere and crust, respectively. Assuming 120 km initial lithospheric thickness, 35-50 km initial crustal thickness, and 60 km and 25 km present-day m e a n lithospheric and crustal thicknesses, respectively, the resulting mantle thinning factor ranges between 2 and 2.43. Mantle thinning factors derived from modelling are higher, because the base of the lithosphere is controlled by the thermal state of the mantle and its shape and depth are modified by the thermal decay during postrift cooling. Therefore the mantle thinning factor derived from the present day lithospheric thickness is a minimum estimation. In the nonuniform extensional model the heat transfer occurs only in vertical direction. The decay of thermal anomaly, caused by the thinning of the lithosphere is faster due to lateral heat conduction. The thickness of the nondeformed lithosphere might have been around 120 km, which is reduced to 60 km. Thus lateral heat transport is significant in the peripheral areas of the basin, in a distance of about 100 km from the edges. In these areas the mantle thinning factors were likely underestimated. In the central areas the one-dimensional approach is acceptable. Thinning factors were calculated from the present day basement depth and heat flow, thus in the areas of outcrops (Mecsek Mts and Apuseni Mrs in Fig. 19) no thinning factors were derived and no temperature was calculated (Fig. 19). It is evident from the heat flow, the extrapolation of mantle thinning factors and temperature in the mantle that the lithospheric mantle was thinned beneath the Mecsek Mts, located inside the basin, and by lesser extent beneath the Apuseni Mts, being in peripheral position. In calculating the thinning factors standard crustal and lithospheric thicknesses were assumed (Table 2). However, the thinning factors strongly depend on the initial conditions. The effect of variation of initial crustal thickness
on the thinning factors was investigated at the Algy6 high, where metamorphic core complexes were found (Fig. 19c). Increased initial crustal thickness has two i m p o r t a n t effects: it results in elevated land surface before rifting due to isostatic compensation, and elevated initial surface heat flow due to the higher concentration of radioactive elements in the crust compared to the mantle (Fig. 20). Initial elevation was calculated assuming local isostatic compensation at the base of the crust, implying crustal and mantle densities given in Table 2. Calculating the initial surface heat flow steady-state was assumed (Appendix A). Thinning factors were derived the same way as for the lithospheric boxes along the sections, except that the present-day basement depth and heat flow, 2301 m and 121 _+ 15% mW m -2, respectively, were fixed according to the observed (corrected) values at the specific location. The initial crustal thickness, elevation and heat flow were varied. With increasing crustal thickness the crustal thinning factor increases, because starting from higher initial elevation larger subsidence is required to attain to the present day basement
Fig. 20. Variation of thinning factors in the function of increasing crustal thickness. Over 50 km of initial crustal thickness the model fails, because the initial heat flow and the crustal thinning factor become too high, thus the predicted heat flow gets higher than the observed (and corrected) heat flow. Mantle thinning factor estimated from the present day thickness of the lithosphere (see equation in text) suggests that mantle thinning factor was higher than two. It is attained when initial crustal thickness is less than 50 km.
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN depth (Fig. 20). The decrease of the mantle thinning factor has two reasons: (1) the initial heat flow is higher, thus less mantle thinning is necessary to obtain the present day heat flow, (2) heat flow depends on both the crustal and mantle thinning factors (Appendix D), and due to higher crustal thinning less mantle thinning results in the same present day heat flow. For initial crustal thickness more than 50 km, the initial heat flow and the crustal thinning factor become so high that crustal thinning alone, without mantle thinning, results in higher present day heat flow than observed, thus the model fails. The same failure occurs for initial crustal thickness more than 45 km, if the lower error limit of the present day corrected heat flow is used. The present day lithospheric thickness (compare Figs 7 & 21) places strong constraints on the minimum mantle thinning factor ([3 > 2). According to Fig. 20, mantle thinning factor larger than two occurs, when the initial crustal thickness is less than 50 km. Therefore, modeling shows that initial crustal thickness higher than 45-50 km, and initial heat flow higher than 75 mW m -2 were not likely.
237
Magmatism Note that, besides the voluminous subductionrelated calc-alkaline Mid-Late Miocene volcanics of the Carpathian region (e.g. Szab6 et al. 1992), there are volcanics of the same age (i.e. >10 Ma; P6cskay et al. 1995) with a seemingly different origin. For example, Badenian/Sarmatian volcanic rocks (i.e. trachytes, shoshonites) in the Styrian and Danube basins (Fig. 1) have a generally more alkaline character than their Carpathian counterparts (Ebner & Sachenshofer 1995; Harangi et al. 1995a). The formation of these enigmatic volcanics is better understood in the context of syn-rift extension (Tari & Horvfith 1995). The rise and decompression of lower crustal material underneath areas characterized by extreme extension (core-complex type) may induce melting of lower and middle crustal rocks. Since systematic trace element studies are not available from these alkaline volcanics, this hypothesis remains to be tested. In contrast, the Late Miocene/Pliocene (i.e. <10 Ma) through Quaternary alkaline volcanics are clearly extension-related (Szab6 et al. 1992, 1995; Harangi et al. 1995b). However, these
Fig. 21. Depth of the ll00~ isotherm in the Pannonian basin calculated from the thermal model. Outside the basin the lithospheric thickness is shown, that part is equivalent with Fig. 7.
238
G. TARI E T A L .
magmas may correspond to a more regional asthenospheric dome than the one beneath the Pannonian basin. Hoernle et al. (1995) recognize the upwelling of the asthenospheric mantle beneath western and eastern Europe. Regardless, helium isotope ratios from groundwater samples clearly show the elevated position of the asthenosphere (Martel et al. 1989). A correlation between the distribution of Pliocene alkaline volcanics and highly extended areas within the Pannonian basin appears to exist. In the SE Pannonian basin most of these volcanics only subcrop (e.g. Fig. 6); however, they were encountered in several wells (e.g. Kecel Basalt of Bal~zs & Nusszer 1987). The outline of these subcrops coincides fairly well with an area where Neogene extension progressed to the development of a narrow rift system (see later).
Neogene evolution of the lithosphere in the Pannonian basin Based on data previously presented, the Pannonian basin cannot be easily interpreted in terms of a narrow or discrete rift using the terminology of Buck (1991) and Ruppel (1995), respectively. Instead, the abundant industry reflection seismic data (e.g. Mattick et al. 1988; Rumpler & Horv~th 1988; Tari et al. 1992; Grow et al. 1994) shows a basin system in which upper crustal extension is distributed over a very broad region, including several subbasins (Fig. 1). Similarly, an overall elevated near-surface heat flow is observed in the same geographic area averaging around 90 mW m -2 (Fig. 16). These basic observations suggest a wide or diffuse style of rifting for the Pannonian basin as a whole, sensu Buck (1991) and Ruppel (1995), respectively. A closer look at certain sub-basins, however, reveals a temporal progression in the style of continental extension. The dominant mode of continental extension is a function of the initial thermal structure and crustal thickness of the extending area (Buck 1991; Fig. 22). It is difficult to evaluate the nearsurface heat flow at the onset of extension (c. 17.5 Ma); however, it was at least as high as the present-day average values in the Pannonian basin (90-100 mW m 2; see above) some 12 Ma after cessation of regional syn-rift extension (Fig. 3). Regarding crustal thickness, relatively unextended areas within the basin system such as the Bakony and Apuseni Mts. (Fig. 1) provide a minimum estimate on the paleothickness of the crust (c. 35 km) just before the Miocene extension (Posgay et al. 1991; Horv~th 1993).
Similarly, a reasonable estimate may be obtained for the maximum thickness of the Early Miocene crust at the western perimeter of the basin (c. 45 kin). These initial values correspond to a core complex type extensional mode in the modified Buck-diagram (Fig. 22), refined by Hopper & Buck (1996). Indeed, the Karpatian formation of the well-documented Rechnitz metamorphic core complex at the boundary of the Styrian and Danube basins (Fig. 2) confirms this mode map (Tari & Bally 1990; Tari 1994; Tari & Horv~th 1995). The Algy6 metamorphic core complex as defined in this work also records this first, Karpatian stage of syn-rift extension (see Table 1). The Karpatian core complex extension was followed by a Badenian wide-rift extension (Fig. 22). The mode map differs from that of Buck (1991) since it also considers the decoupling of the upper crust from the upper mantle in widerift mode (Fig. 22). Although there is at least one possible example of this kind of decoupling in the Pannonian basin (see below) is not clear whether it was the typical scenario during the Badenian. The currently available data are not sufficient to distinguish between a case of straining regions in the crust and mantle migrated together and a case inwhich they are decoupled from one another. At any rate, narrow-rift style extension may have begun during the Sarmatian and the Lowermost Pannonian in certain deep sub-basins of the Pannonian basin, although this extension could not develop into a well-defined riftgraben. The spatial coincidence of anomalously thin crust (c. 22 km), thin lithosphere (c. 45 km), high heat flow (c. 120 mW m -2) and localized alkaline volcanism suggests that in the B6k6s basin a transition from wide rift to narrow rift has occurred during the late Neogene. This narrow rift zone is named Tisza rift after the major river crossing this area close to the junction between Hungary, Serbia and Romania. Regarding the initial crustal thickness and heat flow estimates there is an apparent contradiction between Figs 20 and 22. This is mostly due to the fact that the traditional thermal modelling approach calculates with the cessation of extension at the end of the Badenian whereas in the Buck diagram we assume that local rifting in fact extended well into the Sarmatian/Pannonian (cf. Fig. 3) in the Tisza rift. This resulted in increasing heat flow rather than its exponential decay. Also note that extension occurred typically only in wide rift mode in most part of the P a n n o n i a n basin system, where t h e r m a l modeling based essentially on the model of McKenzie (1978) is indeed viable. In
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
239
Fig. 22. Modes of continental extension tectonics from Buck (1991) and the temporal evolution of extension in the SE Pannonian basin in the framework of the similar, but somewhat refined scheme of Hopper & Buck (1996). The Buck-diagram based on the dry quartz rheology appears to adequately describe the temporal evolution of continental extension. Note, however, the apparent contradiction between the thermal history depicted in Fig. 20 and in this figure. For an explanation see text.
this case Mid-Miocene (Karpatian/Badenian) rifting is followed by the gradual regional cooling of the lithosphere. A n o t h e r area of probable narrow rifting includes the axis of the Danube basin (Tari 1994; Tari & HorvO,th 1995). This D a n u b e rift, however, is not underlain by locally thinned asthenosphere. Instead, it is located on the flank of a larger dome (Fig. 7). Still, local thinning of the crust, relatively high heat flow and pronounced alkaline magmatism may indicate the initial stage of subsequently aborted narrow rifting in this part of the basin. A major difference between the Danube and Tisza rifts is their relative timing. The Danube rift failed earlier since docking of the East
Alpine and West Carpathian thrust-fold belt on the foreland excluded any further extension. The East Carpathian segment of the orogen, however, was active well into the late Miocene/Pliocene, providing space for ongoing extension. This additional extension was accommodated by major transfer faults (Royden 1988; her fig. 9), such as the Dragos-Voda fault (see Gy6rfi et al. this volume) and the South Transylvanian fault (Fig. 2). The spatial evolution of the three distinct extensional styles is summarized in Fig. 23. In addition to the Rechnitz and Algy6 metamorphic core complexes the presence of several others in the basin system is postulated. Similarly, based primarily on the present-day heat
240
G. TARI E T A L .
Discussion
Fig. 23. Spatial evolution of extensional modes in the Pannonian basin, see also Fig. 22. The areas with distinct extensional styles are posted in their presentday position. Note that their correct location in a palinspastic sense would be quite different (see Tari et al. 1995).
flow pattern and crustal thickness, some other deep subbasins also advanced fairly close to the narrow rift stage. However, the Tisza rift appears to be the most developed narrow rift in the Pannonian basin system, even though it still could not advance until the opening of a backarc oceanic basin.
Several open problems complicate the above described scenario of lithospheric evolution. For example, a lithospheric-scale transect across the NW Pannonian basin displays a pronounced 160 km offset between highly strained upper crustal and upper mantle regions (Fig. 24). There are different ways to interpret this geometry and given the available data sets these competing models seem to be indistinguishable. Interpretation A is based on the model of Wernicke (1985) assuming a single throughgoing low-angle normal fault intersecting the whole lithosphere. Interpretation B is based on the regional-scale decoupling of the upper crust from the upper mantle along a ductile lower crust. The interpretation in terms of wide rift decoupling is analogous to the wet quartz case shown in Fig. 22. Note that the distinct zones of extended upper crust and upper mantle are separated by a major strike-slip system, the Mid-Hungarian fault (Fig. 2). The strike-slip component of faulting along this Early Miocene transcurrent system, juxtaposing very different A l p i n e tectonostratigraphic terranes, is on the order of 300 km (e.g. Tari et al. 1995). W h e t h e r this juxtaposition involved only the upper crust (Ratschbacher et al. 1991), the crust (Horvfith 1988) or the entire lithosphere (Posgay & Szentgy6rgyi 1991) is an open problem. The transcurrent m o v e m e n t s may have resulted in a difference in the pre-rift lithospheric rheology of the two 'microplates'. If this is indeed the case, then certain spatial and temporal variations in the mode of extension in the P a n n o n i a n basin system may simply reflect differences in the inherited lithospheric rheology and thermal structure. In fact, the position of pre-existing weakness zones within the crust or in the mantle may have a profound influence on the e v o l u t i o n of extension ( D u n b a r & Sawyer 1989) and may be primarily responsible for the asymmetry of the system (Fig. 24). Cloetingh et al. (1996) suggest an unusally shallow depth (4-6 kin) for the necking level in the P a n n o n i a n basin. This i n t e r p r e t a t i o n assumes a strongest layer controlled system during extension (their fig. 9d). Based on abundant seismic reflection evidence (e.g. Figs 5 and 6) we prefer a detachment controlled system in which necking represents a decoupling surface in the extending crust where extensional detachment faults sole out (e.g. van der Beek et al. 1995). This intracrustal level appears to be coincident with or close to the top of the reflective lower crust, at about 15-20 km.
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
241
Fig. 24. Alternative interpretations of Transect C in the NW Pannonian basin, for location see Fig. 1. The geometry of the transect is adopted from Szafi~inet al. (in press). Based on the currently available data, interpretations in terms of (a) lithospheric simple shear (e.g. Wernicke 1985), or (b) decoupled crustal and mantle weakness zones (e.g. Dunbar & Sawyer 1989) both seem to be viable. Furthermore, Cloetingh et al. (1996) report an extensional strain rate for the Pannonian basin (i.e. 1.58 • 10--15s-a) which appears to be too low. Strain rate is considered a more important factor in the style of extension than pre-rift lithospheric rheology according to these authors. However, the pre-rift rheology inherited from the Alpine orogeny, including thick crust, thin lithosphere and high heat-flow, appears to be more important than the strain rate in the evolution of the Pannonian basin (Tari 1994; Tari & Horvfith 1995). This interpretation also confirms the theoretical models of Buck (1991) and Bassi (1995). Finally, a r e f i n e m e n t of this evolutionary scheme may be expected as more subsurface data become available, such as deep seismic reflection profiles and fission track studies. Note that recent magnetotelluric m e a s u r e m e n t s already suggest a more compex topography of the lithosphere/asthenosphere boundary than depicted in Fig. 7, with local elevations (plumes?) superimposed on the regional mantle diapir (Adfim et al. 1996). Interestingly enough, isotope ratios of Pliocene alkaline basalts also indicate significant local departures from the
lithospheric thickness determined primarily by seismic data (Szab6 et al. 1995).
Conclusions The observations and interpretations of this work may be summarized as follows: (1)
(2)
(3)
(4)
an overthickened crust and warm lithosphere, inherited from the Alpine orogeny, collapsed in m e t a m o r p h i c core complex style at the onset of synrift extension; syn-rift extension occurred in dominantly wide-rift style, covering an area of much larger geographical extent; the failed Tisza rift, including the Mak6 and B6k6s basins in the southeastern Pann o n i a n basin, appears to be the most advanced narrow rift in the entire basin system; simple-shear deformation of the upper crust was decoupled from the upper mantle by a weak lower crust, deforming in a pure-shear manner;
G. TARI E T A L .
242 (5)
(6)
pre-existing weakness zones within the crust a n d m a n t l e m a y be r e s p o n s i b l e for l a t e r a l l y o f f s e t s t r a i n i n g r e g i o n s in t h e lithosphere. t h e p r e r i f t r h e o l o g y a n d initial t h e r m a l structure of the l i t h o s p h e r e are m o r e decisive factors in t h e style of e x t e n s i o n t h a n strain rate.
We are pleased to thank G i b o r Bada, Albert Bally, Liszl6 Csontos, Lfiszl6 Fodor, John Grow, Pall R6dly, /~rp~id Szalay, Csaba Szab6 and G i b o r Vakarcs for discussions on the ideas expressed in this paper. Istvzin Gy6rfi, Z o l t i n Hajnal and John Hopper kindly supplied preprints of their work. Thanks due to Ian Lerche and Laurent Jolivet for providing insightful and constructive reviews. Many thanks to Bernard Durand for his editorial patience.
Appendix A: experimental procedure of fission-track age determinations The samples were treated by the common heavy liquid and magnetic separation processes. For fission-track age determinations the apatite crystals were embedded in epoxy resin, the zircons in FEP-teflon. For apatite 1% nitric acid was used with 2.5-3 min etching time (Burchart 1972). In the case of zircon crystals, the eutectic melt of N a O H - K O H - L i O H was used at the temperature of 200~ (Zaun & Wagner 1985). Neutron irradiations were made at the RISO reactor (Denmark). The external detector method was used (Gleadow 1981), after irradiation the induced fission tracks in the mica detectors were etched by 40% HF for 40 rain. Track counts were made with an Axioskop microscope - computer-controlled stage system (Dumitru 1993), with magnification of 1000. The FT ages were determined by the zeta method (Hurford & Green 1983) using zircon from the Fish Canyon Tuff, Buluk Member Tuff and Tardree Rhyolite and apatite from Durango and Fish Canyon Tuff. Reference ages of 27.8 + 0.2 Ma for the Fish Canyon Tuff, 31.4 + 0.5 Ma for Durango apatite, 16.2 -+ 0.6 Ma for Buluk Member Tuff and 58.7 -+ 1.1 Ma for Tardree Rhyolite has been adopted according to Hurford & Hammerschmidt (1985), Green (1985), Hurford & Watkins (1987) and Hurford & Green (1983). The error was calculated by double Poisson dispersion (Green, 1981). Calculations and plots were made by using the TRACI,:I<Euprogram.
where k is the thermal conductivity tensor and gradT is the temperature gradient. The minus sign appears since heat flows in direction of decreasing temperature. Since the horizontal components of the terrestrial thermal gradient are much smaller than the vertical component, the terrestrial heat flow density can be approximated by:
dT q = kzz d--z
where q is positive upward, and kzz is the vertical component of the thermal conductivity tensor. The terrestrial heat flow density, or shortly heat flow, can not be measured directly at a specific point. It is determined according to eq. B2 by measuring the vertical temperature gradient in deep bore holes and measuring the thermal conductivity of core samples or drill cuttings in laboratory. Due to the errors of thermal conductivity and temperature measurements, the error of heat flow determinations is approximately _+5-20%. There are about 200 heat flow determinations in the Pannonian basin (Hurtig et al. 1992). Additionally, heat flow can be estimated reasonably well, if temperature and the mean thermal conductivity is known. The Neogene basin fill in the Pannonian basin consist of alternating layers of sand, sandstone, clay and marls. Measuring the thermal conductivity of 308 Neogene rock samples, formulas for the change of thermal conductivity with depth for coarse-grained (sand, sandstone) and fine-grained (clay, claystone, shale and marls) sediments were established (D6v6nyi et al. 1983; D6v6nyi & Horvith 1988). Using these formulas heat flow was estimated for wells, where reliable temperature measurements were available and the lithology, especially the ratio of fine-grained to coarsegrained sediments was known from geophysical logs and core samples. The error of the estimated heat flow values cannot be given exactly, but it is estimated to be less than +20% (Dt~v6nyi etal. 1983).
Appendix C: calculation of the thermal effect of sedimentation The effect of sedimentation on the temperature was calculated by solving the heat diffusion equation in vertical direction:
pc(z) ~z : ~z
Appendix B: heat flow calculations The conductive heat flow density q, defined as the flow of heat per unit area per unit time, is determined as follows: q = -kgradT
(B1)
(B2)
k(z,T) -~z
+ A(z)
(C1)
where T denotes the temperature, pc(z) is the specific heat per unit volume, X(z, 7) is the thermal conductivity and A(z) is the volumetric heat production. Boundary conditions were constant temperature at the surface (T = 10~ at z = 0) and at the bottom of the lithosphere (T-- 1333~ at z = 120 km).
L I T H O S P H E R I C S T R U C T U R E OF P A N N O N I A N BASIN Thermal dependence of conductivity was taken into account according to the empirical relationship of Sekiguchi (1984): k(T) = 365.75(k20- 1.84)(
1 - 1--~) + 1.84
(C2)
where k20 is the thermal conductivity at 20~ and T is the actual temperature of rock in K. The initial geotherm corresponds to the steadystate solution of eq. CI:
a (X(z,73 or
0 : ~zz
~z ) + A ( z )
(c3)
The thermal parameters are given in Table 2 (after Kappelmeyer & Haenel 1974 and Zoth & Haenel 1988). Solving eq. C3 with the values of thermal parameters in Table 2, in case of 120 km thick lithosphere and 35 km thick crust, consisting of 15 km upper and 20 km lower crust, the surface heat flow is 58 m W m -2, which equals to the world's average in continents. Equation C1 was solved by finite difference method in Lagrange frame (nodes follow the movement of the material). The lihosphere was divided into nodes numbered from top. Thermal parameters were assigned to each interval between the nodes. Temperature was calculated in the nodes. Sedimentation was simulated by adding a new node at the top of the original grid. The thickness AH of a new element corresponds to the material deposited in the time At. It was assumed that the new layer had been deposited instantaneously at the beginning of the time interval and had surface temperature (7"2-- T1). The temperature at the end of the time interval At was calculated by solving eq. C1 by fully implicit method (Press et al. 1992). Then the process was repeated. The sedimentation of a new layer causes compaction of the sediments below. It was assumed that the sediments compact according to normal porosity-depth trends. Normal porosity-depth trends for coarse-grained (sand, sandstone) and fine-grained (shale, clay, marl) sediments were established by porosity measurements of rock samples in laboratory (D6vdnyi & Horvfith 1988): q~cg(Z) = ~ aiz i
a0 -- 46.90, al = -23.47,
i-0
a2 = 6.71, a 3 = -1.05 q~fg(Z) = ~ b i z i
(C4a)
b0 = 65.00, b 1 =-42.41,
i 0
b2 = -11.56, b3 = 13.90, b 4 = -2.56
(C4b)
where ~Pcg(Z)and q~fg(z) are the normal porosity-depth trend for coarse-grained and fine-grained sediments, respectively.
243
The ratio of fine-grained to coarse-grained sediments in different depositional units was estimated from well logs and the compaction was calculated according to the composition of the unit. Compaction of sediments increases the thermal conductivity, because porosity reduces with depth and the rock matrix is better conductor than pore water. Thermal conductivity of sediments was calculated according to Vacquier (1984) )k(Z) = )km1 - tP(z))kw~P(z)
(C5)
where hm and kw are the conductivities of the rock matrix and water, respectively. Specific heat per unit volume of sediments was given by (Lucazeau & Le Douaran 1984): pC(Z) = pCm(] - r
+ pCwq0(Z )
(C6)
where pcm and pCw are the specific heats per unit volume of rock matrix and water, respectively. Before simulating the sedimentation, the observed sedimentary column was decompacted, that is the thickness of each sedimentary unit with surface porosity was calculated according to: d h ' ( z = O) = d h ( z )
1 -,p(z) 1 - tp(z = O)
(C7)
where dh'(z = 0) is the decompacted thickness at the surface and dh(z) is the compacted thickness at depth z. q~(z = 0) and q~(z) is the porosity at the surface and at depth z, respectively. Sedimentation rates of each depositional unit were calculated for decompacted sediment thickness and the sedimentation was modelled with these sedimentation rates. A t the end of modelling the thickness of the c o m p a c t e d sediments was c o m p a r e d to the observed thickness. The difference in case of 7 km thick sedimentary column was less than 50 m. The model can take into account erosion, too, by removing sediments from the top. The model was checked by comparing the results to analytical solutions. There is analytical solution of eq. C1, if the sedimentation/erosion and the thermal parameters are Constant (Carslaw & Jaeger 1959). The temperature o b t a i n e d from the numerical solution differs less than 3% from the analytical solution. In correcting the heat flow, first the initial (nondisturbed) steady-state heat flow was calculated from eq. C3. At the end of modelling (present day situation) the disturbed heat flow was calculated and the ratio of the disturbed to the nondisturbed heat flow was obtained. The sediment corrected heat flow was received by correcting the observed heat flow with this ratio. Heat flow modelling was made in case of a model with minimum thermal parameters and a model with maximum thermal parameters (Table 2) in the Mak6
244
G. T A R I E T A L .
trough, where the thickness of sediments is some 7 km and 4 km out of it was deposited in the last 6 Ma. The resulting corrected heat flow differs less than 10%: 83 m W m 2 for maximum and 91 m W m 2 for minimum parameters. Equation C3 was also used in calculating the initial surface heat flow for the nonuniform extensional model, when the initial crustal thickness was increased. For any crustal thickness the ratio of the thickness of the upper crust to the thickness of the lower crust was constant and was equal to 3:4. Thermal parameters given in Table 2 were used in calculating the initial heat flow.
q(8,[3,t) = qo [1 + ~ 2x.exp(-nZcr2Kt/aa)]
where q0 is the initial heat flow, K is the thermal diffusivity and [(2/nn-r)(-1)n+qxn is the nth Fourier coefficient, calculated from the temperature distribution after thinning (eq. D1) and xn is:
(-1)"+~ xn -- [(8 - [3)sin(nvH) + sin(n-rrG)] - where
D: description of the non-
uniform extensional model
T:T,,8(1
T = T.[[3(1 - z_) a
-z-)
a
Yci a8
G=I
In the nonuniform extensional model of R o y d e n & Keen (1980) the deformation of the lithosphere is described by different amounts of crustal and mantle stretching. Instead of stretching factors we introduced thinning factors in the following way: the crustal thinning factor was defined as 8 = Yci/Yc, where Yci was the crustal thickness before rifting and Yc was the crustal thickness after rifting, and the mantle thinning factor was defined as [3 = Ymi/Ym, where Ymi was the thickness of the mantle lithosphere before rifting and Ym was the thickness of the mantle lithosphere after rifting. Assuming instantaneous thinning, every infinitesimal volume of the lithosphere keeps its former temperature as it rises to higher position, thus the temperature gradient in the lithosphere rises. Assuming that the geotherm was linear before rifting with T = 0 at the upper surface of the lithosphere, at z = a and T = T a at the bottom, at z = 0, the temperature distribution after thinning i s :
Yci
a8
(D4) [3
(D1) f o r a - Yc - Ym < Z < a - Yc T = 7",, f o r O <_ z < a - yc - y m
In the model local isostasy is assumed, and depending on whether density increase due to crustal thinning or density decrease due to temperature increase is larger, initial subsidence or uplift occurs, respectively. This subsidence (uplift) is contemporaneous with the rifting of the lithosphere, thus it is called syn-rift subsidence (uplift). Following rifting the lithosphere cools and contracts causing further subsidence, which is termed post-rift or thermal subsidence. The formulas given by R o y d e n & Keen (1980) allow to calculate the heat flow and the basement depth any time after rifting as functions of thinning factors. The heat flow, q, is:
(D5)
S(8,[3,t) = E + U + T E
where E is the final equilibrium elevation due to crustal thinning, U is the elevation above the equilibrium elevation, approaching zero with time due to cooling and T E is the topographic elevation before rifting. Positive values indicate uplift, negative values indicate subsidence. E can be written: P m -- P c
E-
e~T,,yd
1
pm(1 _ aTa) yd ( 1 - g)(1 -
2a
)
(D6)
where [am is the mantle density at surface temperature, Pc is the crustal density at surface temperature and oLis the thermal expansion coefficient. U is given by: 4 ~ X(2m+1) 1,2 ..... - , (Tmm+)52 exp[-(2m
Yci + -a- (~ - ~)1
Yci/a
1 -
The subsidence, S, is given by:
ao~ T,, U-l_~
fora-yc
(D3)
n'/T
H=I---
Appendix
(D2)
n 1
+ 1)2peKt/a 2]
(D7)
The model gives the subsidence without water or sediment load. Therefore the effect of sedimentary load on the observed subsidence has to be removed and the resulting 'airloaded subsidence' has to be compared to the model predictions. The 'airloaded subsidence', Sa, was calculated assuming local isostasy:
s,, : s, Om- Ps
(D8)
Pm
where Ss is the depth to the base of Neogene and Pm and Ps are the mantle density and mean sediment density, respectively. The mean sediment density was calculated from the density of the sediment grain and density of pore water assuming normal compaction trend:
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
Ps= I [1-q~(z)]pgdz + I q~(Z)pwdz z-O z 0
(D9)
where pg and Pw are the sediment grain density and pore water density, respectively, and q~(z) is the normal porosity-depth trend (eq. C4a,b). There was no detailed subsidence analysis made along the sections. The thinning factors were calculated by fitting the predicted present day 'airloaded basement depth' (eq. D5) and predicted heat flow (eq. D2) to the observed (corrected) values. The calculation was made by Newton-Raphson iteration, solving the equation system consisting of D2, D3, D4, D5, D6 and D7. The parameters of the model are given in Table 2. Minimum and maximum crustal and mantle thinning factors were calculated from the lower and upper estimated error limits of the corrected and filtered heat flow.
References ADAM,A., SZARKA,L., PRACSER,E. & VARGA, G. 1996. Mantle plumes or EM distortions in the Pannonian Basin? (Inversion of the deep magnetotelluric (MT) soundings along the Pannonian Geotraverse). Geophysical Transactions, 40, 45-78. ANDREESCU,M., BURST,D., DEMETRESCU, C., PENE, M. & POLONIC,G. 1989. On the geothermal regime of the Moesian Platform and Getic Depression. Tectonophysics, 164, 281-286. BALAZS E. & NUSSZER, A. 1987. Unterpannonischer Vulkanismus der Beckengebiete Ungarns. Annals of the Hungarian Geological Institute, 69, 95-104. BASSI, G. 1995. Relative importance of strain-rate and rheology for the mode of continental extension. Geophysical Journal International, 122, 195-210. BgRCZI, I., HAMOR, G., JAMBOR, A. & SZENTGYORGYI, K. 1988. Neogene sedimentation in Hungary. In: ROYDEN,L. H. & HORVATH,E (eds) The Pannonian Basin - a Study in Basin Evolution. American Association of Petroleum Geologists, Memoirs, 45, 57-67. BERCZI-MAKK,A. 1986. Mesozoic formation types of the Great Hungarian Plain. Acta Geologica Hungarica, 29, 261-282. BEzAK, V., SEFARA, J., BmLIK, M. & KUBES, P. 1995. Lithospheric structure in the Western Carpathians: geophysical and geological interpretation. Mineralia Slovaca, 27, 169-178 [in Slovakian with English summary]. BIELIK,M. 1988. A preliminary stripped gravity map of the Pannonian Basin. Physics of the Earth and Planetary Interiors, 51, 185-189. -1991. Density modeling of the Earth's crust in the Intra-Carpathian basins. In: KARAMATA,S. (ed.), Geodynamic Evolution of the Pannonian Basin. Academic Conference of the Serbian Academy of Science and Arts, 62, 123-132. , FUSAN,O., BURDA,M., HOBNER,M. & VYSKOCIL, V. 1990. Density models of the Western Carpathians. Contributions of the Geophysical Institute of
245
Slovakian Academy of Sciences, 20, 103-113. MAJCIN, D., FUSAN,O., BURDA, M., VYSKOCm,V. & TRESL, J. 1991. Density and geothermal modeling of the Western Carpathian Earth's crust. Geologica Carpathica, 42, 315-322. --, MOCANU, V., ADAM, A., DINU, C. &; LILLIE, R. J. 1996. A preliminary study of gravity field in the Eastern and Southern Carpathians and the narrow continental rifts in the Pannonian Basin. In: DECKER,K. (ed.) PANCARDI workshop 1996 - Dynamics of the Pannonian-CarpathianDinaride System. Mitteilungen Gesellschaft. Geologischen Bergbaustudenten Osterreich, 41, 102. BILIBAJKIC, P., MLADENOVIC, M., MUJAG1C, S. & RIMAC, I. 1979. Tumac za gravimetrijsku kartu SFR Jugoslavije-Bouguerove anomalije, 1:500000 [Explanation for the gravity map of SFR Yugoslavia-Bouguer anomalies, 1:500000]. Savezni Geoloski Zavod, Beograd. BLEAHU, M., HAAS, J., KovAcs, S., PER0, Cs., MANTEA, G., BORDEA, S., PANIN, S., BI~RCZI-MAKK, A., STEFANESCU, M., KONRAO, GY., NAGY, E., RALISCHFELGENHAUER, E., SIKIC, K. & TOROK, /k. 1994. Triassic facies types evolution and paleogeographic relations of the Tisza megaunit. Acta Geologica Hungarica, 3% 187-234. - - - , LuPu, M., PATRULIUS,D., BORDE, S., STEFAN,A. & PANIN, S. 1981. The structure of the Apuseni Mountains. In: Proceedings of the Carpathian-Balkan Geological Association Conference 12, Guide to Excursion B3, Bucharest. BLOCK, L. & ROYDEN,L. 1990. Core complex geometries and regional scale flow in the lower crust. Tectonics, 9, 557-567. BUCK, W. R. 1991. Modes of continental lithospheric extension. Journal of Geophysical Research, 96, 163-168. -1993. Effect of lithospheric thickness on the formation of high- and low-angle normal faults. Geology, 21, 933-936. - - , MARTINEZ, E, STECKLER, M. S. • COCHRAN, J. R. 1988. Thermal consequences of lithospheric simple shear: Pure and simple. Tectonics, 7, 213-234. BURCHART,J. 1972. Fission track age determinations of accessory apatite from the Tatra Mountains, Poland. Earth and Planetary Science Letters, 15, 418-422. BUREAU GRAV1MI~TRIQUE INTERNATIONAL 1963. Cartes Mondiales des Anomalies de Bouguer (Berlin-Vienne) 1:1 million. Paris. 1966. Cartes Mondiales des Anomalies de Bouguer (Budapest-Oslo) 1:1 million. Paris. -1971. Anomalies de Bouguer Europe-Afrique, 1:10 million, Paris. CARSLAW, H. S. • JAEGER, J. C. 1959. Conduction of heat in solids. Clarendon Press, Oxford. CLOETINGH, S., VAN WEES, J. D., VAN DER BEEK, P. A. & SPADINI,G. 1995. Role of pre-rift rheology in kinematics of extensional basin formation: constraints from thermomechanical models of Mediterranean and intracratonic basins. Marine and Petroleum Geology, 12, 793-808. CSONTOS,L., NAGYMAROSY,A., KOVAC,M. & HORVATH, --,
246
G. T A R I ETAL.
F. 1992. Tertiary evolution of the intra-Carpathian area: a model. Tectonophysics, 208, 221-241. DALLMEYER, R. D., NEUBAUER, E, HANDLER, R., MOLLER, W., FRITZ, H., ANTONITSCH, W. & HERMANN, S. 1992. 4~ and Rb-Sr mineral age controls for the pre-Alpine and Alpine tectonic evolution of the Austro-Alpine nappe complex, Eastern Alps. In: NEUBAUER, E (ed.) The Eastern Central Alps of Austria. A L C A P A 1992 meeting field guidebook, 47-59. --, PANA, D. & FRITZ, H. 1994. Variscan vs, Alpine tectonothermal evolution within the Apuseni Mountains, Romania: evidence from 4~ mineral ages. Romanian Journal of Tectonics and Regional Geology, 7512, 65-76. DAVIS, G. H. & LISTER,G. S. 1988. Detachment faulting in continental extension: perspectives from the Southwestern U.S. Cordillera. In: CLARK, S. P. JR., BURCHFIEL, B. C. & SUI'PE, J. (eds)Processes in Continental Lithospheric Deformation. Geological Society of America, Special Papers, 218, 133-159. DEMETRESCU,C. & POLONIC,G. 1989. The evolution of the Pannonian Depression (Romanian sector) as derived from subsidence and heat flow data. Tectonophysics, 164, 287-299. -& VEHCIU, S. 1991. Heat flow and lithosphere structure in Romania. In: CERMAK,V. d~;RYBACH, L. (eds) Terrestrial Heat Flow and the Lithosphere Structure. Springer, 187-205. DICEA, O., IANAS, M., LUNGU, A., GHEORGHIU, G., IONESCU, G. & TALOS, D. 1984. Paradigmes structuraux-depositionnels des formations pannoniennes deduits par analyse et interpretation des prospections sismiques de reflection. An. Inst. Geol. Geofiz., 64, 391-400. D0VI~NYI, R 1994. Geophysical investigations of the lithosphere of the Pannonian basin. PhD Thesis, E6tv6s Univ., Budapest. - & HORVATH, E 1988. A review of temperature, thermal conductivity, and heat flow data from the Pannonian Basin. In: ROYDEN, L. H. & HORVATH, F. (eds) The Pannonian Basin - a Study in Basin Evolution. American Association of Petroleum Geologists, Memoirs, 45, 235-255. - - , LIEBE, R, GALFI, J. & ERKI, I. 1983. Geothermal conditions of Hungary. Geophysical Transactions, 29, 3-114. DUMITRU, T. A. 1993. A new computer-automated microscope stage system for fission-track analysis. Nuclear Tracks and Radiation Measurements, 21, 575-580. DUNBAR, J. A. & SAWYER,D. S. 1989. How pre-existing weaknesses control the style of continental breakup. Journal of Geophysical Research, 94, 7,278-7,292. DUNKL, I. 1990. The fission track method and its application for solving geochronological problems. PhD thesis, Technical University of Miskolc. - 1992. Final episodes of the cooling history of eastern termination of the Alps. In: NEUBAUER,F. (ed.) The Eastern Central Alps of Austria. A L C A P A 1992 meeting field trip guidebook, 137-139. - & DEMI3NY,A. 1997. Exhumation of the Rechnitz
Window at the border of the Eastern Alps and Pannonian basin during Neogene extension. Tectonophysics, 272, 197-211. EBNER, E & SACHENSI-IOFER,R. E 1995. Paleogeography, subsidence and thermal history of the Neogene Styrian Basin (Pannonian basin system, Austria). Tectonophysics, 242, 133-150. ERDI~LYI M. 1985. Geothermics and the deep flowsystem of the Hungarian Basin. Journal of Geodynamics, 4, 321-330. F~3LOP, J. 1994. Geology of Hungary, Paleozoic II. Akad6miai Kiad6, Budapest [In Hungarian]. - - , BREZSNYANSZKY,K. & HAAS, J. 1987. The new map of basin basement of Hungary. Acta Geologica Hungarica, 30, 3-20. GERNER, P., BADA, G., DOVI~NYI, P., CLOETINGH, S., ONCESCU,M. & MOLLER,B. 1999. Recent tectonic stress and crustal deformation in the Pannonian region: data and models. This volume. GLEADOW,A. J.W. 1981, Fission-track dating methods: what are the real alternatives? Nuclear Tracks, 5, 3-14. GREEN, P. E 1981. A new look at statistics in fission track dating. Nuclear Tracks, 5, 77-86. GROW, J. A., POGACSAS, GY., Bt~RCZI-MAKK, A., VARNAI, P., HAJD~, D., VARGA, E. & PERO, Cs. 1989. Tectonic and structural conditions of the B6k6s Basin. In Hungarian with English summary. Magyar Geofizika, 30, 63-97. - - , MATTICK, R. E., Bt~RCZI-MAKK,A., PIER0 , Cs., HAJDI), D., POGACSAS,GY., VARNAI, P. & VARGA, E. 1994. Structure of the B6k6s basin inferred from seismic reflection, well and gravity data. In: TELEKI, R G., MATTICK, R. E. & K6KAI, J. (eds) Basin Analysis" in Petroleum Exploration, A case study from the BOkOs basin, Hungary. Kluwer Academic Publishers, 1-38. GYORFI, I. & CSONTOS,L. 1994. Structural evolution of SE Hungary and Neogene basins of the Apuseni Mountains. Rornanian Journal of Tectonics and Regional Geology, 75/1, 19-20. - - , CSONTOS,L. & NAGYMAROSY,A. 1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian basins. This volume. HAJNAL, Z., REILKOFF, B., POSGAY, K., HEGED13S,E., TAKACS, E., ASUDEH, I., ST. MUELLER,ANSORGE, J. & DEIACO, R. 1996. Crustal-scale extension in the Central Pannonian Basin. Tectonophysics, 264, 191-204. HARANGI SZ., VASELLI, O., TONARINI, S., SZAB0, CS., HARANG1, R. & CORADOSS[, N. 1995a. Petrogenesis of Neogene extension-related alkaline volcanic rocks of the Little Hungarian Plain Volcanic Field (Western Hungary). In: DOWNES,H. & VASELLI, O. (eds) Neogene and related volcanism in the Carpatho-Pannonian Region. Acta Vulcanologica, 7, 173-187. - - , WILSON,M. & TONARINI,S. 1995b. Neogene potassic volcanic rocks in the Pannonian Basin: geochemistry, petrogenesis and geodynamical implications. In: DOWNES, H. & VASELLI,O. (eds) Neogene and related magmatism in the CarpathoPannonian Region. Acta Vulcanologica, 7,125-134.
L I T H O S P H E R I C S T R U C T U R E OF P A N N O N I A N BASIN HATCHER, R. D. & HOOPER, R. J. 1992. Evolution of crystalline thrust sheets in the internal parts of mountain chains. In: MCCLAY, K. R. (ed.) Thrust Tectonics. Chapman and Hall, 217-233. HOERNLE, K., ZHANG, Y.-S. & GRAHAM, D. 1995. Seismic and geochemical evidence for large-scale mantle upwelling beneath the eastern Atlantic and western and eastern Europe. Nature, 374, 34-39. HOPPER, J. R. & BUCK,W. R. 1996. The effect of lower crustal flow on continental extension and passive margin formation. Journal of Geophysical Research, 101, 20,175-20,194. HORVATH, E 1988. Neotectonic behavior of the Alpine-Mediterranean region. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin - a Study in Basin Evolution. American Association of Petroleum Geologists, Memoirs, 45, 49-55. -1993. Towards a mechanical model for the evolution of the Pannonian basin. Tectonophysics, 226, 333-357. 1995. Phases of compression during the evolution of the Pannonian Basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology, 12, 837-844. & ROYDEN,L. 1981. Mechanism for the formation of the intra-Carpathian basins: a review. Earth and Evolutionary Science, 3, 307-316. , CLOETINGH, S. & TARI, G. 1995. Neotectonics (Pliocene through Quaternary) in the NW Pannonian Basin. In: HORVATH,E,TARI, G. & BOKOR, Cs. (eds) Extensional Collapse of the Alpine
-
-
-
-
Orogene and Hydrocarbon Prospects in the Basement and Basement Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to Fieldtrip No. 6, Hungary, 47-60. --, DOVI;NY1, P. & LACZ0, I. 1986. Geothermal effects of magmatism and its contribution to the maturation of organic matter in sedimentary basins. In: BUNrEBARTH,G. & STEGENA,L. (eds), Paleogeothermics. Lecture Notes in Earth Sciences, 5, Springer-Verlag, Berlin/Heidelberg, 173-183. HURFORD,A. J. & GREEN, P. E 1983. The zeta age calibration of fission-track dating. Chemical Geology, (Isotope Geosciences), 41, 285-312. - - & HAMMERSCHMIDT,K. 1985.4~ and K/Ar dating of the Bishop an Fish Canyon Tuffs: calibration ages for fission track dating standards. Chemical Geology (Isotope Geosciences), 58, 23-32. & WATKINS R. T. 1987. Fission-track age of the tufts of the Buluk Member, Bakate Formation, Northern Kenya: A suitable fission-track age standard. Chemical Geology, 66, 209-216. - - , FITCH,F. J. & CLARKE,A. 1984. Resolution of the age structure of the detrital zircon populations of two Lower Cretaceous sandstones from the Weald of England by fission track dating. Geol. Mag., 121, 269-277. HURTIG, E., CERMA.K,V., HAENEL, R. & ZuI, V. I. (eds) 1992. Geothermal Atlas of Europe. Haack, Gotha. IANOVICI,W., BORCOS,M., BLEAHU,M., PATRULIUS,D., LuPU, M., DIMITRESGU, R. & SAVU, H. 1976. La -
-
247
gdologie
des Monts Apuseni. Editions de Acadami, Bucuresti. [In Romanian with French summary]. IBRMAJER,J. 1981. Geological interpretation of gravity maps of Czechoslovakia. In: ZATOPEK, A. (ed.) Geophysical Syntheses in Czechoslovakia. Veda, Bratislava, 135-148. JACKSON,J. A. & WHITE, N. J. 1989. Normal faulting in the upper continental crust: observations from regions of active extension. Journal of Structural Geology, 11, 15-36. JORDAN, S. K. 1978. Statistical model for gravity, topography, and density contrasts in the Earth. Journal of Geophysical Research, 83, 1,816--1,824. KAPPELMAVER, O. & HAENEL, R. 1974. Geothermics with Special Reference for Application. Gebrtider Borntraeger, Berlin. KARNER, G. D. &WATTS,A. B. 1983. Gravity anomalies and flexure of the lithosphere at mountain ranges, Journal of Geophysical Research, 88, 10,449-10,477. KILI~NYI, ]~., KROLL, A., OBERNAUER, D., SEFARA, J., STEINHAUSER,P., SZAB0 , Z. & WESSELY, G. 1991. Pre-Tertiary basement contour map of the Carpathian Basin beneath Austria, Czechoslovakia and Hungary. Geophysical Transactions, 36, 15-36. KovAcs, G. & KURUEZ, B. 1984. The pre-Mesozoic formations of the southern part of the Great Hungarian Plain. Hungarian Geological Institute, Budapest [in Hungarian]. KOVACSVOLGYI,S. 1994. Interpretation of gravity and magnetic anomalies in the B6k6s basin in the light of recent data. Magyar Geofizika, 35, 90-94 [in Hungarian with English abstract]. KOROSSY, L. 1992. Hydrocarbon geology of the area between the Duna and Tisza rivers in Hungary. General Geological Review, 26, 3-162 [in Hungarian with English summary]. LABAS, V. 1987. Some structural characteristics of the subsurface in a part of the central zone of the Dinaric gravity low. Nafta, 38, 547-554. --, KRULC, Z., ALIINOVIC, B., BRDAREVIC, N. & LOVRIC,D. 1981. The results of geophysical investigations of Pregibna zona. Nafia, 32, 487--497. LANKREIJER, A., KOVAC,M., CLOETINGH, S., PITONAK, P., HLOSKA, M. & BIERMANN, C. 1995. Quantitative subsidence analysis and forward modelling of the Vienna and Danube basins: thin-skinned versus thick-skinned extension. Tectonophysics, 252, 433-451. LILLIE, R. J. & BIELIK, M. 1992. Crustal development and tectonic models of Western Carpathians from gravity interpretation. Geologica Carpathica, 43, 63-68. LUCAZEAU,E & LE DOUARAN,S. 1985. The blanketing effect of sediments in basins formed by extension: a numerical model. Application to the Gulf of Lion and Viking graben. Earth and Planetary Science Letters, 74, 92-102. MARTEL, D. J., DEAK, J., DOVt~NYI, P., HORV.&TH, E, O'NIONS, R. K., OXBURGH, E. R., STEGENA, L. & STUTE, M. 1989. Leakage of helium from the Pannonian Basin. Nature, 342, 908-912.
248
G. TARI E T A L .
MATTICK, R. E., PHILLIPS,R. L. & RUMPLER, J. 1988. Seismic stratigraphy and depositional framework of sedimentary rocks in the Pannonian basin in southeastern Hungary. In: ROYDEN, n. H. & HORVATH, E (eds) American Association of Petroleum Geologists, Memoirs, 45, 117-145. MCKENZIE, D. 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters, 40, 25-32. MESK0, A. 1983. Regional Bouguer gravity maps of Hungary. Acta Geodinamica Geophisica et Montana Hungarica, 18, 187-200. 1988. Reduced regional Bouguer-anomaly map of Hungary. Acta Geodinamica et Geophisca Montana Hungarica, 23, 89-95. MOCANU,V. I. & RADULESCU,E 1994. Geophysical features of the Romanian territory. In: ALCAPA II "Geological evolution of the Alpine-CarpathianPannonian system" Field Guidebook. Romanian Journal of Tectonics and Regional Geology, 75, 17-40. MOTAS, C. 1983. Nouvelles donn6es sur les rapports structuraux entre les Carpathes M6ridionales et la d6pression Gdtique. Annals of the Romanian Institute of Geology & Geophysics, 60, 147-158. NAGYMAROSY, A. & BALDI-BEKE, M. 1993. The Szolnok unit and its probable paleogeographic position. Tectonophysics, 226, 457-470. NEMESI, L. & STOMFAI,R. 1992. Some supplements to the exploration of the basement of the B6k6sbasin. Magyar Geofizika, 33, 70-79. In Hungarian with English abstract. NI~MEDI-VARGA, Z. 1983. Tectonics of the Mecsek Mountains in the Alpine orogenic cycle. Annual Report Hungary Geological Survey, 1981, 467-484 [in Hungarian with English summary]. OXBURGH, E. R. & ENGLAND,P. C. 1980. Heat flow and the metamorphic evolution of the Eastern Alps. Eclogae Geologicae Helvetiae, 73, 379-398. PAP, S. 1990. Thrust-faulted sequences in Transtisia. Special Publications of the Hungarian Geological Survey, Budapest [in Hungarian]. PI~CSKAY, Z., LEXA, J., SZAKACS, A., BALOGH, K., SEGHEDI, I., KONECNY,V., KOVACS,M., MARTON, E., KALICIAK, M., SZI~KY-FUx, V., POKA, T., GYARMATI,P., EDELSTEIN,0., ROSU, E. & ZEC, B. 1995. Space and time distribution of NeogeneQuaternary volcanism in the Carpatho-Pannonian Region. In: DOWNES,H. & VASELLI, O. (eds) Neogene and related magmatism in the Carpatho-Pannonian Region. Acta Vulcanologica, 7, 15-28. POSGAY, K. & SZENTGYORGYI, K. 1991. Shear zone penetrating the lithosphere at the eastern part of the Pannonian Basin. Magyar Geofizika, 32, 1-15 [in Hungarian with English abstract]. - - , ALBU, MAYEROVA,M., NAKLADOVA,Z., IBRMAJER, I., BLIZKOVSKY,M., ARIC, K. & GUTDEUTSCH, R. 1991. Contour map of the Mohorovicic discontinuity beneath Central Europe. A G U Geophysical Transactions, 36, 7-13. --, BODOKY, T., HEGEDOS, E., KOVACSVOLGYI,S., LENKEY,L., SZAFIAN,P., TAKACS,E., T[MAR, Z. & VARGA, G. 1995. Asthenospheric structure -
-
beneath a Neogene basin in southeast Hungary. Tectonophysics, 252, 467-484. --, HEGEDOS, E. & TiMAR, Z. 1990. The identification of mantle reflections below Hungary from deep seismic profiling. Tectonophysics, 173, 379-385. , TAKACS,E., SZALAY,I., BODOKY,T., HEGEDOS,E., KANTOR, I., T[MAR, Z., VARGA, G., BERCZI, I., SZALAY, /~x., NAGY, Z., PAPA, A., HAJNAL, Z., REILKOFF,B., MUELLER,S., ANSORGE, J., DEIACO, R., & ASUDEH, I. 1996. International deep reflection survey along the Hungarian Geotraverse. AG U Geophysical Transactions, 40, 1-44. POSP[SIL, L. 1980. Interpretation of the gravity field in the East Slovakian Neogene area. Mineralia Slovaca, 12, 421-440 [in Slovakian with English summary]. PRESS, W. H., TEUKOLSKV,S. A., VETrERLIN6, W. T. & FLANNERY, B. P. 1992. Numerical Recipes. Cambridge University Press. RATSCHBACHER, L., FRISCH, W., LINZER, H. G. & MERLE, O. 1991: Lateral extrusion in the Eastern Alps, part 2: Structural analysis. Tectonics, 10, 257-271. RAVNIK, D. 1992. Catalogue of Heat Flow Density Data: Yugoslavia. In: HURTm, E., CERMAK, V., HAENEL, R. & ZuI,V. I. (eds) GeothermalAtlasof Europe, Haack, Gotha. - - , RAJVER,D., POLJAK,M. & ZWCK:,M. 1995. Overview of the geothermal field of Slovenia in the area between the Alps, the Dinarides and Pannonian basin. Tectonophysics, 250, 135-149. RENNER, J. & STEGENA, L. 1965. Gravity research of the deep structure of Hungary. Annales of the University of Sciences Budapest, 8, 153-159. ROYDEN, L. H. 1988. Late Cenozoic tectonics of the Pannonian basin system. In: ROYDEN, L. H. & HORVATH, E (eds) American Association of Petroleum Geologists Memoirs, 45, 27-48. & KEEN, C. E. 1980. Rifting process and thermal evolution of the continental margin of Eastern Canada determined from subsidence curves. Earth and Planetary Science Letters, 51, 343-361. & HORVATH,E (eds) 1988. The Pannonian Basin - a Study in Basin Evolution. In: ROYDEN,L. H. & HORVATH, E (eds) American Association of Petroleum Geologists Memoirs, 45. & DOVI~NYI,P. 1988. Variations in extensional styles at depth across the Pannonian basin system. In: ROYDEN,L. H. & HORVATH,E (eds) The Pannonian Basin - a Study in Basin Evolution. American Association of Petroleum Geologists Memoirs, 45, 235-256. - - , HORVATH, E & BURCHFIEL, B. C. 1982. Transform faulting, extension, and subduction in the Carpathian Pannonian region. Geological Society of American Bulletin, 93, 717-725. - - , NAGYMAROSYA. & STEGENAL. 1983 Evolution of the Pannonian basin system: 2. Subsidence and thermal history. Tectonics, 2, 91-137. ROZLOZSNIK, P. 1936. Die tektonische Stellung der Bihargebirgsgruppe (Muntii Apuseni) im Karpatensystem. Math. und Naturwissenschafl, Anzeiger Budapest, 55, 46-74. -
-
-
-
L I T H O S P H E R I C S T R U C T U R E OF P A N N O N I A N BASIN RUMPLER,J. & HORVATH,E 1988. Some representative seismic reflection lines from the Pannonian basin and their structural interpretation. In: ROYDEN,L. H. & HORVATH,E (eds) American Association of Petroleum Geologists Memoirs, 45, 153-169. RUVVEL,C. 1995. Extensional processes in continental lithosphere. Journal of Geophysical Research, 100, 24 187-24 215. SACHSENHOFER,R. E, LANKREIJER,A., CLOETINGH, S. & EBNER, F. 1997. Subsidence analysis and quantitative basin modelling in the Styrian Basin (Pannonian Basin System, Austria). Tectonophysics, 272, 175-196. SCHEFFER,V. 1960. Some contributions to the geophysical knowledge of the Carpathian basins. Acta Technica Academia Scientifica Hungarica, 30, 423-461. SCLATER,J. G., ROYDEN, L., HORVATH, E, BURCHFIEL, B. C., SEMPKEN, S. & STEGENA, L. 1980. The formation of the intra-Carpathian basins as determined from subsidence data. Earth and Planetary Science Letters, 51, 139-162. SEKIGUCHI, K. 1984. A method for determining terrestrial heat flow in oil basinal areas. Tectonophysics, 103, 67-79. SKORVANEK,M. & BIELA, A. 1993. Density distribution along geological profiles KP-VIII and KP-IX. Contributions to Geophysics at Institute Slovakia, Acadamyof Sciences, 23, 118-125. SPAKMAN,W., VAN DER LEE, S. & VAN DER HILST, R. 1993. Travel-time tomography of the EuropeanMediterranean mantle down to 1400 km. Physics of the Earth and Planetary Interiors, 79, 3-7. STEGENA, L. 1964. The structure of the Earth's crust in Hungary. Acta Geologica Hungarica, 8, 413-431. , G~czY, B. & HORVATH, E 1975. Late Cenozoic evolution of the Pannonian basin. Tectonophysics, 26, 71-90. STEFANESCU,M. (ed.) 1988. Regional geologic profiles across Romania, at the scale of l.'200,O00, section 3-B. Geological Institute of Romania, Bucharest. SZABO, Cs., HARANGI, Sz. & CSONTOS,L. 1992. Review of Neogene and Quaternary volcanism of the Carpathian-Pannonian region. Tectonophysics, 21)8, 243-256. , - - , VASELLI, 0. & DOWNES, H. 1995. Temperature and oxygen fugacity in peridotite xenoliths from the Carpathian-Pannonian Region. In: DOWNES, H. & VASELLI, O. (eds) Neogene and related magmatism in the Carpatho-Pannonian Region. Acta Vulcanologica, 7, 231-239. SZAFIAN,P. & TARI, G. 1995. Preliminary results on the gravity modeling of a crustal transect in the Alpine-Pannonian junction. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6, Hungary, 107-118. - - , HORVATH,F. & CLOETINOH,S. 1997. Gravity constraints on the crustal structure and slab evolution along a trans-Carpathian transect. Tectonophysics, 272, 233-247.
--,
249
TARI, G., HORVATH,E • CLOETINGH,S. in press. Crustal structure of the Alpine-Pannonian junction: combined seismic and gravity study. Geologisches Rundschau. SZALAY,A. 1983. Metasomatic alteration of metamorphic, granitoid rocks in the basement of the Great Plain. Acta Geologica Hungarica, 26, 359-374. SZEPESHAZY, K. 1973. Upper Cretaceous and Paleogene formations of NW-Transtisia. Akad6miai, Budapest [in Hungarian]. TAm, G. 1994. Alpine Tectonics of the Pannonian Basin. Ph.D. thesis, Rice University, Houston, Texas. -& BALLY,A. W. 1990. Metamorphic core complexes at the boundary of the Eastern Alps and the Pannonian basin. Geological Society of America Abstracts with Programs, 22, 97-98. & HORVATH, F. 1995. Middle Miocene extensional collapse in the Alpine-Pannonian transition zone. In: HORVATH,E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of the Alpine Orogene and Hydrocarbon Prospects" in the Basement and Basement Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to Fieldtrip No. 6, Hungary, 75-106. --, BALDI, T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian basin: a geodynamic model. Tectonophysics, 226, 433-455. - - , HORVATH,E & RUMPLER,J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. , -& WEre, G. 1995. Palinspastic reconstruction of the Alpine/CarpathiardPannonian system. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional Collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basement Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to Fieldtrip No. 6, Hungary, 119-131. TELEKI, P. G., MATTICK,R. E. & KOKAI, J. (eds) 1994. Basin Analysis in Petroleum Exploration, a Case Study from the BOk& basin, Hungary. Kluwer Academic Publishers. TOMEK, C. 1988. Geophysical investigation of the Alpine-Carpathian arc. In: RAKUS, M., DERCOURT,J. & NAIRN,A. E. M. (eds) Evolution of the Northern Margin of Tethys, VoL I., M~moires de la SocidtO Gdologique de France, Paris, Nouvelle S6rie, 154, 167-199. - - , IBRMAJER,I., KORAB,T., BmLY, A., DVORAKOVA, L., LEXA, J. & ZBORIL, A. 1989. Crustal structures of the West Carpathians on deep seismic line 2T. Mineralia Slovaca, 21, 3-26 [in Slovakian with English summary]. UNRUG, R. 1982. Geodynamic evolution of the Carpathians. Annales de la SocidtO GOologique de Pologne, 52, 39-66. VACQUIER,g. 1984. Oil fields - a source of heat flow data. Tectonophysics, 103, 81-98. VAKARCS,G., HA RDENBOL,J., ABREAU,V. S., VAIL, P. R., VARNAI, R & TARI, G. 1998. Oligocene-Middle -
-
250
G. TARI E T A L .
Miocene Depositional Sequences of the Central Parathethys and their correlation with regional stages. In: DE GRACIANSKY,P. C., HARDENBOL,J., JACQUIN, T. & VAIL, R R. (eds) Meoszoic and Cenozoic Sequence Stratigraphy of European Basins. Society of Economic Paleontologists and Mineralogists Special Publications, 60, 209-231. , VAIL, P. R., TARI, G., POGACSAS,G., MATTICK,R. E. & SZABO, A. 1995. Third-order MiocenePliocene depositional sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106. VAN DER BEEK, B.A., ANDRIESSEN,EA.M. & CLOETINGI-I,S. 1995. Morphotectonic evolution of rifted continental margins: inferences from a coupled tectonic-surface processes model and fissiontrack thermochronology. Tectonics, 14, 406-421. VASIL[U, I., SOLOMON, A., POPESCU, M. & FOTOPOLOS, S. 1973. Contributions g6ophysiques ~t l'6tude tectonique de la zone Carei-Satu Mare (Roumanie). Revues Roumanian de GOologie, GOophysique et GOographie, Series GOophysique, 17, 223-228. VYSKOCIL, M., BURDA, M., BIELIK, M. & FUSAN, O.
1992. Further density models of the Western Carpathians. Contributions of the Geophysical Institute of the Slovakian Academy of Sciences, 22, 81-91. WARNER,M. 1990. Basalts, water, or shear zones in the lower continental crust? Tectonophysics, 173, 163-174. WERNICKE, B. 1985. Uniform-sense normal simple shear of the continental lithosphere. Canadian Journal of Earth Sciences, 22, 108-125. WORTEL, M. J. R. & SPAKMAN,W. 1992. Structure and dynamics of subducted lithosphere in the Mediterranean region. Proceedings of the Koninklije Nederlandes v. Wetensch., 95, 325-347. ZAUN, P. E. & WAGNER, G. A. 1985. Fission-track stability in zircon under geological conditions. Nuclear Tracks, 10, 303-307. ZOTH, G. & HAENEL,R. 1988. Appendix. In: HAENEL, R., RYBACH,L. & STEGENA,L. (eds) Handbook of Terrestrial Heat-flow Determination with Guidelines and Recommendations of the International Heat Flow Commission. Kluwer Academic Publishers, Dordrecht, 449-466.
Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian Basins ISTVAN GYt3RFI, a Lfi~SZLO CSONTOS 2 & ANDRAS
NAGYMAROSY
2
1EOtvos University, Geophysical Department, Ludovika tdr 2, 1088 Budapest, Hungary 2EOtvOs University, Department o f Geology, Mtizeum krt. 4/A, 1063 Budapest, Hungary
Asbtract: The Mid-Hungarian Lineament is the most important tectonic feature of the Intra-Carpathian area. Its evolution is closely related to the Early Tertiary (Palaeogene-Early Miocene) episodes of basin formation. This paper attempts to explain the structural relationships between the Transylvanian Palaeogene Basin and the Mid-Hungarian Lineament. Microtectonic data collected on the field, combined with the available geological and geophysical data point to the following conclusions. (1) The first compressional event is probably Early Oligocene in age, and is characterized by o-1 oriented ENE-WSW. We assume that the Oligocene basin is flexural in origin. (2) The second phase of compression occurred during Early Miocene times with erl oriented NNW-SSE. The last thrust emplacement accounts for the Ottnangian-Karpatian overthrusting of the Alcapa block on top of the Tisza-Dacia block. This deformational phase could have induced large-scale block rotations in both units. (3) Late Miocene left lateral faulting occurred along the Dragos-Voda fault system, the effects of which can be traced as far as Hungary. This event most probably corresponds to an eastward escape of the Tisza-Dacia block with respect to surrounding terranes.
Based on regional geophysical features (Visarion & Veliciu 1981; Horvfith 1993) two major units can be distinguished within the IntraCarpathian area: the Pannonian Basin system and the Transylvanian Basin. While in the case of the P a n n o n i a n Basin recent g e o d y n a m i c models can explain a d e q u a t e l y the Tertiary structural evolution (Horvfith 1993; Tari et al. 1993), no reasonable explanation has been given for the substantial subsidence which must have affected the Transylvanian Basin. The most important tectonic feature of the I n t r a - C a r p a t h i a n area is the M i d - H u n g a r i a n Lineament (MHL), separating two major tectonic units: the Alcapa and the Tisza-Dacia blocks (Fig. 1). The M H L had a major impact on the evolution of the Pannonian Basin (Csontos et al. 1992), as well as on the evolution of the Maramures and Transylvanian Basins. This very important lineament is covered almost in its entire length by thick Neogene deposits. The only exposure of the M H L is found in Northern R o m a n i a in the Poiana Botizii region. Thus field studies in this region would supply constraints on the kinematics of the M H L , and consequently on the evolution of the Maramures and Transylvanian Basins. The aim of the paper is to reveal different evolutionary stages of the M H L and to establish qualitative Basin models for O l i g o c e n e Miocene times. In the study area (Fig. 1) different stratigraphic
units were examined including both the crystalline basement and the sedimentary cover, the latter ranging from Mesozoic to Early Miocene in age. The technique of quantitative microtectonics was applied in order to determine stress-field evolution of the area. Nevertheless, available palaeomagnetic, stratigraphic and sedimentological data were integrated in the models. The main structural elements can be evidenced in five complementary outcrops, which are presented in detail. Later, we discuss the palaeo-stress analysis, and finally we present the overall tectonic evolution.
Geological setting The r o u n d Transylvanian Basin is situated within the Intra-Carpathian area, and is encircled by the Eastern and Southern Carpathians, and by the Apuseni Mts (Fig. 1). Its basement is composed of crystalline and Mesozoic rocks, which have been tectonically stacked in thrust nappes during Late Cretaceous compressive phases (Ciupagea et al. 1970). These rocks are also exposed at the periphery of the basin. A smaller range of crystalline exposures (Meses and Preluca Mts)separates the Transylvanian Basin from the adjacent Maramures area (Fig. 2). Palaeogene to Early Miocene deposits are mostly confined to the N W part of the Transylvanian Basin and the Maramures area, while Middle Miocene-Pliocene Basin fill is restricted
GYORFI, I. ETAL..1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian Basins. In: DURAND,B., JOLIVET,L., HORVA_~IH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 251-267~
252
I. GYORFI ETAL.
Fig. 1. Geological outline of the Carpathian area indicating the main geodynamic units of this region. The study area is indicated by frame. Arrows indicate palaeomagnetic declinations measured mostly on Late Cretaceous and Palaeogene rocks (Patrascu et al. 1994, with the corrections of Panaiotu pers. comm. 1996).
to its central and southern part. Palaeogene-Lower Miocene sediments of the Transylvanian Basin and the Maramures area are covered by Middle-Upper Miocene volcanic rocks to east and north. The distribution of the Palaeogene-Lower Miocene sequences in the E - S E is rather uncertain, being covered by Middle Miocene-Pliocene sediments and volcanic rocks (Paraschiv 1979). During Palaeogene-Early Miocene times in the study area two major sedimentary domains evolved synchronously (Fig. 2): (1) the shallowmarine epicontinental Transylvanian Palaeogene Basin (TPB) and ( 2 ) t h e deep-marine Maramures Palaeogene (Intracarpathian Flysch) Basin (MPB) (Bombita 1972). Although these basins were closely related, their history is not identical. Our main concern here is to describe the development of these basins.
Transylvanian Palaeogene Basin (TPB) Remains of the TPB can be best studied in the NW margin of the Transylvanian Basin, between Cluj and Poiana Blenchii (Fig. 2). In this sector the basin evolution can be separated to three distinct episodes (Fig. 3). (i) The Eocene Basin was filled by products of two sedimentary cycles including continental, brackish or evaporitic and shallow marine sequences (Popescu 1984). The Late Eocene facies distribution indicates a NW-SE-oriented basin, because continental and shallow-marine facies are found in the SW, near Cluj and in the NE, near the Preluca Mts. Deep-marine finegrained sediments occupy the central, Meses area. According to Popescu (1984), transgressions were from the NW. (ii) A major palaeogeographic change
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER
253
Fig. 2. Schematic geological map of Northern Transylvania (after the geological map of Romania 1:1 000 000 1985). Abbreviations: MM, Maramures basin; DVF, Dragos-Voda fault; SNB, Simleu Neogene Basin; a, deep marine sediments; b, shallow marine sediments.
occurred in the Early Oligocene (NP22-23 boundary) (Fig. 3). Facies belts were shifted northwards, the continental brackish deposits occurring in the Meses area, marine deposits being found in the Preluca area. Sedimentation was characterized by siliciclastics showing moderate subsidence rate (Rusu 1983). The transgression was from the N-NE, indicating that the polarity of the Oligocene Basin changed with respect to the Eocene one (Rusu 189). (iii) Another palaeogeographic change occurred during Early Miocene times (Eggenburgian, NN 2, Fig. 3). Most of the sediments were deposited on the northern margin of the Transylvanian Basin, transport directions being oriented S-SE. Subsidence rates show extremely high values, especially during the Ottnangian-Karpatian period (NN3-4), when thick (up to 4000 m, Paraschiv 1979) siliciclastics were deposited. No major structure has been described previously in this part of the basin. However two important features need to be mentioned: (1) crystalline rocks and granites are thrusted over Palaeogene deposits in the Meses Mts (Popescu 1984; Petrescu & Margarit 1987) and (2) a major anticline of Palaeogene rocks occurs near the Preluca Mts (Popescu 1984; M6szfiros 1989).
Both outcrops were revisited and described in more detail in this paper. Maramures Palaeogene Basin ( M P B )
Palaeogene-Lower Miocene sequences are largely exposed in the Maramures area, from Poiana Botizii to the Calimani Mts north of the Transylvanian Basin (Fig. 2). The basin fill is generally characterized by Upper Senonian to Lower Miocene deposits, with some significant, but rather short hiatuses (Bombita 1972; Antonescu et al. 1975). Three main tectonic units have been described: (1) the Botiza nappe, (2) the Lapus nappe and (3) the autochthonous respectively (Dicea et al. 1980a, b). In each of these units the Senonian-Eocene deposition is dominated by pelagic, often siliciclastic or turbiditic facies, thus accounting for the same of 'flysch basin'. During the Early Oligocene a thick wildflysch sequence was deposited in the Lapus nappe and Autochthonous unit (Fig. 3) with large olistoliths reworking Late Cretaceous to Late Eocene series (Dicea et al. 1980a). These olistoliths were derived from a palaeo-high (Dicea et al. 1980b), located south of the Maramures area, in the northern margin of the TPB, as evidenced by transport directions
254
I. GYORFI E T A L .
Fig. 3, Schematic stratigraphy of the Northern Transylvanian Basin, (after Rusu 1989, Popescu 1984 and Moisescu 1981).
in the wildflysch (Mihailescu & Panin 1962; Sandulescu & Micu 1987) which indicates a southwestern source area. During Late Oligocene-Early Miocene times the deposition was dominated mainly by thick sandy turbidites (Fig. 3) Palaeogeographic reconstructions (Sandulescu & Micu 1989) indicate that the source area of the Oligocene-Lower Miocene sequences was situated in an inner position with respect to the Eastern Carpathians, current marks showing mainly east-directed transport (Jipa 1962). Several thrust sheets have been described in the Maramures area (Sandulescu 1980; Dicea et al. 1980b): * the Botiza nappe consists of Late Jurassic-
Late Cretaceous, Palaeogene, Early Miocene formations which were correlated with the Pieniny Klippen Belt or Magura unit (see later); this is the uppermost structural unit, thrust over the Lapus nappe; the Lapus nappe consists of Late Cretaceous to Palaeogene and Early Miocene formations; the Botiza and Lapus nappes override the Autochthonous unit; the youngest stratigraphic unit involved in the nappe pile is Early Miocene in' age; thrust fronts are curvilinear and mostly dip toward the N-NW. These thrusts were cut and offset by a later, east trending fault, the Dragos-Voda fault (DVF). Left lateral offset along this line is estimated to reach at least 10 km (Dicea et al. 1980a).
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER
Outcrop descriptions In the following several outcrops of key importance will be discussed. We emphasize the structural observations, since the stratigraphy and partly the sedimentology were dealt with by earlier authors (Popescu 1984; Petrescu & Margarit 1987). The most important outcrops will be discussed in more detail, since field data is of key importance for later geodynamic reconstructions.
Brook north of Hodis village The Hodis outcrop is located at the foothills of the Meses Mts (Fig. 2). Here Late Cretaceous granites are thrust over Eocene deposits, which are made up of soft clays and conglomerates (Fig. 4), and forms low hills in the footwall. The grey, massive Cretaceous granite is passing downwards to brecciated, dark grey cataclastite, deformed under brittle-ductile conditions. This is underlain by green, red clays of Eocene age, in a flat, overturned position. In the cataclastic granite the brecciated elements are mostly elongated (Fig. 4a): they form a stack of smaller duplexes. These elements are separated by a finer matrix composed of dark
255
cataclastic material. Small ramps and the main overthrusts show rare slickenfibres (Fig. 4b). These elements indicate a shortening in a W S W - E N E direction. Several E - W trending normal faults dissecting these overthrusts are indicating at least two different extensional phases postdating the main compressional phase (Fig. 4b). Further downwards in the Palaeogene deposits the dips become steeper, but the overturned attitude of the strata is mostly preserved. A large fold developed in the middle part of the section. The fold is composed of Late Eocene marly limestone. Layer-parallel slips and smaller thrust faults dissecting some layers are indicating shortening in a W S W - E N E direction. Orientation of bedding planes and of the fold axis (Fig. 4c) indicates that the fold was most probably generated synchronously with the thrust emplacement of the Cretaceous magmatics rocks on top of the Eocene sequence. Right-lateral and normal faults cutting through earlier compressional structures were observed immediately adjacent to the fold. They represent two distinct phases: one phase indicates NNE-SSW-oriented compression, the other being characterized by NW-SE-directed extension. The age relationship between them is
Fig. 4. Section of the Pestera valley at Hodis. (a) Detail of the overthrust. (b) Fault data near the overthrust. (c) Bedding plane dips, fold axes and related faults measured near the fold. (d) All fault data from the valley, separated into distinct phases. All plots on lower hemisphere Schmidt projection.
256
I. GY(3RFI ETAL.
uncertain. However, based on a single superposed striation, the youngest phase seems to be the NW-SE-directed extensional phase. In this outcrop the relative chronology of structural events (Fig. 4d) is as follows: (1) E N E - W S W - o r i e n t e d compression creating major overthrusts and folded structures; (2) N N W - S S E compression and perpendicular subhorizontal extension creating smaller thrusts, strike slip faults and normal faults; (3) N W - S E extension generating normal and oblique slip faults (most probably reactivated older structures).
Gorges of Poiana Blenchii This set of outcrops is located near the Preluca Mts (Fig. 2), and makes part of a regional structure, the Glod-Gostila anticline of several kilometres wavelength with a rough N N E - S S W axis (Popescu 1984; M6sz~iros 1989). The core of the anticline is cut by a spectacular gorge in Eocene limestones (Fig. 5), on the flanks of which softer Oligocene clays and sandstones build up a smooth landscape. This anticline develops in the footwall of the Poiana Botizzi overthrust. The structure of the gorge is more complicated than a simple anticline. In fact it is made
up of a number of a decametric-hectometric folds associated with smaller and larger thrust faults (Fig. 5). Individual layers were often cut by small thrust surfaces with small offsets, but major ramps crosscutting several layers in the core of major folds were also observed and measured. The scale of folds did not permit measurement of axes directly, and the complicated ramp pattern inhibits axis construction from bedding data. The fault-slip data are dominated by thrust movements; several strike-slip faults were also measured (Fig. 5a). At the southern entrance of the gorge, rare normal faults display a conjugate pattern. However normal faults are otherwise generally absent. We found two major sets of directions within slickenslide lineations of thrust surfaces (Fig. 5a): one with N E - S W direction and another, with N N W - S S E direction. The first set was proved to be relatively older on five composite fault surfaces. The second phase shows moderate directional spread of transport lineations (dispersion of about 25~ Normal faults are indicating N W - S E extension and are probably the youngest structures. Similar structural phenomena were observed in a nearby valley cutting through Oligocene formations. Here black shales of Early Oligocene
Fig. 5. Section of the Poiana Blenchii Gorge. (a) All faults measured in the gorge, separated into distinct phases. (b) Outcrop of Early Oligocene Ileanda shales. (c) Fold axes and related faults. All plots on lower hemisphere Schmidt projection.
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER age are cut by flat thrust surfaces (Fig. 5b). Rare slickenslide lineations show a southward transport direction. Smaller folds are associated to these thrust faults, their axes are oriented E-W. Both ductile and brittle structures indicate N-S shortening (Fig. 5c). Relative chronology of the observed structures is as follows: (1) N E - S W shortening generating smaller ramps within layers and strike-slip faults cross-cutting them; (2) N-S to N N W - S S E shortening inducing major and minor folds including the big anticlinorium, thrust surfaces and eventually strike-slip faults; (3) N W - S E extension either creating or reactivating earlier slip surfaces. Poiana Botizii
The third important outcrop area is that of Poiana Botizii, in the Maramures region (Fig. 2). Here Jurassic-Early Cretaceous, mostly calcareous deposits form thin tectonic slivers between repeatedly exposed red-grey Senonian marls, calcareous marls and dark Eocene turbidites
257
(Antonescu & Popescu 1975). The uppermost imbricated structure (Botiza nappe) is thrust upon Palaeogene turbidites (Lapus nappe), then on less-deformed Palaeogene and Lower Miocene deposits of the Maramures Basin ('Autochtonous') (cf. Dicea et aL 1980a, b). The stratigraphic similarity between the Botiza Klippen and the Pieniny Klippen belt (PKB) has been noticed by several earlier authors (i.e. Bombita 1972; Antonescu et al. 1977; Birkenmajer 1977; Sandulescu 1980; Dicea et al. 1980a, b; Sandulescu et al. 1981; etc.). A synthetic section (Fig. 6) is presented here based mainly on the outcrops of the Valea Varastina. In the whole section dips are medium to steep, usually to the NW, but in some places, mostly in the southern part of the section, southerly dips also occur. Three parts of the section are particularly important. In the brook of the main valley a fold develops in Eocene turbidites (Fig. 6a). A small thrust fault is at the origin of the internally tight, t h e n open fold. Smaller ramps restricted to single or few layers are also
Fig. 6. Cross-section of the Poiana Botizii region. Stratigraphy after Bombita (1976), Dicea et al. (1980a). (a) Fold in Eocene turbidites in the main valley. (b) Bedding planes, fold axes and related sets of faults. The first fault set is synchronous with the fold, the second representing a later phase. (c) Boudin necks, mullionlike rods and bedding poles in Jurassic allodapic limestones and cherts. The roughly E-W semi-ductile elongation is thought to be synchronous with the first phase of deformation. (d) Superposed slickenslide lineations in Jurassic marl. (e) All fault data from the valley, separated into phases. All plots on lower hemisphere Schmidt projection.
258
I. GYORFI E T A L .
observed on the fold limb. Striations on these faults are perpendicular to the fold axis (Fig. 6b). Both the axis and the related striations are at present oblique: the fold and the associated thrust planes seem to be tilted by a later deformation. A set of reverse and strike-slip faults without slickenslides (Fig. 6b), but with pronounced offsets was also measured at the same site. These structures delineate a compression with subhorizontal o-1 oriented NNW-SSE. We argue that the whole sequence was tilted toward N - N W during the second compressional phase. The a m o u n t and direction of tilting roughly corresponds to the general dip of the strata (320/40). Tilting back earlier structures (fold axis, related slickenslide lineations), NNE-SSW-directed s h o r t e n i n g can be deduced for the first phase of deformation. In the Valea Varastina a larger exposure of Jurassic nodular limestones, Aptychus marls, is found at the waterworks. Red Senonian marl is covering this series. In both rocks numerous generations of slickenslide lineations were observed and measured (Fig. 6d). Two main thrust directions were observed on several fault surfaces, thus two shortening directions can be deduced from these measurements. In four cases the superposition criteria showed that the N E - S W - t r a n s p o r t i n g thrusts are older, the NNW-SSE-transporting ones are younger. In one case we found opposite relative ages. In the same Jurassic limestones and marls numerous strike-slip faults were found. The measured faults can be assigned to two conjugate sets, with a N E - S W and NNW-SSE shortening direction. The first seems to be the older. This fits the shortening directions deduced from the thrust faults proving the activity of the strike-slip faults during compression. Dispersion of the fault slip data can be observed during the second deformational phase. In a little quarry immediately above, the exposed Jurassic sequence is made of red-black radiolarites, often sandy, cherty, allodapic limestones and calcareous olistostromes. The sequence has a moderate, uniform dip. Variations in the dip and thickness of strata are observed along strike. This is due to a boudinage of the layers, which are often truncated, disrupted. Boudin necks, rods (very similar to mullions), related to incipient boudinage were measured in this outcrop. All elements were oriented N-S to N N E - S S W (Fig. 6c) indicating E - W extension. The observed features show that stretching occurred probably during the second compressional phase. The relative chronology of the observed structures in the Poiana Botizii area is as follows: (1)
N E - S W shortening generating smaller folds and thrust surfaces; these structures are generally tilted. ( 2 ) N N W - S S E to N-S shortening, creating semi-ductile extension, smaller thrusts and strike-slip faults. This latter event may be held responsible for the generation of the map-scale nappes in the region. ( 3 ) N E - S W shortening creating normal and strike slip faults. Structures and the generating stress fields of the first and third phases are very similar, so we cannot clearly separate them. However, superposition criteria suggest that one deformation event with N E - S W shortening directions preceded the N N W - S S E shortening. Because of potential confusion of structures no separate diagram was given for the third phase (Fig. 6e).
C h e n d r e a road cut This outcrop is located in the northwestern part of the Transylvanian Basin (Fig. 2 ) n e a r the village of Chendrea, where the Early Miocene (NN3-4) Hida sandstone is exposed for several hundred metres along a major road cut. Here the Hida sandstone is built up of conglomerates and sandstones. This outcrop, in the vicinity of d e f o r m e d P a l a e o g e n e - E a r l y Miocene formations shows a uniform, shallow dip to the SE. Traces of ductile or brittle deformation are absent: not a minor fault was found. Several other outcrops of the same formation show practically no trace of deformation apart from minor normal faulting or landslide. This indicates, that most of the deformation described so far was taking place prior to, or synchronously with the deposition of this formation.
I z a valley This outcrop is located on the northwestern b o r d e r of the R o d n a Mts (Fig. 2). H e r e massive, coarse grained Borsa turbiditic sandstones are exposed in a roadcut. Although the outcrop is relatively small, several high angle faults were identified. These faults are oriented E-W, and the striation on faults indicates a leftlateral motion. Since the age of the Borsa sandstone, based on biostratigraphic data, is Late Oligocene-Early Miocene (Dicea et al. 1980a), the left lateral displacement is younger. In the close vicinity of this outcrop Late Eocene rocks with strike slip faults are exposed. All these outcrops indicate that E N E - W S W shortening occurred synchronously with N N W - S S E extension and E - W oriented left lateral strike slip faulting (see later Fig. 9). This event could account for the map-scale offsets along the Dragos Voda fault.
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER
259
Fig. 7. Faults related to the first deformational phase (Early Oligocene). Ages given below the stereograms indicate the age of outcrop. Numbers refer to the localities given by the same number on the map. All plots on lower hemisphere Schmidt projection. Black arrows and marks indicate the principal stress directions computed Marks: o-1, star; o-2, diamond; o-3, triangle. Abbreviations: Gm, Gilau Mountains; AM, Apuseni Mountains; ME, Meses Mountains; SNB, Simleu Neogene Basin; PM, Preluca Mountains; RM, Rodna Mountains; EC, East Carpathians; DVF, Dragos Voda fault; LN, Lapus Nappe; BN, Botiza Nappe.
260
I. GYORFI E T A L .
Structural analysis In this section all the m e a s u r e d data are assessed. Measurements include fault attitude, slickenslide orientation, sense of slip on the fault, and where observed, chronology of superposed structures. Data were processed iteratively in several steps (manually and by computer program) in order to separate different phases. The coherence of a fault set included in one phase was examined by the 'right-dihedra m e t h o d ' of A n g e l i e r & Mechler (1977). In
several cases manual methods were preferred to provide better control on the computation. All outcrops show several episodes of compressional deformation, with different shortening directions and thus with different principal stress orientations (Figs 7, 8, 9). These directions remain constant in the study area, and the relative chronology of events, when visible, was also similar. Therefore we assume, that the area was tectonized during several phases. In other words, the recorded structural data indicate different, large-scale structural events.
Fig. 8. Faults related to the second deformational phase (Early Miocene). Ages given below the stereograms indicate the age of outcrop. Numbers refer to the localities given by the same number on the map. All plots on lower hemisphere Schmidt projection. Black arrows and marks indicate the principal stress directions computed by the reverse Angelier method (19??). Marks: o-1, star; 0-2, diamond; 0-3, triangle. Abbreviations: Gm, Gilau Mountains; AM, Apuseni Mountains; ME, Meses Mountains; SNB, Simleu Neogene Basin; PM, Preluca Mountains; RM, Rodna Mountains; EC, East Carpathians; DVF, Dragos Voda fault; LN, Lapus Nappe; BN, Botiza Nappe.
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER
261
Fig. 9. Faults related to the third deformational phase (Late Miocene-Pliocene). Ages given below the stereograms indicate the age of outcrop. Numbers refer to the localities given by the same number on the map. All plots on lower hemisphere Schmidt projection. Black arrows and marks indicate the principal stress directions computed by the reverse Angelier method (Angelier & Mechler 1977). Marks: o.1, star; o-2, diamond; o-3, triangle. Abbreviations: Gin, Gilau Mountains; AM, Apuseni Mountains; ME, Meses Mountains; SNB, Siinleu Neogene Basin; PM, Preluca Mountains; RM, Rodna Mountains; EC, East Carpathians; DVF, Dragos Voda fault; LN, Lapus Nappe; BN, Botiza Nappe.
According to the age relationships of microtectonic data and to the stress-tensor computations the first major event characterized a compressive stress-field, with crl oriented E N E - W S W to N E - S W (Fig. 7). Structures related to this deformation were found in Mesozoic rocks of the Poiana Botizii area and in Early to Late Eocene rocks throughout the area. It was later followed by deformation caused by another compressional stress-field of N W - S E to N N W - S S E o-1 direction (Fig. 8). The structures related to this stress-field were recorded in Mesozoic, Eocene and Oligocene rocks. The Early Miocene Borsa sandstones at Fiad were also affected by this deformation. The final recorded d e f o r m a t i o n can be assigned to a strike-slip type stress field, with E N E - W S W 0-1 and N N W - S S E 0-3 directions (Fig. 9). This latter event seems to be restricted to the Maramures
area, along the Dragos Voda fault (Antonescu & Popescu 1979; Antonescu et al. 1980). Related faults occur in crystalline rocks and Eocene, Oligocene to E a r l y Miocene formations. A fourth d e f o r m a t i o n a l event was also found, which was characterized by N E - S W extension. This phase is recorded only locally and seems to postdate the compressions. However no relative age constraints are available with respect to the third described deformation. Several levels of i n h o m o g e n e i t y were observed in the fault sets. Deformation assigned to the first phase proved to be r e m a r k a b l y homogeneous within subareas (Fig. 7). Outcrops in the western border of the Transylvanian Basin (stations 1, 2, 3, 4) show no substantial deviation in the computed stress tensor axes. The stress tensors derived from the northern margin of the TPB and Maramures region (stations 5-11) are
262
I. GYORFI E T A L .
also homogeneous. However, the main stress directions of these two sub-areas deviate considerably with respect to each other. We speculate these deviations result from the second compressional phase since northern outcrops of TPB were strongly affected by this phase in the vicinity of the Poiana Botizii overthrusts. These later deformations could well include rotations, which may have changed the original fault orientations and thus the principal stress directions. The second phase of deformation shows a remarkable areal homogeneity (Fig. 8). Outcrops from all previous sub-areas show the same principal computed stress directions. On the other hand, there is a relative inhomogeneity within the same site: in the Poiana Blenchii and Poiana Botizii sites (stations 5, 8) the average shortening is indeed NNW-SSE, but this shortening is composed of sets of N-S, SSW and SSE verging thrusts and related ductile structures. Two hypotheses may be presented to account for these changes. (1) In the first hypothesis a rotating external stress field was assumed to interact with a non-rotating block. In this situation the stress-field should start from a N-S compressional direction and evolve to a NW-SE compressional direction, since the Early Miocene Fiad site (station 12) records only the NNW-SSE compressional structures. More ductile structures are characteristic for N-S shortening direction (Fig. 6). The Fiad outcrop could in fact record the last increments of this deformation. If the Tisza-Dacia block was fixed, then the stressfield should have rotated counterclockwise, from a N-S direction to a NW-SE direction. (2) In the second hypothesis, stable stressfield and a rotating Tisza-Dacia block are assumed. The same arguments as those listed above suggest the presently N-S compressional structures predate the NW-SE compressional ones. This temporal relationship may be accounted for by clockwise rotation of the major block under a stable NW-SE compressional stress-field. Since palaeomagnetic measurements indicate a clockwise rotation of the Tisza-Dacia unit we consider the second hypothesis as more realistic, than the first one. These deformations cannot be precisely dated, because we investigated mainly E o c e n e - E a r l y Miocene formations. Accordingly, we suppose that the Early Oligocene constitutes the lower age limit for the recorded deformations. Since the Hida beds of Early Miocene age are not deformed, the main compressive deformations took place prior to its deposition as mentioned by Dicea et aL (1980a). However considering the different shortening episodes (phase 1 and 2) and the polyphase
structures (e.g. the tilted fold) not all the shortening and thrusting should have taken place during the Early Miocene. Most likely the observed deformation events spread over a much wider time interval, from the Early Oligocene until the Early Miocene. Overthrusts of Early Miocene rocks in the Poiana Botizii region seem to date only the last shortening episodes. Another constraint on the timing of tectonic events can be derived from the analysis of sediments in adjacent basins. The drastic Early Oligocene palaeogeographic change in the TPB, and the synchronous appearance of wildflysch in the MPB (Fig. 3) both indicate, a major change in the structural regime and relate to active deformations. On the other hand, the deposition of the Hida beds also follows a marked change in the sedimentary environment, with coeval compressional tectonic activity in the nearby Poiana Botizii area. This basin was rapidly subsiding, but otherwise not deformed. Thus the final overthrusts were probably synchronous with the deposition of the Hida beds. The E-W running Dragos Voda fault cuts across older, Early Miocene thrust faults in the Maramures Basin (Fig. 2). North of Poiana Botizii Badenian and Sarmatian (Middle Miocene) rocks are also truncated by this major fault (Antonescu et al. 1981). Although Romanian authors already assigned a Mid-Late Miocene age to this deformation we think its dating can still be refined. Thus a series of deep pull-apart basins related to left-lateral displacements were evidenced in NE Hungary (Rumpler & Horvfith 1988). Adjacent to the Hungarian border similar strike-slip-related structures can be identified on seismic profiles in the Romanian part of the Pannonian Basin (Dicea et al. 1984). According to the seismic evidence, left lateral master faults of these basins are always running ENE-WSW, and their activity is Early Pannonian (Late Miocene; 10-7 Ma) in age. These master faults extend eastward and connect to the Dragos Voda fault. Therefore we speculate that the Dragos Voda fault system can be traced west as far as the Hungarian border beneath or across the volcanic rocks of the Gutii Mrs, the activity of this left lateral shear zone being restricted to the Late Miocene (Fig. 1). The volcanic rocks dated as 11.5-8.5 Ma old (Edelstein et al. 1992) also seem to be offset in a left lateral way by this fault system.
G e o d y n a m i c interpretation
,
Although the Maramures and the Northern Transylvanian Basins were thoroughly mapped
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER and analysed in detail (Dicea et al. 1980a, b; Sandulescu 1980; Rusu 1983; Popescu 1984), its structural evolution is poorly constrained. The confidentiality of high-quality data (seismics, boreholes) persists, thus no quantitative models are available. Evaluation of surface structural data can offer a qualitative a p p r o a c h for the Tertiary evolution of the area, but the internal architecture of these basins, the amount of subsidence and shortening remains still unknown. Before discussing the geodynamic significance of the microtectonic data we shall summarize additional regional constraints provided by other sources. As shown earlier (Fig. 3) the lithofacies of the Botiza nappe and its Mesozoic basement are different from the rocks of the Transylvanian-Dacian unit. Palaeogeographic differences persisted until the mid-Palaeogene, implying that these units are in a closer contact only since the Early Miocene. Because the Botiza Mesozoic series either relate to the Pieniny Klippen belt, or to the Magura nappe (both parts of the Alcapa block), the Botiza nappe must also derive from the same block. Palaeomagnetic measurements (Fig. 1) from the Carpathian area (M~rton 1987; M~rton et al. 1992; Patrascu et al. 1992, 1994) show that the Alcapa and Tisza-Dacia blocks underwent opposed rotations during the Early Miocene. In the Alcapa block the bulk of rotation occurred between Ottnangian and Badenian (NN3-5) times. In the Tisza-Dacia block the exact age of rotations is not known, but predates the Badenian and postdates the Palaeogene, being thus synchronous with the rotations of the Alcapa block. The Tisza-Dacia unit was rotated 60-70 ~ clockwise during this period (estimates of Patrascu et al. 1994), while the Alcapa block underwent 50-60 ~ of anti-clockwise rotation at least (M~rton & Fodor 1995). The deformation events discussed in this paper are predating these major rotations. This is obviously the case for the first compressional event, supposed to be of Early Oligocene age. The second compression of Early Miocene age could have also taken place prior, or partly synchronously with the rotations (see previous chapter). The third, Late Miocene (Pannonian) phase post-dates major rotations, although some minor rotations are still observed in the volcanites of Late Miocene age in NE Hungary and in the Maramures region (Patrascu et aL 1994; Fodor et aL this volume; Mfirton pers. comm.). (1) The first major compressional event occurred during the Early Oligocene and is characterized by a o.1 oriented E N E - W S W and (Fig. 10.1). The o-1 directions for this period are
263
very similar to the o-1 directions of the Palaeogene compressional phase observed in the Southern Carpathians (Ratschbacher et al. 1993). Considering that the rotation of the Tisza-Dacia unit occurred during Early Miocene, the Palaeogene palaeogeography can be restored by rotating back this unit of 60-70 ~ counterclockwise. The restored o-1 directions show roughly N-S compression. This stress-field is remarkably consistent with that obtained in stable Europe for the same period (Bergerat 1985). Even better, it is practically identical to the reconstructed stress-field of the Alcapa block (Fodor 1995). This compressional event uplifted the Apuseni Mts and the Transylvanian Eocene Basin resulting in a northward shift of the facies boundaries and in domination of siliciclastic sediments. During this time the Transylvanian Basin was elongated in a N W - S E direction. A roughly parallel basin was located in the Maramures region and is extending probably southeastward beneath the Late Tertiary volcanic rocks. This basin is more external, but located on top and behind the Crystalline-Mesozoic ranges of the East Carpathians (Sandulescu & Micu 1989). An uplift zone then developed between the two basins separating the near-shore and shallow marine TPB from the deep-water turbiditic (wildflysch) sequences of the MPB. This elevated ridge has been evidenced by geophysical prospecting and drillhole data (Dicea et al. 1980a). Considering the variations of thickness and transport directions of Early Oligocene sediments in the MPB (Sandulescu & Micu 1989), this ridge was probably created by a major N-NE-moving thrust fault, and represented the source area for the wildflysch sequences. A tentative reconstruction of the IntraCarpathian blocks by the Early Oligocene is found on Fig. 10(1). However the relative position of the two blocks remains unclear. They could have been more distant being separated by an unknown palaeographic domain, but because of lack of evidence and to preserve the simplest solution we put them in contact. The N-S-oriented o-1 directions resulted in the development of major foredeeps in front of the two blocks in a subduction-type setting. The Maramures and Transylvanian Palaeogene Basins were located on the upper plate, more internally relative to the Carpathian arc. These Basins (and probably the ridge separating them) are compressional in origin. (2) The second compressional phase (Fig. 10.2) occurred during Early Miocene times with o-1 oriented NNE-SSW. This phase was also recognized in other parts of the Tisza-Dacia unit
264
I. GYORFI E T A L .
Fig. 10. Plate dynamics and location of the studied basins. Palaeogeographic reconstructions and schematic cross sections for three successive periods: 1, Early Oligocene; 2, Late Oligocene-Early Miocene; 3, Late Miocene.
(Ratschbacher et al. 1993; Gy6rfi & Csontos 1994). This phase generated major deformations within the MPB and on the northern margin of the TPB. The completely exotic nature of the Poiana Botizii klippen involved in the uppermost nappe (Botiza nappe) was recognized and interpreted as the frontal thrust of the Alcapa block overriding the Tisza-Dacia unit (Balla 1984; Csontos et al. 1992). Although the Botiza nappe belongs to the Alcapa block, its microtectonic data and history of deformation are very similar to those of nearby exposures of the Tisza-Dacia block. During this second deformation phase the Botiza nappe was thrust over the Tisza-Dacia unit and, although initially part of the Alcapa block, was involved in the rotations of the Tisza-Dacia block. However the present contact of the Botiza nappe with the Alcapa block is uncertain. Towards the NW in the basement of the very deep East Slovakian Neogene Basin (Fig. 1) deep structural units related to the Alcapa block such as Penninic (Sot~ik et al. 1993) or Magura (Csontos & V6r6s in preparation) units were found. This would indicate a considerable thinning and coeval subsidence of the Alcapa block in this sector. A NW-SE-directed extension is suggested also by the map view of the Pieniny Klippen belt and the Outer
Carpathian nappes. This thinning and extension could relate to the conjugate rotations of the Tisza-Dacia block and Botiza nappe (CW 60-70 ~ in one hand, and of the Alcapa unit (CCW 50-60 ~ on the other hand. This would imply a pole of relative rotation for the Tisza-Dacia unit SW of the Poiana Botizii sector. In Late Oligocene-Early Miocene times a more uniform basin developed in the area deepening to the north, and trending E - W to NE-SW. This basin had its depocentre in the Maramures region but is traced by boreholes as far west as the Szolnok Basin. A subsequent Early Miocene Hida Basin was oriented E-W, its depocentre being located further to the south. While the Late Oligocene-Early Miocene turbiditic formations (Borsa sandstone) are folded and imbricated, the Miocene clastics belonging to the Hida Basin are only gently tilted. Overthrusting in the Poiana Botizii region could have generated a foredeep basin, with turbidite deposition in the Maramures-Szolnok area. Advancing thrust fronts and resulting tectonic loads could have resulted in a shift of the basin axis towards the Transylvanian Basin. Thus the Early Miocene Maramures and the slightly younger Hida Basins constitute successive foredeep (or flexural) basins which developed
STRUCTURAL EVOLUTION OF PANNONIAN-TRANSYLVANIAN BASINS BORDER during the overthrusting of Alcapa block over the Tisza-Dacia unit (Fig. 10,2). This model suggests that probably due to larger dextral motion along the Periadriatic line during the Late Oligocene-Early Miocene, the Alcapa block collided with the Tisza-Dacia block (Fig. 10,2). This process could have resulted in the observed stress field with E-W-oriented o.1. The overthrust of the Alcapa block over the Tisza-Dacia unit and related shortening could have induced the rotations observed in these blocks. The slight differences in stress-field orientations between the Maramures area and the Transylvanian Basin, as well as the continuous change of shortening directions during the second deformational phase suggest the onset of rotation was synchronous with the thrusting. (3) During the third phase (Fig. 10,3) o.1 was oriented toward E N E - W S W , whereas palaeomagnetic measurements do not show major rotations. This stress-field is similar to those measured in the Pannonian Basin and adjacent regions (Csontos et al. 1991; F o d o r 1995; Csontos 1995). In other words the change in principal stress directions (o-1) seems to be a real reorientation of the regional stress-field, corresponding to the shift of the deformationfront in the Outer Carpathians. This deformation (Fig. 10,3) resulted in the eastwards shift of the Tisza-Dacia block, along N E - S W and E - W oriented left lateral, and W N W - E S E oriented right lateral faults bordering the northern and southern margins of the block, respectively (Ratschbacher et al. 1993; Linzer 1996). The left-lateral displacement in the northern zone is also recorded by the shape of the East Carpathians, which shows a r e e n t r a n t in the sector immediately east of the Bogdan Voda fault zone (Fig. 2). This eastward escape was possible because the last available space to be consumed was found in the East Carpathians (e.g. Csontos et al. 1992). The general strike-slip-type stress-field was creating shortening in the N E - E Carpathians and extension in the East Slovakian Basin. It was also imposing left lateral m o t i o n along the Dragos Voda fault, which connects with the shear zone extending beneath the Great Hungarian Plain.
Conclusions Based on field data and considering other sources in the literature we propose a new qualitative evolutionary model for the studied area. In the same time we attempt to integrate these results into the geodynamic frame of the Pannonian-Carpathian region.
265
This study shows at least three distinct deformational phases. An upper limit for the timing of the first two phases is given by the late Early Miocene age of the Hida beds, which are not deformed. The third phase occurred during Late Miocene. The first compressional event is probably Early Oligocene in age, and is characterized by o.1 oriented toward ENE-WSW. The resulting Oligocene Basin is flexural in origin, with o.1 orientations being perpendicular to the basin axis. The second phase of compression occurred during Early Miocene times. Coeval shortening accounts for thrust nappes, east trending folds and microstructures indicating o.1 oriented N N W - S S E . Late O l i g o c e n e - E a r l y Miocene sediments were deposited in a rapidly subsiding flexural basins adjacent to the d e f o r m a t i o n fronts. The last overthrusts took place during the deposition of the Hida beds, in O t t n a n g i a n Karpatian times. The Alcapa block was thrust over the Tisza-Dacia block. This deformational phase could have also induced large block rotations within the Alcapa and Tisza-Dacia blocks. Late Miocene left lateral normal faulting along the Dragos Voda fault system can be traced as far west as Hungary. It most probably corresponds to an eastwards escape of the Tisza-Dacia block with respect to its surroundings. However this escape was of limited distance, in the order of 10 km. Special thanks are due to A. Hosu, Z. Szilveszter and F. Wanek, from the Cluj University who shared their experience of the local geology. Helpful criticism and comments of F. Roure (IFP), G. Tari (Amoco), P. Trdmoli~res (ENSPM)are gratefully acknowledged. The study was financially supported by the Integrated Basin Studies (Pannonian Basin) project of the EC with the direction of Frank Horvfith and by a grant of the Hungarian Ministry for Culture and Education (MKM 338/94, and OTKA no. 15976).
References ANGELmR, J. & MECHLER,R 1977. Sur une m6thode graphique de recherche des contraintes principales 6galement utilisable en tectonique et en sdismologie: la m6thode des diddres droits. Bulletin de la SocidtO Gdologique de France, VII, 19, 1309-1318. ANTONESCU,E & POPESCU,A. 1975. Contributions h la connaissance de la g6ologie de la region Dragomiresti-Botiza. D. S. Inst. Geol. Geofiz., 61(5), 37-60 (in Romanian). , MITREA, G., MOROSANU, I. & DUTU, C. 1977.
Asupra pozitiei stratigrafice a stratelor de Valea Carelor (Bazinul Maramures). D. S. Inst. Geol. Geofiz., 63, 31-45.
266 --,
I. GYORFI E T A L .
-& NATASE,A. 1975. Contributions ~ la connaissance de la stratigraphie et de la tectonique de la zone de Poiana Botizei-Fiad (Maramures). D. S. Inst. Geol. Geofiz., 61(5), 3-35. -& PoPEscu, A. I. 1979. Contributions a la connaissance de la stratigraphie et de la tectonique du Mioc6ne de la zone de Vadu IzeiBirsana-Botiza (Maramures). D. S. Inst. GeoL Geofiz., 66(5), 5-20. & -1981. Contributii la cunoasterea stratigrafiei si tectonicii Miocenului din reginuea vadu Izei-Birsana-Botiza (Maramures). [Contributions to the knowledge of the stratigraphy and tectonics of the Miocene from the Vadul izeiBirsan-Botiza Region (Maramures)], D. S. Inst. Geol. Geofiz., 66, 5-23. , POPESCU,T., DUTU, C. & Rosu, V. 1980. Contributions h la connaissance de la stratigraphie et de la tectonique de la r6gion de Singeorz-Bai-Lunca Ilvei. D. S. ale sed., 65(4), 87-110. BALLA, Z. 1982. Development of the Pannonian basin basement through the Cretaceous-Cenozoic collision: a new synthesis. Tectonophysics 88, 61-102. BERGERAT, E 1985. Ddformations cassantes et champs de contrainte tertiaires dans la plate-forme europ~enne. M6moires du Sciences de la Terre, Universit6 P. & M. Curie, Paris, 85--07. B1RKENMAJER, K. 1977. Jurassic and Cretaceous lithostratigraphic units of the Pieniny Klippen Belt, Carpathians, Poland. Studia Geologica Polonica, 45, 1-158. BOMBITA, G. 1972. I~tudes g6ologiques dans les Monts Lapus. An. Inst. GeoL Geofiz., 39, 1-108. CIUPAGEA,D., PAUCA,M. & ICHIM,T. 1970. The geology of the Transylvanian Depressions. Ed. Acad. Bucuresti [in Romanian]. CSONTOS, t . 1995. Tertiary tectonic evolution of the Intra-Carpathian area: a review. In: DOWNES, H. VASELLI,O. (eds) Neogene and related magmatism in the Carpatho-Pannonian region. Acta Vulcanologica, 7/2, 1-13. --, NAGYMAROSY,A., HORVATH, E & KOVAC, M. 1992. Tertiary evolution of the Intra-Carpathian area: a model. Tectonophysics, 208, 221-241. - - , TARI, G., BERGERAT,E & FODOR, L. 1991. Structural evolution of the Carpatho Pannonian area during the Neogene. Tectonophysics, 199, 73-91. DICEA, 0., DUTESCU, P., ANTONESCU, F., MITREA, O., BOTEZ, R., DONOS, I., LUNGU,V. & MOROSANU,I. 1980a. Contributions h la connaissance de la tectonique de la zone transcarpathique de Maramures. D. S. Inst. Geol. Geofiz., 65(5), 35-53. -& -1980b. Contributions /~ la connaissance de la stratigraphie de la zone transcarpathique de Maramures. D. S. Inst. Geol. Geofiz., 65(4), 21-86. , IANAS,V. & LUNGU, V. 1984. Paradigmes structuraux-d6positionelle de la secteur du NO du Basin Pannonique. An. Inst. Geol. Geofiz., 66, 391-400. EDELSTEIN,0., B ERNND,A., KovAcs, M., CRIHAN,M. & PI~CSKAY,Z. 1992. Preliminary data regarding the K-Ar ages of some eruptive rocks from the
Baia-Mare Neogene volcanic rocks. R~vues Roumanien de Gdologie, 36, 45-60. FODOR, L. 1995. From transpression to transtension: Oligocene-Miocene structural evolution of the Vienna basin and the East Alpine-Western Carpathian junction. Tectonophysics, 242, 151-182. - - - , CSONTOS,L., BADA, G., GYORFI, I. & BENKOVICS, L. 1999. Tertiary tectonic evolution of the Pannonian basin system and neighbouring orogens: a new synthesis of palaeostress data. This volume. GYORFI, I. & CSONTOS,L. 1994. Structural evolution of SE Hungary and Neogene basins of the Apuseni Mountains (Romania). Romanian Journal of Tectonics and Regional Geology, 75, 19-20. JIPA, D. 1962. Directii de aport in Gresia de Borsa (Maramures). Com. Acad. R. P. R., 12,1363-1368. LINZER, H. G. 1996. Kinematics of retreating subduction along the Carpathian arc, Romania. Geology, 24, 167-170. MARTON,E. 1987. Paleomagnetism and tectonics in the Mediterranean region. Journal o f Geodynamics, 7, 33-57. --, PAGAC, P. & T(JNYI, I. 1992. Palaeomagnetic investigations on Late Cretaceous-Cenozoic sediments from the NW part of the Pannonian Basin. Geologica Carpathica, 43, 363-369. MESZAROS, N. 1989. Possibilities of hydrocarbon accumulation in the Salauta Valley (CosbucTelciu) and east of Magoaja. In: PETRESCU,I. (ed.) The Oligocene from the Transylvanian basin, Romania. Cluj-Napoca, 000-000. MIHAILES('U,N. & PANIN,N. 1962. Directii de curent in depozitele Eocene-Oligocene din regiunea Telciu-Romuli (Maramures). Com. Acad. R. P. R., 12, 1357-1362. MOlSESCU,V. 1981. Oligocene deposits of Transylvania and their correlation in Paratethys. Rfvues Roumanien de Gdologie, Gdophysique et Gdographi, ser. Geologique, 25, 161-169. PARASCHIV,P. 1979. Romanian oil and gas fields. Institute of Geology and Geophysics. Technical and Economical Studies, A13, Bucharest 382. PATRASCU,S., BLEAHU, M., PANAIOTU,C. & PANAIOTU, C. E. 1992. The palaeomagnetism of Upper Cretaceous magmatic rocks in the Banat area of South Carpathians: tectonic implications. Tectonophysics, 213, 341-352. --, PANAIOTU, C., SECLAMAN, M. & PANAIOTU, C. E. 1994. Timing of rotational motion of the Apuseni Mountains (Romania): palaeomagnetic data from Tertiary magmatic rocks. Tectonophysics, 233, 163-176. PETRES('U, I. & MARGARIT, G. 1987. Possibilit6s de la formation de charbons dans l'l~ocene du nordouest de la Roumanie. In: PETRESCU, I. (ed.) The Eocene from the Transylvanian basin, Romania. Cluj-Napoca, 000-000. POPESCU, B. M. 1984. Lithostratigraphy of cyclic continental to marine Eocene deposits in NW Transylvania, Romania. In: POPESCU, B. M. (ed.) The Transylvanian Paleogene Basin. Geneva, 37-73. RATSCHBACHER, L., LINZER, H.-G., MOSER, E, STRUSIEVICZ, R.-O., BEDELEAN, H., HAR, N. &
S T R U C T U R A L E V O L U T I O N OF P A N N O N I A N - T R A N S Y L V A N I A N BASINS B O R D E R MOGOS, E-A. 1993. Cretaceous to Miocene thrusting and wrenching along the central South Carpathians due to a corner effect during collision and orocline formation. Tectonics, 12, 4. RUMPLER,J. • HoRVATH, E 1988. Some representative seismic reflection lines from the Pannonian basin and their structural interpretation. In: ROYDEN,L. H. & HORVATH,E (eds) The Pannonian Basin - a Study in Basin Evolution. American Association of Petroleum Geologists Memoirs, 45, 153-169. Rvsv, A. 1983. Remarks on Oligocene chrono- and biostratigraphy in Transylvanian (Romania). An. Inst. Geol. Geofiz., 59, 229-238. 1989. Problems of correlation and nomenclature concerning the Oligocene formations in NW Transylvania. In: PETRESCU, I. (ed.) The Oligocene from the Transylvania basin, Romania. ClujNapoca, 112-147. SANDULESCU,M. 1980. Sur certains probl6mes de correlation des Carpathes Orientales Roumaines avec les Carpathes Ukrainiennes. D. S. Inst. Geol. Geofiz., 65(5), 163-180. -& MICE, M. 1989. Oligocene palaeogeography of -
-
267
the East Carpathians. In: PETRESCU, I. (ed.) The Oligocene from the Transylvanian basin, Romania. Cluj-Napoca, 000-000. , KRAUTNER, H. G., BALINTONI, I., RUSSO-SANDULESCU,D. & M~CU,M. 1981. The structure of the East Carpathians (Moldavita-Maramures area). Guide to the excursion/B1, XIInd CarpathoBalkan Geological Congress, Bucharest. Inst. Geol. Geofiz., ???. SOTAK, J., SPISIAK,J. & BIRON, A. 1993. Metamorphic sequences with "Biindnerschiefer" lithology in the pre-Neogene basement of the East Slovakian Basin. Mittelungen Osterreichen Geologisches Gesellschaft, 86, 111-120. TARI, G., BALDI, T. & BALDI-BEKE, M. 1993. Palaeogene retroarc flexural basin beneath the Neogene Pannonian basin: a geodynamic model. Tectonophysics, 226, 433-455. VISARION,M. & VELICIU,S. 1981. Some geological and geophysical characteristics of the Transylvanian Basin. Earth and Evolutionary Sciences, 3-4, 212-217.
Recent tectonic stress and crustal deformation in and around the Pannonian Basin: data and models P. G E R N E R 1, G. B A D A 2, P. DOVI~NYI 1, B. MI~ILLER 3, M. C. O N C E S C U 3, S. C L O E T I N G H 4 & F. HORV/i, T H 1
1Department of Geophysics, EOtvOs University, Budapest, Hungary 2Department of Applied and Environmental Geology, EOtvOs University, Budapest, Hungary ~Geophysical Institute, Karlsruhe University, Karlsruhe, Germany 4Faculty of Earth Sciences, Vrije Universiteit, Amsterdam, The Netherlands Abstract: Recent (active) tectonics of the Pannonian Basin and its surroundings has been investigated using data from over 900 earthquake focal mechanism solutions, 200 borehole breakout analyses, some in-situ stress measurements and by applying finite element modelling technique. We have established a database for indicators of recent stress, and analysed the stress state of the region by the methods of the World Stress Map project. The alignments of the largest horizontal stresses have been mapped and the tectonic regimes were also determined. We present a map of seismoactive faults and seismic energy release combining historical and modern seismicity data and results of local seismotectonic studies. The pattern of earthquake slip vectors and the style of faulting are summarised in order to characterise the active deformations. Our results show that the alignment of the largest horizontal stress exhibits a radial pattern around the Adriatic sea. In the Southern Alps and northwestern Dinarides the largest horizontal stress (SH) is aligned N-S and thrust faulting is dominant. Along the southern Dinarides and the Dalmatian coast thrusting with strike-slip component can be observed. Here the trajectories of SH are aligned NE-SW. E - W aligned SH trajectories and normal faulting are characteristic of the R h o d o p e Massif. Thrust faulting of the Vrancea region seems to be distinct from the compressive regime around the Adriatic sea. In the Pannonian Basin borehole breakout analyses show that the direction of largest horizontal stress is changing from N-S in the western part to N E - S W in the east. Most of focal mechanisms and available hydraulic fracturing measurements indicate strike-slip and thrust faulting inside the basin. The lack of normal faulting mechanisms indicates that the extension of the basin has been terminated and a new compressive stress regime prevails. The crustal deformation of the area is controlled by the counterclockwise rotation of Adria with respect to Europe around a pole at the 45~ latitude and 6-10~ longitudes, which is inferred from satellite geodesy and supported by earthquake slip vectors. This m o v e m e n t can explain the shortening of the Southern Alps, and squeezing eastward the region between the Adriatic sea and the M u r - M u r z line. Rotation of Adria generates thrusts along the Dalmatian coast, and this compressive deformation extends into the land far from the coastline, and leads to squeezing of the Pannonian Basin from the southwest. The seismicity pattern in the Pannonian Basin shows that earthquakes are restricted to the crust and the control by pre-existing (mostly Miocene) fault zones is strongly masked by random activity due to general weakness of the lithosphere. Although earthquakes are of small to m e d i u m magnitude (M < 6), the cummulative energy release is remarkably higher than in the surrounding Carpathian arc. The Vrancea zone is the only exception, where high energy release in the crust and down to 200 km depth is associated with a relict subducted slab. Finite element stress modelling has been performed in order to simulate the observed stress pattern and, hence, to understand the importance of different possible stress sources in and around the Pannonian Basin. The observed radial stress pattern of the region can be well explained by the counterclockwise rotation of the Adriatic microplate as a firstorder stress source. Additional boundary conditions, such as the active deformation at the Vrancea zone and the role of rigid crustal blocks at the Bohemian Massif and the Moesian Platform, can significantly effect the style of deformation and the alignment of the largest horizontal stress. Furthermore, our calculations show that differences in the crustal thickness and the presence of large scale fault zones in the Pannonian region have only local influence on the model results.
GERNER, P., BADA, G., DOVI~NYI,P., MI)LLER, B., ONCESCU, M. C., CLOETINGH, S. & HORVATH, E 1999. Recent tectonic stress and crustal deformation in and around the Pannonian Basin: data and models. In: DURAND, B., JOLIVE% L., HORVATH, E & SERANNE, M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 269-294.
270
P. GERNER E T A L .
In this paper the state of recent stress and the style of active deformation of Pannonian region are investigated using earthquake focal mechanism solutions, well bore breakout determinations and in-situ stress measurements, as well as seismicity data and finite element stress modelling. The location map of the studied can be seen in Fig. 1 and it is defined by geographical coordinates 42-52~ and 12-28~ The Pannonian Basin system was formed by the extensional collapse of the Alpine mountain belt during the Early to Mid-Miocene shortly after the eastward escape of the Eastern Alps (Ratschbacher et al. 1991; Horvfith 1993). Normal and strike-slip faulting dominated during the Mid-Miocene and the synrift extension was compensated by the shortening at the Carpathian belt. Post-rift thermal cooling resulted in a considerable subsidence in a tensional stress regime, although the general subsidence was interrupted by a few compressive
tectonic events during the basin evolution (Horv~th 1995). The Quaternary subsidence history, however, deviates from the general and exponentially decaying subsidence rate of a thermally relaxing basin, as it manifests an accelerated subsidence in the central parts and uplift on the flanks of the basin. Horv~ith & Cloetingh (1996) explained this feature by lithospheric buckling due to compressive far field stress, leading to basin inversion and lithosphere bending. Using geological and geophysical data Horv~th & Cloetingh (1996) concluded that the late stage evolution of the Pannonian Basin can be understood in terms of progressive build-up of a compressive stress regime since the Pliocene. The present state of stress in the area has been summarized first by D6v6nyi & Horv~ith (1990) and Mtiller et al. (1992) recognized that earthquakes inside the basin do not indicate normal faulting, in spite of the extensional origin of the area. Philip (1987)
Fig. 1. Key map showing the main tectonic units of the studied area, the Pannonian region. Inserts show selected areas in Fig. 7B-H. 1, Molasse belt; 2, Flysch belt; 3, internal pre-Tertiary units; 4, Bohemian Massif; 5, Neogene volcanic rocks; 6, Pieniny Klippen Belt. TB, Transsylvanian Basin; TR, Transdanubian Range; VB, Vienna Basin.
RECENT TECTONICS, PANNONIAN BASIN and Reba'f et al. (1992) published interpolated stress maps for the Pannonian area, which indicate a N-S to NE-SW directed maximum horizontal stress, and thrust and strike-slip faulting regime for the Pannonian Basin. In this paper, first we review the data we have derived for the determination of the state of recent stress: earthquake focal mechanism solutions, borehole breakout analyses and in-situ stress measurements. We construct a stress data base according to the World Stress Map regulations and determine the pattern of largest horizontal stress and the characteristics of the stress regimes. Then, we present a new map of seismoactive faults and summarize the characteristics of the focal mechanism solutions. We also determine the pattern of earthquake slip vectors in order to characterize the kinematics of active deformation of the studied area. Finally, the results of finite element stress modelling is presented with the aim of understanding the importance of different stress sources which may effect the Pannonian region.
State of recent stress in and around the Pannonian Basin On the basis of a few microtectoflic studies in the Pannonian Basin, Bergerat et al. (1983) suggested that the maximum and minimum principal stresses were aligned to north-south and east-west during the Miocene, respectively. This pattern has been refined by Csontos et al. (1991) and Fodor (1995) using more fault sliP data. They inferred a rotation of the largest horizontal stress from NW-SE in the Early Miocene to NE-SW in the Late Miocene. An up-to-date review of all the accessible data and their interpretation is given by Fodor et al. (this volume). The recent tectonic stress of the region was first investigated by earthquake focal mechanism solutions in the vicinity of the Alps and in the Balkan peninsula (Scheidegger 1967; Ritsema 1974). Grtinthal & Stromeyer (1986) compiled many recent stress data including in situ stress measurements, and provided some possible explanation for the orientation of the largest horizontal stress by finite element modelling. In spite of these efforts, the state of stress inside the Pannonian Basin remained poorly known because of the little number of stress indicators inside the basin. A breakthrough has occurred since the elaboration of borehole breakout analysis (Bell & Gough 1979), and the study of crustal stresses in the Pannonian Basin by this method has been performed since the late eighties. D6v6nyi &
271
Horv~ith (1990) published the first results of insitu overcoring stress measurements and well bore breakout analyses carried out at various sites in Hungary. Compiling the European stress data base Mtiller et al. (1992) investigated the state of the recent stress using mostly earthquake focal mechanism solutions and well bore breakouts. RebaY et al. (1992) analysed the modern stresses of the Alpine-Mediterranean region, including the Pannonian Basin. A detailed presentation of the Hungarian overcoring stress measurements and its tentative interpretation has been given by Becker (1993). In this paper we use the stress data of the European part of the World Stress Map project (Mtiller et al. 1992) for overcoring stress measurements and also breakout data outside of Hungary, in addition to our new determinations. Because the methods of stress determinations are of different type and value, a quality ranking system was developed by Zoback & Zoback (1989,1991) to assess reliability and permit comparison of different data. The ranking criteria includes accuracy and number of measurements, the length of depth interval and the ability of methods to record fully or incompletely the stress tensor. Five classes are used in ranking from A to E, where A stands for the most reliable information and E for useless data (Zoback 1992). In this paper we follow this quality ranking system, but it is slightly modified in a few cases. In our ranking system the number of first motion polarity are also considered in earthquake focal mechanisms, for example a B quality requires 50 or more first motion polarity data, and D is used for less than 25, in addition to the criteria of Zoback & Zoback (1991). If the quality ranking coming from the number of first motions yielded different value than that of the World Stress Map, always the lower quality was accepted. The modification in our breakout quality ranking is that we do not apply the technique of Mardia (1972) for the evaluation of the mean breakout direction, and we used a sixgrade scale instead of five (Table 1). Earthquake focal mechanism solutions
The data set of focal mechanisms used in this study, which is available from the Society Library and the British Library Document Supply Centre, Boston Spa, Wetherby, West Yorks LS23 6BQ, UK as Supplementary Publication No. SUP 18129 (37pp) consists of published focal mechanism solutions collected from literature and new solutions calculated from the P wave first motion polarities. Centroid-moment tensor (CMT) solutions are taken from
272
P. GERNER ETAL.
Table 1. Relation of quality ranking system of stress data according to Zoback & Zoback (1991) and of this paper
FM BO
A
B
C
D
5
n > 50 4
50 > n > 25 3 and 2
n > 25 1 and 0
FM, focal mechanism, n is the number of first motions observed for an earthquake. BO, borehole breakout, Numbers from 0 to 5 relate to our classification of borehole breakouts. Dziewonski et al. (1992) and m o m e n t - t e n s o r inversions from Sipkin & N e e d h a m (1992). The solutions found in the literature are very different in quality and style; we applied a checking procedure for the orthogonality of the nodal planes as well as the P, B and T axes. Where some of the fault plane parameters were missing in the primary source, we computed them from the original data. The values of strike, dip and rake of nodal planes are standardized after the convention of Aki & Richards (1980). We have arrived at 58 new fault-plane solutions for events from January 1980 to D e c e m b e r 1993. The computation was based on the first arrival collection of the International Seismological Centre Bulletins (from January 1980 to August 1987), the National Earthquake Information Centre (NEIC) Earthquake Determination Reports (from May 1991 to December 1993) and the Istituto Nationale di Geofisica (ING) Databank (from January 1990 to December 1993). A l t o g e t h e r 207 earthquakes were available having 20 or more first motion polarities, and only those were used to calculate fault plane solution. The take-off angles were determined using the IASP91 Earth-model (Kenett & Engdahl 1991) for velocity-depth function and the TAU p r o g r a m package of Bulland & Chapman (1983) for the computation of slowness values. The fault-plane solutions were computed by the FPFlT program of Reasenberg & O p p e n h e i m e r (1985). After the computation, the quality of the solutions were checked, and only those were accepted, of which the P and T axes and the fault planes are well constrained, there was no multiply solutions, the inconsistent polarities were less then 30% and the arrivals
are well distributed on the focal sphere. This procedure resulted in 58 new fault plane solutions, mainly south of the Pannonian Basin. Altogether 914 focal mechanism of 379 earthquakes are included into the catalogue but we used only 362 focal mechanisms with consistent focal parameters. The directions of principal stress axes were considered to be the P, B and T axes of focal mechanism solutions. B o r e h o l e b r e a k o u t analysis We selected over 140 hydrocarbon exploration wells having adequate logs for breakout analysis in Hungary. The elongations in wellbore diameters in two orthogonal directions (caliper log) were measured by four-arm dipmeter during routine diplogging surveys. The breakout analysis was based on these data and carried out by using our own computer program. This program displays the two components of the caliper logs and computes the breakout direction taking the criteria of Plumb & Hickman (1985). This procedure resulted in a total of 72 directions of the maximum horizontal stress (SH), from which 60 were determined in individual wells and 12 were average of nearby wells. In several cases the calculation was impossible due to the lack of definite breakouts or ambiguity in selection of the appropriate direction from two characteristic directions. Namely, two perpendicular directions of borehole diameter elongation in one well can occur (site 4 in Fig. 2). One is indicating the direction of the least horizontal stress we are interested in. The other is generated by natural hydrofracturing during drilling activity and mud circulation, and according to theory, aligns normal to the least horizontal stress (Moos & Zoback 1990). In this case, it is difficult to decide in a single well which of the borehole diameter elongation corresponds to the breakout direction. If more wells are available in a relatively small area, it is easier to make a distinction or occasionally, the shape of the elongation is so markedly different that the choice is obvious. In the well at site 4 in Fig. 2. the hydrofrac and breakout directions can be distinguished. It can be seen that breakouts are quite wide and flat, while the hydrofracs situated 90 ~ apart are longer and narrower.
Fig. 2. Some representative plots of breakout analyses in the Hungarian part of the Pannonian Basin. The circle on each figures show the horizontal section of well, the dots around it display the diameters recorded in every 0.2 meter during the logging tool was pulled upwards. The breakout zones correspond to those places around the wall of the borehole, where the dots are grouped into two clusters 180 degrees apart. The azimuth of these clusters shows the direction of minimal horizontal stress, the largest stress in the horizontal plane is 90 degrees apart. In each plot the name of the well (site), the number of diameter records (N), the azimuth of the largest horizontal stress (a) and the depth intervals of data records (interval selection) are shown.
RECENT TECTONICS, PANNONIAN BASIN
273
274
P. GERNER ETAL.
Borehole televiewer data are available from one hydrocarbon exploration well in southeastern Hungary. The data was recorded over an interval of 3351 to 3594 m in metamorphic rocks. The breakouts identified indicate that the predominant direction of the smallest horizontal stress is aligned N W - S E . Some representative borehole breakout diagrams are shown in Fig. 2. for the Hungarian part of the Pannonian Basin. These plots were selected to characterise various depth intervals, regions and style of breakout. The figures show dots around a circle (tool size) recorded by the arms of the logging device, and the average breakout direction can be clearly determined if the dots are grouped in a characteristic direction. For example, the Bagam6r-1 well exhibits a large and typical breakout diagram. The alignment of the largest horizontal stress is the small axis of an ellipse best fitting all the data. Five small depth intervals were selected in Baj~nsenye-3 well, and we d e t e r m i n e d the average orientation. In Baktal6rfinth~za-1 well, one narrow and small breakout zone can be found, which indicates E - W oriented largest horizontal stress. The plot of caliper data of well Csan~dalberti-E-2 show two perpendicular directions of borehole diameter elongation. One (135-165 ~ and 315-345 ~) is indicating the direction of least horizontal stress (Sh), while the other (70-250 ~) is generated by natural hydroffacturing. Kiskunhalas-D-I/4 well exhibits a wide and shallow breakout pattern in the depth interval of 2890-3258 m. In K6rOslad~ny4 well a typical breakout zone in N-S direction can be seen. In Ruzsa-23 well some very wide and only few meter long breakout zones can be found. V6gegyh~za-Ny-1 well shows a well recognisable breakout pattern.
In-situ stress m e a s u r e m e n t s The in-situ stress measurements in the Pannonian Basin have been based on overcoring in 5-8 m drill holes and registering the strain release of the overcored section by doorstopper and triaxial strain cell (Becker 1993). To avoid gravitationally or thermally induced stresses, a careful selection of the test sites was carried out. Altogether six places were found suitable in Mesozoic limestones and dolomites for overcoring measurements. The stress determination at a site resulted from the average of several m e a s u r e m e n t s of strain combined with the elastic parameters of the rock (Young's modulus and Poisson ratio). This m e t h o d allows the calculation of the stress magnitudes as well, however it is always affected by the inaccuracy
of the m e a s u r e m e n t of elastic parameters. Because we applied the overcoring technique in very shallow holes only, the quality of these data is poor. Two hydraulic fracturing have been carried out in southeastern H u n g a r y to increase the yield of two deep h y d r o c a r b o n reservoirs. These offer first direct deep data on horizontal stress magnitudes of the P a n n o n i a n Basin. A t depth of 3274-3293 m and 3319-3349 m the closure pressure, which is equal to and counteracts the least stress in the rocks (Nolte 1982), was 66.0 and 65.4 MPa, respectively. Unfortunately, the largest horizontal stress can not be determined since the tubing in the well enables only the m e a s u r e m e n t of Sh. These values of the smallest horizontal stress are just slightly below the weight of the total o v e r b u r d e n , which can be supposed to be about 75 MPa using 2300 kg m -3 average density down to the depth of 3300 m.
I n f e r r e d stress directions a n d regimes A total of 579 stress data have been collected, in which 63% earthquake focal mechanisms, 32% borehole breakouts, 4% overcoring and 1% hydrofracs can be found. Based on these data we determined the directions of the largest horizontal stress (Fig. 3), and estimated the magnitude of the horizontal stresses relative to the vertical one. We used the simple assumption that one principal stress direction is vertical and hence, the other two are in the horizontal plane. The borehole breakout analysis and the overcoring technique give directly the direction of the largest horizontal stress, but an interpretation is needed to determine the SH direction from focal mechanisms. In this case we applied the method of World Stress Map project, which takes the azimuth of P axis as the largest horizontal stress if P is close to the horizontal, otherwise the B axis is usually considered (Zoback 1992). The algorithm of Hansen & Mount (1990) has been used to derive smoothed pattern of the m a x i m u m horizontal stress from individual points of stress determinations (Fig. 4). We pay attention that on those regions, where there are no any stress data, this smoothed pattern shows only interpolated data. The alignment of the largest horizontal stress trajectories shows a radial pattern of SH directions around the Adriatic sea. For example, N N W - S S E and N-S aligned largest horizontal stresses are typical in the Southern Alps. Northwards, it changes to N W - S E , which is the dominant stress direction in Western Europe (Mtiller et al. 1992). In the western Pannonian Basin N-S
RECENT TECTONICS, PANNONIAN BASIN
275
Fig. 3. Direction of the largest horizontal stress in and around the Pannonian basin. The calculation of stress direction was carried out according to the convention of Zoback (1992). We note that only data of A-D quality are shown. The insert in the upper right corner represents the four stress regimes of the region. A, Western European regime. B, Pannonian basin regime. C, Alpine-Dinarides regime. D, Balkan regime. 1, intra-montain basins; 2, Neogene volcanic rocks; 3, Molasse belt; 4, Flysch belt; 5, Internal pre-Tertiary units; 6, Vardar-zone; 7, Rhodope Massif; 8, Tauern window; 9, FMS stress data - normal faulting; 10, FMS stress data strike-slip faulting; 11, FMS stress data - thrust faulting; 12, stress directions based on borehole breakout data; 13, stress directions based on overcoring data; 14, stress directions based on hydrofracts. -
and N N E - S S W SH directions can be found. The central end eastern part of the basin is characterized by E N E - W S W and E - W SH directions, indicating a remarkable deviation from the West European trend. It is obvious that considerable disturbances can be observed in the northwestern and the central part of the Pannonian Basin. North to it the data indicating N-S-aligned SH. Along the Dinarides, the alignment of the largest stress is N N E - S S W and NE-SW. The southern border of the studied area is characterized by E N E - W S W and E - W directed maximal stress, which tendency is typical along the Southern and Eastern Carpathians. Unfortunately, there are only two reliable measurements of stress magnitude in the Pannonian Basin from deep hydraulic fracturing in Southeastern Hungary. The stress magnitude
data resulted from shallow overcoring measurements are uncertain. Therefore, we used indirect data from focal mechanisms to infer the relative stress magnitudes. Z o b a c k (1992) suggested using the plunges of P, B and T axes to infer the relative stress magnitudes in the horizontal and vertical directions from focal mechanism solutions. The stress regime categories are based on Anderson's (1951) classification: normal faulting, thrust faulting and strike-slip faulting regimes. Additionally, two transitional regimes are defined and there is a sixth category for unknown regime if the data do not yield clear information about the stress regime (Zoback 1992). This m e t h o d gives an opportunity 'to compare the vertical, the maximum and the minimal horizontal stress magnitudes without the knowledge of absolute values.
276
P. GERNER E T A L .
Fig. 4. Smoothed direction of the largest horizontal stress as derived from single stress data shown in Fig. 3 using the algorithm of Hansen & Mount (1990). Note that this map reflects the general pattern of the horizontal stress.
Inside the Pannonian Basin almost all of the focal mechanisms indicate strike-slip and thrust faulting, and the lack of the normal faulting events is characteristic (Fig. 3). This implies that the determined SH directions could well correspond to the direction of maximum principal stress. In central Italy, most of the earthquakes suggest normal faulting regime, usually with strike-slip component. Thrust faulting can be found mainly in northern Italy. In the Adriatic sea there are also normal faulting events along the Italian coast, but the horizontal P axis is characteristic along the coast of the Dinarides, and the mechanisms are both thrusting and strike-slip faulting. In the Southern Alps and the Dinarides thrust and strike-slip faulting is dominant. Normal faulting events appear in the land from Albania to Bulgaria indicating a continuation of the Aegean extensional stress regime.
Other indirect information about stress magnitudes can be obtained from values of overpressure in the Pannonian Basin. Overpressure appear frequently in Neogene shales and also fractured crystalline basement rocks deeper than 2.5 to 3 km (Horv~ith 1995). For example, it follows from the fact that high overpressure values are maintained in the basin that the minimum principal stress must exceed this overpressure, which can be used as lower bound of SH. Similarly, considering the general law for frictional slip on faults, we can estimate the upper bound on maximum stress value supposing thrust faulting regime (see Engelder 1993). The measured minimum stresses (64.5 and 66.0 MPa) yield 204.0 and 205.9 MPa largest stresses, which are highly above the vertical stress and can be taken as the upper limits for SH in faulting regimes (Fig. 5).
RECENT TECTONICS, PANNONIAN BASIN
Fig. 5. Estimates of stress magnitudes in the SE part of the Pannonian Basin. The minimal stresses are measured by hydraulic fracturing technique. The vertical stress is the weight of the overburden taking an average density of 2200 kg m-3 to 2400 kg m-3. The maximum stress is estimated from the relation of general law for frictional slip on faults (Engelder 1993).
277
Pannonian Basin. In spite of that, the seismotectonic models account for some gross features of the seismicity, but the main drawback has been that internal deformation of blocks was usually ignored. As a consequence of the more intense seismicity to the south and east of the Pannonian Basin, most of the focal mechanism studies concentrated on those areas (Constantinescu et al. 1966; McKenzie 1972; Ritsema 1974; Cagnetti et al. 1976; Gasparini et al. 1985; Anderson & Jackson 1987; Oncescu 1987; Eva et al. 1988; Slejko et al. 1989; Shanov et aL 1992; Muco 1994). Only few solutions were published for the Pannonian Basin (Gangl 1975; T6th et al. 1989) and for the Carpathians to the north of the basin (Gibowitz et aL 1982, 1989; Grassl et al. 1984; Grtinthal et al. 1990). After the pioneering work of Ritsema (1974), only a few solutions have been published and a complete catalogue of focal mechanism solutions is still missing. Recently, a large data base of stress indicators has been collected for Europe in the frame of the World Stress Map Project (Mt~ller et al. 1992). This data set comprises many focal mechanism solutions, and our data base can be considered as a completion of this catalogue. S e i s m i c i t y o f the P a n n o n i a n B a s i n
S e i s m o t e c t o n i c s a n d crustal d e f o r m a t i o n in a n d a r o u n d the P a n n o n i a n B a s i n
High quality and reliable seismotectonic models were published for the Alps and Northern Dinarides, where the quantity and quality of seismic stations enabled precise determinations of hypocentral parameters, and the high seismicity allowed clear definition of seismoactive structures (Anderson & Jackson 1987; Slejko et al. 1989; Carulli et al. 1990; Favalli et al. 1992; Console et al. 1993). The seismotectonics of the Pannonian Basin were explained, however, by more simple models by delineating seismically active zones and a few rigid blocks (Horwith 1984; Schenk et al. 1986; Gutdeutsch & Aric 1988). The delineation of seismoactive zones was rather subjective because of the dispersed seismicity inside the Pannonian Basin. The scattered pattern is thought to be an apparent feature of seismicity due to the short time interval of earthquake instrumental records in a region of high earthquake recurrence rate. It means, that the amount of observed earthquakes is insufficient and not really representative to characterize the seismicity of the
The earthquake activity of the region is restricted to the crust, with the only exception of the Vrancea region, where all the deep earthquakes of the region are concentrated. The observed seismicity is quite low in and north to the Pannonian Basin. There is no detectable seismicity in the investigated area of the East European platform. In Bohemian Massif some earthquake swarms occurred in certain regions, for example in the Eger-graben (Grianthal et al. 1990). The observed activity is thought to be mostly mining-generated tremors in the Ostrava-Silesia region of southern Poland and the northeastern part of the Czech Republic (e.g. Gibowitz et al. 1982). Moderate seismicity can be observed along the southeastern border of the Vienna Basin and the Mur-Murz line in the Eastern Alps (Gutdeutsch & Aric 1988). In the Pannonian Basin, the pattern of seismicity is quite scattered; however, some zones of increased earthquake activity have been delineated (Horv~ith 1984). The seismicity is low in the Moesian Platform, but some moderate earthquakes occurred more to the south, in the Rhodope Massif. The high seismicity of the Southern Alps and the Dinarides indicates considerable active deformation in the southern part of the studied
278
P. GERNER E T A L .
area. The Adriatic sea was characterized as rigid block with low seismicity (Anderson & Jackson 1987), but some seismic activity has occurred since 1986 (Favalli et al. 1992; Console et al. 1993), which suggests higher seismic energy release. In any case, the earthquake activity along the Alps and Dinarides is clearly associated with the m o v e m e n t of the Adriatic microplate relative to Europe (Ward 1994). The high seismicity at the southeastern corner of the Carpathians (Vrancea zone) seems to be isolated and can be related to the presence of a lithospheric slab remnant from the Miocene subduction (Oncescu 1984). Using the seismicity data and various geological and geophysical observations, a map of seismoactive faults in the studied area has been constructed. In the Pannonian Basin structural i n f o r m a t i o n from seismic reflection surveys was also t a k e n into consideration, because seismoactive faults c a n n o t be d e l i n e a t e d clearly due to the disperse seismicity. In the
Dinarides and Alps, the significantly higher level of seismicity has enabled to elaborate detailed seismotectonic models (e.g. Skoko et al. 1977; Cornea et al. 1980; Slejko et al. 1989; Del Ben et al. 1991; Stanishkova & Slejko 1991; Pospisil et al. 1992). We also estimated the spatial pattern of the total released seismic energy in order to asses the amount of active deformation (Fig. 6). We applied the surface wave magnitude versus scalar moment relation (Anderson & Jackson 1987) to obtain the seismic energy. The locations of maximum values on the map are dominated by large earthquakes. The Southern Alps and the Northwestern Dinarides, Central Italy and the Dalmatian coast show high energy release. The Vrancea region is clearly isolated, and also shows high activity. The Pannonian Basin is characterized by higher energy release than the surrounding area except the Dinarides, which implies more intense recent tectonics. The northern part of the Adriatic sea indicates low
Fig. 6. Seismic moment, seismoactive faults and historical and instrumental seismicity (M > 3) from 1091 to 1994. The seismic moment are estimated using the surface wave magnitude versus scalar moment relation of Anderson & Jackson (1987). Seismoactive faults are printed in blue.
RECENT TECTONICS, PANNONIAN BASIN energy release, but around the 42~ increased seismicity can be found.
latitude
Active deformation To characterize the active deformation of the Pannonian region, we investigated the style of active faulting and horizontal m o v e m e n t by analysing the focal mechanisms along fault zones (Fig. 7a). Comparing the strike of nodal planes of the focal mechanisms and the alignment of the fault zone, the plane of faulting and the sense of movement during an earthquake
279
can be deduced. Along the M u r - M u r z line and the southeastern border of the Vienna Basin (Fig. 7b) the focal mechanisms indicate left lateral strike-slip motion (Gutdeutsch & Aric 1988). The earthquakes close to the fault of Lavant and Metnitz valleys exhibit clear strikeslip movements, and the sense of motion seems to be right lateral, but the P axis is roughly parallel to the fault. The horizontal movement on the Medvednica fault zone was determined as left lateral (Skoko et al. 1977), which is in good agreement with the now available focal mechanisms. In the Southern Alps thrusting can be observed along the Periadriatic lineament (Fig.
Fig. 7. Representative focal mechanism solutions. (a) Focal mechanisms for the whole area of interest. Here most of the available solutions are shown, independently from its stress quality ranking. The location of the selected areas can be seen in Fig. 1. (b) Alps-Dinarides-Pannonian Basin junction; (c) Southern Alps, the Friuli region; (d) Dalmatian coast; (e) Montenegro; (f) Rhodope Massif and the southern Moesian Platform; (g) Banat region in SW Romania; (h) the Vrancea region in the Eastern Carpathians
280
P. GERNER E T A L .
/\ j ~
.
(b)
4~
'~ 17~
47~
(c)
12oE
44~
(d)
14~
~
~--
42~ 15~
18~
RECENT TECTONICS, PANNONIAN BASIN
--~
(e)
,o
281
\
42~ 18~
'~/
37~
Moesianplatform
I
(f)
(g)
27~
45~
~ (h)
23~
20~
M/oe si an/p 1at f ~
44oN 25~
28~
282
P. GERNER E T A L .
7c). Similarly, the Fella-Sava line indicates thrusting on the central and eastern segments, but further to the southwest, the Valsugana line shows left lateral strike-slip movement. In general, right lateral strike-slip faulting can be recognized on the Idrija and Rijeka lines, but thrusting can also occur locally. The focal mechanisms along the Dalmatian coast of the Adriatic sea from Split towards Albania indicate thrusting on faults parallel to the coast line (Fig. 7d). Right lateral movement is observable on faults striking perpendicular to the Dinarides, like the Scutari-Pec line (Fig. 7e). Also strike-slip motion was determined for a few e a r t h q u a k e s around Skopje ( A n d e r s o n & Jackson 1987; Fig. 7f), but eastward normal faulting is more dominant eastwards (Muco 1994). Similarly, some normal faulting mechanism can be found around the Bjelasnica mountains and the valley of River Neretva in the Dinarides. The continuation of the A e g e a n extensional stress regime towards the Balkan can be recognized in the R h o d o p e Massif, between the Fore-Balkan line and the Vitosa line, where many reliable focal mechanisms indicate normal faulting (Kiratzi & Papazachos 1995). In Southern Serbia thrusting and strikeslip faulting can be recognized at the Kopaonik mountains (Dokovic et al. 1990). Left lateral strike-slip movement along E-Wdirected faults or more plausibly right lateral motion on N-S-striking faults is the style of faulting to the SW of Timisoara at the rear side of the Southern Carpathians (Fig. 7g). Some poorly constrained focal mechanisms suggest normal faulting along the Southern-Carpathians, but the Vrancea region (Fig. 7h) is clearly dominated by thrusting (Oncescu 1987). In order to determine the dominant faulting mechanism of a tectonic unit, we used the triangle diagram of Fr6hlich (1992). In this diagram the pure strike-slip, thrust and normal faults are plotted to the corners of a triangle, and mixed mechanisms are situated along the borders or inside the triangle. Based on the seismicity pattern of the studied area we selected seven regions to characterize the faulting mechanisms separately (Fig. 8). Inside the Pannonian Basin the focal mechanisms are strikeslip, thrusting and their combination. In the Southern Alps and northwestern Dinarides thrust faulting is dominant, and some earthquakes exhibit strike-slip component. In the Adriatic plate the focal mechanisms display the combination of pure mechanisms. Along the Southern Dinarides and the Dalmatian coast two groups of focal mechanisms can be observed. There is a group of earthquakes which
exhibits thrusting with strike-slip component, and there is an other group with mixed strikeslip and normal faulting mechanisms. Normal faulting becomes more characteristic in the R h o d o p e Massif indicating the continuation of the extension in the Aegean. In the Vrancea region thrust faulting is obviously the dominant mechanism We projected the horizontal components of the earthquake slip vectors to maps, using parameters of focal mechanism solutions. For each event the preferred plane of faulting is chosen either by comparing its strike and dip to geological structures or supposing same slip directions for the closest events. The horizontal components of the slip vectors are shown in Fig. 9, while the smoothed intra- and extrapolated version can be seen in Fig. 10. We applied the algorithm of H a n s e n & Mount (1990) for smoothing, which enabled us to use the earthquake magnitudes as the weight of the slip vectors. Our analysis shows that the horizontal slip vectors are roughly perpendicular to the Southern Alps and the Dinarides indicating the convergence between the Adriatic plate and Europe. This controls the recent deformation pattern in the southern part of the Pannonian Basin. The Adriatic microplate seems to be rotating around a pole to the west of the Southern Alps, but the northwestern trend of the slip vectors of the Friuli region suggest that the pole is below or close to the 45~ latitude. The constrains on the longitudinal position of the pole are much weaker, the pole is somewhere between the 6-10~ longitudes.
Finite element modelling of the recent tectonic stress in and around the Pannonian Basin The comparison of stress observations and the results of stress modelling offers a good possibility to improve our understanding about tectonic processes, which generates stress in the crust. Several sets of finite element models were constructed to analyse the nature and the relative importance of the different sources of recent tectonic stress in the East AlpineC a r p a t h i a n - P a n n o n i a n - D i n a r i c region. The modelled area is a bit smaller than the region where stresses and focal mechanisms have been determined, because we disregarded the extensional province in the Rhodope Massif (compare Figs 4 and 7a to Fig. 11). It was done because this smaller area is characterized by the dominance of strike-slip an6 thrust faulting which suggest that the horizontal maximum stress (SH) is
R E C E N T TECTONICS, P A N N O N I A N BASIN
SS
NF O
~
X Dinarides
(
~
Adriatic plate
)
TF
SS
NF
TF
Pannonian basin
NFO
SS
0
Rhodope-
TF
,(~}~(~}
TF
SS 0
{~) TF
NFO
Vrancea region
li~
SS
NFO Central Italy
0
283
SS 0
NF 0
'
TF
~
SS
massif
9 9
01-
lkm 5km
O
5 - 20 k m
(!) 20- 35 km Q) greaterdepth
NFO
[23 A (5) •
B C
+ D
TF
Fig. 8. Triangle diagrams showing the general style of faulting of selected areas. Pure strike-slip, thrust and normal faulting are projected to the corner of each triangle, while the solutions of mixed components are situated along the borders or inside the triangle.
equal or very close to the maximum principal stress (~1). We simulated several tectonic forces and boundary conditions which can induce the observed SH pattern in the studied area.
The modelled area was treated as an elastic 2D plate. The Poisson ratio was taken uniformly v -- 0.25. The Young's modulus for the relatively weak internal sectors was chosen to be E = 50
284
P. GERNER E T A L .
Fig. 9. Directions of horizontal slip vectors derived from focal mechanism solutions. Dashed lines show seismoactive faults.
GPa, for the external regions we adopted a value of E = 70 GPa, while for the Bohemian Massif and the Moesian Platform a value of E = 100 GPa was assumed. For meshing, we chose an average element size of 50 X 50 km which is very close to the applied grid density of the smoothed stress map. This allows the convenient comparison of observed and calculated stress data for the modelled area. Plain stress approximation is adopted for the calculation of the regional horizontal stress field. In the modelling procedure the stress calculations were carried out by applying different displacements at the edges of the studied area (Fig. 11). These represent the present-day deformation of different tectonic units in the area, which were derived from and constrained by both focal mechanism solutions (Anderson & Jackson 1987) and the latest results of space geodesy (e.g. Cenci et aL 1993; Mueller & Kahle
1993; Smith et al. 1993). These results confirm that the Adriatic microplate drifts independently from the African plate. In our model we adopted the values suggested by Ward (1994): the A d r i a - E u r o p e Euler pole was taken at 46~ 6~ which is very close to the pole calculated by Jackson & McKenzie (1988). Therefore, for the movement of the Adriatic microplate an counterclockwise angular displacement of 0.1 ~ was utilized around a pole at 46~ 6~ This represents a time period of 0.5 Ma applying an angular velocity of 2~ Ma. The model edge at the Bohemian Massif and the Vrancea zone was deformed with an order of magnitude less (200 m), while the boundary at the Moesian Platform was kept immovable. A fixed framework was applied in a distance of 500 km from the edges of the modelled area in order to avoid any kind of edge effect. Therefore, the 'free' model edges are allowed to be deformed only to a certain
RECENT TECTONICS, PANNONIAN BASIN
285
Fig. 10. Smoothed direction of horizontal slip vectors inferred from the data shown in Fig. 9. The algorithm of Hansen & Mount (1990) was applied, where the weight the vectors is the magnitude of the earthquake.
limit. These b o u n d a r y conditions have been combined with other tectonic elements such as the changing crustal thickness of the area (Horvfith 1993) and two main fault zones separating the main structural units of the Pannonian region (Csontos et al. 1992). Our modelling results show that the general features of the recent stress field in the Pannonian region can be simulated by applying relatively simple and geologically reasonable boundary conditions. The models comprise the deformation of crustal blocks with different geometry and strength in an overall convergent setting, given by the ongoing Africa-Eurasia collision. The best fit between the observed and the calculated stress field (see Figs 12 and 13) was achieved when the counterclockwise rotation of the Adriatic plate was taken as a firstorder stress source. It appears that this tectonic unit is moving independently from both the
African and Eurasian plates. Additional and second-order stress-generating boundary conditions are the active shortening and compression at the Vrancea zone, the fairly moving Bohemian Massif, and the immobile Moesian Platform. They all induce stress in the modelled elastic plate superimposed on the effect of the Adriatic microplate. Other stress sources, such as the changing crustal thickness and the presence of two main fault zones in the studied area, have only minor influence on the stress pattern, giving rise to 10-15 ~ local perturbation of stress directions.
Discussion Earthquake focal mechanisms are used to obtain information on stresses acting in the lithosphere, and it is assumed that P and T axes are the principal stress axes. They represent, however, the
286
P. GERNER E T A L .
Fig. 11. Model geometry and boundary conditions of the finite element modelling procedure. A framework in a 500 km distance from model edges was created to avoid edge effects or errors. The Adria-Europe rotation pole was taken from Ward (1994).
direction of maximum shortening and maximum extension, respectively. The principal stresses are very close to the P and T axes during failure only in homogeneous rocks (Anderson 1951). W h e n an earthquake occurs by slip on pre-existing faults the only restriction is that the greatest principal stress must lie in the dilatational quadrant (McKenzie 1969) and it can be quite different from the P axis. In spite of this theoretical limitation, it is the experience of regional studies that the average orientation of the P, B and T axes correlates well with the principal stress directions (Zoback 1992). Moreover, slip vectors from focal mechanisms can be also inverted for the regional stress tensor (e.g. Gephard & Forsyth 1984). Borehole breakout analysis and overcoring technique give directly the direction of the largest horizontal stress, and can be used to control the stress data derived from focal mechanisms. In seismotectonic studies the connection
b e t w e e n mapped faults and e a r t h q u a k e hypocentres is usually somewhat ambiguous because faults are not defined down to the depths where seismicity take.s place. A n o t h e r uncertainty comes from the fact that the location of epi- and hypocentres is always associated with an error of a few kilometres. A further difficulty is given by the inherent duality of focal mechanism solutions in the sense that during the selection of the fault plane there are two equal choices. If additional tectonic information is available, then the selection of the probable fault plane is based on the comparison of the strike of the surface fault and the nodal planes. All of these can lead to serious problems on a local scale, which can be overcome if larger areas, more data and adequate tectonic knowledge is available. We believe that this is mostly the case in the Pannonian Basin and its surroundings. Generalized stress trajectories were published
Fig. 13. The best fitting modelling result showing the combined effect of the applied boundary conditions, the changing crustal thickness and the presence of two fault zones (for outline see Fig. 11).
R E C E N T T E C T O N I C S , P A N N O N I A N BASIN
Fig. 12. Smoothed direction of the m a x i m u m horizontal stress (SH) for the available stress data in the modelled area (for outline see Fig. 11). The smoothing and extrapolating algorithm of Hansen & M o u n t (1990) was applied.
287
288
P. GERNER E T A L .
for the investigated area by Philip (1987), Grtint h a l & Stromeyer (1986, 1992) and Reba'f et al. (1992). Although these studies were based on much less stress data their results are basically consistent with ours. The large amount of data allowed us to achieve better resolution and reliability inside the basin and also in its surroundings. Nevertheless, our stress map has also some discrepancy, for example at the Little Hungarian Plan. Here we would expect a NW-SEaligned largest horizontal stress considering the regional stress and d e f o r m a t i o n patterns. However, using the available focal mechanism and breakout data our smoothing program indicates NE-SW-directed SH. Taking into account the quality and distribution of data we conclude that our stress map does not reflect the local stress here. A seismotectonic model for the area was presented by Gutdeutsch & Aric (1988), assuming an eastward moving rigid Pannonian block. The delineation of block boundaries was based on geological evidence and seismicity pattern. It should be kept in mind that the time period of earthquake records available in the Pannonian Basin seems to be much shorter than the earthquake recurrence period. Therefore, the seismicity catalogue only reports a short period of the whole seismic history, and it does not necessarily indicates all the block boundaries. The existence of more block boundaries is supported by Csontos et al. (1992), who suggested three main tectonic units inside the Pannonian Basin. The seismic activity and seismic energy release of the Pannonian Basin can be compared to its surroundings (Fig. 6), and it requires to suppose more internal deformation. Comparing the strength of the lithosphere and stress magnitudes also suggests that conditions for small to medium energy earthquakes are favourable in the Pannonian Basin. The high heat flow of the basin (D6v6nyi & Horv~ith 1988) implies that the strength of the lithosphere are considerable lower then that of the surrounding area (Brace & Kohlstedt 1980). Furthermore, stresses in the crust are amplified beneath the basins, because the crust and the effective elastic thickness are reduced which leads to larger stress magnitudes beneath the basins (Kusznir & Bott 1977). This is especially the case for the Pannonian Basin, where large contrasts of crustal and lithospheric thickness between the basin and the surrounding mountain ranges can be found (Horv~ith 1993). T h e r e f o r e we suggest that the observed pattern of seismicity in the Pannonian Basin can be better explained by the release of accumulated stress at the many existing zones of weak-
ness, rather than underestimate the small to moderate magnitude events and assume rigidity of a P a n n o n i a n crustal block. The weakness zones correspond to older tectonic structures, for example fault zones or boundaries of crustal blocks, or to any other site of weakness due to compositional or thermal reasons. Stress concentrations can be observed at fault tips or beneath the subbasins of the Pannonian Basin, where the lithosphere is extremely thin and hot, like the B6k6s Basin in SE Hungary (Posgay et al. 1995). On the other hand, it can be seen from stress modelling that changes in the crustal thickness of the Pannonian Basin do not significantly effect the orientation of the maximum principal stress. This might partly be the consequence of the fact that our 2D finite element model ignores stresses arising from changes of topography and the lateral density inhomogeneities of the lithosphere. Although this type of stress is obviously superimposed on the observed stress field, the effect of density induced stress is supposed to be second order compared to those generated by the rotation of the Adriatic plate. According to Fleitout & Froidevaux (1982) and Ranalli (1992), density contrasts in the lithosphere can generate stresses at the order of tens of MPa, while stress sources acting at plate boundaries induce hundreds of MPa (e.g. Turcotte & Schubert 1982; Cloetingh & Wortel 1986; Richardson et aL 1992). Nevertheless, the isometric shape of the Pannonian Basin system may also explain that the changing crustal thickness (i.e. pure geometrical inhomogeneity) has only minor effect on the stress orientation in the area. The reactivation of older fault zones during earthquakes depends on the orientation of the local stress field with respect to the strike of the fault plane. It is an interesting possibility that the style of faulting may change along an older regional fault zone due to significant changes of the stress directions relative to the strike of this older fault zone. Our model implies that earthquakes can occur inside the crustal blocks too, if the stress magnitude becomes larger than the strength of rocks, and do not require the existence of large crustal fault at any earthquake. Nevertheless, if once earthquake occur at a given place, it will lead to the development of fault, and the time interval of fault growth depends on the rate of stress accumulation. In this manner, we can determine the main characteristics of the kinematics the general pattern of the recent crustal deformation in and around the Pannonian Basin (Fig. 14). The convergence of the Adriatic plate and Europe is clearly indicated by thrust faulting around the
RECENT TECTONICS, PANNONIAN BASIN Adriatic see from the Southern Alps to the Dinarides. This feature is governing the recent deformation of Pannonian Basin itself, and has much more influence on the recent tectonics of this area as it was previously suggested. The style of earthquake focal mechanisms of this region, the horizontal stress directions and slip vectors, as well as the seismicity and seismic energy release pattern support our explanation. However, the normal faulting of the Southern Carpathians and the Rhodope Massif argues for that the high seismicity of the Vrancea zone is independent feature of the area, and is separated from the Adriatic convergence. The low seismicity north and east of the Pannonian Basin indicates that most of the energy supplied by the relative motion of Adriatic plate is consumed by the Dinarides and the Pannonian Basin. On the other hand, our modelling results allow us to delineate the main characteristics of
289
the stress sources of the area as well (Fig. 15). These suggest the primary importance of the deformation of crustal blocks with different geometry and strength in an overall convergent setting associated with Africa-Europe collision. The most important stress source appears to be the counterclockwise rotation of the Adriatic microplate at the southwestern boundary of the P a n n o n i a n region ('Adria-push'). The focal mechanism solutions (e.g. Anderson & Jackson 1987), the latest results of space geodesy (Mueller & Kahle 1993; Ward 1994) and the modelling results confirm the notion that the Adriatic crustal block moves i n d e p e n d e n t l y from both the European and the African plates. Additional boundary conditions with an order of magnitude less, such as deformation at the Bohemian Massif, can also significantly effect the modelling results. The compressional stresses at the Vrancea zone are at least partly
Fig. 14. Kinematics of recent crustal deformation in the Pannonian region. In- and outward pointing arrows indicate shortening and extension, respectively. The sense of movements along seismoactive faults is also shown.
290
P. GERNER E T A L .
originating as a results of the final detachment of the subducted lithospheric slab b e n e a t h the sharp bend of the E a s t e r n Carpathians (Spakman 1990). Shortening at the Bohemian edge reflects the compressive effect of the Western European stress province (MOiler et al. 1992) on the westernmost sectors of the Pannonian region. The presence of two main fault zones in the central part of the modelled area and the changing crustal crustal thickness have only locally comparable importance with the applied boundary conditions. These tectonic factors can induce only 10 to 15 ~ perturbations of stress directions. The recent increase of horizontal stress magnitude in the Pannonian Basin is associated with an accelerated subsidence in the central part, while the basin flanks and the Transdanubian Range have been uplifted since late Pliocene (Horvfith & Cloetingh 1996). They explain this anomalous subsidence pattern in terms of stress induced lithospheric deflection, i.e. the increasing intraplate stresses cause largescale bending of the lithosphere. Our data on the
recent state of stress and crustal deformation of the Pannonian region and the results of finite element stress modelling confirm the notion that compressive stresses in and around the PannonJan Basin are capable to produce such a largescale deformation of the lithosphere in this area.
Conclusions The investigated area around the Pannonian Basin can be described by five different stress provinces. In the Apennines, there is a normal faulting regime with N-S and N W - S E directed largest horizontal stress. This regime extends to the coastal region of the Adriatic sea, but here the stress directions are not well constrained by the data. The Northwestern Dinarides, the Southern and Eastern Alps and the Bohemian Massif belong to the Western European stress regime (Mailer et al. 1992), with a N-S to NW-SE-directed horizontal stress and strikeslip and thrust faulting regime. The Dinarides can be characterized by a strike-slip and thrust faulting regime, the direction of the largest
Fig. 15. Cartoon summarizing the main stress sources affecting the Alpine-Carpathian-Pannonian-Dinaric system applied in our models (after Bada et al. 1998). Buttresses are meant as rigid crustal blocks indenting their surroundings. Dashed lines represent faults that were included during modelling.
RECENT TECTONICS, PANNONIAN BASIN horizontal stress ( N E - S W ) is roughly perpendicular to the mountain range, but a clockwise rotation of 40-50 ~ can be observed from the Southern Alps to the Hellenides. The northeastern coastline of the Adriatic sea also belongs to this regime. The influence of the A e g e a n extension can be observed in the normal faulting regime of the Rhodope Massif and in the southeastern Dinarides. In this regime, the largest horizontal stress is E-W-directed. This pattern of the largest horizontal stress manifests a radial configuration around the Adriatic sea, which implies, that the general pattern of the largest horizontal stress is determined by the relative motion of the Adria plate respect to Europe. Most of the focal mechanisms in the Pannonian Basin show strike-slip and thrust faulting, only 2 of the 32 solutions indicate normal faulting. The lack of the normal faulting events in the basin involves that the largest stress lies in the horizontal plane. The overwhelming strike-slip and thrust faulting mechanism of earthquakes inside the basin suggests that extension of the Pannonian Basin has been finished, and it suggests that the structural inversion of the Pannonian Basin is underway. We note, however, that all the pure thrust faulting focal mechanisms can be found at the basin flanks, and inside the basin dominantly strike-slip faulting has occurred, in some cases with a thrusting component. The result of the hydraulic fracturing also indicates strike-slip deformation, but the magnitude of the least horizontal stress is very close to the vertical stress. In the western Pannonian Basin N-S and N N E - S S W greatest stress can be found, while the central end eastern part show mainly E N E - W S W and E - W directed maximal stresses, indicating a remarkable deviation from the West European trend, and fit well into the radial stress pattern around the Adriatic plate. The recent crustal kinematics of the investigated area is governed mainly by the rotation of Adria relative to Europe around a pole at the 45~ latitude and 6-10~ longitudes. This movement is resulting in the compressive deformation of the Southern Alps and the right lateral strikeslip motion on the NW-SE-aligned faults in the Northwestern Dinarides. The region between the Adriatic sea and the M u r - M u r z line seems to be squeezed out eastwards, causing left lateral strike-slip movement on the Medvednica line and the M u r - M u r z line, and right lateral motion along the Idrija fault. The motion of Adria is producing thrusts along the Dalmatian coast, and this compressive deformation extends into the land far from the coast line. It can be observed along the Sava though, and it is resulted in the strike-slip motions of the Banat
291
region. This far field effect of the convergence between Adria and the Dinarides is squeezing the Pannonian Basin from south and southwest. Compression at the Vrancea zone is separated from these general compressive pattern by the normal faulting of the Southern Carpathians and the Rhodope Massif, which implies that those seismicity is not generated by the motion of Adria. This is in good agreement with the results of our finite element calculations, which confirmed that the recent crustal stress and deformation in the Pannonian region is governed by distinct tectonic factors with different origin and relative importance. Clearly, we can conclude from stress observations, seismicity studies and stress modelling that the most important stress source appears of to be the counterclockwise rotation of the Adriatic microplate. We suggest that the seismicity pattern of the P a n n o n i a n Basin can be better u n d e r s t o o d taking into consideration of the finite strength of lithosphere and tectonic stresses, rather than assuming movement along the boundaries of rigid crustal blocks. The earthquakes of the basin are generated by failure of suitably aligned weakness zones in the recent stress field of the lithosphere. These weakness zones are usually reactivated older fault systems, but earthquakes can also occur at any sites of lower strength due to thermal or compositional reasons. It means that earthquakes can occur anywhere, if the arrangement of the weakness zones in the stress field is favourable and the stress magnitudes are large enough to induce faulting. Our model does not require the existence of large crustal fault at any e a r t h q u a k e hypocenter, but e a r t h q u a k e activity will lead to fault development. The authors are grateful to T. Zsfros, K. Aric, V. Karnik, H. Bungum, J. Smriglio and M. Zirbes for providing seismicity data. D. Slejko, P. Favali and C. Gasparini are thanked for focal mechanism data. The computation of the fault plane solutions was carried out using the programs of D. Oppenheimer and R. Bulland, who are acknowledged for sending program codes and user guides. This work was supported by the joint TEMPUS project of the E6tv6s University of Budapest, the Vrije Universiteit of Amsterdam and the University Fridericiana of Karlsruhe (JEP-1506), and the IBS (Integrated Basin Studies) of the European Community (contract JOU2-CT92-0110). Their financial help is gratefully acknowledged.
References AKI, K. & RICHARDS,P. G. 1980. Quantitative Seismology. Vol. 1. Freeman, San Francisco. ANDERSON,E. M. 1951. The Dynamics of Faulting and Dyke Formation With Application to Britain. Oliver and Boyd, Edinburgh.
292
P. G E R N E R E T A L .
ANDERSON, H. & JACKSON,J. 1987. Active Tectonics of the Adriatic Region. Geophysical Journal of the Royal Astronomical Society, 91, 937-983. ARTYUSHKOV, E. V. 1973. Stresses in the lithosphere caused by crustal thickness inhomogenities. Journal of Geophysical Research, 78, 7675-7708. BADA, G., CLOETINGH, S., GERNER, P. & HORVATH, E 1998. Sources of recent tectonic stress in the Pannonian region: inferences from finite element modelling. Tectonics, 134, 87-102. BECKER, A. 1993. Contemporary state of stress and neotectonic deformation in the CarpathianPannonian region. Terra Nova, 5, 375-388. BELL, J. S. & GOUGH, D. I. 1979. Northeast-southwest compressive stress in Alberta: Evidence from oil wells. Earth and Planetary Science Letters, 45, 475M82. BEROERAT, E, GEYSSANT, J &. LEPVRmR, C. 1983. Etude de la fracturation dans le bassin pannonien: mdcanismes et 6tapes de sa crdation. In: Tectonique cassante et distension en coulissement. Annales de la Socidt~ GOologique du Nord, 103, 265-272. BRACE, W. E & KOHLSTEDT, D. L. 1980. Limits on lithospheric stress imposed by laboratory experiments. Journal of Geophysical Research, 85, 6248-6252. BULLAND, R. & CHAPMAN,C. H. 1983. The computation of seismic travel times. Bulletin of the Seismological Society of America, 73, 1271-1302. CAGNETTI,V., PASQUALE,V. & POLINARI,S. 1976. Focal mechanisms of earthquakes in Italy and adjacent regions. CNEN RT/AMB Reports, 76. CARULLI,G. B., NICHOLIC, R., REBEZ, A. & SLEJKO,D. 1990. Seismotectonics of the Northwest External Dinarides. Tectonophysics, 1"/9, 11-25. CENCI, A., FERMI, M., SCIARRETTA, C. • DEVOTI, R. 1993. Tectonic motion in the Mediterranean area from laser ranging to LAGEOS. In: SMITH,D. E. & TURCOTrE, D. L. (eds) Contributions of space geodesy to geodynamics: crustal dynamics. A G U Geodynamics Series, 23, 347-358. CLOE'nNGH, S. & WORI'ZL, M. J. R. 1986. Stress in the Indo-Australian plate. Tectonophysics, 132, 49-67. CONSOLE, R., DI GIOVAMBATTISTA, R., FAVAIA, P., PRESGAVE, B. W. & SMRIGLIO,a. 1993. Seismicity of the Adriatic microplate. Tecwnophysics, 218, 343-354. CONSTANTINESCU,L., RUPRECHTOVA,L. & ENESCU,D. 1966. Mediterranean-Alpine Earthquake Mechanisms and their Seismotectonic Implications. Geophysical Journal of the Royal Astronomical Society, 10, 347-368. CORNEA, I., POLON1C, G. & STEFLEA,V. 1980. Seismotectonic studies in the Baia Mare-Halmeu area. Studii si Cercetari de Geologie, Geofizica, Geografie. Seria Geofizica, 19, 52-60. CSONTOS L., NAGYMAROSI,A., HORVATH, E & KOVAC, M. 1992. Tertiary evolution of the IntraCarpathian area: a model. Tectonophysics, 208, 221-241. , TARI G., BERGERAT, E & FODOR, L. 1991. Evolution of the stress field in the Carpatho-
Pannonian area during the Neogene. Tectonophysics, 199, 73-91. DEL BEN, A., FINETTI, I., REBEZ, A. &; SLEJKO,D. 1991. Seismicity and seismotectonics at the Alps-Dinarides contact. Bollettino di Geofisica Teorica ed Applicata, 33, 155-176. DOKOVIC, I., MAROVIC, M., STOJNIC, N. & NEDELJKOV1C, S. 1990. Seismotectonics, seismic energy and danger in Kopaonik seismic area. Beograd. DOVI~NYI P. & HORVATH,E 1988. A review of temperature, thermal conductivity, and heat flow data for the Pannonian basin. In: ROYDEN, L.H. & HORVATH,E (eds) The Pannion Basin - a Study in Basin Evolution. American Association of Petroleum Geologists Memoirs, 45, 195-233. & -1990. Determination of contemporary crustal stress regime in Hungary. Acta Geodinamica Geophisica Montana, Hungarica, 25, 257-266. DZIEWONSKI, A. M., EKSTROM, G. & SALGANIK,M. R 1992. Centroid-moment tensor solutions for October-December 1991. Physics of the Earth and Planetary Interiors, 74, 89-100. ENGELDER, W. 1993. Stress regimes in the lithosphere. Princeton University Press, Princeton, New Jersey. EVA, C., RIUSCETrl, M. & SLEJKO,D. 1988, Seismicity of the Black Sea Region. Bollettino di Geofisica Teorica ed Applicata, 30, 53-65. FAVALI, R, FUNICELLO, R., MATT1ETrl, G., MELE, G., MONTONE, P., SAININI., E & Tozzl, M. 1992. Seismotectonic identity of the Southern Adriatic Area. In: CARMIC;NANI, L. & SASSI, E P. (eds) Contributions to the Geology of Italy with special regard to the Paleozoic basements. IGCP 276 Newsletter, 5, 339-343. FI.Errotn, L. & FROIDEVAUX, C. 1982. Tectonics and topography for a lithosphere containing density heterogeneities. Tectonics, 1, 21-56. FODOR, L. 1995. From transpression to transtension: Oligocene-Miocene structural evolution of the Vienna basin and the East Alpine-Western Carpathian junction. Tectonophysics, 242, 151-181. --, CSONTOS, L., BADA, G. & GYORFk I. 1999. Tertiary tectonic evolution of the Pannonian basin system and neighbouring orogens: a new synthesis of palaeostress data. This volume. FROHLICH, C. 1992. Triangle diagrams: ternary graphs to display similarity and diversity of earthquake focal mechanisms. Physics of the Earth and Planetary Interiors, 75, 193-198. GAN(;L, G. 1975. Seismotectonic Investigations of the Western Part of the Inneralpine Pannonian Basin (Eastern Alps and Dinarides). In: Proceedings of XIV. General Assembly, European Seismological Commission, Trieste, Sept 16-22, 1974, Nationalkomitee fur Geodasie und Geophysik, Berlin, 409-410. GASPARINI, C., IANNACCONE, G. & SCARPA, R. 1985. Fault-plane solutions and seismicity of the Italian Peninsula. Tectonophysics, 117, 59-78. GEPHART, J. & FORSYTH, D. W. 1984. An improved method for determining the regional stress tensor -
-
RECENT TECTONICS, PANNONIAN BASIN using earthquake focal mechanism data: application to the San Fernando earthquake sequence. Journal of Geophysical Research, 89, 9305-9320. GmowIrz, S. J., GUTERCH, B., LEWANDOWSKAMARCINIAK, H. & WYSOKINSKI,L. 1982. Seismicity induced by surface mining: The Belchatow, Poland, earthquake of 29. November 1980. Acta Geophysica Polonica, 30, 193-219. , NIEWlADOMSK[,J., WIEJACZ, P. 8z DOMANSKI, B. 1989. Source study of the Lublin, Poland, Mine tremor of 20. June 1987. Acta Geophysica Polonica, 37, 111-132. GRASSL, S., GROSSER,H. & GRUNTHAL,G. 1984. Microand Macroseismic Studies of the Leipzig Earthquake of February 20,1982. Gerlands Be#rage zur Geophysik, 93, 173-184. GRIJNTHAL, G. & STROHMAYER,D. 1986. Stress pattern in Central Europe and adjacent areas. Gerlands Be#rage zur Geophysik, 443-452. & -1992. The Recent Crustal Stress Field in Central Europe: Trajectories and Finite Element Modelling. Journal of Geophysical Research, 97, 11 805-11 820. - - , SCHENK,V., ZEMAN, A. & SCHENKOVA,Z. 1990. Seismotectonic model for the earthquake swarm 1985/86 in the focal area Vogtland/West Bohemia focal area. Tectonophysics, 174, 369-383. GUTDEUTSCH,P. & ARIC, K. 1988. Seismicity and neotectonics of the East Alpine-Carpathian and Pannonian area. American Association of Petroleum Geologists' Memoirs, 45, 183-194. HANSEN, K. M. & MOUNT,V. S. 1990. Smoothing and extrapolation of crustal stress orientation measurements. Journal of" Geophysical Research, 1155-1165. HORVATH, E 1984. Neotectonics of the Pannonian basin and the surrounding mountain belts: Alps, Carpathians and Dinarides. Annales Geophysicae, 2, 147-154. -1993. Towards a mechanical model for the formation of the Pannonian basin. Tectonophysics, 226, 333-357. 1995. Phases of compression during the evolution of the Pannonian basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology, 12, 837-844. - - & CLOETINGH,S. 1995. Stress-induced anomalous subsidence and uplift in the late-stage evolution of the Pannonian basin. Tectonophysics, in press. & GERNER, R 1995. The neotectonics of the Pannonian basin: a change from extension to compression. Acta Geologica Hungarica, in press. JACKSON, J. & MCKENZIE, D. 1988. The relationship between plate motions and seismic moment tensors and the rates of active deformations in the Mediterranean and Middle East. Geophysical Journal, 93, 45-73. KENNETT,B. L. N. & ENGDAHL,E. e . 1991. Traveltimes for global earthquake location and phase identification, Geophysical Journal International, 105, 429-465. KIRATZI, A. A. & PAPAZACHOS, C. B. 1995. Active crustal deformation from the Azores triple junction to the Middle East. Tectonophysics, 243,1-24. 95,
-
-
95,
-
-
-
-
293
KUSZNIR, N. J. & BOTT, M. H. R 1977. Stress concentration in the upper lithosphere caused by underlying visco-elastic creep. Tectonophysics, 43, 247-256. MARDIA,K.V. 1972. Probability and Mathematical Statistics. Academic Press, London. MCKENZlE, D. E 1969. The relation between fault plane solutions for earthquakes and the direction of principal stresses. Bulletin of Seismological Society America, 59, 591-601. 1972. Active Tectonics of the Mediterranean Region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. Moos, D. & ZOBACK,M. D. 1990. Utilisation of observations of well bore failure to constrain the orientation and magnitude of crustal stresses. Journal of Geophysical Research, 95, 9305-9325. Muco, B. 1994. Focal mechanism solutions for Albanian earthquakes for the years 1964-1988. Tectonophysics, 231, 311-323. MUELLER, ST. & KAHLE, H.-G. 1993. Crust-mantle evolution, structure and dynamics of the Mediterranean-Alpine region. In: SMITH, D. E. & TURCOrrE, D. L. (eds) Contributions of space geodesy to geodynamics: crustal dynamics. AGU Geodynamics Series, 23, 249-298. MOLLER, B., ZOBACK, M. L., FUCHS, K., MASTIN, L., GREGERSEN, S., PAVONI, N., STEPHANSSON, 0. ~; LJUN66REN, C. 1992. Regional pattern of tectonic Stress in Europe. Journal of Geophysical Research, 97, 11 783-1l 803. NOLTE, K. G. 1982. Fracture design considerations based on pressure analysis. Society of Petroleum Engineers, SPE 10911. ONCESCU, M. C. 1984. Deep structure of the Vrancea region, Roumania, inferred from simultaneous inversion for hypocenters and 3D velocity structures. Annales Geophysicae, 2, 23-28. 1987.On the Stress Tensor in Vrancea Region. Journal of Geophysics, 62, 2-65. PHILIP, H. 1987. Plio-Quaternary evolution of the stress field in Mediterranean zones of subduction and collision. Annales Geophysicae, 5B, 301-320. PLUMB, R. A. ~; HICKMAN, S. H. 1985. Stress induced borehole elongation: a comparison between the four-arm dipmeter and the borehole televiewer in the Auburn geothermal well. Journal of Geophysical Research, 90, 5513-5521. POSGAY,K., BODOKY,T., HEGEDISS,E., KOVACSVOLGYI, S., LENKEY,L., SZAFIAN,R, TAKACS,E., T[MAR, E. & VARGA, P. 1995. Asthenospheric structure beneath a Neogene basin in Southeast Hungary. Tectonophysics, 252, 467-484. POSPISIL, L., BUDAY,T. &: FUSAN,O. 1992, Neotectonic movements in the West Carpathians. Zapadni Karpaty, 16, 65-84. RATSCHBACHER, L., FRISCH, W. • LINZER, H. G. 1991. Lateral extrusion in the Eastern AIps, part 2: structural analysis. Tectonics, 10, 257-271. RANALLI, G. 1992. Average lithosphereic stresses induced by thickness changes: a linear approximation. Physics of the Earth and Planetary Interiors, 69, 263-269. REASENBERG, P. A. & OPPENHEIMER,D. 1985. FPFIT, -
-
-
-
294
P. G E R N E R ETAL.
FPPLOT and FPPAGE: Fortran computer programs for calculating and displaying earthquake fault-plane solutions. USGS Open File Report. REBAI, S., PHILIP,H. & TABOADA,A. 1992. Modern tectonic stress field in the Mediterranean region: evidence for variation in stress directions at different scales. Geophysical Journal of the Royal Astronomical Society, ll0, 106-140. RICHARDSON, R. M., SOLOMON, S. C. & SLEEP, N. H. 1992. Tectonic stress in the plates. Reviews in Geophysics and Space Physics, 17, 981-1019. RITSEMA,A. R. 1974. The Earthquake Mechanisms of the Balkan Region. Koninklijk Nederlands Meteorologish Instituut UNDP/UNESCO report on survey of the seismicity of the Balkan region, project REM/70/172. SPAKMAN,W. 1990. Tomographic images of the upper mantle below central Europe and the Mediterranean. Terra Nova, 2, 542-553. SCHEIDEGGER,A. E. 1967. The tectonic stress in the vicinity of the Alps. Zeitschriftfiir Geophysik, 33, 167-181. SCHENK, V., SCHENKOVA,Z., POSPISIL,L. & ZEMAN, A. 1986. Seismotectonic model of the upper part of the Earth's crust of Czechoslovakia. Studia Geophysica et Geodaetica, Ceskoslovenska Akademie ged., 30, 321-330. SHANOV,S., SPASSOV,E. & GEORGIEV,T. 1992. Evidence for the existence of a paleosubduction zone beneath the Rodopean massif (Central Balkans). Tectonophysics, 206, 307-314. SIPK1N,S. A. • NEEDHAM, R. E. 1992., Moment-tensor solutions estimated using optimal filter theory: global seismicity, 1990. Physics of the Earth and Planetary Interiors, 70, 16-21. SKOKO, D., CVIJANOVIC, D. & PRELOGOVIC, E. 1977. Seismic activity based on the neotectonics. In: Proceedings of the Symposium on the Analysis of Seismicity and on Seismic Risk. Czechoslov. Acad. Sci. Libice., 17-22 October 1977, 81-91. SLEJKO, D., CARULLI, O. B., NICHOLIC, R., REBEZ, A.,
ZANFERRARI, A., CAVALLIN, A., DOGLIONI, C., CARRARO, O., CASTALDINI, D., ILICETO, V., SEMENZA,E. d~;ZANOLLA,C. 1989. Seismotectonics of the Eastern Southern-Alps: a review. Bollettino di geofisica teorica ed applicata, 31, 109-136. SMITH, D. E., KOLENKIEWICZ,R., NEREM, R. S., DUNN, R J., TORRENCE, M. H., ROBBINS,J. W., KLOSKO,S. M.,WILLIAMSON, R. O. & PAVLIS,E. C. 1993. Contemporary global horizontal crustal motion. Geophysical Journal International, 119, 511-520. STANISHKOVA,I. & SLEJKO, D. 1991. Some seismotectonic characteristics of Bulgaria. Bollettino di geofisica teorica ed applicata, 33, 187-210. SZABO, Cs., HARANGI, Sz. d~; CSONTOSL. 1992. Review of Neogene and Quternary volcanism of the Carpathian-Pannonian region. Tectonophysics, 208, 243-256. TARI, G., BALDI,T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian basin: geodynamic model. Tectonophysics, 226, 433-456. TOTn, L., MONUS, R & ZS~ROS,T. 1989. The Berhida (Hungary) earthquake of 1985. Gerlands Beitrage zur Geophysik, 98, 312-321. WARD, S. N. 1994. Constrains on the seismotectonics of the central Mediterranean from Very Long Baseline Interferometry. Geophysical Journal International, 117, 441-452. ZOBACK, M. D. & ZOBACK, M. L. 1991. Tectonic stress field of North America and relative plate motions. In: SLEMMONS, B. Er Air. (eds) The Geology of North America, Decade Map vol. 1, Neotectonics of North America. GSA, Boulder Colorado, 339-366. ZOBACK, M. L. 1992. First- and second-order patterns of stress in the lithosphere: the World Stress Map Project. Journal of Geophysical Research, 97, 11 703-11 728. -& ZOBACK,M. D. 1989. State of stress of the conterminous United States. Memoirs of the Geological Society of America, 172, 523-539.
Tertiary tectonic evolution of the Pannonian Basin system and neighbouring orogens: a new synthesis of palaeostress data LASZL0
FODOR
l, L A S Z L 0
C S O N T O S 2, G A B O R LfitSZL0
BENKOVICS
B A D A 1, I S T V A N
GYORFI
3&
4
1Department of Applied and Environmental Geology, EOtvOs University, Budapest, 1088 M6zeum krt 4/a, Hungary 2Department of General and Historical Geology, EOtvOs University, Budapest, 1088 M6zeum krt. 4/a, Hungary SDepartment of Geophysics, EOtvOs University, Budapest, 1085 Ludovika tOr 2, Hungary 4Department of Engineering Geology, Technical University of Budapest, Stoczek u. 2, Hungary Abstract: Compilation of a microtectonic observation data base for most of the data measured in the Pannonian Basin and surrounding orogens permits a detailed reconstruction of the Tertiary stress field evolution. Combination of tectonic observations, borehole, gravity and seismic data, palaeogeographic and stratigraphic information led to an understanding of fault kinematics and description of the structural evolution in seven major tectonic episodes. The first two episodes depict the kinematics of the two major separated blocks, the Eastern Alpine-Western Carpathian-Northern Pannonian (Alcapa) and the Southern Pannonian-Eastern Carpathian (Tisza-Dacia) microplates. A Mid-Eocene to Early Oligocene N-S compression led to contractional basin formation both in the foreland (Western Carpathians) and hinterland (Hungarian Palaeogene basins) of the orogenic wedge. Due to oblique convergence, the Palaeogene basins are generally asymmetric and often dissected by dextral tear faults. Northward advance of the Adriatic promontory initiated the separation of the Alcapa from the Southern Alps and its eastward extrusion. This process probably started during latest Oligocene and reached its climax during the Early Miocene. The main displacement was accommodated by dextral slip along the Periadriatic and Mid-Hungarian shear zones and during and after this tectonic episode Alcapa suffered 50 ~ CCW rotation. At about the same time period the Tisza-Dacia block also experienced rotation of 60-80% but clockwise. These opposite rotations resulted in the marked actual deviation of earlier compression axes, which are now N or NW in the Eastern Alps, WNW-ESE in the Western CarpathianPannonian domain and NE-SW in the Tisza-Dacia domains. Termination of rotations can be considered as the time for final amalgamation of the two separate blocks and the beginning of extensional tectonics in a single Pannonian unit. The Pannonian Basin system was born by rifting of back-arc style during the late Early and Mid-Miocene time. Extension was controlled by the retreat and roll-back of the subducted lithospheric slab along the Carpathian arc. Two corners, the Bohemian and Moesian promontories formed gates towards this free space. At both the northern and southern corners, broad shear zones developed. The initial NE-directed tension was gradually replaced by a later E- to SE-directed tension as a consequence of the progressive termination of subduction roll-back along the arc from the Western Carpathians towards the Southern Carpathians. There is growing evidence that an E-W-oriented short compressional event occurred during the earliest Late Miocene but during the most of the Late Miocene extension was renewed. Starting from the latest Miocene roll-back terminated everywhere and a compressional stress field has propagated from the Southern Alps gradually into the Pannonian Basin, and resulted in Pliocene (?) through Quaternary tectonic inversion of the whole basin system.
D u r i n g the past 15 years, the u n d e r s t a n d i n g of the tectonic evolution of the P a n n o n i a n Basin has increased considerably. This i m p r o v e m e n t h a d several sources. O n e i m p o r t a n t c o n t r i b u t i o n has c o m e f r o m the increased precision in biostratigraphy (e.g. Bfildi 1986), and r a d i o m e t r i c
dating of m a g m a t i c rocks (P6cskay et aL 1995). A n o t h e r factor was the publication of several seismic sections f r o m the basin system including seismic stratigraphic results (Kil6nyi et aL 1991; L6rincz & Szab6, 1993; Pogficsfis et al. 1988, 1994a,b; R u m p l e r & Horvfith 1988; Tari et al.
FODOR, L. & CSONTOS L., BADA, G., GYORFI,I. & BENKOVICS,L. 1999. Tertiary tectonic evolution of the Pannonian basin system and neighbouring orogens: a new synthesis of palaeostress data. In: DURAND,g.,
JOLWET, L., HORVATH,E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 295-334.
296
L. FODOR E T A L .
1992; Horvfith et al. 1995). The third source of information has been the systematic structural investigation, i.e. a great number of tectonic measurements and related palaeostress calculations have been carried out in areas where brittle rocks crop out. The main goal of the present paper is to review and interpret this palaeostress data set. Tectonic measurements started in the frame of a Hungarian-French co-operation. The first results seemed to fit the general extensional-transtensional character of the Pannonian Basin (Bergerat et al. 1983, 1984). The following research gradually involved different authors in other areas of the basin system and its surroundings (Poljak 1984; Bergerat & Csontos 1988; Nem~ok et al. 1989; Fodor et al. 1990) and permitted the first basin-scale synthesis (Csontos et al. 1991). By the beginning of the 1990s numerous groups started their researches in the Western Carpathians, Southern and Eastern Alps. Their results describe the local structural geology and often modify ideas on the general structural evolution of the whole area
(Kovfi6 et al. 1989; Marko et al. 1990, 1991; Nem~ok & Lexa, 1990; Decker et al. 1993, 1994; Linzer et al. 1995; Nemes et al. 1995a,b; Ratschbacher et al. 1993a,b; Kovfi6 & H6k, 1993; Vass et al. 1993, GyOrfi et al. this volume). Recent attempts have been also made for regional synthesis using stress data (Nem6ok 1993; Decker et al. 1993; Linzer et aL 1995; Marko et al. 1995). This paper reviews the results of tectonic measurements carried out in the Pannonian Basin system and neighbouring parts of the Alpine-Carpathian-Dinaric system. Interpretation of measurements has been improved by the analysis of map-scale structures, stratigraphic research, borehole data, seismic sections and gravity maps. Incorporation of palaeomagnetic data in tectonic reconstruction is also inevitable because they show large and complex during the Tertiary (Mfirton 1987; Mfirton & Mauritsch 1990; Mfirton & Mfirton 1996). Structural models based on map analysis or analogue modelling (Balla 1984; Ratschbacher et al. 1991a,b) also contributed to our geodynamic models. Using our palaeostress database and the above mentioned
Fig. 1. Major units and pre-Neogene outcrops of the East Alpine-Pannonian-Carpathian region. Palaeomagnetic data after Mfirton & Mauritsch (1990); Mfirton & Mfirton (1996); Krs et al. (1982, 1991); Bazhenov & Burtman (1980); Bazhenov et al. (1993); Pfitrascu et al. (1992, 1994).
PANNONIAN BASIN PALAEOSTRESS DATA other structural data, we give a detailed description of the Tertiary stress field and structural evolution than we attempt to arrive at a geodynamic model which connects the stress field evolution and the kinematics of the area.
Geological setting The Carpathian mountains and the Dinarides encircle an inner lowland. This Intra-Carpathian Basin is not uniform, but is subdivided by lowaltitude inselbergs, e.g. the Transdanubian Range (TR), Bakk, Mecsek, Villfiny, Papuk and Apuseni mountains (Fig. 1). The major subbasins are: the Graz Basin, the Little Hungarian Plain-Danube Lowland, the Great Hungarian Plain, which form together the Pannonian Basin. The Transylvanian, Vienna and East Slovakian Basins are situated close or above the contact of the inner Eoalpine nappes and the Outer Carpathian Flysch Belt (Fig. 1). The basins were formed by stretching of continental lithosphere during the Neogene, synchronously with the compressional deformation of the Outer Carpathian flysch nappes (Horvfith & Royden 1981; Horvfith & Rumpler 1984). The gross geological structures of the Carpathians and the Dinarides are similar. Both are formed by nappes verging outward from the Pannonian centre. The external units are composed of a Late Cretaceous-Early Neogene flysch wedge, deformed during the Tertiary and thrust on their foredeep and foreland (Fig. 1). The internal units are made up of less continuous exposures of Mesozoic rocks and their crystalline basement, deformed during different periods of the Jurassic and Cretaceous. A narrow, strongly deformed zone of Mesozoic rocks, the Pieniny Klippen Belt is situated between the Inner West Carpathians and the Flysch belt (Birkenmajer 1985). The Neogene volcanic belt follow the outlines of the chains on their internal side. Large amounts of volcanic rocks are buried beneath sediments of the Pannonian Basin. Modern seismic and borehole investigations revealed that the substrata of the Neogene Basins is composed of Alpine nappes (Papp 1990; Horvfith 1993; Tari 1994), which are the continuations of outcropping units around the basin system. Based on palaeobiogeographical contrast in the northern and southern inselbergs (TR and Mecsek-Villfiny Mts respectively) G6czy (1973, 1984) recognized a major faunal discrepancy: the TR showed an affinity to a Tethyan palaeobiogeographical province, while the Mecsek and Villfiny Mts to a European palaeobiogeographical province during the Early Jurassic. This interpretation is corroborated by a number of
297
subsequent faunal studies (V6r6s 1993). This implies that the substratum of the Pannonian Basin is not uniform, but is composed of at two major blocks of different original paleogeographic position (Fig. 2). Palaeomagnetic measurements also support this tectonic subdivision. The declinations of the TR, B~ikk Mts, together with the Inner West Carpathians show Tertiary counterclockwise rotation (Mfirton 1987; M~rton & Mfirton 1996). In contrast, data from the Mecsek, Villfiny and Apuseni Mrs, together with the East and South Carpathians show Tertiary clockwise rotation (Mfirton & Mfirton 1978; Balla 1987; P~trascu et al. 1992, 1994; Bazhenov et al. 1993; Fig. 1). Because the amount of rotation is quite uniform within the two units, it is probable that the two major blocks moved as separate microplates during the Tertiary (Balla 1984, 1988; Pfitrascu et al. 1994). These continental microplates are called the Alcapa and Tisza-Dacia units (Balla 1984; Csontos et al. 1992; V6r6s & Csontos 1992; Fig. 2). They had significantly different Mesozoic position and were juxtaposed in the late Tertiary. The boundary between the two units is a tectonic zone, the Mid-Hungarian fault system, which is buried for most of its length below the Neogene volcanic and sedimentary rocks of the Pannonian Basin. The Alcapa block (Fig. 2) is located north of the Mid-Hungarian fault system. Its original northern limit is the Pieniny Klippenbelt (Fig. 1). This northern limit gradually shifted to more external nappe boundaries, as during evolution of the Alcapa, more and more Outer Carpathian flysch nappes were accreted to the internal Mesozoic unit. The western boundary of the Alcapa can be found in the Eastern Alps along the low-angle normal fault at the eastern limit of the Tauern window.
Rotations derived from palaeomagnetic data In structural analysis and stress field reconstruction in particular, the amount and time of rotations of different units are crucial. We briefly discuss the most important palaeomagnetic data, because description of stress field history is not possible without this information. Senonian rocks within the Northern Calcareous Alps show slight clockwise rotation (Mauritsch & Becke 1987; Mauritsch & Mfirton 1995). The rest of the Alcapa exhibit larger clockwise rotations (Mfirton 1993). In the northern Pannonian Basin and the Western Carpathians
298
L. FODOR E T A L .
Fig. 2. Major tectonic units of the study area. This division is mainly valid for the late Palaeogene-Early Miocene period.
rotation attained 80 ~ CCW (M~irton & M~irton 1996; Tt~nyi & M~irton 1996; M~rton et al. 1996). This happened in two phases: an early 45-50 ~ CCW and a later 30-35 ~ CCW rotation can be distinguished. The age of the first rotation is well-constrained, having occurred in the late Ottnangian-early Karpatian (18-17 Ma; M~lrton & M~rton 1996). The second phase ended before the Mid-Miocene because Badenian and younger rocks did not show anomalous palaeodeclination (M~rton & M~irton 1989). The only exception occurs in the northeastern part, near the East Slovakian Basin, where 30 ~ CCW rotation occurred as late as the end of the Sarmatian (Mfirton & P6cskay 1995; Orlicky 1996). The northern Carpathian flysch belt also shows 45-60 ~ CCW rotation (Krs et al. 1982, 1991) but this is less constrained in time. Senonian rocks in the southern part of the Eastern Alps show moderate (30-60 ~ counterclockwise rotation. A similar 30 ~ of CCW rotation was d e m o n s t r a t e d close to this site in northern Slovenia on Badenian rocks (M~rton & Jelen in preparation). All these data suggest that the Alcapa terrane rotated consistently in
counterclockwise sense, but by different amounts, which indicate internal deformation of Early to Mid-Miocene age. The Tisza-Dacia unit underwent opposite rotation with respect to the northern unit. Data from southern Hungary (Mecsek) and the Apuseni Mts, Southern Carpathians derived from Palaeogene and earliest Miocene rocks show clockwise rotation of about 90~ (Mfirton & Mfirton 1978; Surmont et aL 1990; P~trascu et al. 1992, 1994). Because volcanics of 15 Ma and younger age do not show any important rotation, it must have happened during the early Miocene. In conclusion, rotation of the Alcapa and Tisza-Dacia units with different sense occurred at a similar time, between 19 and 15 Ma.
Stress field derived from tectonic observations We carried out systematic tectonic studies at the margins of the P a n n o n i a n Basin system, in Hungary, Austria, Slovakia, Czech Republic,
PANNONIAN BASIN PALAEOSTRESS DATA Romania and Slovenia. The field observations consisted of measurements of brittle structures, mainly faults, but joints and folds were also taken into account. Sense of the fault slip was determined using kinematic indicators occurring on and along the fault planes (Angelier 1979a; Hancock 1985; Petit 1987). A theoretically well-known equation defines the relationship between the different type of faults and the stress field (Anderson 1951; Bott 1959). The three orthogonal principal stress axes: the maximum (0-1), the medium (0"2) and the minimam axes (0-3) can be derived from the observed brittle failure pattern. Concerning their mutual position with respect to horizontal, three main type of stress state can be determined in a given point: compressional, strike-slip and tensional stress states if 0"3, 0"2 or o-1 are vertical, respectively. The method of palaeostress calculation from fault slip (striae) observation was elaborated and extensively discussed by Angelier (1979b, 1984, 1990). At several outcrops faults were not formed under one stress field. In such cases two types of separation into h o m o g e n e o u s subsets were carried out. First, the software of Angelier & Manoussis (1980) allows an automatic phase separation. Parallel to this, faults were separated manually, taking into account A n d e r s o n ' s simple geometric assumptions. Then the manual grouping of faults was tested by computer calculations. In most cases, the 'automatic' and 'manual' phase separation yielded similar or identical stress axes. A palaeostress data base was set up from the calculated stress data. This contains the site name, coordinates, age of deformed rock, the direction of the main stress axes, the type of calculation method and the number of observations. To cover the whole study area, we completed the data base for the surrounding mountain ranges from publications (see Appendix for sources of the stress data). Stress calculation methods used by different authors were, unfortunately not the same. In some cases the stress axes were only estimated by graphical methods of Alexandrowski (1985) or Angelier & Mechler (1979). The majority of authors used computer calculation of the stress axes, however they applied different softwares (e.g. Nem6ok 1993; Ratschbacher et al. 1994). This difference can hardly be avoided and some discrepancy in stress data can be attributed to this fact. In our analysis special attention was given for the timing of each determined stress state. This timing was based on observations at several levels. First, the age of the brittle deformation was interpreted from the outcrop itself, using
299
synsedimentary, syndiagenetic structures, the age of the deformed and the overlying undeformed rocks, relative chronology between structures belonging to separate phases. Then outcrop-scale structures and the calculated local stress field was compared to map-scale structures, to the occurrence of thick sedimentary or volcanic sequences. For example, structures often pre- or post-date the tilting (or folding) of beds, the age of which can be determined from local data. All these data gave lower and upper time constraints for the d e f o r m a t i o n and palaeostress field for a smaller area. The basinwide synthesis is composed from such constrained age determinations. Figure 3 shows an example of this work. The presented Als6told section is composed of tilted layers of Karpatian to Badenian in the footwall, and Pannonian in the hanging wall of an important, map-scale fault (Noszky 1940). Part of the small faults within the footwall block is tilted, they are characterized by N E - S W tension (Fig. 3b, lower left corner). Hence, this stress field is older than the tilting. Conjugate normal and strike-slip faults were formed by E S E - W N W tension, perpendicular to the axis of the tilt. Part of them are symmetrical to bedding thus they are c o n t e m p o r a n e o u s with the tilting, while others (mainly in the P a n n o n i a n ) are posttilting. The deformed Mid-Miocene to Pannonian rocks suggest continuous Sarmatian to Pannonian age for the tilting within the same ESE tension. W N W - E S E compression is indicated by few data; their relation to tilting is not clear, but they do not seem to suffer the full amount of tilt. Finally, N N W - E S E compression postdates the tilting, thus represents a Pontian to Q u a t e r n a r y episode. Seismic data from the nearby Zagyva trough impose a younger lower time constraint for this deformation, around the beginning of the Pliocene. Different durations were attributed to the separated tectonic phases (palaeostress fields) in different sub-areas of the basin system and surrounding mountain chain. In fact, these variations partly reflect the dating possibility within the sub-areas, but also reflect real inhomogeneity of the stress field. Except for clear cases (e.g. improvement in dating after publication of the data), the inferred timing of the original authors was kept. Fault slip data and stress field calculations were used for kinematic analysis of Tertiary structures known from earlier works or established during this study. A t first, we made attempts to directly determine the kinematics of some map-scale faults by field observations. G o o d outcrop-scale kinematic observations
300
L. FODOR E T A L .
were often extrapolated to nearby faults where direct determination of their nature was not possible. Furthermore, the knowledge of the stress field permitted an estimation of kinematics of faults below the thick Q u a t e r n a r y or Neogene sedimentary cover in the deeper parts of the Pannonian Basin. In such areas, we used also published tectonic data (i.e. structural maps, description of folds and faults without stress calculations). In addition, seismic reflection sections, borehole data, gravity map, information on offset paleogeographic boundaries, depositional environments and geometry of the basin fill were also taken into consideration. The results are presented in a series of maps, together with the stress field data. Reflecting our present knowledge, the structural maps show different levels of accuracy and details. Because of the different early Tertiary tectonic history, we discuss the observed stress field and fault p a t t e r n separately for the Alcapa and Tisza-Dacia units for the Palaeogene to the Mid-Miocene. Late Early Miocene was the time of major and opposite sense rotations of these units, and from latest Early Miocene one can consider a more or less uniform Carpathian-Pannonian domain.
(a)
Pre-rift events (Eocene to Early Miocene) Alcapa: repeated dextral transpression E o c e n e to early O t t n a n g i a n Eocene to early Oligocene transpression (Figs 4a, 5a) The oldest Tertiary stress field was determined in the Alcapa terrane. The orientation of the stress axes are similar in the different subareas, o-1 was horizontal and oriented to N W - S E in the Eastern Alps (Nemes et al. 1995b); (Decker et al. 1993, 1994; Linzer et al. 1995). In the Transdanubian Range 0-1 varies slightly from W N W - E S E to N W - S E (Fodor et al. 1992; Bada et al. 1996) (Fig. 5a). This direction was also determined in the westernmost Carpathians (Marko et al. 1990, 1995), and similarly oriented 0.Hmax (with vertical o-1) is postulated in the Central C a r p a t h i a n Paleogene Basin of the Western Carpathians (Jablonsk3~ et al. 1994). In the easternmost Alps the characteristic structures are E N E to N E trending thrust or fold axes. NW trending dextral faults could reflect oblique thrusting (Linzer et al. 1995). Such contractional d e f o r m a t i o n started in the Cretaceous, but the deformed Senonian to Lower Eocene sediments demonstrate the continuation during the early Palaeogene.
PANNONIAN BASIN PALAEOSTRESS Late KarpatianMid-Badenian 17 - 15.5 M a
DATA
Late Badenian Sarmatian
Latest SarmatianEarly Pannonian
Late Miocene
15.5 - 11 M a
11 - 9 M a
9 - 6 Ma Post-tilt e x t e n s i o n
301 PlioceneQuaternary 6 - 0 Ma Second inversion
First i n v e r s i o n
S e c o n d rifting & transtension
'
"
.~
g .
second 15 ~ tilt
;::=
,-...a
First rifting p h a s e
first 15 ~ tilt
(b) Fig. 3. E x a m p l e of separation of tectonic phases (Als6told, Cserhfit hills, N E Hungary). (a) Geological crosssection, partly after the map of Noszky (1940). N o t e the eastward tilt of the sequence. (b) Stereographic projections of m e a s u r e d faults and calculated stress axes. Pre-tilt (original) and post-tilt (recent) positions are equally shown. D o t s represent slickenside lineations on fault planes; d o u b l e arrows: strike-slip; centrifugal arrows: normal faults (motion of hanging wall); o-1, 0-2, 0-3 stress axes: stars with 5, 4, 3 branches.
302
L. FODOR E T A L
In the Buda and Gerecse Hills the best documented map-scale structures are E-W-trending dextral strike-slip faults, the Nagykovficsi and Buda6rs zones (Maros 1988; Balla & Dudko 1989), whose separation was partly accommodated by the blind Buda imbricate stack (Fodor et al. 1994) and by reverse faults and folds east of the Danube (Csontos & Nagymarosy 1998). Some en echelon normal faults are also related to strike-slip faults. On the basis of fault slip data, dextral reactivation was postulated just north of the Gerecse, along the HurbanovoDidsjen6 line. In these areas the original palaeogeography, structural pattern was reconstructed on the basis of sedimentary dykes, synsedimentary faults and tilting, different sediment thickness and facies, fault-related talus cones, bioperforated fault planes, soft-sediment deformations, different types of gravity flows (F~iy-Tfitrai 1984; Fodor et al. 1992, 1994; Bada et al. 1996; Sztan6 & Fodor 1997). Some new observations suggest that faulting can slightly predate Mid-Eocene sedimentation which started on an already dissected topography (Kercsm~ir 1996). The observations were extrapolated to the central and southwestern Transdanubian Range where measurements are scarce but W N W - E S E compression was determined in some Senonian sites (Fig. 5a). The NE-trending occurrence of Eocene depressions were probably bordered by reverse faults and segmented by E - W dextral faults (Fodor et al. 1992). Activity of NE-trending (reverse) faults might be indicated by northwestward thickening Eocene sediment wedges (Mindszenty et al. 1988). Dextral transpression occurred also in the westernmost Carpathians (Plagienka 1991; Marko et al. 1990). Geological maps and some scattered microtectonic observations suggest a similar scenario in the Stil'ov Palaeogene Basin (Salaj 1995; Marko & Fodor unpublished data 1991) while the Central Carpathian Palaeogene Basins were probably characterized by WNWtrending normal faults (Kovg~, Marko, Fodor unpublished data 1994). This phase of transpression started before the Eocene (Plagienka 1991) and the main phase occurred during the Oligocene (Marko et al. 1990). L a t e s t O l i g o c e n e to Early M i o c e n e escape tectonics (Figs 4b, 5b). The Early Miocene (late
Egerian-early Ottnangian) stress axes did not deviate significantly from the Eocene ones in the main part of the Western Carpathians and in the Pannonian region (Figure 5b). N W - S E or W N W - E S E o"1 and perpendicular o-3 reactivated E-W (ENE-WSW) trending dextral,
NNW-SSE- to NNE-SSW-trending sinistral and reverse faults. The separation of this event from the previous one is possible because of good synsedimentary evidence for the older event or, locally by the existence of marked erosion before Eggenburgian times. In the Eastern Alps a new, strike-slip-type stress field was established, whose o1 was oriented around N-S (locally slightly deviating to NNW-SSE or NNE-SSW). Corresponding to this, conjugate set of strike-slip faults developed (Decker et al. 1993). The main structures are NE- to ENEtrending sinistral fault zones within the Northern Calcareous Alps and northward thrusting in the marginal Flysch and molasse belts. (Fig. 5b). However, the time constraint is relatively poor concerning the beginning of this stress field (Decker et al. 1993; Linzer et al. 1995). The eastern margin of Inner Western Carpathians was also characterized by a different 0-1 direction, N-S to NNE-SSW (Fig. 5b) (Nem~ok et al. 1993). This stress field might induce dextral transpression along the eastern Pieniny Klippen Belt (Ratschbacher et al. 1993a) and thrusting within the Carpathian flysch belt. However, it is difficult to date the stress data, and the phase could have started later (during the MidMiocene). The main deformation of the whole area is due to the dextral (re)activation of the Periadriatic line. Its general dextral character is well documented in the Western Alps (Laubscher 1988; Schmid et al. 1989). Recently, Nemes et al. (1995a) and Fodor et al. (1998) offered further structural evidence for dextral shear at the eastern end of the fault system in Slovenia. To follow the continuation of the Periadriatic line into the Pannonian area, one has to reconstruct the large scale CCW rotation of the Alcapa domain. Before Ottnangian time the southern border of the Alcapa, the MidHungarian shear zone was oriented NW-SE, thus forming a straight eastward continuation of the Periadriatic fault system. The kinematics and amount of dextral separation and timing of the motion of this formerly unique fault system was extensively discussed by Balla (1988), Balla & Dudko (1989), Csontos et al. (1992), Tari (1994) and Tari et al. (1995). The key point in their reconstruction is the similarity between the Dinaric and Bt~kk-type late Palaeozoic, the South Alpine and Transdanubian Mesozoic, the Hungarian and Slovenian-Italian Palaeogene developments. The similarity of these presently distant localities were already discussed by Ldczy (1913) and more recently by Premru (1981), Kfizm6r & Kovfics (1985), Ebner et al. (1991), Schmidt et al. (1991), Csontos et al.
PANNONIAN BASIN PALAEOSTRESS DATA
Fig 4a
303
304
Fig 4b
L. F O D O R E T A L .
PANNONIAN BASIN PALAEOSTRESS DATA
Fig 4c
305
306
Fig 4d
L. F O D O R E T A L .
P A N N O N I A N BASIN P A L A E O S T R E S S DATA
Fig 4e
307
308
Fig 4f
L. F O D O R E T A L .
PANNONIAN BASIN PALAEOSTRESS DATA (1992) and Haas et al. (1995). The former unity of the Slovenian and the North Hungarian Palaeogene Basins was inferred by Bfildi (1986) and Nagymarosy (1990). The strong deformation by subsequent Miocene phases overprints the Early Miocene internal structure of the Mid-Hungarian shear zone. The stress data are, however, in accordance with dextral slip along this fault system. Some other kinematic data are known along the northern border of the shear zone. South of the Velence hills dextral faults were mapped in the subsurface which displaced Eocene rocks (Dudko et al. 1989). These structures are supposed to continue in the Tdalmfis zone (Csontos & Nagymarosy 1998) which determines the location of an Egerian-Eggenburgian delta (Sztan6 1994). South of the Lake Balaton exotic duplexes of the Palaeogene Basin were found (see on Fig. 4a, after Sztr~kos 1975; Balla et al. 1987; Nagymarosy 1990; K6r6ssy 1990). These duplexes were formed during the separation of the formerly unique Hungarian-Slovenian Palaeogene Basin. Both surface observations (in Slovenia) and borehole data (in Hungary) suggest that the Slovenian and Hungarian Basin fragments were separated before the Karpatian, due to Early Miocene dextral slip (Balla & Dudko 1989; Fodor et al. 1998). Within the inner part of the Transdanubian Range the activity of E - W trending dextral faults was demonstrated. In the Buda and Gerecse hills dextral slip reactivated Eocene faults (Fodor et al. 1994; Bada et al. 1996). Such faults separate areas with thick and thin upper Oligocene series, suggesting younger, Early Miocene displacement (Nagykovficsi line, Balla & Dudko 1989). Upper Oligocene sandstones were dragged into tectonic lenses within fault zones during Early Miocene slip (Gyalog 1992). Normal faults belonging to later rifting phase displace some dextral strike-slip faults, thus their activity ended by the Karpatian. In the Western Carpathians N W - S E compression influenced the Eggenburgian sedimentation (Kovfi~ et al. 1989). In fact, the main boundary faults of the depressions seem to reactivate Oligocene transpressional structures, but a marked discordance indicates Early Miocene reactivation, too (Kovfi~ et al. 1990). In contrast to dextral transpression, different kinematics occurred along the Darn6-zone.
309
There anticlines and reverse faults were mapped in Egerian to Eggenburgian rocks and demonstrated on seismic lines (Fig. 4b), (Telegdi-Roth 1951; Sztan6 & Tari 1993). In addition to shortening, the en echelon geometry and oblique orientation of structures to the compression may suggest slight sinistral component of slip. The deformation is well-dated along the zone. The first step might be indicated by the erosion of the thick Kiscellian series in the hanging wall of the reverse faults before the beginning of Eggenburgian sedimentation, during late Egerian (Bfildi 1986; Nagymarosy 1990). On the downthrown side the depositional geometry of the Eggenburgian sediments suggest that the zone represented the easternmost boundary of an embayment (Sztan6 1994). The tectonically controlled shore is marked by talus cones, fan deltas, liquefactions and gradual displacement of the source area of the fan deltas (Sztan6 & Jdzsa 1996). In the northern part, similar mismatch between the source area and pebble material of deltas was demonstrated (Szentp6tery 1988). Slices of Eggenburgian rocks incorporated into faults suggest postsedimentary continuation of the motion. However, the bulk of the displacement predates the first rhyolite tuff horizon (19 Ma; Rad6cz 1966) and the Ottnangian coal series seals the compressional structures.
T h e T i s z a - D a c i a unit: E o c e n e t h r o u g h Early Miocene compression
The earliest Tertiary deformations in the Tisza-Dacia block were found in northern Transylvania. Here Eocene to Early Miocene rocks are exposed along the margins of Mesozoic to crystalline outcrops. These sediments are considerably deformed at the marginal parts, while they are flatlying towards the centre of the basin. The deformed outcrops show polyphase tectonism, first characterized by an E N E - W S W to N E - S W compression, followed by a NNW-SSE to N W - S E compression (Fig. 5). Thrust faults are dominant, but strike-slip faults are also abundant. In some outcrops folds are also seen. Eocene to Early Oligocene rocks are affected by both phases, while Early Miocene rocks record only the second phase. Local map-scale structures range from folds to thrust faults. Two
Fig. 4. Structural evolution and basin formation in the Pannonian Basin and adjacent mountain ranges through six episodes. Structural geometry is more detailed in the Pannonian part and schematic within the Alpine and Carpathian ranges. Main sedimentary and volcanic areas are stippled. References for structural geometry are discussed in the text.
310
Fig. 5(a)
Fig. 5(b)
L. FODOR
ETAL.
PANNONIAN BASIN PALAEOSTRESS DATA
Fig. 5(c)
Fig. 5(d)
311
312
Fig. 5(e)
Fig. 5(f)
L. FODOR E T A L .
PANNONIAN BASIN PALAEO STRESS DATA
313
Fig. 5(g)
Fig. 5. Stress data of the Pannonian-Alpine-Carpathian region. Only the maximal horizontal stress axes are represented. Large arrows indicate the general trend of stress type (compressional, strike slip, tensional). Geological outlines after Mahel et al. (1984). Sources: Decker et al. (1993); Fodor (1995); Nem~ok (1993); Nem~ok et al. (1993); Nem~ok & Lexa (1990); Ratschbacher et aL (1993a, b); Fodor et al. (1992, 1994); Vass et al. (1993); Csontos & Bergerat (1992); Gy6rfi et al. (this volume); and own unpublished data.
major thrusts are known: the Meses thrust putting Late Cretaceous granites on top of overturned Eocene, deposits and the Botiza area, where an imbricate sheet of Mesozoic to Eocene rocks is thrust u p o n Late O l i g o c e n e - E a r l y Miocene rocks. The Meses thrust is an oblique transpressional zone related to the first stressfield, while the Botiza imbricate stack was created by the second deformation (Fig. 4a, b). Timing of the deformation is partly ascertained using superposition, partly from rocks not showing earlier fault sets. Dating was complemented by considering the sedimentological features of the Transylvanian and Maramures basins (Gy6rfi et al. this volume). We inferred that the N W - S E - o r i e n t e d Transylvanian Palaeogene Basin was created by compression. One potential thrust responsible for the flexure of the basin is found at the Meses Mts, while other supposed faults remain hidden beneath the thick Mid- to Late Miocene fill of the Transylvanian Basin. A n Early Oligocene age was suggested for this deformation. The E - W oriented Late Oligocene-Ottnangian turbiditie
Borsa and Early Miocene Ottnangian-Karpatian coarse clastic Hida Basins were supposedly created by the Botiza thrusts and the second, N N W - S S E compressional phase. The age of this deformation should be therefore Latest Oligocene to Early Miocene.
Rifting of the Pannonian Basin: latest Early to Mid-Miocene (Figs 4c, d, 5c, d) Easternmost
Alps
Late Early and Mid-Miocene times represent the major rifting period in the Pannonian region (Horvfith & Royden 1981; Tari 1994). The onset of rifting corresponds to a pronounced change in stress field and tectonic evolution was complex in different sub-basins. In contrast, the easternmost Alps experienced a relatively simple evolution: the same stress field established for the latest Oligocene and earliest Miocene seems to continue during the rifting of the Pannonian Basin. In the Eastern Alps ~rl was around N-S
314
L. FODOR E T A L .
and o-3 was typically horizontal (Figs 4c, d, 5c, d). However, the earlier frontal thrusting onto the European foreland ceased and the new structures are conjugate strike-slip faults and related transtensional and transpressional features (Ratschbacher et al. 1989, 1991a, b; Decker et al. 1994; Linzer et al. 1995; Nemes et al. 1995a, b). The main subsidence is Karpatian-early Badenian within the large Styrian (Ebner & Sachsenhofer 1995) and other small pull-apart basins (Ratschbacher et al. 1991a). Pannonian
Basin-Carpathians
East of the Vienna and Styrian Basins, the Pannonian Basin system and the internal part of the Carpathian orogen were characterized by a tensional, locally transtensional stress field. The direction of the minimal stress axis (03) varied in time and space between NE-SW and NW-SE (or even NNW-SSE). These varying directions can be generally separated into two distinct events. NE-SW- to ENE-WSW-tension-generated structures, which are systematically older than those characterized by E - W to SE-NW tension. Recent research by Fodor et al. (in prep.) in North Hungary demonstrates that the change between the two stress fields did not occur at the same time in different basins, thus the stress field remained inhomogeneous during the Mid-Miocene. First, we describe these two stress fields and then discuss their timing within the sub-areas. N E - S W to E N E - W S W tension: stress .field a n d fault characteristics (Figs 4c, d, 5c, d). The first
phase of rifting was characterized by NE-SW to E N E - W S W tension or strike-slip type stress field with O'HmaxNNW-SSE to NW-SE throughout the whole Pannonian area (Fig. 5c). The main structures are normal and normal-oblique faults with orientations varying from NW-SE to NNE-SSW, but the main direction changes from one sub-basin to another. E-W- to WNW-ESEtrending dextral and N-S- to NE-SW-trending sinistral faults are also common. In the area of the Danube Basin and its southeastern rim, the normal faults are trending NW-SE or NNE-SSW. The larger ones are lowangle, determined by seismic sections (Tari et al. 1992). Fission track ages from the Rechnitz and Inovec hills demonstrate that exhumation of the footwall and low-angle normal faulting were coeval (Dunkl & Dem6ny 1997; Kov~i~ et aL 1994). Transition from ductile to brittle deformation around 20 Ma (Ratschbacher et al. 1990) also corroborates this relationship. These data argue for the formation of the Rechnitz metamorphic core complexes due to simple
shear extension (Tari et al. 1992; Gy0rfi 1992; Tari 1996). Although the connection of the exhumation of metamorphic rocks and lowangle normal faulting was not clearly demonstrated at other locations of the study area, we tentatively suggest the existence of several similar core complexes (Fig. 4c). High-angle normal faults often border a series of tilted blocks or synsedimentary half grabens, mainly trending NW-SE. Such blocks constitute the Zala Basin, the southern and central Bakony, the main part of the Gerecse and Buda hills, the subsurface G6d6116 area, the Etes graben and a number of depressions within the Western Carpathians (Fig. 4c). NW-trending border faults limit the depressions between the Btikk hills and the Tokaj hills. The Eastern Slovakian Basin itself may also be such a large graben, although moderately dipping boundary faults might have dextral component too (Tomek & Thon 1988; Kovfi6 et al. 1995). Highangle normal faults probably flatten to low-angle detachment faults, like near the Rechnitz windows (Gy6rfi 1992; Tari 1996). The system of large, NW-SE-oriented normal faults and rotated half-grabens continued south of the Mid-Hungarian zone, both in the buried areas and in the exposed ranges. Larger and very deep basins are found in SE Hungary, in the Danube-Tisza interfluve and immediately east of the Tisza river (Posgay et al. 1995). Here the H6d and B6k6s Basins (Fig. 4c) reach at present 7 km depth (however large part of the subsidence occurred during the post-rift phase). Basins of similar orientation and bordered by N W - S E normal faults flank the western Apuseni Mts (Gytirfi 1993). These basins are bordered by NE-dipping low-angle normal faults (Gy6rfi & Csontos 1994), the footwalls of which were uplifted synchronously (fission track data of Dunkl pers. comm. 1993; Posgay et al. 1995). In the southwestern Pannonian Basin, the Drava and Sava grabens can also belong to the system of tilted blocks (Prelogovid et al. 1995; Tari & Pami6 1998). Border faults often have strike-slip or obliqueslip thus the depressions have transtensional character (Fig. 4c). One of the best documented example is the Vienna Basin where fast subsidence occurred between sinistral and normal border faults of a rhombohedral pull-apart depression (Royden 1985; Fodor 1995). The sinistral nature of the southeastern border faults is also indicated by displaced talus cones (Vass et al. 1988b). In the Southern Carpathians, near the margin of the Tisza-Dacia block E - W trending dextral shear zone developed (Ratschbacher et aL 1993b; Rabagia & Fiil6p 1994).
PANNONIAN BASIN PALAEOSTRESS DATA Within the Pannonian Basin, some other fault zones had strike-slip or oblique-slip character. A set of en echelon half grabens are situated within the sinistral Darn6 shear zone of North Hungary (Bergerat et al. 1984; M~irton & Fodor 1995). The Bakony hills are dissected by WNW-ESEtrending dextral faults (M6sz~iros & T6th 1981; M6sz~ros 1982) which border small Miocene Basins near Herend and V~irpalota (K6kay, 1976; Balla & Dudko 1989). The western segment of the Hurbanovo zone could also have dextral character, suggested by nearby mapscale faults (Bence et al. 1991) and fast subsidence in the Zeliezovce sub-basin (Lankreijer et aL 1995). E - W to S E - N W tension (Figs 4d, 5d). The second phase of rifting is mainly characterized by tensional or strike-slip type stress field. The direction o-3 is variable, from E - W to SE-NW (Fig. 5d). This difference is interpreted as variation within a slightly inhomogeneous stress field. This stress field created new faults but often reactivated with different kinematics the faults of the first rifting phase. The NW-SE-trending faults became dextral strike-slips, the NNE-SSWtrending strike-slips changed to normal or normal-oblique faults. One of the best documented examples is the Vienna Basin which lost its pull-apart character when the stress field and, thus, the strike-slip kinematics of the border faults changed (Fig. 4c,d; Fodor 1995). A similar phenomenon was demonstrated in basins of the westernmost Carpathians (northern branches of the Danube Basin) which were essentially bordered by N-S to NE-SW-striking normal faults ( K o v ~ et al. 1990; Kovfi~ & Barfith 1995). Faults of the second phase are clearly superimposed on the first generation from the Gerecse to Bt~kk hills (Fodor et al. in prep). The separation along the NW-SE-trending older faults decreased, they are reactivated as dextral oblique faults or are cut by the new, N-S- to NNE-SSW-trending fault sets (Fig. 4d). The slip along the border faults resulted in a second tilt episode in ESE or WNW directions. Along the Mid-Hungarian shear zone depressions are mainly trending ENE-WSW. Seismic sections show a number of criteria which is typical for strike-slip fault. Tari (1988) used such criteria and demonstrated sinistral strike-slip faults in the northern Vatta-Maklfir trough which is in good agreement with surface measurements (Csontos 1988; M~irton & Fodor 1995). The system of ENE-trending sinistral strikeslip faults and rhomb-shaped depressions frequently occur in the southern Pannonian Basin.
315
Their pull-apart origin was already suggested by earlier publications (Rumpler & Horv~ith 1988). In the western side of the Apuseni Mts the N W - S E extensional stress-field did not create new major normal faults but the older N W - S E faults were reactivated as strike-slip faults. Microtectonic data are recorded in Badenian sediments (Gy/3rfi 1993). In the Mecsek hills a strike-slip type stress field was recorded in Early Miocene rocks. It is characterized by N-S o-1 and perpendicular o"3 directions. Major structures are N E - S W sinistral and N N W - S S E dextral faults. Sinistral faults are connected to E - W en echelon folds (Benkovics 1997).
T i m i n g a n d tentative separation o f rifting phases In this section we analyse the time constraints for the onset of the first rifting episode and its change to the second tensional stress field. This overview is presented from west to east, starting from Vienna Basin. During the Karpatian and Early Badenian continued thrusting in the northern Outer Carpathians permitted the sinistral opening and displacement transfer from its southeastern to the northwestern boundary fault. However, from the middle Badenian onward, thrusting ceased in the western sector of the Western Carpathians (Ji~'i6ek, 1979), thus sinistral shift was maintained only along the SE boundary fault. The change in stress field occurred contemporaneously with this migration of thrust activity, after Early Badenian (Fodor 1995; Marko et al. 1995). The Ottnangian initiation of sedimentation around the Rechnitz windows is in good agreement with the time of its exhumation (20-18 Ma). The formation of the Styrian and Sopron basins can thus be connected to the first phase of rifting (Tari 1994). In fact, at the western edge of the Danube Basin, only the Ottnangian sediments seem to be affected by E N E - W S W tension, the lower Badenian rocks show only the effect of the younger stress fields (Fodor 1995). A pronounced unconformity occurs within the Badenian series of the Danube and Zala Basins, on the hanging wall of the large low-angle detachment fault above the Rechnitz windows (Horvfith et al. 1995). This unconformity can be correlated to the change in the stress field during the middle Badenian. Stress field change is well dated in the central and northeastern Pannonian Basin and Western Carpathians. The oldest, synsedimentary activity is indicated by different thickness and
316
L. FODOR E T A L .
facies development within the Darn6 zone and the Etes trough, starting already in the Ottnangian (H~imor 1985; M~irton & Fodor 1995). Differential subsidence continued through Karpatian and earliest Badenian. Early Badenian initiation of sedimentation is somewhat younger in the Zeliezovce depression of the Danube Basin (Kov~i6 et al. 1993; Lankreijer et al. 1995). Outside these zones, the initiation of the N W - S E grabens are less well constrained, because they are superimposed on Egerian to Eggenburgian rocks and Ottnangian to Karpatian strata are missing. The earliest indication of change in the stress field occurs in the subsurface southern Gerecse and Buda hills (Fig. 4c, d). Here Karpatian(?)-Lower Badenian clastics are situated in grabens which are discordantly covered by Middle to Upper Badenian sediments (K6kay 1989). These latter sediments were already deposited in new N-S-trending grabens (K6kay 1990). Thus the rotation of tension from NE to E had already occurred before or within the Middle Badenian (15-14 Ma). Synsedimetary tectonism of the neighbouring Budajen6 trough and south Buda hills give also good time constraint for the onset of the second tension. Both the Upper Badenian and Sarmatian limestones are dissected by synsedimentary dykes and microfaults (Bergerat et al. 1983, Bada et al. 1996). Sediment transport direction points in each case to the direction of gravity lows, from the master faults or tilted block edge toward basin centres. This suggests that the shape of the depressions were formed by late Badenian tectonism and kept during the Sarmatian. Considerable thickness variation of nearshore limestone versus basin marl indicate amplification of master faults during Sarmatian sedimentation. Eruptions of voluminous stratovolcanoes of the B6rzsOny, M~tra hills and the Kremnicka Planina highland started during Early Badenian (Kone6ny & Lexa 1984; Vass et al. 1979) and can be connected to the first rifting episode. Some sites of these volcanoes are still deformed by the earlier N E - S W tension (Bence et al. 1991; Kov~i6 & H6k 1993; Vass et al. 1993) while other sites bear the trace of only the younger deformation. The 15 Ma Karancs subvolcanic andesite body (Orlicky et al. 1995) is deformed by NE-SW tension. Intrusion of the NW-SE-trending dykes of the Cserh~t (14 Ma, P6cskay et aL 1995) may have also been facilitated by N E - S W tension. However, they are dragged by E - W sinistral faults in a ductile manner immediately after intrusion, before complete cooling (Balla 1989). Looking to the coeval sediments, the latest Early
Badenian limestone ('lower Leithakalk') has the trace of both stress field, while Upper Badenian to Sarmatian limestones were already affected only by the second tension. Seismic sections (Benkovics 1991; Tari et al. 1992) demonstrate that the Zagyva and Tura master faults were active during the late Badenian-Sarmatian sedimentation (Fig. 4d). Borehole data and surface thickness and facies (H~imor 1985) suggest that late Badenian to Sarmatian sediment wedges are progressively thickening toward the master faults while their facies is generally changing from carbonate rocks on the edge of the tilted blocks to (fine) clastics basinward. All these data suggest that between the Gerecse hills and Darn6 zone the change in stress field can be placed at the beginning of or within Late Badenian (14-13 Ma). In the central Western Carpathians NE-SW tension was replaced by E - W tension during the Late Badenian and Early Sarmatian. This tension changed again to N W - S E tension at the end or after the Sarmatian (Nem6ok & Lexa 1990; H6k et aL 1995). The first change of tensional direction is coeval with similar changes just south of this region, in North Hungary. The timing markedly changes in the Tokaj-Slanec hills. Here the stress evolution has three stages. The maximal horizontal stress axis of the first event was oriented WNW-ESE, then NNW-SSE, and finally NNE-SSW. The first stress field was strike-slip type, while the two younger were tensional ones. The lower part of the volcanic suite (upper Badenian to lower Sarmatian) was deformed by three different stress fields, while the upper volcanic suite (upper Sarmatian-lowermost Pannonian) was deformed only by the youngest tension. Consequently, a change in stress field occurred during the first half of the Sarmatian, and another one only at the end of this period. Similar statements can be made using dykes: the older ones are trending N W - S E , while progressively younger ones changed in direction to N-S then to N E - S W (Gyarmaty et al. 1976; Moln~r 1994). One uppermost Sarmatian site shows synsedimentary faults trending NE, fitting to a latest Sarmatian onset of the youngest, NE-SW tension. Palaeomagnetic data suggest 30-35 ~ CCW rotation which occurred within the Sarmatian (M~rton & P6cskay 1995; Orlicky 1996). We tentatively correlate this rotation with the older change of stress field. This change can be only apparent, because by eliminating the rotation, the measured W N W - E S E maximal stress axes turned back to original NNW-SSE. Then the same NNW-SSE compression and perpendicular tension continued, without rotation. The best
PANNONIAN BASIN PALAEOSTRESS DATA structural expression of the stress field is the fast subsidence along NW-SE faults in the neighbouring East Slovakian Basin (Fig. 4d), (Vass et al. 1988a; Tomek & Thon 1988; Rudinec 1989; Kovfi~ et al. 1995). All data suggest that the first rifting phase terminated gradually from the western to the northeastern Pannonian Basin. The transition occurred at the beginning of or during the MidBadenian west of the Danube, during the Late Badenian east of the Danube, around the Zagyva graben, and only in late Sarmatian in the Tokaj-Slanec hills and East Slovakian Basin. Within the Tisza-Dacia unit no such detailed chronology can be given. In seismic sections of the Great Hungarian Plain the change from NE-SW tension to NW-SE tension can be put somewhere in the Badenian. Syn-rift deposits in NW-SE-oriented grabens are pre-Mid-Badenian in age, while syn-rift deposits of NE-SW tilted grabens and pull-apart sub-basins are Badenian (Gy6rfi 1993; GyOrfi & Csontos 1994). In the western flanks of the Apuseni Mts the change in the stress-field orientations occurs in the Mid-Badenian. Older deposits are affected by both tensional phases, while Mid- to Late Badenian and Sarmatian sediments are affected by N W - S E and NNW-SSE tension (GyOrfi 1993).
Early stage basin inversion (earliest Late Miocene) (Fig. 5e) ENE-WSW or E - W compression at about 10 Ma was determined in several sites of the Eastern Alps (Decker et al. 1993; Linzer et al. 1995; Peresson & Decker 1997). The stress field was clearly demonstrated and separated from other phases in the eastern border of the Vienna Basin (Fodor et al. 1990). Some conjugate strike-slip faults in Hungary are also characterized by N E - S W to ENE-WSW oriented o-1 (Fig. 5e). The demonstration is relatively clear in sites where both the ESE-WNW tension and ENE-WSW compression are present (Fig. 3.). Where only NE-SW oriented o.1 was calculated, the faults might be attributed to this phase but they may also be considered as local deviation of the MidMiocene E S E - W N W tension. In this case, the occurrence of such stress data would be attributed to variation in magnitudes of the stress axes (change from tensional to strike-slip or even compressional stress field). The main microtectonic features are the inversion of sense of NNE-SSW-trending strike-slip faults, from sinistral to dextral. However, this inversion was
317
never large enough to inverse offsets on mapscale. Steep reverse faults are sometimes associated to these strike-slip faults. We ranged such determinations into an individual inversion phase (with E S E - W N W o-1) only when additional time constraints, mainly relative timing with respect to other stress fields were present. Such examples from South Biikk occurred obviously in tilted position, which suggest a pre-Pannonian age. In the Cserh~it this phase occurs after the bulk of tilting (Middle Miocene, Fig. 3.). This inversion phase is relatively clearly visible on seismic sections in the southern Pannonian Basin and south of Budapest, where inversion of the former basins occurs (Horvfith 1995). This inversion is coupled by the erosion of the Sarmatian deposits and non-deposition of the Early Pannonian on the emerged blocks. In the Eastern Alps this phase was determined only in Mesozoic and Senonian rocks thus the time constraint is not really good (Decker et al. 1993; Peresson & Decker 1997). In the eastern Vienna Basin and northern Hungary, Badenian and occasionally Sarmatian sediments were also deformed (Fodor et al. 1990). The relative chronology of this stress field with respect to mid-Miocene stress field and tilting would suggest post-Mid-Miocene age in the central Pannonian area. The general lack of this deformation in Upper Miocene rocks and some relative chronological criteria with respect to Late Miocene deformation, bracket this phase around the turn of Mid- and Late Miocene, at latest Sarmatian or earliest Pannonian. However, some sites might allow other timing as well, e.g. a site in the eastern Vienna Basin would suggest considerably younger, Pontian age (Sauer et aL 1992; Peresson & Decker 1997). In the eastern Pannonian Basin and western Transylvanian Basin the orientation of o-1 did not change during the late Mid-Miocene-Late Miocene, thus this event does not appear as a separate phase. Also, in the central Western Carpathians the NE-SW maximal horizontal stress axes (strike-slip type stress) were not changed through the Middle-Late Miocene boundary.
Renewed extension and strike-slip faulting (Late Miocene) (Figs 4e, 51') After the main Mid-Miocene extension, the Pannonian Basin system was influenced by thermal subsidence (Horv~th & Royden 1981; Royden et aL 1982). However, E - W to SE-NW tension was determined in Sarmatian and Pannonian
318
L. FODOR E T A L .
sedimentary rocks around several sub-basins in the western and central Pannonian Basin and Western Carpathians (Fig. 50 . This stress field is similar to that of the Mid-Miocene and demonstrates the renewal of tectonic activity after the main phase of rifting. Master faults of the western and central Pannonian Basins were reactivated mainly as normal faults with oblique component. Some seismic sections, Tari et al. (1992) suggest that faults frequently do not cut through the lower Pannonian sequence as a single plane, but the deformation was distributed within and sealed by the soft young sediments. In some other places growth in Late Miocene deposits along listric normal faults is observed. Faulting also occurred along newly formed depressions, where no previous sedimentation took place (depressions of eastern V6rtes, western Gerecse, the Aggtelek hills; Grill et al. 1984). Some NE- to ENE-trending sinistral faults were also renewed. Such reactivation is indicated by important sediment thickness and en echelon geometry of depressions in the Vienna Basin (Wessely 1988; Hamilton et al. 1990; Fodor 1995). Strike-slip faulting of the southern boundary zone of the North Pannonian unit was also renewed and was associated with transpressional accommodation structures (Fig. 4e). The timing of motion is relatively precise here: Pannonian (T6almfis: Csontos & Nagymarosy 1998), Pannonian to Pontian (Vatta-Maklfir: Tari 1988; Csontos et al. 1991), late Miocene (Szolnok: L6rincz & Szab6 1993). Further southwards, a sinistral transtensional shear zone was demonstrated from Paks to Szolnok (Pogfics~s et al. 1989). Locally deep pull-apart basins were described in the Derecske region (Rumpler & Horvfith 1988), in the southern part of the Danube-Tisza interfluve. All these basins were created by E N E - W S W - o r i e n t e d left-lateral shear zones. In North Transylvania an important E-W-oriented left-lateral fault can be observed (Dicea et aL 1980) with Late Miocene activity (Gy6rfi et al. this volume). Sarmatian to Pannonian volcanic rocks are deformed and partly intrude the fault. Due to later deformation, Late Miocene structures are poorly identified in southern Transdanubia. The important thickness variation of Pannonian sediments suggests continuing tectonic subsidence in the Drava and Sava grabens.
Late-stage basin inversion (Pliocene through Quaternary) (Figs 4f, 5g) The direction of the maximal stress axes are varying from NW-SE to N-S (Fig. 5g). The
stress field is strike-slip type or compressional at the Alpine-Dinaric junction and in the northern Pannonian Basin and mainly tensional in the eastern-central part. This stress field is similar to the recent one (Gerner 1992; Gerner et aL this volume). The transpressional character of the stress field and structures of this phase are well expressed in the southern Pannonian Basin, from NE Slovenia to the Mecsek hills. The WNW (NW) trending dextral and ENE (NE) trending sinistral faults form obtuse angle. Strike slip was associated with open to tight folding, reverse faulting and with steepening of beds, all structures striking parallel to, or situated en echelon with respect to strike-slip faults. This geometry suggests that the strain was probably partitioned between strike-slip and reverse faults or folds, which is typical for transpressional deformation. Dextral transpression was documented along the Periadriatic zone and connected to the Sava folds region (Polinski & Eisbacher 1992; Poljak, 1984; Premru 1976; Fodor et al. 1998). Similar transpressional tectonics was recently documented in the Papuk Mts by Jami6i6 (1995) and in other parts of northern Croatia (Prelogovi6 et al. 1995). Here the reverse slip probably reactivated the earlier normal boundary faults of the Drava and Sava grabens (Tari & Pami61998). Along the parallel WNW trending Bab6csa high (K6r6ssy 1989) partitioning of NNE-vergent reverse and strike-parallel dextral slip is postulated. In the region, where the Periadriatic line changes their strike to E-W, folding and reverse reactivation of earlier normal faults was described (Rumpler & Horv~ith 1988; Pogfics~is et al. 1994). Between Zagreb and the Lake Balaton Middle to Upper Miocene sediments are folded with ENE-WSW-oriented fold axes. Folding affected Pannonian to early Pontian sediments below the Holocene mud of the Lake Balaton (Sacchi et al. this volume). Strike-slip and reverse faults are well documented on both the northern and southern boundary of the Mecsek hills (Vad~isz 1935; Wein 1967; Csontos & Bergerat 1992; Tari et al. 1992). Accordingly, the whole Mecsek Hills seem to represent one transpressional, inverted pop-up structure (Csontos & Bergerat 1992; Tari 1992). Seismic reflection sections demonstrate that these fault zones are continuing north-eastwards up to the eastern part of the Pannonian Basin (Pog~ics~is et al. 1989). Structures become more transtensional in the northern Pannonian region. Here E - W to E N E - W S W tension was determined in Pliocene basalts and Pannonian sediments (Bergerat & Csontos 1987; Dudko et
PANNONIAN BASIN PALAEOSTRESS DATA al. 1992; Bada et al. 1993; Vass et al. 1993). The
corresponding 0"1 was sometimes horizontal but mainly vertical. There are few strike-slip faults in the central, northern Pannonian Basin and Western Carpathians which are clearly documented during this stage. One example is the southern boundary zone of the Vienna Basin where transtensional and transpressional segments are connected (Marko et al. 1991; Hubatka & Pospigil 1990). In its continuation south-vergent reverse faults were postulated within the Western Carpathians, south of the Tatra Mts (Tomek 1988; Tomek et al. 1989). Some observations on synsedimentary tectonics suggest that the transpression and inversion may have (at least locally) started in the Pannonian (Wein 1967). However, the deformed nature of Pannonian and Pontian sediments support the amplification of deformation toward the end of the Late Miocene and the beginning of the Pliocene. Along the Periadriatic zone and within the Drava graben borehole and seismic data show that Pliocene strata seal the bulk of the deformation (Vrabec 1994; K6r0ssy 1989, 1990). However, the thick, slightly deformed basin fill of the small Velenje Basin in Slovenia demonstrates the continuation during and after the Pliocene (Brezigar et al. 1987). The eventual connection between the localization of basalt volcanoes, volcanic lines and faulting would suggest Pliocene to early Quaternary age of tectonism. Where Pliocene sediments are lacking, the upper time constraint of the deformation is not good. However, the general lack of Pliocene sediments, the strongly eroded character of Miocene rocks, the close connection of subsurface structure with recent morphology, and earthquakes would suggest that inversion deformation has continued through the Pliocene and Quaternary. The most recent period of the structural history is discussed by Gerner et al. (this volume).
Discussion and geodynamic model The middle Eocene to early Oligocene stress field reflects the Mesoalpine convergence of the European and Adriatic plates and intervening crustal blocks. This convergence resulted in the dextrally oblique nappe stacking of the Northern Calcareous Alps (Linzer et al. 1995). From the Late Eocene, this nappe stack and the Inner Western Carpathian nappes were thrust onto the Alpine and Carpathian Flysch Belt. In the hinterland of the thrust and fold belt, the Hungarian Palaeogene Basin system was formed possibly as a retroarc flexural basin (Tari et al.
319
1993). Due to the oblique convergence, this basin system was also dissected into isolated depressions by E - W dextral tear faults and associated en echelon normal faults (Fig. 6a). Taking into account the Early Miocene rotations, the original orientation of o'1 was N-S in the Pannonian-Western Carpathian part and was different from the Alpine N W - S E direction (Fig. 6a). This reconstruction indicates a slightly divergent pattern of the stress trajectories which can probably be explained by different boundary conditions along the orogenic area: with a relatively free interface east from the Western Carpathian unit. Reconstruction of the rotation of the Tisza-Dacia block also yields an original N-S compression, parallel to that of the eastern Alcapa 0"I directions (Fig. 6a). The Tisza-Dacia block had flysch troughs on its northern and northeastern external parts. The compressive stress field generated the Palaeogene flysch trough of the East Carpathians, which were overridden by the Dacides from the south. Palaeogene deposits between the Apuseni Mts and the East Carpathians can be interpreted as flexural basins on top of and in the hinterland of the prograding East Carpathian arc. On major part of the Tisza-Dacia unit no Palaeogene deposits can be found. There are two exceptions. A small, but relatively deep continental basin near the Mecsek (W6ber 1985) and the more voluminous marine deposits of the Szolnok flysch trough. The former indicates that there was erosion and continental environment on most of the area. The Szolnok flysch is thought to have formed at the contact zone of the Alcapa and Tisza-Dacia units (Fig. 1.). One of the most important Tertiary tectonic process was the juxtaposition of the Alcapa and the Tisza-Dacia terranes. This event was traditionally considered as an effect of the eastward escape (extrusion) of the wedge-shaped Alcapa unit (Fig. 6b), (Balla 1984; Kfizm6r & Kov~ics 1985; Csontos et al. 1992). Alcapa was bordered by a sinistral strike slip zone in the north, a dextral one in the south and a low-angle normal fault at the western side, along the eastern edge of the exhuming Tauern window (Neubauer & Genser 1990; Ratschbacher et al. 1991a). Due to the eastward extrusion, the Alcapa block was thrust upon the Tisza-Dacia unit, observed along the Botiza area (Gy6rfi et al. this volume). The stress trajectories (corrected for the palaeomagnetic rotations) show relatively homogeneous N-S compression in the Alcapa. A fan shaped 0"1 trajectories may occur only at its eastern boundary along the contact with the southern unit (Fig. 6b).
320
L. F O D O R E T A L .
(a)
(b)
PANNONIAN BASIN PALAEOSTRESS DATA
(c)
(d)
321
322
L. F O D O R E T A L .
PANNONIAN BASIN PALAEOSTRESS DATA
323
(g) Fig. 6. Evolution of major blocks of the Pannonian area: tentative model for the connection of stress field and deformation. Compare with Figs 4 and 5. Kinematic reconstruction of major blocks is partly after Balla (1984); Csontos (1995); Kov~ et al. (1994), Ratschbacher et al. (1991a). Numbers along the thrust front indicate the age of the last thrust in Ma, after Ji~f~ek (1979).
The final exhumation of the Tauern window can be connected with this process (Ratschbacher et al. 1989; Neubauer & Genser 1990), which could have started in the latest Oligocene (late Egerian). However, the bulk of displacement occurred only at the end of Eggenburgian or in the early Ottnangian, indicated by the disruption of the formerly unique Hungarian and Slovenian Palaeogene Basins (Csontos et al. 1992; Tari et al. 1995). Sinistral strike-slip displacement along the northern boundary zone of the Eastern Alps is less than 100 km which is in striking contrast with the more important apparent dextral separation along the Periadriatic line, between the Southern Alps and the North Pannonian area. This is estimated between 300 to 400 km by Tari (1994). The solution is probably given by Tari et al. (1995) who emphasized that the dextral shift is the sum of the northwestward motion of the Adriatic promontory, the eastward extrusion of the Alcapa and the extension of the Pannonian area. During and after the escape/extrusion event
(around 18-17 Ma) the Alcapa and Tisza-Dacia units experienced important rotations, counterclockwise and clockwise sense, respectively (Mfirton & Mauritsch 1990; Mfirton & Mfirton 1996). Following Balla (1984) and Csontos (1995) our tectonic reconstruction incorporate the assumption of rigid-body rotations despite that small-scale domino-type block rotations are not excluded at some regions (Marko et al. 1991). Due to these rotations, the formerly linear P A L - M H Z was bent and the subsequent Mid-Miocene dextral shift along the southern boundary of the extruding Eastern Alps was transferred from the Periadriatic zone to other NW-SE oriented shear zones within the Dinarides. On the other hand, the time of the rotation of the Alcapa unit is very close to the exhumation of the Rechnitz window (17 + 1.9 Ma; Dunkl & Dem6ny 1997). The differential rotation between the Alps and Pannonian domain resulted in the disruption of the formerly unique Alcapa block and the separation of the East Alpine and Pannonian-Carpathian domains.
324
L. FODOR E T A L .
The b o u n d a r y can be put to the Rechnitz detachment fault of Tari (1994) at mid-crustal level. One consequence of the rotation is the decoupled stress field which was markedly different in the Alps and in the Pannonian Basin from the onset of rifting. The opposite r o t a t i o n of the Alcapa and Tisza-Dacia rises a serious problem, also encountered by Balla (1984). The area between the two blocks has to be reduced considerably (see triangle on Fig. 6b). Since there is no proof or indication of subduction in this sector, we infer that the area, i.e. the portions close to the Mid-Hungarian Zone underwent very intensive deformation. The area in question had to be shortened in N W - S E direction and elongated in a perpendicular direction. We speculate that at least part of the very deep basins along the MidHungarian shear zone were born as a result of major elongation, with the help of NW-SEoriented normal faults. After the rotation, the unified Carpathian-Pannonian block was still bordered by strike-slip fault zones on the north and south, and low angle normal fault at the west (Rechnitz window), so one can speculate on continuing extrusion tectonics (e.g., Decker & Peresson 1996). The rifting of the Pannonian Basin can be explained by the interplay of several driving mechanisms. One c o m p o n e n t is the convergence of the Adriatic microplate and Europe. The effect of this is clearly seen in the relatively stable stress field orientation of the Eastern and Southern Alps and along the Periadriatic fault system. The resulting tectonic process is the continuation of the uplift of the Eastern Alps and the formation of conjugate sets of strike-slip faults which led to the model of Miocene extrusion of the Eastern Alps (Ratschbacher et al. 1989; N e u b a u e r & Genser 1990). Conjugate strike-slip faults permitted only minor N-S contraction while more important shortening was maintained along the transpressional Periadriatic zone and in southvergent thrusting of the Southern Alps (Boccaletti et aL 1990). The other, probably d o m i n a n t driving mechanism of rifting is the subduction roll-back along the outer Carpathians (Royden et al. 1982; Csontos et al. 1992; Royden 1993). The onset of calc-alkaline volcanism around 19 Ma was associated with the subduction. The migration of volcanic centres toward the thrust front was long-time connected to the retreat of the subducting plate toward the foreland (Balla 1981, 1984; Szab6 et al. 1992; Lexa et al. 1995). The third factor which might have been contributed to rifting is given by the CCW rotation
of the North Pannonian Unit. It is because, this occurred not only relative to a fix geomagnetic frame, but there was also differential rotation between North Pannonia and the rest of the Alcapa unit, i.e. the Eastern Alps. This implies that the transition zone between the Eastern Alps and North Pannonia bent and the 'hinge zone' coincide with the demonstrated metamorphic core complexes. Therefore, we speculate that they are genetically related (Csontos 1995; Fodor et al. 1998). Two main directions of tension can be differentiated within the majority of the Pannonian Basin system, a N E - S W and an E - W to S E - N W one. The main tectonic subsidence can be connected to only one tensional phase in some basins while in others the combined effects of both tensions were important. The first phase of rifting was in general oriented east-northeast or eastward. At least part of the determined direction suffered 25 ~ to 35 ~ CCW rotation, reconstruction of which brings the original orientation to E - W (Fig. 6c). This suggest that the Pannonian lithosphere was extended east or east-northeast during the Karpatian to Early Badenian first rifting phase. This orientation points in general towards the subduction front, which, at this time, incorporated the whole Carpathian front (Fig. 6c). The thinning of the lithosphere occurred by low angle normal and accompanying high angle faults (Tari et al. 1992). High-angle normal faults are widespread far from the metamorphic core complexes. They probably developed in 'wide rift mode' (Tari 1994). The oppositely tilted halfgrabens, the sudden change of extended and n o n - d e f o r m e d blocks need the existence of transfer faults. The largest one could be the MidHungarian shear zone but several smaller ones are also documented (Tari et al. 1992; Csontos 1995). While they are subparallel to the tensional direction, their kinematics is not always evident. Smaller sinistral and dextral separations are also documented along minor N-S to NNE-SSW and E - W faults, respectively, but strike-slip faulting seems to be minor with respect to normal faulting (Figs 4c, 6c). At both the northwestern and southeastern ends of the front, at the junction with the Alpine and Balkan thrust-and-fold belts, important strike-slip zones developed to accommodate the steps of the Carpathian thrust front (Fig. 6c). At the southeastern corner earlier, sometimes curved dextral faults were reactivated (Balla 1984) and Oligo-Miocene Basins were inverted (Ratschbacher et al. 1993b). On the northwestern corner the sinistral opening of the Vienna Basin accommodated the northeastward step of
PANNONIAN BASIN PALAEOSTRESS DATA the Carpathians with respect to the Alpine thrust belt. Strain partitioning occurred along this oblique front: at the Vienna Basin margin sinistral displacement occurred while N from the sinistral shear zone, in the external Flysch and Molasse zones thrusting was almost perpendicular to the arcuate thrust front (Fodor et al. 1995). At both corners the tensional stress field is replaced by strike-slip type stress, having fan shaped 0-1 trajectories (Fig. 6c). After the phase of the first rifting, 20-30 ~ CCW rotation occurred, affecting part of, or the whole North Pannonian-West Carpathian unit. The age of rotation is well-defined as earliest Mid-Miocene (early to middle Badenian) in central North Hungary (Mfirton & Fodor 1995). However, N E - S W tension was observed in rocks younger than the age of rotation, mainly in Badenian andesitic rocks of central North Hungary where rotation is proved to be absent or minor (Mfirton & Mfirton 1996). This suggests that the duration of the first rifting event overlaps the time span of the rotation, at least in the central and northeastern Pannonian Basin. The suggested scenario can be the following; initial E - W tension, rotated to N E - S W during the Early Badenian, renewed N E - S W tension (already non-rotated) during the Mid-Badenian. After this rotation the stress field changed in the western Pannonian Basins (Vienna and Danube Basins) and the minimal stress axis (o-3) rotated to ESE-WNW. The tension was still oriented toward E or NE in the central and eastern Pannonian Basins (Fig. 6d). This change in tensional direction was gradually younging eastward; Late Badenian in the central northern Pannonian Basin and late Sarmatian in the Tokaj-Slanec volcanic chain and in the East Slovakian Basin (compare Fig 6c, d & f). To explain this inhomogeneous tensional direction, we suggest that the northeastward suction effect of the subducted slab beneath the northeastern and eastern Carpathians was modified by the lateral boundary conditions of the overriding slab. Spakman (1990) suggested a laterally propagating slab detachment beneath the Carpathians. The detached portion of the slab could exert no more drag effect on the overriding plates. In consequence, the drag was exerted on shorter and shorter sections and the orientation has also changed. The northeastward drag ceased after the Mid-Miocene and only the drag effect toward the Eastern Carpathians remained. The other consequence of the gradual cessation of thrusting is that the overriding Western C a r p a t h i a n - N o r t h e r n Pannonian unit could not slip more (north)eastward and its northwestern boundary became gradually fixed. In
325
consequence, the drag effect appeared along a curved path between the retreating (moving) and the fixed front of the Western Carpathians and northern Pannonian Basins (Fig. 6c, d, f). The centre of this curved path changed, explaining the rapidly changing tensional directions. The directional variation is more expressed in the central Western Carpathian depressions because they were situated closer to the freezed lateral boundary: here three episodes of tension could be differentiated (Nem~ok et al. 1993; Hdk et al. 1995). The E - W tension occurred during the latest Badenian and Sarmatian, when thrusting (and probably slab retreat) was active east of this region. From the latest Mid-Miocene the tension was oriented SE probably corresponding to the drag effect from the East Carpathians (Fig. 6f). In the northern central Pannonian Basins only two tensional direction were clearly separated. During the Late Miocene the tension was similar to that of the Sarmatian, because the ESE 0-3 direction approximates the probable direction of the drag (Fig. 6f). Back-arc extension is well-expressed in the fast subsidence of the East Slovakian Basin during the Late Badenian and Sarmatian (Figs 4d & Fig. 6d). Because the thrust front was fixed just north of this basin, the extension was associated with moderate counterclockwise rotation of rocks and the thrust front itself. This rotation was observed in the Tokaj hills and in the flysch belt (Bazhenov & Burtman 1980; Balla 1984; Mfirton & P6cskay 1995; Orlicky 1996). The rotation was fast, occurring within the late Sarmatian (12-11 Ma). Because the area west of the Bt~kk hills did not rotated (Fig. 1), the area affected by extension and rotation had triangle shape, narrowing northward (Balla 1984). The rotation affected the earliest faults of the Tokaj hills. The original E N E - W S W tension (and perpendicular compression) rotated to NNE-SSW orientation. A younger E N E - W S W tension is already non-rotated and expresses directly the drag of the retreating subduction zone. The other explanation for the inhomogeneous stress field may be the collision of the overriding plates with buoyant, non-subductable material of the European margin. This caused halting of the thrusts and forelandward propagation of the orogene, consequently the hampering of backarc extension. The westernmost Carpathians were blocked first in the Badenian, therefore the back-arc region could extend only in an E - W to S E - N W direction. In the eastern Pannonian Basin no major difference occurs between the first and second phase of rifting (Fig. 6d, f). The southern
326
L. FODOR E T A L .
boundary zone of the Tisza-Dacia unit continued the dextral slip during the M i d - L a t e Miocene. The retreating subducted slab b e n e a t h the E a s t e r n Carpathians was far enough to create local differences, arcuate strain trajectories developed only close to thrust fronts and the lateral boundaries of the eastward moving unit (Linzer 1996). The tensional deformation was coeval with, or interrupted by, strike-slip or true compressional d e f o r m a t i o n occurring mainly within the external thrust belt. Such a d e f o r m a t i o n is natural in fold and thrust belts, but well-dated stress data are scarce (Figs 5c, d, e and 6c, d, e). One example can be the stress field along the Pieniny Klippen belt (Ratschbacher et al. 1993a), where the timing is poor. In the Pannonian region the N E - S W to E - W compressional stress data can also be regarded as a far field effect of compression in the n o r t h e a s t e r n Carpathians. Although generally not well-dated, we tentatively connect this event with the termination of subduction along this front (Fodor et al. 1990; Peresson & Decker 1997). Since this termination of thrusting was time-progressive, the same might be the case for the strike-slip or compressional event. However, the resolution of timing is not good enough to support this idea. The best documented inversion structures are located in the central-southern Pannonian Basin and the strike of the structures would indicate N-S compression. In that way the or1 trajectories would show an arcuate pattern, starting from NE or E N E direction in the external thrust belt and rotate gradually to N-S approaching the Dinarides (Fig. 6e). The age of this inversion is latest Sarmatian or early Pannonian (11-8? Ma). From the latest Miocene the continuing northwestward convergence of Europe and Africa induced N-S to NNW-SSE compression in the main part of the Pannonian Basin (Fig. 6g). This effect was amplified during the latest Miocene-earliest Pliocene and continued up to recent times (Gerner et al. this volume). Important shortening occurred in an E - W trending zone, from the Venetian Alps to the Southern Carpathians. This shortening was propagating in time through the Miocene, from the tip of the Adriatic promontory, from the L o m b a r d i a n Alps (Massari 1990). This suggests that this compressive belt was gradually propagating eastward. Dextral shift of Adria and its deformed foreland with respect to the Pannonian area was maintained by dextral transpression along the Periadriatic zone, Sava folds and other NWstriking dextral faults within the Dinarides (Fig. 6g). Large-scale thrusting seems to lack in the Pannonian Basin, indicating eastward decrease
in the magnitude of inversion. Folding and thrusting are progressively replaced by sinistral and dextral transpressional reactivation of a number of earlier faults. The subduction finished all along the Eastern Carpathians and the Pannonian Basin has undergone inversion and tectonic reactivation since the Pliocene (Horvfith 1995; Gerner et al. this volume). In addition to the IBS project, field work of the study was supported by the grants OTKA (Hungarian Scientific Research Fund) F 014186 for L. F., T 015 976/95 for L. C. and OMFB/Tt~T Balaton Project F-49/96 for L.B. The work benefitted from the help of a great number of colleagues (mainly cited in the reference list) in Hungary and the neighbouring countries in the form of field assistance and discussions. We thank them all very much. The manuscript was reviewed by L. E. Ricou whose comments are appreciated. Drawings were partly made by L. N6meth. We specially thank the editorial work of F. Horvfith which significantly helped to make this paper readable. The editorial patience and pressure of B. Durand is also acknowledged here.
Appendix: references for stress and structural data, including maps Eastern
A lps (Austria)
whole: Fuchs (1984); Neubauer & Genser (1990); Ratschbacher et al. (1989, 1991a, b). Structural data - regional: Decker et al. (1993, 1994); Fodor, (1995); Fltigel & Neubauer (1984) (1:200 000); Fltigel et al. (1988) (1:200 000); Fuchs & Grill (1984) (1:200 000); Hamilton et al. (1990); Linzer et al. (1995); Nemes et al. (1995b); Polinski & Eisbacher (1992); Ratschbacher et al. (1990); Schmid et al. (1989). Palaeostress - calculations: Decker et aL (1993, 1994); Fodor (1995); Fodor et al. (1990); Linzer et al. (1995); Nemes etal. (1995a, b). Palaeostress - estimations, microtectonic data: Polinski & Eisbacher (1992); Sauer et al. (1992). Structural d a t a -
Western Slovak
Carpathians Republic,
(Czech
Republic,
Poland)
Structural data - whole: Fus~in et al. (1967; 1:500 000); (1987; 1:500000); Ji~f6ek (1979); Mahel et al.
(1984) (1:500 000); geol. maps of 1:200 000. data - regional: Birkenmajer (1985); Kone6ny & Lexa (1984) (1:50 000); Kov~i~et al. (1989, 1990, 1994, 1995); Lankreijer et al. (1995); Plagienka, (1991); Ratschbacher et al. (1993a); Salaj (1995). Palaeostress - calculations: Fodor (1995); Fodor et al. (1990,1995); Marko et al. (1991); Nem6ok (1993); Nem~ok et aL (1993); Ratschbacher et al. (1993a).
Structural
PANNONIAN BASIN PALAEOSTRESS DATA Palaeostress - estimations, microtectonic data: H6k et al. (1995); Kov~6 et aL (1989, 1990, 1995); Kov~i6 & H6k (1993); Nem6ok et al. (1989); Nem6ok &
Lexa (1990). Eastern Carpathians
(Romania)
Structural data - whole: Sfindulescu et al. 1981 Structural data - regional: Berza et al. (1988a, b). Palaeostress - calculations: Ratschbacher et al. (1993b); Linzer 1996; Gy6rfi et al. this volume. Southern Alps-Dinarids
(Slovenia-Italy)
Structural data - whole: 1:500 000 geological map of
Yugoslavia. Structural data - regional: Boccaletti et al. (1990);
Massari (1990); 1:200 000 sheets of Yugoslavia; Premru (1976). Structural data - local: Brezigar et al. (1987); Vrabec (1994). Palaeostress - calculations: Fodor et al. (1998); Nemes et al. (1995a). Palaeostress - estimations, microtectonic data: Poljak (1984). Southern Pannonian
basin (Croatia)
Structural data - whole: Dragi6evid et al. (1983) Prelogovid et aL (1995); Tari & Pami6 (1998). Structural data - regional: Jami6id (1983, 1995); Pamid
(1995). Pannonian
basin (Hungary)
Structural data - whole: Balla (1984,1988); Brezsny~inszky & Haas (1990) (1:500 000); Csontos et al. (1992); Csontos (1995); Kildnyi et al. (1991); Szafifin (unpublished); Tari (1994); Tari et al.
(1992); Maps of 1:200 000. Structural data - regional: Bada et al. (1996); Balla (1989); Balla & Dudko (1989); Balla et aL (1987);
Balogh (1964) (1:100000); Benkovics (1991); Csontos (1988); Dudko et al. (1989, 1992); Fodor et al. (1992, 1994); Grill et al. (1984); Gyarmati et al. (1976) (1:50 000); Gy6rfi (1992, 1993); Gy6rfi & Csontos (1994); Hfimor (1985); Horvfith et al. (1998); K6r6ssy (1989, 1990); L6rintz & Szab6 (1993); M6szfiros (1982); M6sz~iros & T6th (1981); Rad6cz (1966) (1:50 000); Rumpler & Horvfith (1988); Tari et al. (1993); Tari (1988, 1992); Telegdi-Roth (1951); Vass et al. (1979, 1993); Vad~isz (1935); Wein (1967). Structural data - local: Bence et al. (1991); Bergerat et al. (1983); Csontos & Nagymorosy (1998); K6kay (1976, 1989, 1990); Maros (1988). Palaeostress - calculations: Bada et al. (1993); Benkovics (1991; 1997); Bergerat (1989); Bergerat et aL (1984); Bergerat & Csontos (1987, 1988); Csontos (1988); Csontos & Bergerat (1992); Csontos et al. (1991); Fodor et al. (1992, 1994); GyOrfi (1993); Gy6rfi & Csontos (1994); Maros (1988); Mfirton & Fodor (1995); Sztan6 & Fodor (1997).
327
R e f e r e n c e s
ALEXANDROVSKI,P. 1985. Graphical determination of principal stress directions for slickenside lineation populations: an attempt to modify Arthaud's method. Journal o f Structural Geology, 7, 73-82. ANDERSON, E .M. 1951. The dynamics o f faulting. Oliver & Boyd, Edinburgh, 2nd ed., 206 p. ANGEI~ER, J. 1979a. Ndotectonique de l'Arc EgOen. Soci6t6 G6ologique du Nord, Publications 3. 1979b. Determination of the mean principal directions of stresses for a given fault population. Tectonophysics, 56, 17-26. 1984. Tectonic analysis of fault slip data sets. Journal o f Geophysical Research, 8, B7, 5835-5848. 1990. Inversion of field data in fault tectonics to obtain the regional stress -III. A new rapid direct inversion method by analytical means. Geophysical Journal International, 103, 363-373. & MANOUSSIS, S. 1980. Classifcation automatique et distinction de phases superpos6e en tectonique cassante. Comptes Rendus de l'AcadOmie des Sciences, Paris, 290, 651~554. - - & MECHI~ER,R 1977. Sur une m6thode graphique de recherche des contraintes principales 6galement utilisable en tectonique et en s6ismologie: la mdthode des di~dres droits. Bulletin de la SociOt~ GOologique de France, 719, 1309-1318. BADA, G., FODOR, L., SZI~KELY,B. & T1MAR, G. 1996. Tertiary brittle faulting and stress field evolution in the Gerecse Mts., N. Hungary. Tectonophysics, 255, 269-289. - - , ILLt~S, K., KOROKNAI, B., LELKES, B., Nt~METH, K. & TOTI4, T. 1993. [Summary report on the -
-
-
-
-
-
geological-geophysical
field
work
of
1993].
Manuscript, E6tv6s University, Budapest [in Hungarian]. BALDI, T. 1986. Mid-Tertiary Statigraphy and Paleogeographic Evolution o f Hungary. Akad6miai Kiad6, Budapest. BALLA,Z. 1984. The Carpathian loop and the Pannonian basin. A kinematic analysis. Geophysical Transactions, 30, 313-353. 1987. Tertiary palaeomagnetic data for the Carpatho-Pannonian region in the light of Miocene rotation kinematics. Tectonophysics, 139, 67-98. 1988. On the Origin of the structural pattern of Hungary. Acta Geologica ttungarica, 31, 53-63. 1989. A Di6sjen6i diszlok~ci6s 6v fijra6rt6keldse [Reevaluation of the Di6sjen6 dislocation zone]. -
-
-
-
Annual Report o f the EOtvOs L. Geophys. Inst. for 1987, 45-57 [in Hungarian]. -
-
& DUDKO, A. 1989. Large-scale Tertiary strikeslip displacements recorded in the structure of the Transdanubian Range. Geophysical Transactions, 3~54. , & REDLER-TATRAI, M. 1987. A K6z6pDumintt~l fiatal tektonikfija f61dtani 6s geofizikai adatok alapjfin [Young tectonics of Mid-Transdanubia based on geological and geophysical data]. Annual Report o f the EOtvOs L. Geophys. Inst. for 1986, 74-94 [in Hungarian]. 35,
L. F O D O R ETAL.
328
BALOGH, K. 1964. Die geologische Bildungen des Btikk-Gebirges. Year Book of the Hungarian Geological Institute, 48. BAZHENOV, M. L. & BURTMAN, g. 1980. About the nature of the northern part of the Carpathians. Doklady Akadamya Nauk, SSSR, 255,681-685 [in Russian]. -- - , g. & SANDULESCU,M. 1993. Paleomagnetism of the Upper Cretaceous rocks and its bearing on the origin of the Romanian Carpathian arc. Romanian Journal of Tectonics and Regional Geology, 75, 9-14. BENCE, G., CSASZAR,G., DARIDA-TICHY,M., DUDKO,A., GALOS, M., GANGL, G., KERTI~SZ,P., KORPAS, L. & Z~ER, C. 1991. Geologische und ingenieurgeologische Beschreibung der Donaustufe bei Nagymaros. In: LOBITZER, H., CSASZAR, G. (eds) Jubiliiumsschrifi 20 Jahre geologische Zusammenarbeit Osterreich- Ungarn, Geologisches Bundesanstalt, Wien, 385-400. BENI~OVlCS,L. 1991. Neogene tectonical development of the Zagyva half graben (NE Hungary). In: 1st International Meeting of Young Geologists, Budapest, Abstracts, 69. -1997. Etude structurale et gdodinamic des Monts Buda, Mecsek, et Villdny (Hongrie). PhD. thesis, Univ. Lille, France. BERGERAT, E 1989. From pull-apart to the rifting process: the formation of the Pannonian Basin. Tectonophysics, 157, 271-280. -& CSONTOS, L. 1987. Pr6sence de microfailles hydroplastiques de compaction dans ies lignites mioc6nes et plioc6nes des massifs de M~tra et du Btikk (Hongrie): Interpr6tation dans le contexte extensif du Bassin Pannonien. Comptes Rendus de l'Acad~mie des Sciences, Paris, 305, 791-796. & -1988. Brittle tectonics and paleo-stress field in the Mecsek and Vill~ny Mts. (Hungary): Correlation with the opening mechanism of the Pannonian Basin. Acta Geologica Hungarica, 31, 81-100. , GEYSSANI; J. & KAZMER, M. 1983. Une tectonique synsedimentaire originale du Mioc6ne moyen des environs de Budapest, marqueur de l'extension du Bassin Pannonien. Comptes Rendus de l'Acad~mie des Sciences, Paris, 296, 1275-1278. & LEPVRIER, C. 1984. Neotectonic outline of the Intra-Carpathian basins in Hungary. Acta Geologica Hungarica, 27, 237-251. BERZA,T., SEGHEDI,A. & DRAGANESCU,A. 1988a. The Danubian units in the Northern Vilcan Mountains (South Carpathians). D{Iri de Seama Institutul Geologia si Geofizica, 72-73, 5, 23~42. -& STANOIU,I. 1988b. Danubian units in the Eastern part of the Re_ezat Mountains (South Carpathians). D(tri de Seama Institutul Geologia si Geofizica, 72-73, 5, 5-22. BIRKENMAJER, K. 1985. A model of structural evolution of the Pieniny Klippen Belt, Carpathians. In: Proceedings Rep. XIIIth CBGA Congress, Guide to Excursion, 2. Geological Society of Poland. BOCCALETTI, M., CALAMITA,E, DEIANA, G., GELATI, -
-
-
-
-
-
R., MASSARI,E, MORATTI, G. & RICCI LUCCH[, E 1990. Migrating foredeep-thrust belt system in the northern Appennines and southern Alps. Palaeogeography, Palaeoclimatology, Palaeoecolgy, 77, 3-14. BOTT, M. H. P. 1959. The mechanism of oblique slip faulting. Geological Magazine, 96, 109-117. BREZIGAR, A., OGORELEC, B., RIJAVEC, L. & MIOC, P. 1987. Geologic setting of the Pre-Pliocene basis of the Velenje depression and its surroundings. Geologija (Ljubljana), 30, 31-65 [in Slovenian with English abstract]. BREZSNYANSZKY,K. & HAAS, J. 1990. Tectonic map of Hungary, 1: 500000. Hungarian Geological Institute, Budapest. CSONTOS,L. 1988. Etude gOologique d'une portion des Carpathes Internes, le massif du Bftkk (Nord-est de la Hongrie), (stratigraphie, structures, mOtamorphisme et gOodinamique). PhD Thesis, Univ. Lille Flandres-Artois, no. 250. 1995. Tertiary tectonic evolution of the IntraCarpathian area: a review. Acta Vulcanologica, 7, 1-13. BERGERAT, E 1992. Reevaluation of the Neogene brittle tectonics of the Mecsek-Vill~ny area (SW Hungary). Annales Univ. Scientiarum Budapestiensis Section Geologica, 29, 3-12. & NAGYMAROSY,A. 1998. The mid-Hungarian line: a zone of repeated tectonic inversions. Tectonophysics, 297, 51-71. , - - , HORVATH, E & KOvAC M. 1992. Tertiary evolution of the Intra-Carpathian area: a model. Tectonophysics, 208, 221-241. - - - , TARI, G., BERGERAT,E & FODOR, L. 1991. Structural evolution of the Carpatho-Pannonian area during the Neogene, Tectonophysics, 199, 73-91. DECKER, K. & PERESSON~H. 1996. Tertiary kinematics in the Alpine-Carpathian-Pannonian system: Links between thrusting, transform faulting and crustal extension. In: LIEBL, W. & WESSELV, G. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. E A G E Special Publications, 5. The Geological Society, 69-77. - - - , MESCHEDE, M. & RING, U. 1993. Fault slip analysis along the northern margin of the Eastern Alps (Molasse, Helvetic nappes, North and South Penninic flysch, and the Northern Calcareous Alps). Tectonophysics, 223, 291-312. , PERESSON, H. & FAUPL, P. 1994. Die miocfine Tektonic der 6stlichen Kalkaplen: Kinematik, Pal~iospannungen und Deformationsaufteilung wfihrend der "lateralen Extrusion" der Zentralalpen. Jahrbuch der Geologischen Bundesanstalt, 137, 5-18. DICEA, 0., DUTESCU, R, ANTONESCU, E, MITREA, G., BOTEZ, R., DONOS, LUNGU, g. & MORO~ANU, I. 1980. Contributions ~ la connaissance de la tectonique de la zone transcarpatique de Maramureg. Dhri de Seama lnstitutul Geologia si Geofizica, 65, 35-53. D RAGICEVIC,B., MARINOVI(",D., STANKOVIC,BATUSI(~, V. & NIKOLIC D. 1983. Geological properties and oil and gas bearing capacity of deeper parts of the -
-
-
-
PANNONIAN BASIN PALAEOSTRESS DATA Pannonian basin in Yugoslavia. Nafta, 37, 369-380. DUDKO, A., BENCE, G. & SELMEChI. 1992. The tectonic origin of Miocene basins on the south-western edge of the Transdanubian Central Range.
Annual Report of the Hungarian Geological Institute from 1990, 107-124. , DARIDA-TICHY,M., MAJKUTH,T. & STOMFAI,R. 1989. A Kelet-velencei paleovulkfin szerkezete [Structure of the East-Velence palaeovolcanoe]. Altaldnos FOldtani Szemle, Budapest, 135-147. DUNKL, I. & DEMENY, A. 1997. Exhumation of the Rechnitz Window at the border of Eastern Alps and Pannonian basin during Neogene extension. Tectonophysics, 272, 197-211. EBNER, E & SACHSENHOFER,R. 1995. Paleogeography, subsidence and thermal history of the Neogene Styrian Basin (Pannonian basin system, Austria). Tectonophysics, 242, 133-150. - - , KOVACS S. & SCHONLAUB,H. R 1991. Das klassische Karbon in (3sterreich und Ungarn - ein Vergliech der sediment/~ren fossilftihrenden Vorommen. In: LOBn'ZER, H. & CSASZAR, G. (eds) Jubiliiumsschrift 20 Jahre geologische Zusammenarbeit Osterreich-Ungarn, 1, Geol. Bundesanstalt, Wien, 263-294. FAY-TATRAY,M. 1984. Contribution to the lithology of the reworked clastic dolomite complex of the Southern Gerecse forelands (Transdanubia, Hungary). Annales Univ. Scientiarum Budapestiensis Section Geologica, 24, 151-166. FLI21GEL,W. & NEUBAUER, E 1984. Geologische Karte der Steiermark 1:200000. Geol. Bundesanstalt, Wien. - - , KROLL, A. & WEBER, E 1988. Geologische Karte
des priitertiiiren Untergrund, Steierisches BeckenSiidburgenlandische Schwelle. Geol. Bundesanstalt, Wien. FODOR, L. 1995. From transpression to transtension Oligocene-Miocene tectonic evolution of the Vienna Basin and the East Alpine-Western Carpathian junction. Tectonophysics, 242,151-182. - - , FRANCO,J., KREJ~I, O. & STRAN[K,Z. 1995. Paleogeographic and tectonc evolution of the Carpathian flysch Belt of the southern Moravia (Czech Republic). In: Proceedings' of X'V. Carpathian-Balkan Geological Congress, Special Publications, Geological Society of Greece, 31-33. - - , JELEN, M., MARTON, E., SKABERNE,D., CAR, J. & VaABEC, M. 1998. Miocene-Pliocene tectonic evolution of the Periadriatic line in Slovenia implications for Alpine-Carpathian extrusion models. Tectonics, 17, 690-709. , MARKO, E & NEM~OK, M. 1990. Evolution microtectonique et paldo-champ de contrainte dans le Bassin de Vienne. Geodynamica Acta, 4, 147-158. , KAzIVIs M., MAGYARI,A. & FOGARASI,A. 1992. Gravity - flow dominated sedimentation on the Buda paleoslope (Hungary). Record of Late Eocene continental escape of the Bakony Unit. Geologisches Rundschau, 82, 695-716. , MAGYARI A., FOGARAS1A. & PALOTASK. 1994. Tertiary tectonics and Late Paleogene
329
sedimentation in the Buda Hills, Hungary. A new interpretation of the Buda line. FOldtani KOzlOny, 124, 12%305. FUCHS, W. 1984. Grosstektonische Neuorientung in den Ostalpen und Westkarpaten unter Einbeziehung plattentektonischer Gesichtspunkte. Jahrbuch der Geologischen Bundesanstalt, 127, 571-632. -& GRILL, R. 1984. Geologische Karte von Wien und Ungebung 1:200 000. Geol. Bundesanstalt, Wien. FUSAN, O., BIELY, A., IBRMAJER, J., PLANCAR, J. & ROZLOZNiK, L. 1987. Basement of the Tertiary of the Inner West Carpathians. Geol. Ustav D. Sttlra, Bratislava. - - , KODYM, O., MATEJKA, A. & URBANEK, L. 1967.
Geological map of Czechoslovakia, 1:500 000. Geological Survey of Czechoslovakia, Prague. GALACZ, A. & VOROS,A. 1972. Jurassic history of the Bakony Mts. and interpretation of principal lithological phenomena. FOldtani KOzlOny, 102, 122-135 [in Hungarian, with English abstract]. G~;czu B. 1973. Plate tectonics and paleogeography in the East-Mediterranean Mesozoic. Acta Geologica Hungarica, 17, 421-428. 1984. Provincialism of Jurassic ammonites: examples from Hungarian faunas. Acta Geologica Hungarica, 27, 379-389. GERNER, P. 1992. Recent stress field in Transdanubia (Western Hungary). FOldtani KOzlOny, 122, 89-105. - - , BADA, G., DOVI~NYI,P., HORVATH,E, M~LI,ER, B., ONCESCU, M. & CLOETING, S. 1999. Recent tectonic stress and crustal deformation in the Pannonian Basin: data and models. This volume. GRILL, J., KovAcs, S., LESS, GY, RET~, ZS, ROTn, L. & SZENTPI~TERY,I. 1984. Geology and evolution of the Aggtelek-Rudabfinya Mts. FOldtani Kutatds (Budapest), 27, 49-56 [in Hungarian]. GYALOG, L. 1992. [Contributions to the structural geological knowledge of the Vhrgesites area]. Annual
Report of the Hungarian Geological Institute from 1990, 69-74 [in Hungarian with English abstract]. GYARMATI, P., PERLAKI, E. & PENTELI~NYI, L. 1976. Geological map of the Tokaj Mts. 1:50 000. Hungarian Geological Institute. GYORFI, I. 1992. The simple shear extensional model of the Pannonian basin. Terra Nova, Abstract Supplement 2, 4, 26. -1993. DK-Magyarorszdg ~s az Erd~lyi-KOzdphe-
gys~g neogdn medenc~inek szerkezetfejlO'd~se [Structural evolution of the Neogene basins of SE Hungary and the Apuseni Mts.]. Master's thesis, --
E6tv6s University, Budapest. & CSONTOS,L. 1994. Structural evolution of SE Hungary and Neogene basins of the Apuseni Mountains (Romania). Romanian Journal of Tectonics and Regional Geology, 75, suppl. 1, 19-20. , -& NAGYMAROSY, A. 1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian basins. This
volume. HAAS J., KovAcs S., KRYSTYN,L. & LEIN R. 1995. Significance of Late Permian-Triassic facies zones in
330
L. F O D O R ETAL.
terrane reconstructions in the Alpine-North Pannonian domain. Tectonophysics, 242, 19-40. HAMILTON, W., Jlfl[~EK, R. & WESSELY,G. 1990. The Alpine-Carpathian floor of the Vienna Basin in Austria and CSSR. In: MINAR~KOVA, D. & LOBITZER, H. (eds) Thirty years of geological
cooperation between Austria and Czechoslovakia, Geol. Bundesanstalt, Wien, 46-55. HAMOR, G. 1985. Geology of the N6grdd-Cserhdt area. Geologica Hungarica, sex geol., 22. HANCOCK, P. L. 1985. Brittle microtectonics: principles and practice. Journal of Structural Geology, 7, 437--457. HOK, J., SIMON, L., KOVA~, E, ELE~KO, M., VASS, D., HALMO, J. & VERBICH, E 1995. Tectonics of the Hornonitrianska Kotlina depression in the Neogene. Geologica Carpathica, 46, 191-196. HORVATH, E 1993. Towards a mechanical model for the formation of the Pannonian basin. Tectonophysics, 225, 333-358. - 1995. Phases of compression during the evolution of the Pannonian Basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology, 12, 837-844. - & ROYDEN, L. 1981. Mechanism for the Formation of the Intra-Carpathian Basins: a Review. Earth Evolution Science, 3, 307-316. - & RUMPLER, J. 1984. The Pannonian basement: extension and subsidence of an alpine orogene. Acta Geologica Hungarica, 27, 229-235. - - , TAm, G. & BOKOR, CS. (eds) (1995) Extensional collapse of the Alpine orogene and Hydrocarbon prospects in the Basement and Basin Fill of the Western Pannonian Basin. In: AAPG Inter-
national Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6, Hungary. HUBATKA, E & Posplgm, L. 1990. Evidence for horizontal shifts in the Vienna Basin. Mineralia slovaca, 22, 63-67. JABLONSI~, J., MIg[K, M., SOTJ~K,J., MARSCHALKO,R., ZAGORgEK, K., KOHLER,E., GROSS,P. & MARKO,F. 1994. Cretaceous and Paleogene paleogeography
and geodynamics of the Alpine-Carpathian-Pannonian region. Excursion guidebook, Comenins Univ., Bratislava. JAMICI(~ D. 1983. About tangential movements in the region of the Slavonian Mountains. Mafia, 34, 685-691. - 1995. The role of the sinistral strike-slip faults in the formation of the structural fabric of the Slavonian Mts. (Eastern Croatia). Geologica Croatica, 48, 155-160. JIMCEK, R. 1979. Tectonic development of the Carpathian arc in the Oligocene and Neogene. In: MAHEL, M. (ed.) Tectonic profiles through the West Carpathians. Geol. Ustav D. St0ra, Bratislava, 205-214. KAZMr~R,M. & JOZSA, S. 1989. The missing volcanic arc of the Paleogene Alpine subduction. Terra Abstracts 1, 58. -& KOVACS, S. 1985. Permian-Paleogene Paleogeography along the Eastern part of the InsubricPeriadriatic Lineament system: Evidence for continental escape of the Bakony-Drauzug Unit. Acta Geologica Hungarica, 28, 71-84.
KERCSMAR,ZS. 1996. Syntectonic sedimentation in the marine Eocene of Tatab~inya, Hungary. In: Sedi-
ment'
96,
Kurzfassungen der Vortrage und Poster.
Univ. Wien, 74. KILENYI 1~., KROLL, A., OBERNAUER, D., SEFARA, J., STEINHAUSER, P., SZAB0, Z. & WESSLY, G. 1991. Pre-Tertiary basement contour map of the Carpathian Basin beneath Austria, Czechoslovakia and Hungary. Geophysical Transactions, 36, 15-37. KONECNY,W. & LEXA, J. 1984. Geological Map of the
Central Slovakia Neogene Volcanic Field. Regional geological maps of Slovakia, 1:50 000. Geol. Survey of Czechoslovakia, Bratislava. KovAC M. & BARATH, I. 1995. [Tectono-sedimentary development of the Alpine-Carpathian-Pannonian junction zone during the Miocene]. Mineralia slovaca, 28, 1-11 [in Slovak with English summary]. - - , HOLICKY, I., MARKO, E & Tt)NYI, I. 1989. Basin opening in the lower Miocene strike slip zone in the SW part of the Western Carpathians. Geologica Carpatica, 40, 37-62. - - , KRAL, J., MARTON,E., PLAgIENKA,D. & UHER, P. 1994. Alpine uplift history of the Central Western Carpathians: geochronological, palaeomagnetic, sedimentary and structural data. Geologica Carpathica, 45, 83-96. - - , KovAC P., MARKO, E, KAROLI, S. & JANOCKO,J. 1995. The East Slovakian Basin - A complex back-arc basin. Tectonophysics, 252, 453-466. - - , MARKO,E & NEMCOK,M. 1990. Neogene history of intramontane basins in the Western part of the Carpathians. Rivista Italiana di Paleontogia i Stratigrafia, 9 6 , 381-404. KovAC R & HOK, J. 1993. The Central Slovak fault system - the field evidence of a strike slip. Geologica Carpathica, 44, 155-159. KOKAY, J. 1976. Geomechanical investigation of the southeastern margin of the Bakony Mts. and the age of the Lit6r fault line. Acta Geologica Hungarica, 20, 245-257. 1989. A m6ny-zs6mbOki teriilet bddeni ~s
szarmata kOpzddmOnyei [Badenian-Sarmatian formations of the Mfny-Zsdmb~k area]. Manuscript, EOtvOs Univ. Budapest [in Hungarian]. 1990. ]Middle Badenian rocks in Budapest].
Annual Report of the Hungarian Geological Institute from 1988, 101-108 [in Hungarian with English abstract]. KtSROSSY, L. 1989. A Drfiva medencei k6olaj 6s f61dgfizkutatfis f61dtani eredmdnyei [Hydrocarbon geology of the Drfiva Basin in Hungary, in Hungarian]. _Adtaldnos FOldtani Szemle (Budapest), 24, 3-121. 1990. A Ddlkelet-Dunfintfil k6olaj - 6s f61dgfizkutatfisfinak f61dtani eredmdnyei [Geological results of hydrocarbon prospecting in the SE half of Transdanubia]. Altaldnos FOldtani Szemle (Budapest), 25, 3-53 [in Hungarian]. KRS, M., KRSOVA,M., CHVOJKA,R. & POTFAJ,M. 1991. Palaeomagnetic investigations of the flysch belt in the Orava region, Magura unit, Czechoslovak Western Carpathians. Geologickd Prdce, 92, 135-151.
PANNONIAN BASIN PALAEOSTRESS DATA --,
MUSKA, P. & PAGAn, P. 1982. Review of palaeomagnetic investigations in the West Carpathian of Czechoslovakia. GeologickO Prdce, 78, 39-58. LANKREIJER, A., KOVA~, M., CLOETINGH, S., PITON.~K, P., HLOgKA, M. & BIERMANN, C. 1995. Quantitative subsidence analysis and forward modeling of the Vienna and Danube Basin. Tectonophysics, 252, 433-452. LAUBSCHER, H. P. 1988. Material balance in Alpine orogeny. Geological Society of America Bulletin, 100, 1313-1328. LEXA, J., KONE~NY, V., KorAn, M., NEM~.OK, M. 1995. Relationship among convergence, back-arc extension and volcanism in the Western Carpathians during Neogene. In: Europrobe
workshops - Pancardi, Star(t Lesna, Slovak Republik, Abstracts, 65-66. LINZER, H. G. 1996. Kinematics of retreating subduction along the Carpathian arc, Romania. Geology, 24, 167-170. --, RATSCHBACHER,L. & FRISCH, W. 1995. Transpressional collision structures in the upper crust: the fold-thrust belt of the Northern Calcareous Alps. Tectonophysics, 242, 41-62. L0czY, L. 1913. Balaton k6rny6k6nek geol6gi~ja ~s morfologifija [Geology and Geomorphology of the Balaton area]. Balaton Tudomdnyos Tanulmdnyozdsdnak Eredm~nyei, I, Budapest. L6RINCZ, K. D. & SZAB0, P. 1993. Seismic Analysis of Multi-Phase Tectonism in the Central Part of the Pannonian Basin in Hungary. In: SeENCEP,,A. M. (ed) Generation, Accumulation and Production of Europe's Hydrocarbons IlL European Association of Petroleum Geoscientists, 3, 311-323. MAHEL, M., KODYM, O. & MALKOVSKY,M. 1984. Tectonic map of Czechoslovakia, 1:500 000. Geol. 121stav D. Stt~ra, Bratislava. MARKO, E, FODOR, L. & KovA6 M. 1991. Miocene strike-slip faulting and block rotation in the Brezovsk6 pohorie Mts., Western Carpathians. Mineralia slovaca, 23, 189-200. --, Kovx4, M., FODOR, L. & SUTOVSKA, K. 1990. Deformations and kinematics of a Miocene shear zone in the northern part of the Little Carpathians (Bukovg Furrow, Hrabn'k Formation). Mineralia slovaca, 22, 399-410. --, PLAS~ENKA,D. & FODOR L. 1995. Meso-Cenozoic tectonic stress fields within the AlpineCarpathian transition zone: a review. Geologica Carpathica, 46, 19-27. MAROS, GY. 1988. Tectonic survey in the Vit~ny-vfir area, W Hungary. Annual Report of the Hungarian Geological Institute from 1986, 295-310 [in Hungarian with English abstract]. MARTON, E. 1987. Palaeomagnetism and tectonics in the Mediterranean region. Journal of Geodynamics, 7, 33-57. 1993. Paleomagnetism in the Mediterraneum from Spain to the Aegean: a review of data relevant to Cenozoic movements. In: BOSCHI, E., MANOTVANI,E. & MORELLI,A. (eds) Recent evol-
ution and seismicity of the Mediterranean region. NATO ASI Series, C, 402, 367-402.
331
& FODOR, L. 1995. Combination of palaeomagnetic and stress data: a case study from North Hungary. Tectonophysics, 242, 99-114. -& MARTOY, R 1978. On the difference between the palaeomagnetic poles from the Transdanubian Central Mountains and the Villainy Mountains, respectively. Magyar Geofizika, 19, 129-136. & -1989. A compilation of palaeomagnetic results from Hungary. Geophysical Transactions, 35, 117-133. & -1996. Large scale rotation in North Hungary during the Neogene as indicated by palaeomagnetic data. In: MORRIS,A., TARLING,D. H. (eds) Paleomagnetism and Tectonics of the Mediterranean Region. Geological Society, London, Special Publications, 105, 153-173. & MAUR1TSCH,H. J. 1990. Structural applications and discussion of a palaeomagnetic post-Paleozoic data base for the Central Mediterranean. Physics of the Earth and Planetary Interiors, 62, 46-59. -& PI~CSKAY, Z. 1995. The Tokaj-Vihorl~it-OasIgnis Triangle: Complex Evaluation of Palaeomagnetic and Isotope Age Data from Neogene Volcanics. In: Proceedings of XE. C.B.G. Congress, Special Publications, Geological Society of Greece, 30. , VASS,D. & Tt?NvI, I. 1996. Rotation of the South Slovak Paleogene and Lower Miocene rocks indicated by palaeomagnetic data. Geologica Carpathica, 47, 31-42. MASSARI, E 1990. The foredeeps of the Northern Adriatic Margin: Evidence of diachroneity in deformation of the Southern Alps. Rivista Italiana di Paleontologia i Stratigrafia, 96, 351-380. MAURITSCH, H. J. & BECKE, M. 1987. Palaeomagnetic investigation in the Eastern Alps and the southern border zone. In: I~OGEL, W. & FAUPL,P. (eds) Geodynamics of the Eastern Alps. Deuticke, Vienna, 282-308. -& MARTON, E. 1995. Escape models of the Alpine-Carpathian-Pannonian region in the light of palaeomagnetic observation. Terra Nova, 7, 44-55. MESZAROS,J. 1982. [Maj or horizontal tectonic dislocation as a guide to mineral prospectors in the western Bakony Mountains]. Annual Report of -
-
-
-
the Hungarian Geological Institute from 1980, 517-536 [in Hungarian with English abstract]. & TOTH, I. 1981. Vizszintes eltoldd~isok Ajka t6rs6g6ben 6s gyakorlatijelent6s6gtik. [Strike-slip dislacement and their implications in the region of Ajka]. Altaldnos FOldtani Szemle, Budapest 16, 25-34 [in Hungarian]. MINDSZENTY,A., SZINTAI,M., TOTH, K., SZANTNER,E, NAGY, T., GELLAI, M. & BAROSS, G. 1988. Sedimentology and depositional environment of Csabpuszta bauxite (Paleocene/Eocene) in the South Bakony Mts. (Hungary). Acta Geologica Hungarica, 31, 339-370. MOLNAR, E 1994. [Reconstruction of hydrotermal processes accompained by precious-metal enrichment in the area between Sfitoraljatijhely-Rudabfinyficska and Vfigfishuta, Tokaj Mts., NE -
--
332
L. F O D O R ETAL.
Hungary]. FOldtani KOzlOny, 124, 25-42 [in Hungarian with English abstract]. NAGYMAROSY,A. 1990. Paleogeographical and paleotectonical outlines of some Intracarpathian Paleogene basins. Geol. Zbornik-Geologica Carpathica, 41, 259-274. NEMr M. 1993. Transition from convergence to escape: field evidence from the Western Carpathians. Tectonophysics, 217, 117-142. & LEXA, J. 1990. Evolution of the basin and range structure around Ziar mountain range. Geol. Zbornik-Geologica Carpathica, 41, 229-258. , HOK, J., KOVAe, E, MARKO, E, MADARAS, J. & BEZAK, V. 1993. Tectonika Z~ipadnych Karpatoch. In: RAKI3S, M., VOZAR, J. (eds) Geodynamicky Model a Hlbinnd Stavba Zdpadnych Karpdlt. Geol. l.)stav D. Sttira, Bratislava, 263-267. , MARKO, E, KOVAQ, M. & FODOR, L. 1989. Neogene Tectonics and Palaeostress changes in the Czechoslovakian Part of the Vienna Basin. Jahrbuch der Geologischen Bundesanstalt, 132, 443-458. NEMES, E, NEUBAUER, E & WANG, X. 1995a. Deformation Partitioning along Major Strike-Slip Zones in the Eastern Alps: The Periadriatic Line and Salzachtal-Ennstal-Line. In: Europrobe W o r k s h o p s - Pancardi, Stard Lesna, Slovak Republik, Abstracts, 13-14. --, PAVHK, W. & MOSER, M. 1995b. Geologie und Tektonik im Salztal (Steiermark) - Kinematik und Pal~iospannungen entlang des Ennstal-Mariazell-Blattverschiebungssystems in den N6rdlichen Kalkalpen. Jahrbuch der Geologischen Bundesanstalt, 349-367. NEUBAUER, E & GENSER, J. 1990. Architectur und kinematik der 6stlichen Zentralalpen - eine fibersicht. Mitteilungen des naturwissenschaftlichen Vereins far Steiermark, 000-000. NoszKY, J. 1940. Das Cserh(u-gebirge. Geologishce Beschreibung Ungarischer Landschaften, Hungarian Geological Institute. ORLICKY, O. 1996. Paleomagnetism of neovolcanics of the East-Slovak Lowlands and Zemplinske vrchy Mts.: A study of the tectonics applying the palaeomagnetic data (Western Carpathians). Geologica Carpathica, 47, 13-20. , Tt~NYLI. & VASS,D. 1995. Palaeomagnetic properties of the Siator andesites and the sandstones. Mineralia slovaca, 27, 243-250. PAMIC,J. 1995. Tertiary Volcanic rocks from the Southern Pannonian Basin, Croatia. Acta Geologica Hungarica, 37, 259-283. PAPP, S. 1990. Felpikkelyezett rOtegsorok a KOzdpTiszdntitlon. Special Publications of the Hungarian Geological Institute, Budapest. PATRASCU,S., BLEAHU,M., PANA1OTU,C. & PANAIOTU, C. E. 1992. The paleomagnetism of the Upper Cretaceous magmatic rocks in the Banat area of the South Carpathians: tectonic implications. Tectonophysics, 213, 341-352. , PANAIOTU,C., ~ECLEMAN,M. & PANAIOTU,C. E. 1994. Timing of rotational motion of Apuseni Mountains (Romania): palaeomagnetic data from Tertiary magmatic rocks. Tectonophysics, 233, 163-176. -
-
PERESSON, H. & DECKER, K. 1997. Far-field effects of Late Miocene subduction in the Eastern Carpathians: E - W compression and inversions of structures in the Alpine-Carpathian-Pannonian Region. Tectonics, 16, 38-56. PETIT,J. R 1987. Criteria for the sense of movement on fault surfaces in brittle rocks. Journal of Structural Geology, 9, 597-608. PI~CSKAY, Z., LEXA, J., SZAKACS, A., BALOGH, K., SEGHEDI, I., KONE~NY, V., KovAC M., MARTON, E., KALIQIAK, M., SZt~KY-Fux, V., POKA, T., GYARMATI, P., EDELSTEIN, 0., ROSU, E. & ZEC, B. 1995. Space and time distribution of NeogeneQuaternary volcanism in the Carpatho-Pannonian Region. Acta Vulcanologica, 7, 15-28. PLA~IENKA, D. 1991. Mesozoic tectonic evolution of the epi-Variscan continental crust of the Central Western Carpathians - a tentative model. Mineralia slovaca, 23, 447-457. POGACSAS, GY., LAKATOS,L., I~JJSZ,~SZI,K., VAKARCS, G., VARKONYI, L. & VARNAL R 1988. Seismic fades, electro facies and Neogene sequence chronology of the Pannonian basin. Acta Geologica Hungarica, 31, 175-205. --, MATTICK, R. E., TARI, G. & VARNAI, P. 1994. Structural Control on Hydrocarbon accumulation in the Pannonian Basin, Hungary. In: TELE~I E eT AL. (eds) Basin Analysis in Petroleum Exploration. Kluwer Academic Publishers, 221-235. - - , VAKARCS,G., BARVITZ,A. & LAKATOS,L. 1989. Postrift strike-slip faults in the Pannonian Basin and their role in the hydrocarbon accumulation. In: Abstracts and Papers of the Technical Program - 34th International Geophysical Symposium, 34, 601-611. POLINSKI,R. K. & ElSBACHER,G. H. 1992. Deformation partitioning during polyphase oblique convergence in the Karawanken Mountains, southeastern Alps. Journal of Structural Geology, 14, 1203-1213. POIAAK, M. 1984. Neotectonic investigations in the Pannonian Basin based on satellite images. Advances in Space Research, 4, 139-146. POSGAY,K., BODOKY,T., HEGED1)S E., KOVACSVOLGYI, S., LENKEu L., SZAFIAN,P., TAKACS,E., T[MAR, Z., & VARGA, G. 1995. Asthenospheric structure beneath a Neogene basin in southeast Hungary. Tectonophysics, 252, 467-484. PRELOGOVI(:, E., JAMIQIQ, D., ALJINOV1Q,B., VELIQ, J., SAVH(;, B. & DRAGA$, M. 1995. Dinamika nastanka struktura j uznog dij ela Pannonskog bazena. Proceedings of the First Croatian Geological Congress, Opatija, 481-486. PREMRU, U. 1976. Neotectonic Evolution of Eastern Slovenia. Geologija (Ljubljana), 19, 211-249. 1981. Tectonic evolution of Slovenia during the time interval from the upper Cretaceous to the Tertiary period (in Slovenian with English abstract). In: Symposium on problems of Danian in Yugoslavia, Proceedings, Ljubljana, 147-154. RABAGIA, T. • FOLOP,A. 1994. Syntectonic sedimentation history in the Southern Carpathians foredeep (Getic depression). Romanian Journal of Tectonics and Regional Geology, 75, suppl. 1, 48. RADOCZ, GY. 1966. Geological map of Helvetian
-
-
PANNONIAN BASIN PALAEOSTRESS DATA
formations around the Biikk Mts., 1:100 000. Hungarian Geol Survey, Budapest. RATSCHBACHER, L., BEHRMANN, J. & PAHR, A. 1990. Penninic windows at the end of the Alps and their relation to the intra-Carpathian basins. Tectonophysics, 172, 91-105. , Frisch, W., LINZER, H. G., & MERLE, O. 1991a. Lateral extrusion in the Eastern Alps, part 2.: structural analysis. Tectonics, 10, 257-271. --, SPERNER, B., MESCHEDE, M., DECKER, K., NEM~OK, M., NEM~OK, J. & GRYGAR, R. 1993a. The Pieniny Klippen Belt in the Western Carpathians of the northeastern Slovakia: structural evidence for transpression. Tectonophysics, 226, 471-483. - - , NEUBAUER,E, SCHMID,& NEUGEBAUER,J. ]989. Extension in compressional orogenic belts: The Eastern Alps. Geology, 17, 404-407. - - , MOSER, E, STRUSIEVICZ,R-O., BEDELEAN, H., HAR, N. & MOGOS, P-A. 1993b. Cretaceous to Miocene thrusting and wrenching along the Central South Carpathians due to a corner effect during collision and orocline formation. Tectonics, 12, 4, 855-873. , MERLE, O., DAVY, PH. & COBBOLD, P. 1991b: Lateral extrusion in the Eastern Alps, Part 1: Boundary conditions and experiments scaled for gravity. Tectonics, 10, 245-256. - - , SeERNER, B., MESC~EDE, M. & FmSCH, W. 1994.
Computer technics and applications: a program library for quantitative structural analysis. Tiibinger Geowissenschaftliche Arbeiter, Reide A, 21. ROYDEN, L. H. 1985. The Vienna Basin: a thin-skinned pull-apart basin. In: BIDDLE, K. T., CHRISTHIEBLICK, N. (eds) Strike-slip deformation, sedimentation, basin formation. SEPM Special Publications, 319-338. -1993. The tectonic expression of slab pull at continental convergent boundaries. Tectonics, 12, 303-325. , HORVATH,E & BURCHFIELB.C. 1982. Transform faulting, extension and subduction in the Carpathian-Pannonian region. Geological Society of America Bulletin, 73, 717-725. RUDINEC, R. 1989. New view on to the paleogeographic development of the Transcarpathian depression during the Neogene. Mineralia slovaca, 21, 27-42. RUMPLER, J. & HORVATH, E 1988. Representative Seismic Reflection Lines and Their Structural Intrpretations. In: ROYDEN, L. H. & HORVATH,F. (eds) The Pannonian Basin. A A P G Memoirs, 45, 153-170. SACCHI,M., HORVATH,F., & MAOYARI,O. 1999. Role of unconformity-bounded units in stratigraphy of continental records: a case study from the late Miocene of Western Pannonian basin, Hungary.
This volume. SALAJ, J. 1995. Geology of Middle Vfih valley Klippen and Periklippen belt, Stilov Paleogene and Mesozoics of northern part of Strfizovsk6 vrchy hills - part 3 [in Slovak with English abstract], Zemni Plyn a Nafta, Hodonfn, 40, 3-52. SANDULESCU, M., KRAUTNER, H. G., BALINTONI, I.,
333
RUSSO-SANDULESCU, D. & MICU, M. 1981. The structure of the East Carpathians (MoldaviaMaramures area). In: Guide to excursion B1, 12th Carpatho-Balkan Geological Congress. Institute of Geology & Geophysics of Romania, Bucharest, 92. SAUER, R., SEIFERT,R & WESSELY,G. 1992. Guidebook to Excursion in the Vienna Basin and the Adjacent Alpine-Carpathian Thrustbelt in Austria. Mitteilungen der Osterreich&chen Geolog&chen Gesellschaft, 85. SCHMID, S. M., AEBL[, H. R., HEILER, E & ZINGG, A. 1989. The role of the Periadriatic line in the Tectonic evolution of the Alps. In: COWARD, M. R, DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 153-171. SCHMIDT, TH., BLAU, J. & KAZM~R, M. 1991. Largescale strike-slip displacement of the Drauzug and the Transdanubian Mountains in early Alpine history; evidence from Permo-Mesozoic facies belts. Tectonophysics, 200, 213-232. SPAKMAN,W. 1990. Tomographic images of the upper mantle below central Europe and the Mediterranean. Terra Nova, 2, 542-553. SURMONT, J., SANDULESCU, M. & BORDEA, S. 1990. Mise en evidence d'une r6aimentation tiMCrdtac6 des S6ries M6sozoiques de l'unit6 de Bihor (Monts Apuseni, Roumanie) et de sa rotation horaire ultdrieure. Comptes Rendus de l'AcadOmie des Sciences, Paris, 310, 213-219. SZAB0, CS, HARANGI, Sz. & CSONTOS,L. 1992. Review of Neogene and Quaternary volcanism of the Carpathian-Pannonian region. Tectonophysics, 208, 243-256. SZENTPI~TERY,I. 1988. Oligocene and Lower Miocene formations of the Rudabfinya Mts. and their neighbourhood. Annual Report of the Hungarian Geological Institute from 1986, 121-128 [in Hungarian with English abstract]. SZTANO, O. 1994. The tide-influenced P6tervfisfira Sandstone, Early Miocene, Northern Hungary: sedimentology, paleogeography and basin development. Geologica Ultraiectina, 120, 1-155. & FODOR, L. 1997. Bathyal slope deposits in the Paleogene Basin: A case study of the Upper Eocene Piszke Marl (Nyergest~jfalu, Sfinc Hill, Hungary). FOldtani KOzlOny 127, 267-290. & JOZSA, S. 1996. Interaction of basin-margin faults and tidal currents on nearshore sedimentary architecture and composition: a case study from the Early Miocene of northern Hungary. Tectonophysics, 266, 319-341. & TARI, G. 1993. Early Miocene basin evolution in Northern Hungary: Tectonics and Eustacy. Tectonophysics, 226, 485-502. SZTRAKos, K. 1975. Wiederuntersuchung der Palaogenschuchten yon Kar~id-Buzs~ik. FOldtani KOzlOny, 105, 488-494 [in Hungarian with German abstract]. TARI, G. 1988. Strike-slip origin of the Vatta-Maklfir trough, northeastern Hungary. Acta Geologica Hungarica, 31, 101-109. 1992. Neogene transpression in the Northern Thrust Zone, Mecsek Mrs., Hungary. Annales
-
-
-
-
-
-
-
-
334
L. F O D O R E T A L .
Univ. Scientiarum Budapestiensis Section Geologica, 29, 165-187. -1994. Alpine tectonics of the Pannonian basin. PhD thesis, Rice University, Houston. - 1996. Extreme crustal extension in the R~ba river extensional corridor (Austria/Hungary). Mitteilungen der Gesellschaft der Geololgie und Bergbanstudenten in Osterreich, 41, 1-18. , BALDI, T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian Basin: a geodynamical model. Tectonophysics, 226, 433-455. , HO~VATH, F., & RUMPLER, J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. & WEIR, G. 1995. Palinspastic reconstruction of the Alpine/Carpathian/Pannonian system. In: HORVATH, E, TARI, G. & BOKOR, Cs. (eds) Extensional collapse of the Alpine orogene and Hydrocarbon prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to fieldtrip No. 6., Hungary, 119-132. TELEGDI-ROTH, K. 1951. Enseignements gdologiques de la prospection et de ia production du p6trole Biikksz6k. Year Book of the Hungarian Geolological Institute, 40, 1-21 [in Hungarian with French abstract]. TOMEK, C. 1988. Geophysical investigation of the Alpine - Carpathian Arc. In: Evolution of the northern margin of Tethys. International geological correlation program, project 198. Univ. of South Carolina, Occasional Publications ESRI New Series. & THON, A. 1988. Interpretation of Seismic Reflection Profiles from the Vienna Basin, the Danube Basin, and the Transcarpathian Depression in Czechoslovakia. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin. A A P G Memoirs, 45, 171-182. - - , IBRMAJER, I., KORAB,T., BIELY,A., DVOI~AKOVA, L., LEXA, J. & ZBORIL,A. 1989. Crustal Structures of the West Carpathians on deep reflection seismic line 2T. Mineralia slovaca, 21, 3-26. T(JNYI, I. & MARTON,E. 1996. Indications for large Tertiary rotation in the Carpathian-Northern
Pannonia region outside of the North Hungarian Paleogene Basin. Geologica Carpathica, 47, 43-50. VADASZ, E. 1935. Das Mecsek-gebirge. Geologische Beschreibung Ungarischer Landschaften, Hungarian Geological Institute, I. VASS, D., HOK, J., KovA4, R & ELE(~KO,M. 1993. [The Paleogene and Neogene tectonic events of the Southern Slovakia depression in the light of the stress-field analysis]. Mineralia slovaca, 25, 79-92 [in Slovak with English summary]. , KONE~NY, V. & SEFARA, J. 1979. Geologickd stavba Ipelskej kotliny a Krupinskej planiny. Geol. Ustav D. Stt~ra. , KovA~, M., KONE~NY, V. & LEXA, J. 1988a. Molasse basins and volcanic activity in the West Carpathian Neogen: its evolution and geodynamic character. Geologica Carpathica, 39, 539-561. --, NAGY, A., KOHUT, M. & KRAUS, I. 1988b. The Devinska Nova Ves Member: Olistostromes on the southeastern border of the Vienna basin. Mineralia slovaca, 20, 109-122. Vo~os, A. 1993. Jurassic microplate movements and brachiopod migrations in the western part of the Tethys. Palaeogeography, Palaeoclimatology, Palaeoecology, 100, 125-145. -& CSONTOS, L. 1992. Mesozoic plate-tectonic reconstruction of the Alpine-CarpathianPannonian region, part II: plate-tectonic reconstruction. Terra Nova Abstract Supplement 2, 4, 69. VRABEC, M. 1994. Some Thoughts on the Pull-apart Origin of Karst Poljes along the Idrija Strike-Slip Fault Zone in Slovenia. Acta Carsologica (Ljubljana), 23, 155-167 [in Slovenian with English abstract]. WEIN, Gu 1967. Uber die Tektonik StidostTransdanubiens. FOldtani KOzlOny, 97, 371-395. WESSELY, G. 1988. Structure and development of the Vienna Basin in Austria. In: ROVDEN, L. H. & HoRVATH, F. (eds) The Pannonian Basin. A A P G Memoirs, 45, 333-346. WEBER, B. 1985. Paleogene beds in the vicinity of Szigetvgr (South Hungary). FOldtani KOzlOny, 1 1 5 , 1-21.
Late Neogene sedimentary facies and sequences in the Pannonian Basin, Hungary E. J U H A S Z 1, L. P H I L L I P S 2, P. M U L L E R 1, B. R I C K E T T S 3, A. T O T H - M A K K 1, M. L A N T O S 1 & L. O. K O V A C S 1 1Geological Institute o f Hungary, 1143. Budapest, Stef(mia (tt. 14, Hungary 2USGS Branch o f Pacific Marine Geology, 345 Middlefield Road, M S 999 Menlo Park, CA 94025, USA 3Geological Survey o f Canada, Cordilleran Division, 100 West Pender Street, Vancouve~ B C V6B 1R8, Canada
Abstract: Detailed sedimentoiogical, facies and numerical cycle analysis, combined with magnetostratigraphy, have been made in a number of boreholes in the Pannonian Basin, in order to study the causes of relative water-level changes and the history of the basin subsidence. Subsidence and infilling of the Pannonian Basin, which was an isolated lake at that time occurred mainly during the Late Miocene and Pliocene. The subsidence history was remarkably different in the individual sub-basins: early thermal subsidence was interrupted in the southern part of the basin, while high sedimentation rate and continuous subsidence was detected in the northeastern sub-basin. Three regional unconformities were detected in the Late Neogene Pannonian Basin fill, which represent 0.5 and 7.5 Ma time spans corresponding to single and composite unconformities. Consequently two main sequences build up the Late Neogene Pannonian Basin fill: a Late Miocene and a Pliocene one. Within the Late Miocene sequence there are smaller sedimentary cycles most probably corresponding to climatically driven relative lake-level changes in the Milankovitch frequency band. Considering the periods, the estimated values for precession and eccentricity in this study (19 and 370 ka) are close to the usually cited ones. In the case of obliquity the calculated period (71 ka) slightly deviates from the generally accepted number. Based on the relative amplitudes of oscillations, precession (sixth order) and obliquity (fifth order) cycles had the most significant impact on the sedimentation. Eccentricity caused cycles (fourth order) are poorly detectable in the sediments. The longer term (third order) cycles had very slight influence on the sedimentation pattern. Progradation, recorded in the Late Miocene sequence, correlates poorly in time within the basin. The dominant controls of this process probably were changes of basin subsidence rate and the very high sedimentation rate. The slow, upward trend of silt and sand bed thickness as well as that of the grain size also reflects the local progradation.
From Oligocene to Pliocene time the Paratethys extended from the N o r t h e r n Alpine molasse basins to the Aral sea. The Pannonian Basin, as a part of the Central Paratethys, was situated between the Eastern Alps, the Dinaride chain and the Carpathian Mountains. Faunal and palaeogeographic evidence indicate that in the early Late Miocene (Sarmatian, Fig. 1) the Pannonian inland sea was finally disconnected from its n e i g h b o u r i n g basins and b e c a m e a lake (Jfimbor 1987; MUller & Magyar 1992a). Migration patterns of aquatic molluscs and stable isotope data of shells (Mfity~is et al. 1996) suggest that the lake had no outflow for the first few million years of its history (from about 12.0 Ma to 7.5 Ma). During the second half of the lake history (from about 7.5 Ma to 5.5 Ma or
slightly later), intermittent outflows occurred toward the neighbouring Dacian lake (South R o m a n i a and North Bulgaria, Mt~ller & Magyar 1992b). At that time the catchment area of the Pannonian lake was much smaller than that of the m o d e r n Pannonian Basin (H~mor 1988). Early stratigraphic analyses of the Pannonian lake sediments were based on facies d e p e n d e n t benthic molluscs (Bartha 1971). Like in other isolated basins of the world, the endemic fauna has posed difficulties for extrabasinal correlation. Progress has b e e n m a d e by applying m a m m a l biostratigraphic zonal schemes (Steininger et al. 1990), as well as other independent methods like radiometric and magnetostratigraphic dating (Pog~cs~s et aL 1988; Elston et al. 1990; K6kay et al. 1991). A fairly reliable
JUHASZ,E., PHILLIPS,L., MI~ILLER,P., RICKETTS,B., TOTH-MAKK,z~., LANTOS,M. & KOVACS,L. O. 1999. Late Neogene sedimentary facies and sequences in the Pannonian Basin, Hungary. In: DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 335-356.
336
E. JUHASZ E T A L .
Ma
Ser. P L I o
4-
c N E
Standard
Parathethyan
stages Piacenzian
stages
p
Romanian
A
Zanclean
Dacian
N
Mammal
zones M N 15
M N 14
N
Messinian 6-
M N 13 o
M
Pontian N
I
M N 12
I A
O
8-
N#
M N 11
C Tortonian I0-
E
s.
N
L.
Pannonian
M N 10
MN 9
E
Serravallian
Sarmatian
MN 8
12according to Jhmbor 1987
Fig. 1. Late Neogene standard time scale and
correlation with stages in the Central Paratethys.
date is available for the beginning of the Pannonian lake, i.e. the start of the Pannonian s.l. stage: 12.6 Ma (K6kay et al. 1991) and 11.6 Ma (Steininger et al. 1990). The top of the Pannonian s.1. stage is, by definition, the end of Pliocene. Sedimentological analysis of samples from 16 boreholes of the Hungarian Geological Institute was carried out with the aim of defining the sedimentary architecture of the basin, reconstructing the history of the basin-fill processes, interpreting the causes of the relative waterlevel changes of the P a n n o n i a n lake and analysing the basin subsidence. This study is based on integrated sedimentological and stratigraphic analysis (core and well-log evaluations, high-resolution stratigraphy), magnetostratigraphy and biostratigraphy. The Pannonian Basin consists of several subbasins, some containing more than 5000 m of Neogene sediments. The subsidence analyses shown here come from the shallower parts of the basin, where the whole Pannonian strata were penetrated by the borehole. However in the deeper sub-basins development of strata can be more complete, and deposition spanned a longer period. Sequence stratigraphy of the Pannonian Basin has been studied by Pogficsfis et al. (1988, 1992,
1993), Tari et aL (1992) and Vakarcs et al. (1995), using mostly reflection seismic data. They determined four third-order sequences in the Late Miocene strata and concluded, based on a comparison with the sea-level curves of Haq et al. (1987), that water-level 'was considerably affected by the Upper Miocene-Pliocene events that generated the global (eustatic) sea-level changes'. The results of this study, based on integrated magnetostratigraphic and sedimentological investigations, is to substantiate and quantify the above statement. Oscillation of the water level of the Pannonian lake was recognized quite early on the basis of classic (Bartha 1971; Jfimbor 1980), and later on modern, stratigraphic investigations (Korpfis-H6di & Pog~ics~is 1992; Pog~icsfis et al. 1988; Juhfisz Gy. 1993; Vakarcs et al. 1995). However, none of the studies dealt with the numerical characterization of the cyclicity and the cause of water-level changes. This study will also answer this question and suggests a possible cause for the lake-level oscillations together with the classification of the most characteristic cyclic components in the lake-level changes. The high-resolution, bed-by-bed-scale sedimentological study of the cores points to problems in correlating significant events in the basin. Methods
Sedimentological
method
We relied on samples from 15, continuously cored wells, located in Hungary (Fig. 2, Table 1). Sedimentological and lithological features together with the palaeontological records were observed using detailed logging of the cores and then interpreted in terms of sedimentary facies. Following the genetic stratigraphical method of Homewood et al. (1992) first the genetic depositional units were identified. Each unit is characterized by its most distant and closest facies to the shoreline. Graphically, the facies change from the deepest water (most distant facies) to the shallowest (closest facies) gives a triangle for each unit. The succession of the genetic units and their stacking pattern shows the changes or the stability of the palaeoenvironments (Fig. 3, see also Figs 9-11). Using magnetic polarity data, the stacking pattern diagrams for each borehole were reliably dated. The variation of the stacking pattern reflects the migration of the lake shoreline, i.e. the relative water-level changes at the site of the borehole. Numerical cycle analysis
Numerical cycle analysis aimed to reveal regularities in the thickness of lithological units: sand, silt, clay and
LATE NEOGENE
SEDIMENTS, PANNONIAN BASIN
337
/./i?
, z
_ Ck><, ., "\ .
t\
,," ~,,ox ~ t}
7
'i" "J
/
'~/ r
"~ ~
•
I
"
Im
; ~ i '?li,~--~
~l//IZ
~
,
/
7~.
- -~
/---\__J/
-
L?5~
~'7;?~
'
=ffj-)
Bh-3
2 ,-- 2-"
s-
. . . .
To-2
~.__)'
/ / Bacs-I
Fig. 2. Isopach m a p of the P a n n o n i a n basin fill of L a t e M i o c e n e age m H u n g a r y with the location of the studied b o r e h o l e s . L e g e n d : 1, b o r e h o l e s ; 2, thickness of P a n n o n i a n s.l. strata (in km); 3, o u t c r o p s of p r e - N e o g e n e rocks; 4, shelf edge (after Vakarcs 1997) in the m a r k e d time-interval.
T a b l e 1. Main data o f the boreholes Dating N a m e of the well
Short name
Bottom depth
B e d r o c k of Pannonian
Nagyl6zs-1 Szombathely-II Iharosber6ny-I Igal-7 Berhida-3 Tolnan6medi-2 Tengelic-2 Paks-2 Paks-3 Bficsalmfis-1 Jfinoshalma-1 Kaskantyti-2 Kecskem6t-3 Szirfik-2 D6vavfinya-2 V6szt6-1 Tiszapalkonya-1
Nlt-1 Szh-II Ib-I Ig-7 Bh-3 To-2 Te-2 Pa-2 Pa-3 Bfics-1 Jh-1 Kas-2 Kecsk-3 Szi-2 Dv-2 V-1 Tp-1
1335.2 2063.2 2000.0 1419.6 690.0 1200.0 1183.9 1593.0 625.0 1195.0 538.8 1387.5 700.0 2000.0 1200.0 1200.0 1987.8
Sarmatian Sarmatian Badenian Sarmatian Sarmatian Jurassic Sarmatian Sarmatian L. B a d e n i a n Cretaceous Sarmatian Sarmatian Sarmatian -
Biostrat.
Wells
Magneto
Studied
x x x
x x x x x x x x x x x x x x
x x x x x x x x x x x x x
x
Referred
338
E. JUHASZ E T A L . SEDIMENTOLOGICAL
Magnetostratigraphy
FEATURES
.. m 3 ~ i l~ ~ m - -q ~ ~ ~ [
~~
1
347 i i.i".~
[
.
.
~-~- I
:[:!i::i:i:
-
,
.
.
.
Legend:
~
i
4 1][][]
I
355 ~'. " : ' : ' i ! l
3
,
7
365
':""~':--::' ~",1,
,'.',','.
377
~
~
;
-
9
Z ~
i~ ~
13
,/W//-
I
Fig. 3. Lithological and sedimentological features, the stacking pattern of the genetic units and the facies in borehole Iharosberdny (Ib-I). The genetic units, which thickness ranges from 12 m to 5 m, were formed in delta plain environment with clear aggradational stacking pattern. Note the decrease of thickness of genetic units and the relative increase of sand comparing to other lithologies up-section. Legend: 1, silt; 2, silty clay; 3, marly silt; 4, lignite; 5, sand; 6, sandy silt; 7, sandy clay; 8, wood pieces in horizontal position; 9, leaves in horizontal position; 10, gastropods; 11, roots; 12, bivalves; 13, bioturbation; 14, trough cross-beds; 15, ripple marks; 16, large scale cross-beds; 17, conchoidai cracks; 18, organic-rich silt/clay.
coal layers. The series of thickness values were smoothed with running windows along the sections (Fig. 4). Also polynomials of different degree were fitted to the data series. Curves of low-degree polynomials reflect trend-like changes, while high-degree polynomials reveal high frequency rhythms. For categorizing of the interpreted sedimentary cycles, in terms of orders, the classification of Mitchum & Van Wagoner (199i) was used.
For dating of the Pannonian sedimentary rocks magnetostratigraphic study in eight boreholes have been performed (see Table 1). The palaeomagnetic samples, collected at 0.5 m intervals at the drill site, were measured in cryogenic magnetometers, mainly at the joint laboratory of the H u n g a r i a n Geological Institute and E6tvOs Lorgnd Geophysical Institute in Budapest, while samples of Tiszapalkonya-1 and Kaskantyfi-2 at the US Geological Surveys palaeomagnetic laboratory in Flagstaff (Arizona). Samples from the D6vawlnya-2 and V6szt6-1 boreholes were processed by Cooke et al. (1979). Following measurement of the natural remanent magnetization, series of pilot samples representing different lithologies, depths and inclinations were selected for progressive AF demagnetization. The remaining samples were demagnetized at one to three steps in 15-30 (40) mT. Geological and palaeomagnetic studies indicate that the sediments accumulated rapidly, were promptly buried and have remained undisturbed and unexposed since burial. Therefore the strata contain only minor secondary magnetizations that disappeared at 10-30 mT (Elston et al. 1990, 1994; Lantos & Elston 1995). The majority of inclinations exhibited no hint of different polarities near the threshold level of stability. The polarity zones of the boreholes were originally correlated with the geomagnetic time scale of Berggren et al. (1985), employing radioisotopic ages, and results of the lithostratigraphy, palaeontology and seismostratigraphy. Magnetostratigraphic correlations commonly are anchored to the long normal polarity interval of Chron C5n. Details of the correlation have been published elsewhere (Elston et al. 1990, 1994; K6kay et al. 1991; Lantos & Elston 1995; Pogficsfis et al. 1994); only a short summary is given here. Only a few K/Ar age determinations can be directly used as tie points. An age of 12.6 + 0.5 Ma was determined from a dacite tuff layer at the base of the Pannonian deposits in the Berhida-3 borehole (Balogh in K6kay et al. 1991). Seismic stratigraphic profiles connect magnetostratigraphic test holes and also boreholes that contain K/Ar radioisotopic ages in the basin. By way of illustration, an age of 9.61 + 0.38 Ma was reported for a basalt layer in the Kiskunhalas Ny-3 drill hole (Balogh et al. 1986). The seismostratigraphic correlation between the Kiskunhalas Ny-3 and Kaskantyti-2 drill holes indicates that the long interval of normal polarity in the lower part of the Kaskantyfi correlates
LATE N E O G E N E SEDIMENTS, PANNONIAN BASIN
339
silt thickness in m 15-
iI
oo oo oI |o i o o o
data ol
running average
',I
ct ii iI 11
10-
o n
oO
o e
','o
,
a
At.
_
.
,,
0 1000
i
e
i
I
I
I
I
800
600
400
200
15-
running average polynomial (10) fit 10-
"o
_ ....... m~
~
,,
'
'
"
l'I;'i
b
s
,
.'- '
e
0 1000
h,; "
o
,
800
I
I
I
600
400
200
15-
........
polynomial (10) fit polynomial (2) fit
10-
C
_
1000
I
1
I
I
800
600 depth in m
400
200
Fig. 4. Methods used for revealing trends and regularities in thickness values along sections. In order to visualise the oscillations of the values, discrete data points (vertices) are linked with a continuous curve. Running average (a) exposes small-scale changes, high-degree polynomials (b) capture rhythmic waves, lowdegree polynomials (c) display trend-like variations.
340
E. JUH/~SZ ETAL.
with chron C5n (Pogficsfis et al. 1994). The ages d e t e r m i n e d from m a g n e t o s t r a t i g r a p h y and K/Ar measurements were assigned to seismic horizons, then extended by tracing seismic horizons across the basin (e.g. Pogficsfis et al. 1992, 1994). The Limnocardium praeponticum mollusc zone is the lowermost sublittoral unit of the Pannonian lake sequence, and occurs between 11.5 and 10 Ma (Jfimbor et al. 1987). The L. praeponticum zone is present in the lower part of the Nagyl6zs-1 and Szombathely-II Pannonian sections (Korpfis-H6di 1992). Because of its presence, the lowermost long interval of normal polarity in these boreholes is correlated with chron C5n. Additionally, the Congeria banatica mollusc zone coincides broadly with chron C5n (Elston et al. 1990, 1994). The C. banatica zone was determined in the Berhida-3, Kaskantyfi-2, Nagyl6zs-1, Szombathely-II and TiszapalkonyaI boreholes. All magnetostratigraphic records were recently correlated with the new time scale of Berggren et al. (1995). The results of the integrated studies in the basin are internally consistent. As no radioisotopic data were available in stratigraphically higher parts of the sections, the age of strata at the top of the correlated sections is somewhat uncertain. The ages determined from magnetostratigraphy are summarized in Fig. 5.
Sequence stratigraphic results
sedimentary rocks in each borehole. At the top of the Miocene succession, yellow and white mottles, calcareous nodules, abundant root casts indicate subaerial exposure (Fig. 7). The thickness of the altered zone (palaeosol) below the unconformity ranges from 1 to 10 m. The Pliocene sequence starts with coarse channel sand and flood plain marl facies. There is a definite colour and grain-roundness difference between the Miocene and the Pliocene flood plain sediments: the colour of the Miocene layers is grey, while that of the Pliocene is multicoloured, brown, red, grey and green. The grains of the Miocene sand are very well and well rounded, and those of the Pliocene are subangular. In the Transdanubian part of the basin the Pliocene is poorly developed or missing (Fig. 6, boreholes Nlt-1, Szh-II, Bh-3, Ib-3) compared to that of the Great Hungarian Plain (Fig. 6, boreholes Tp-1, Kas-2). In the Iharosberdny-I well it is 21 m, in the studied boreholes of the Great Hungarian Plain it is maximum 140 m, but in some other wells (like in D6vavfinya-2 and Vdszt6-1) it can be a few hundred metres thick. Between the Miocene and the Pliocene sequences a time gap of about a 2.0-2.5 Ma is estimated.
Boundary 1 (SB1). A subaerial erosional boundary is present between the Pliocene and Pleistocene strata. In most of the Transdanubian region SB1 and SB2 appear together, composing one super unconformity, separating Quaternary and Miocene strata (Fig. 6, boreholes Nit-l, Szh-II, Bh-3, Ib-3).
Unconformities Three major unconform boundaries have been observed in the Late Neogene Basin fill (Fig. 6).
Boundary 3 (SB3). P a n n o n i a n strata unconformably overlie older Miocene and preNeogene rocks in the studied wells (except for the subbasin at the well Berhida-3). Magnetostratigraphic data show that the age of the basal strata ranges from 9.7 Ma to 11.2 Ma. In the different subbasins Pannonian strata cover the eroded surface of older formations: Upper Cretaceous schist in borehole Bficsalmfis-1, Upper B a d e n i a n red algal limestone in b o r e h o l e Iharosber6ny-I, Sarmatian sandstone in borehole Jfinoshalma-1, calcareous silt, sandstone and conglomerate in borehole Szombathely-II and Sarmatian mollusc-bearing, coarse-grained limestone in borehole Kaskantyfi-2. Boundary 2 (SB2). A significant unconformity is observed between the Miocene and Pliocene
Sedimentary units Between SB3 and SB2 an apparently continuous sequence was developed in the studied continuously cored boreholes including four units with distinct stacking pattern.
Transgressive (retrogradational) unit. In most of the studied boreholes, the base of the Pannonian sequence contains thin (0-10 m, exceptionally 23 m), poorly sorted sandstone and conglomerate (Fig. 8). The most common types of clasts are derived from the underlying Middle and Lower Miocene or pre-Neogene rocks, but rip-up clasts of mud or marl are also present. The maximum diameter of the usually rounded clasts is 4 cm. Only subtle stratification can be observed in the conglomerate layers. They are capped by welllaminated marl which contains molluscs, most commonly Congeria and Paradacna. This unit is called Basal Conglomerate in the deeper basins (Phillips et al. 1994).
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN co
co
co,
d,
4
,4
341
4
.+ ~S o ~
~-o
~/ ~/
~1/ oo/ rl/o~ / /
III
o o
/
~./ st I~~
/~
ool
~j
Isl/
L i
PJ/ J/ ,; c.I ~-
9
o
oo
/o
\ co ~
o
~4
.a
(i
9
.........
o
II
9
m o
9
9
8 m
o;
~;8 "S
o\
~ \
I I
\
\ .
.
.
/oo\ \ \
t
\
~s 9
.
.,...
Zl~2S atucl sa.rl~mou~ r
I
t
i
I
.....
I
t
I
1 .....
I
t
I -+
I
I
342
E. JUH/~SZ ETAL.
Fig. 6. Presentation of the unconformities and the changes of the palaeoenvironment v. time in the studied borehole sections. Legend: SB1-3, sequence boundaries; t, terrestrial; fl, fluvial; sh, shallow lacustrine; o, open lacustrine. Dotted pattern indicates stratigraphic gaps. L o w e r aggradational unit. The lower aggradational unit overlies the transgressive layers (retrogradational unit). Most of the dark grey marl, calcareous marl, clayey silt and silt beds are structureless and/or bioturbated, but locally are laminated. The thickness of the basin (offshore) marl varies from 200 to 650 m in the boreholes. In each borehole the marls are interbedded with 10-15 cm thick, graded sand beds, which have irregular but sharp bases. The aggradational unit is made up of 1-6 smallerscale cycles, each cycle containing facies which oscillate within the offshore, or between the offshore and shoreface. The thickness of the individual cycles range from 40 m to 210 m. The overall stacking pattern of the strata is aggradational (Figs 8 and 9). The lower aggradational unit includes the basal marl and the prodelta lacustrine turbidites of the B6k6s Basin, one of the deepest sub-basins of the Pannonian Basin (Phillips et al. 1994). Progradational unit. Progradational units (Fig. 9, and see also Fig. 6) with a thickness of 70-200 m occur in the middle part of the late Miocene sequence. It is made up of claystone, abundant coarse siltstone and fine sandstone. Strata dip as much as 7 ~. Graded bedding, alternation of siltstone and sandstone laminae, small-scale crossbedding, rip-up clasts in the base of sand-
stone beds and bioturbation are the most characteristic sedimentary features observed in these layers. The progradational units contain 1-4 smaller-scale cycles. Individual cycles range from 10 to 150 m in thickness. The environment of deposition changed from offshore to shoreface or to delta plain, showing a strong progradational stacking pattern (Fig. 9). The most characteristic facies here is shoaling beach; however in places where the sedimentation was dominated by deltas, delta-front and delta-slope deposits, with mouth bar, are common. The magnetostratigraphic data give approximately 0.3~).4 Ma duration for the deposition of the transitional unit. Second aggradational unit. In the upper part of the Late Miocene sequence flood plain facies sediments occur with aggradational stacking pattern (Fig. 10). These deposits represent a variety of palaeoenvironments, such as channel, lake, pond, marsh and flood plain. Large and small scale cross-bedding in the upward-fining sandstone, planar lamination or strong bioturbation of siltstones, shells or shell fragments, lignite beds and mottled clay and siltstone, calcareous nodules and root structures representing palaeosols are the most characteristic sedimentary features in this unit. The thickness of the upper aggradational unit
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN S E D I M E N T A R Y TEXT. & STRUCT.
343
CO1
O
L e g e n d :
CLASTICS s
a
n d
_z
day
(m) .Z-~
.-Y
i
...-.....-. . . . . , . - --~-
...
m
\
20s 4-i ~4 14.
.
.
.
.
.
210
.
.
.
6
.
. . . . . . . . . . . . . . . . . . . . . . . . 9
,,
'__-~
. . . . . .
7~oo
.
.
. . . . . . . .
9~ f
9
.
I0
.
. . . . .
.
.
II
. . . . 9
.
.
n
. . . . 215
-
' - -
.
.
.
.
.
. 9
12--
" .
.
.
.
.
i .
i
_ _
13 ~ ' ~
. . . . . . . . . . '7~i --'' . . . . .... , .... 220
.
. - -
.
. -
-
J4 f i
---i 77
----.i ...... --.4 . . . . . . . . ...... ..-i . . . . . . . . ,~,,:, - - - - , . . -----i
- -
-
too
H
-
i
,
H
. . . . 225-
_:--
18 l
z
" . . . . .
Z
19
SB2
!
X ~
H
It
~ : . . ~ ~ ~
~10o
Fig. 7. Unconformity, representing about 1 Ma gap between the Upper Miocene and Pliocene strata in borehole J~inoshalma-1. Concretions made of calcium carbonate, root casts and mottled structure indicate subaerial conditions along SB2. Legend: 1, sand; 2, sand with silt; 3, silty sand; 4, silt; 5, clayey silt; 6, clay; 7, rip up dolomite clasts in sand; 8, coalified plant debris on bedding plain; 9, root casts; 10, calcium carbonate concretions; 11, huminite; 12, planar bedding or lamination; 13, cross-bedding with the angle of cross strata; 14, bioturbation; 15, burrows along bedding plane; 16, pale colour; 17, medium colour; 18, dark colour; 19, sequence boundary.
varies from 30 to 1300 m. The thickness of the smaller-scale upward-fining cycles ranges from 3 to 22 m. In most cases, the sediment is of delta plain or flood plain facies, but in the case of the basinal areas (Kaskantyti-2, Iharosber6ny-I) shoreface facies are also present. The stacking
pattern of the cycles is aggradational with slight progradation, i.e. the thickness of the sand layers increases upward. Within this unit, coal and fine sand bed bearing cycles prove the oscillations of the lake level.
E. JUHASZ ETAL.
344
Ma 1 m -.~'---.=1 5 5 0 ~ ~"
MaI m 1650-)v-~.L_"U.g:3
L~U:
~='~
1700"~
|
2
I \
[ \
LOWER AGGRADATIONAL
9,74 __ 1 6 0 0.':. ~:"."".i ". _-?-Z§ ~ _ _ : _ar
~ R A
DATIONAL
UNIT
1650~~ 2~-Z z cs
1700C = : ~
.~."... 9 , ~ " 9 ~b
~_~~11,05 __
~ G ,
i
\ ~ ~
i
LOWER AGGRADATIONAL UNIT
~SGI~SSIVE SB2
Legend: 8 f/" 9 ~ l0 G H ~ 12 9 13 ~ 14 ~" 15 (~ 16 ~ 17 _/~ ISHCS 19~SB2
Fig. 8. Transgressive and lower aggradational stacking pattern of genetic units from borehole section Szombathely-II. Legend: 1, sandy conglomerate; 2, sand alternating with silt; 3, marly, sandy silt; 4, marly silt; 5, sand alternating with marly silt; 6, sandy, silty, marly conglomerate; 7, gastropods; 8, coalified plant debris; 9, strong bioturbation; 10, ichnofossil; 11, shell; 12, Characeae oogonium; 13, coalified leafs; 14, burrows; 15, CaCO 3concretion; 16, concretion; 17, slumping; 18, hummocky crossbedding; 19, sequence boundary.
Third (Pliocene) aggradational unit. A n additional aggradational unit is present between sequence boundaries SB1 and SB2 (Fig. 6). Its thickness in the studied boreholes varies from 20 m to 140 m, but in the basinal area it can exceed 500 m (see Fig. 5, Dv-2 borehole). It consists predominantly of coarse, fining-upward channel sand, and subordinate finer grained sediment with calcareous nodules, fluvial and terrestrial fauna (Viviparus, Helicidae). The entire unit contains one to five smaller-scale cycles with a maximum cycle thickness of 50 m. The facies oscillates within the fluvial regime in the marginal areas, while in the basinal areas between the delta plain and fluvial facies (Fig. 10). Cycle analysis Statistical analysis, carried out on four boreholes (5800 m of core), shows that despite the differences in relative abundance of the lithological
1850~ ~
Fig. 9. Aggradational (lower) and progradational stacking pattern of genetic units from borehole Tiszapalkonya-1. For legend see Fig. 8. The thick black line represents a coal layer.
units, variations in the silt and clay thickness show the same trend, and also the sand and the coal curves are very similar (Fig. 1 la, b). The two sets of curves are shifted by about one fourth of a wavelength (Fig. l l a ) . A slow, trend-like increase of the grain size is expressed along the sections. Depending on the complexity of variations, the smoothed curves and curves from fitted polynomials revealed several thickness oscillations with different wave lengths. The time control was given by magnetostratigraphic dating in each well. The following average periods were detected: c. 19 ka, c. 71 ka, c. 370 ka (Fig. 12a, b and c). Also a longer term >1 Ma cycle (Fig. 12c) was observed. From measuring the average distance between the extreme points of each curve, rough estimates were obtained for the amplitudes of the oscillations of the layer thickness: c. 0.5 m for the >1.0 Ma, c. 0.3 m for
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN
345
Kas.2 Maim
5oi
I
I SB1
i JPLIOCENE ~AGGRADATIONAL U N ~ SB2
//////PPER AGGRADATIONAL UNIT
r
Legend:
2~
77~ Fig. 10. The upper aggradational set of genetic units with elementary cycles from borehole Kaskantyti-2. together with the Pliocene sequence. For legend see Fig. 8. Additional legend: 1, bedded; laminated silt; 2, large-scale cross-bedding; 3, lost section; 4, shell fragments; 5, bioturbation; 6, cycle number.
the c. 370 ka, c. 1 m for the c. 71 ka, and c. 1.5 m for the c. 19 ka cycle. From this, the relative importance of the cyclic events on sedimentation can be estimated (Juhfisz et al. 1997).
Inferences for basin subsidence Thermal subsidence, beginning about 13 Ma ago, followed an earlier phase of rifting (Royden
1988). Pronounced differential subsidence also resulted in several sub-basins, some containing more than 5000 m of sediment. A l t h o u g h chronostratigraphic control is poor for the deeper portions of the sub-basins, deposition probably spanned a similar period to that represented by the four boreholes studied here, attesting to very high subsidence and sedimentation rates in the different parts of the basin.
346
E. JUHASZ E T A L .
----
5 -
4
~aoa] (Tp-i)
-
i
..=
coal
~3._h /
\
/
2 -
~,~
J /
9 \ \ N ,~
~ 'D ~D"D " Q' mD.b q,
"
1
0 1600
/
9 o 9149
o
~
-
~.
\
/ / /
/ /
"
I
I
I
I
I
i
I
1400
1200
1000
800
600
400
200
depth in m
5
-
4
-
! !
0!
(Szh-II) o o "*%
~3-
9 # S #
~**
o0
1
| I
I I
t
%%
J I I I
J
-
I 1000
I 800
I
I
600 depth
400
I 200
I 0
in m
Fig. 11. Thickness trend of the lithological units (sand, silt, clay and coal), revealed by numerical cycle analysis, in borehole Tp-1 and Szh-II. (a) The sand and coal curves and also the silt and clay curves move together and show the same pattern through the borehole. The degree of fitted polynomials is 5. There is a one fourth wavelength shift between the two sets of curves. (b) Sand and coal thickness trends in borehole Szh-II, represented by ten-degree polynomials. For more explanation see text.
The subsidence analyses shown here therefore r e p r e s e n t only the 'shallower' parts of the Pannonian Basin. In subsidence analysis, back-stripping a column of sediment separates the isostatic effects of sediment and water load, from the effects of tectonic subsidence (Steckler & Watts 1978). The data used to calculate tectonic subsidence, according to the techniques outlined by Sclater & Christie (1980), and Bond & Kominz (1984) are outlined below. The software used in these analyses was developed by the Basin Analysis G r o u p at the Free University of A m s t e r d a m (S. Cloetingh & R. Stephenson, pers. comm.).
(1) Stratigraphic time-depth data for the Pannonian Basin are derived from logged drill core and m a g n e t o s t r a s t i g r a p h y (discussed in this paper), and a few K - A r dates on felsic volcanic rocks and tufts. Stratigraphic-depth intervals, for which bracketing age dates are known, and the proportions of main lithological constituents were determined directly from continuous core in the deepest well (summarized in Table 2). The chronostratographic position of unconformities and their corresponding sequence boundaries are shown in Fig. 6. (2) Porosity-depth data, used to correct for changes in compaction and cementation, are not directly available in the boreholes: the standard
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN 6
6th order', average period=19 ka
~
347
~
E
"~
4
a
depth in m
0
i
I
700
I
I
600
i
I
500
i
I
400
I
200
9.44 5 -
I
300
9 . 3 4 a g e in M a
5th order: average period=71 ka
E~ 4 ._q 3-
b
2m
1
I
--
depth in m
0
I
1200 4
-
9.54
I
1000
I
i
I
800
I
I
I
600
I
I
400
200
T
i
0
9.31age in Ma
4th order: average period=370 ka 3rd order
3- 4 0 0 ka
~
1 -
( T ~ l -1 ? 800
[
1
depth into
i
i
i
I
i
i
700
600
500
400
300
200
7. 74
7.0
age
m Ma
Fig. 12. Cyclicities of different order shown by smoothed sand or silt thicknesses. (a) c. 19 ka cycles in borehole Szh-II, the curve was gained by a nine point running average; (b) c. 71 ka cycles revealed by fitting a nine-degree polynomial in borehole Szh-II; (c) c. 370 ky and >1 My cycles represented by a nine-degree and a three-degree polynomials, respectively, in borehole Tp-1. The age control is given from magnetostratigraphic measurements.
exponential p o r o s i t y - d e p t h relations and material parameters of Sclater & Christie (1980) were used. (3) The problem of determining palaeobathymetry, perhaps the most difficult parameter to establish in any basin analysis, is exacerbated in the Pannonian Basin by a highly endemic molluscan fauna. In the analysis we have used conservative estimates of palaeodepths based on facies analyses a n d
extensive palaeoecological studies (MiJller & Magyar 1992a,b). Palaeodepths range from zero where subaerial exposure is indicated (fluvial deposits, palaeosols), 0-10 m for shoreface, delta plain and lagunal-estuarine deposits, up to 50 m for shelf-like and outer delta plain deposits, to a maximum of 200 m for deeper basin and delta slope facies. Backstripping of four of the boreholes (Fig. 13) indicates two patterns of subsidence, again
E. J U H A S Z ETAL.
348
Table 2. Summary of data used in the back-stripping analyses D e p t h to base of unit (m)
Age (Ma)
Sand
Silt
Clay/ Shale
Carbonate
Water depth
Kaskantygt-2 50 MPU 151 SBI 241 SB2 300 410 490 580 700 830 910 1160 SB3 1185
0.1 1 1.9 3.2 3.8 5.5 5.9 6.5 6.7 7.4 8.2 8.9 9.2 11 12.8 14
0.34
0.5
0.16
0
0
0.8
0.1
0.1
0
0
0.5
0.3
0.2
0
0
0.15 0.43 0.17 0.08 0.21 0.13 0.08 0.08
0.85 0.57 0.83 0.8 0.7 0.7 0.62 0.3
0 0 0 0.12 0.09 0.17 0.3 0.47
0 0 0 0 0 0 0 0.15
0
0
0
1
20
0.1 1 1.9 3.2 3.8 5.3 8.8 9.8 12.8
0.86
0.14
0
0
0
0.9
0.07
0.03
0
0
0.47
0.53
0
0
0
0.65 0.01
0.35 0.79
0 0.2
0 0
200 10
0.1 1 1.9 3.2 3.8 6.2 6.4 6.7 7.4 7.9 8.2 8.9 10
0.34
0.5
0.16
0
0
0.71
0.06
0.23
0
0
0.42
0.22
0.36
0
0
0.45 0.62 0.53 0.25 0.48 0.38 0.13
0.28 0.23 0.4 0.61 0.47 0.52 0.77
0.27 0.15 0.07 0.14 0.05 0.1 0.1
0 0 0 0 0 0 0
0 5 5 20 20 50 200
0.1 1 1.9 3.2 3.8 5.2 8.8 9.6 10
0.29
0.53
0.18
0
0
0.44
0.23
0.33
0
0
0.35
0.58
0.07
0
0
0.21 0.1 0.11
0.69 0.1 0.1
0.07 0.8 0.7
0.03 0 0.09
10 20 50 100 100 100 100 200
Jdnoshalma-1 92 MPU 165 SB1 218 SB2 335 538 SB3
Tiszapalkonya-1 50 MPU 128 SB 1 266 SB2 400 510 870 1060 1220 1570 1987
Bglcsalmas-1 50 MPU 145 SB 1 190 SB2 280 495 532
200 200 200
D e p t h correspond to the base of each time unit and measure from the top of each well. Ages also correspond to the base of each unit and unconformity (sequence boundaries SB1, SB2 and SB3, and the middle Pleistocene unconformity MPU, see Fig. 15). The proportion of principal lithologies (sandstone, siltstone, shale, carbonate) are averaged over the thickness of each unit. Water depths (metres) are below sea level.
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN illustrating the differential nature of the basin dynamics. All four wells indicate rapid subsidence between 10 and 9 Ma, corresponding to the basal transgressive, and subsequent deep water deposits of the succeeding highstand. H o w e v e r there are significant differences between the Bficsalmfis-l-Jfinoshalma-1 boreholes, and the Kaskantyti-2-Tiszapalkonya-1 boreholes (note the scale difference in Fig. 13). The actual amount of accommodation space created in the latter two boreholes during this time period (tracked by the basement curve), is almost double that of the Bficsalmfis-1 Jfinoshalma-1 holes. Furthermore, the significant uplift recorded in both the tectonic and basement subsidence curves in the Bficsalmfis-lJfinoshalma-1 boreholes between about 6 and 4 Ma, is not observed in the other two boreholes: in the latter the basement and tectonic curves are markedly divergent, suggesting an additional component of accommodation space to that formed by thermal-isostatic subsidence. After about 4 Ma the pattern of subsidence in the Bficsalm~is-l-Jfinoshalma-1 wells is similar to that in the Kaskantyti-2 and Tiszapalkonya-1 wells. The disparities among the subsidence curves and the sea-level curve further illustrate the
peculiar nature of Pannonian Basin dynamics and the corresponding creation or loss of stratigraphic accommodation space (Royden 1988; Juhfisz 1991). Early subsidence during the Pannonian (s.l.) was interrupted in the Bficsalmfisl-J~inoshalma-1 successions because of differential tectonic uplift and reduction in sediment influx, perhaps related to strike slip movement on fault bounding subbasins (Horvfith 1984). During the same time interval, basement subsidence and high sedimentation rates continued the Kaskantyfi-2-Tiszapalkonya-1 successions. After 4 Ma, thermo-isostatic subsidence seems to have dominated in all four successions. In all four wells, the 6-4 Ma interval corresponds to the SB2 (subaerial) unconformity. Clearly, the dominant factors that determined the style and geometry of Pannonian Basin fill during the Late N e o g e n e were tectonism (including thermo-isostatic subsidence) and sediment influx.
Discussion Subsidence and infilling of the Pannonian Basin occurred mainly during the Late Miocene and Pliocene. The studied boreholes, with full core 0 . _
KASKANTYU-2 g:O.2 v
o 0.6
349
E v
-
o
-
TISZAPALKONYA ~._~
u \
"r~cto~lc +--+
~
I #
+--'
73 m 1.0-
I
1
I
I
I
t
-~
~ "x_~Basementl -I J.- :I ,1 ~I
10
5
time
Base
1.8 10
0
i
i
i
i
i
i
9
8
7
6
8
4
Ma
'1
3
i,
-~----~.
2
1
0
t i m e Ma.
O
I
E v
~
....
JANOSHALMA-1
E 0.2
O r-9
r-
73 "~
_Q -.j
"12
0.4
-~04
o~
I 10
9
8
7
6
5
time Ma.
4
3
2
l
I
I
t 10
I
I
t
[ 5
time Ma.
~
}
I
l 0
Fig. 13. Back-stripped subsidence curves for the Kaskantyti-2, Tiszapalkonya-1, Bficsalm~is-1and Jfinoshalma-1 boreholes. Note the scale differences in the graphs. The tectonic curve tracks the basin subsidence with the isostatic effects of sediment and water loads removed.
350
E. JUHASZ ETAL.
recovery, are sufficiently scattered over the area of the basin to depict a general picture of its evolution, except for the evolution of the eastern margin and the deepest southeastern subbasins. Sequence stratigraphical remarks are summarized on Fig. 6. On the figure the generalized depositional environments are pictured, as function of time, for each section. The curves illustrate the shoreline-shifts, which give information for the relative lake level changes. The major subaerial unconformities and the sequences between them are presented as well. For comparison, the eustatic sea level curve is added (Haq et al. 1987).
Unconformities Magnetostratigraphic data were applied to some of the borehole sections. At SB3 the basal Pannonian sedimentary rocks have a magnetostratigraphic signature that corresponds to an age of 11.8 and 9.7 Ma; the unconformity represents therefore a significant hiatus. The Berhida section is exceptional in this respect, since the sedimentation seems to be continuous between the Sarmatian and the Pannonian. The three unconformities (SB1, SB2, SB3), identified in the basin fill represent significant erosion and a minimum 0.5 Ma and maximum 7.5 Ma time gap. They are interpreted as regional or simple (SB3), and super or composite unconformities (SB2 and SB1). In terms of sequence stratigraphy they can bound third and/or second order sequences (see Fig. 6). Therefore in the studied boreholes two sequences are present: a Late Miocene, which, with the duration of its formation seems to exceed the third-order scale, and a Pliocene one. The unconformity at the Miocene/Pliocene boundary (SB2) seems to correlate well with the Messinian salinity crisis recognized by Hsti (1978).
Sedimentary features The succession of the changing sedimentary environments records the relative water level changes in the Pannonian lake. Between SB3 and SB2 an apparently continuous sequence developed, including four units with distinct stacking pattern: a basal transgressive, a lower aggradational, a middle progradational, and an upper aggradational units. The last mentioned three units may be recognized also on seismic profiles. An idealised profile of the Pannonian Basin fill (Fig. 14) shows the main sedimentological features, the facies distribution, the stacking pattern with the observed thickness
conditions, the time control, derived from the magnetostratigraphic measurements, the bounding major unconformities, and the relative lake-level changes based on the genetic stratigraphic analysis. Above SB3 a relatively rapid transgression is expressed by the landward shift of facies. Subsequently, deep-water environments were established and a distinctive aggradational unit accumulated. The striking basinward shift of the facies, i.e. a drop in relative water level, is a characteristic part of each lake level curve. The second aggradational unit, up to the Miocene-Pliocene boundary (SB2), records an equilibrium between sedimentation rate and basin subsidence (i.e. there were no major changes in the accommodation potential). In the studied boreholes we have not found any sign of deeper water facies capping shoreface or delta front facies, except the thin basal conglomerates and sandstones. The Pliocene sequence is situated between the SB2 and SB3 unconformities. The predominantly fluvial character is strikingly different to the upper part of the Late Miocene sequence. In the studied boreholes, in terms of sequence stratigraphy, there are no complete sequences in either order. The lowstand deposits are lacking or not recognizable by sedimentological methods in deepwater facies. Only thin transgressive and thick highstand deposits are evident. But, as all the studied boreholes are located in the shallow, marginal part of the Pannonian Basin, we can not exclude the occurrence of complete sequences in the deep sub-basins.
Correlation Six boreholes are presented (Fig. 6) for the intrabasinal correlation. They were dated by magnetostratigraphy, using biostratigraphical and radiometric tie-points, enabling to make a good estimation of the age of the sedimentary cycles (Fig. 5). In the case of the Pliocene and Quaternary deposits, the time span of deposition was estimated mostly by biostratigraphical correlation. The recognized facies were assembled into four main facies groups: terrestrial, fluvial, shoreface and offshore sediments. The temporal change of environment at the studied sites was plotted against time in the diagrams (Fig. 6). The diagram illustrates also the global (eustatic) sea level changes as proposed by Haq et al. (1987). These diagrams may be regarded as a rough estimation of relative lake level variations in the former environs of the studied boreholes. The stacking pattern of each borehole shows a
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN
Lithologic and sedimentary features
Age
Ma
Depositional environment
RADAT,ONA ]A
3.2-1.8 -" ~ S B 1
6.2-3.9 --
Shoreline shift and stacking pattern
sea I
mud. ] sd. [gr.
E
land
~
~
2
351
SBI
ALLUVIAL PLAIN
UNIT 40-140m
SB2
~
SB2
AGGRADATIONAL UNIT 90-1300m
DELTA PLAIN
~z BEACH ! !
DELTA FRONT
PROGRADATIONAL UNIT 70-200m
i BASIN & SLOPE
11.2-9.8~
3
AGGRADATIONAL UNIT 100-400m
SB3 TRANSGRESSIVE UNIT 2-10m
Fig. 14. Complex presentation of the stratigraphic, lithological, sedimentological and facies characteristics for the eastern part of the Pannonian basin fill. The cycle analysis was carried out on the Upper Miocene sequence (except the basin and slope facies sediments). The upper part of the Upper Miocene sequence is cyclic; however the cycle thickness and cycle stratigraphy can be variable. The stacking pattern of the genetic units and the shoreline shift indicate two dramatic changes in the life of the Pannonian lake: right above SB3 a rapid and intense transgression, which gave rise to the lake, and the progradational unit with a distinct shoreline shift towards the sea. The ages of the sequence boundaries were determined by magnetostratigraphic measurements. The lithological and sedimentologicalfeatures were identified on the continuous cores of the studied boreholes. For legend see Figs 8 and 9.
general regressive, upward-shoaling trend. The shift of the site of deposition from terrestrial to offshore, then from offshore to shoreline, or terrestrial facies groups, is time-transgressive in the basin. Thus, the shift from offshore to shoreface (or even to fluvial and to terrestrial) facies
groups reflects the prograding delta- and i n t e r d e l t a infilling process. Considering the magnetostratigraphic data, it is evident, that the p r o g r a d a t i o n a l phases in each well do not coincide in time within the basin. The dominant controls of the p r o g r a d a t i o n a l process,
352
E. JUH/i, SZ E T A L .
probably, were the different subsidence in the individual subbasins and the very high sedimentation rate, which overprinted the influence of the water level changes in this scale in the studied area. Most probably the signals of low frequency lake level changes are overprinted by the high sediments influx. Very faint records of low frequency lake level changes were detected only in few cases (see Cycle analysis and Relative water level sections). Relative water level
Numerical cycle analysis revealed similarities and differences in the thickness change of the lithological components. The same thickness trend of the silt + clay and sand + coal, and the shift between the two pairs (see Fig. l l a and b) leads to the question of interpretation of low and high water level positions. Concerning lake-level variations, most literature supposes climatic changes as one of the most important factors. They determine low lake level as a response of aridification which causes coarse bedload at nearshore areas and chemical precipitation in the basin; while in the case of high lake level due to humid climate it is not expected to load coarse and abundant sediment, because of the reduction of the stream gradient and the stabilisation of the bank by vegetation (Swann 1964; Picard & High 1972, 1981; McGowen et al. 1979; Galloway & Hobday 1983; Allen & Collinson 1986; Talbot & Kelts 1989; Dam & Surlyk 1992, 1993; Surlyk et al. 1993 and others). Opposite interpretations of coarse grained sediments to be products of high water level and the result of increased runoff was suggested by Brough (1928) and Olsen (1984), who also proposed mainly chemical precipitation during low lake level. Data from the Pannonian lake sediments support the latter interpretation. The same thickness trend of the sand and coal layers suggests that they both were formed during rising lake-level conditions (see Fig. l l a , b). A number of papers suggested that coal is produced in great thickness in transgressive environments (Galloway & Hobday 1983; Einsele 1992; Riegel 1991). In the case of the Pannonian lake, however, more sand was transported into and deposited in the lake during transgression and high lake-level. The periods of the detected higher order cycles in the Pannonian sediments are very close to, or are equivalent with the climatic cycles discovered by Milankovitch (1940) and Bacsfik
(1944). The values of the Milankovitch frequency band are: 14-18 ka for precession (Berger 1980, 1988), 41 ka for obliquity (Berger et al. 1989), and 100 and 400 ka for eccentricity (Imbrie & Imbrie 1979 in Fischer & Bottjer 1991). In the early Pliocene (between 5 Ma and 2.8 Ma), which is very close in time to the studied Pannonian sequence, Tiedeman et al. (1994) found the obliquity period (41 ka) as the dominant record, while later (3-1.5 Ma) the precessional periodicities (19-23 ka) were dominant. The average period values of the cycles identified in the Pannonian sediments are: c. 19 ka caused by precession; c. 71 ka caused by obliquity and c. 370 ka is the longer period of eccentricity (Fig. 12c). The >1.0 Ma (see Fig. 12c) cycle, which is very faintly developed in the Pannonian sediments, can be an analogue to the third-order global climatic cycle, which causes eustasy. This is expressed by the lower order curve (see Fig. 12c). The rhythm of the oscillation of the lake is very similar to that of the seas, but no evidence of direct or indirect correlation has been found. For a partly similar situation, an inverse correlation was suggested by Semenenko (1987). Studying the records of the past 300 ka of the Caspian sea, he found, that in those periods when the water-level of the world oceans was high, the water-level of the Caspian sea was low and vice versa (Semenenko 1987). The relative amplitudes of the oscillations, i.e. the impact of the cyclic climatic event on the layer thickness suggests that the major part of the regular changes in the layer-thickness was controlled by the precession and obliquity driven climate changes (fifth and sixth order cycles). In other words, among the climatic factors the higher order cycles had the greatest impact on the sedimentation pattern of the Pannonian lake. The fourth (eccentricity) and third order climatic cycles (eustatic in the world oceans) had only a very faint influence on the sedimentation of the Pannonian lake.
Conclusions (1) Basin subsidence curves indicate the different nature of Pannonian Basin dynamics and the corresponding creation or loss of stratigraphic accommodation space. Early subsidence during the Pannonian (s.l.), between about 5.3 and 4.0 Ma, was interrupted in the southern part of the Great Hungarian Plain, while during the same time interval, basement subsidence and high sedimentation rates continued in the northern sub-basins. After 4 Ma, thermo-isostatic subsidence seems to have dominated in all four successions.
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN The 6-4 Ma interval corresponds to the SB2 (subaerial) unconformity, whereas the H a q et al. (1987) curve in the same period shows transgression and sea-level highstand. Clearly, the dominant factors that determined the style and geometry of Pannonian Basin fill during the Late Neogene were tectonics (including thermoisostatic subsidence) and sediment influx. Although the subsidence analysis alone cannot discount possible eustatic effects, it does illustrate that eustasy played no significant role in the sequence-stratigraphic architecture of the Pannonian Basin. (2) On the basis of the detailed sedimentological studies three regional unconformities were detected in the Late Neogene Pannonian Basin fill, which can correspond to third and second order sequence boundaries. Consequently two sequences build up the Late Neogene Pannonian Basin fill: a Late Miocene, which in terms of Vail et al. (1991) can be interpreted as a second order, and a Pliocene, which is a third order sequence. Based on palaeomagnetic data, it is obvious that neither the transgression nor the progradation, recorded in the Late Miocene sequence, coincide in time within the basin. The dominant controls of the latter process probably were the rapidly changing basin subsidence and the very high s e d i m e n t a t i o n rate. The general slow increase in frequency and thickness of silt and sand beds along the sections would fit the idea that the Pannonian lake was infilled by the basinward migration of the marginal facies. (3) There are no similarities between the sealevel curve (Haq et al. 1987) and that of the Pannonian lake level in second or third order scale, suggesting that the sea level fluctuations had no direct influence on the lake level. However, the high frequency (>1 Ma magnitude) water-level oscillations of the Pannonian lake, which were faintly observable in the sedimentation, could have followed the r h y t h m of the eustatic changes of the world oceans. This points to the careful interpretation of the sedimentary cycles of an endorheic lake in terms of sequence stratigraphy, and also to the problems of direct correlation b e t w e e n lacustrine and marine sequences. The Late Miocene fill of the Pannonian Basin shows a distinct, high-frequency cyclicity caused by climatically driven relative water-level changes. In terms of sequence stratigraphy, the series sedimented during these time intervals can correspond to fourth, fifth and sixth order sequences. The water-level fluctuations are evidently caused by cyclic climatic changes of the Milankovitch frequency band, mainly due to
353
variations in precession and obliquity. Eccentricity caused cycles (fourth order) are poorly detectable in the sediments. The longer term climatic cycles (third order) had a very slight influence only on the sedimentation pattern of the Pannonian lake. (4) The SB2 boundary, at the top of the Late Miocene sequence, seems to reflect a major global or Mediterranean event. It may be correlated in time, and probably, causally to the Messinian salinity crisis and Lago Mare event, as well. A similar conclusion was drawn by Csat6 (1992; 1995). The authors express their deep gratitude to the Hungarian Science Foundation (OTKA T 7372 and T 019679), the US-Hungarian Joint Fund J.F.No.329. and the Integrated Basin Studies EC project which supported this research. The study was completed in the framework of a co-operation between the Geological Institute of Hungary and the Geological Survey of Canada. We are grateful to S. Cloetingh and R. Stephenson (Free University, Amsterdam) for permission to use the backstripping software.
References ALLEN, P. A. & COLLINSON, J. D. 1986. Lakes. In: READING H. G. (ed.) Sedimentary environments and.facies. Blackwell Science, Oxford 63-94. BACSAK, G. V. 1944. Die Erdgeschichte der Letztverftossenen 600 000 Jahre. Commentarii de Studiis
Disputationum Instituti Geologici Regii Hungarici, 6 (5). BALOGH, K., ARVA-SOS,E. & PI~CSKAY,Z. 1986. K/Ar dating of post-Sarmatian alkali basaltic rocks in Hungary. Acta Mineralogica-Petrographica, 28, 75-93, Szeged. BARTHA, E 1971. A magyarorsz~igi pannon biosztratigrgfiai vizsg~ilata. [Biostratigraphic investigation of the Pannonian layers, Hungary]. In: GOCZANE & BENKO J. (eds) A magyarorszdgi pannonkori k@z6dm~nyek kutat6sai. Akad6miai Kiad6, Budapest 11-172 [In Hungarian]. BERCZI, I. & PHILHeS,R. L. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, Southeast Hungary. In: KIL~NYI t~. • TELEKI P. G. (eds) Geophysical Transactions, Special Edition. EOtv6s Lor~ind Geophysical Institute of Hungary, 31, 55-74 (also US Geological Survey Open-File Report, 85-360). BERGER, A. 1980. Milankovitch astronomical theory of paleoclimates: a modern review. Vistas in Astronomy, 24, 103-122. - - 1988. Milankovitch theory and climate. Reviews of Geophysics, 26, 624-657. , LOUTRE, M. E & DEHAN2; V. 1989. Influence of
the changing lunar orbit on the astronomical fiequences of pre-Quaternary insolation patterns. Paleoceanography, 4, 555-564. BERGGREN, W. A., KENT, D. V., FLYNN, J. J. & VAN
354
E. JUH/~SZ E T A L
COUVERING, J. A. 1985. Cenozoic geochronology. Geological Society of America Bulletin, 96, 1407-1418. - - , SWISHER, C. C. & AUBRY, M.-R 1995. A revised Cenozoic geochronology and chronostratigraphy. - In: BERGGREN,W. A., KENT, D. V., AUBRY, M.-R & HARDENBOL, J. (eds) Geochronology time scales and global stratigraphic correlation. SEPM Special Publications, 54, 129-212. BOND, G. C. & KOMINZ, M. A. 1984. Construction of tectonic subsidence curves of the early Paleozoic miogeocline, southern Canadian Rocky Mountains: implications for subsidence mechanisms, age of breakup, and crustal thinning. Geological Society of America Bulletin, 95, 155-173. BROUGH, E. 1928. On rhythmic deposition in the Yordale Series. Proceedings of the University Durham Philosophical Society, 8, 116-126. COOKE, H. B. S., HALL, J. M. & RONAI, A. 1979. Paleomagnetic, sedimentary and climatic records from boreholes D6vav~inya and Vdszt6, Hungary. Acta Geologica Hungarica, 22, 86-109. CSAT0 ,I. 1992. The Messinian water-level change and the connected stratigraphic traps in the Pannonian basin, Hungary. In: Abstracts, EAPG 4th Conference, Paris 1-5 June 1992, 110. -1995. Sequence stratigraphic interpretations and modelling in lacustrine rift basins; southern Dead Sea basin, Israel and Pannonian Basin, Hungary. PhD thesis, University of South Carolina, Columbia. DAM, G. & SURLYK, E 1992. Forced regression in a large wave - and storm-dominated anoxic lake, Rheatian-Sinemurian Kap Stewart Formation, East Greenland. Geology, 20, 749-752. & -1993. Cyclic sedimentation in a large wave- and storm-dominated anoxic lake; Kap Stewart Formation (Rhaetian-Sinemurian), Jameson Land, East Greenland. Special Publications of the International Association of Sedimentology, 18, 419--448. EINSELE, G. 1992. Sedimentary Basins. SpringerVerlag. ELSTON, D. R, LANTOS, M. & H/~MOR, T. 1990. Az AIf61d Pann6niai (s.I.) k6pz6dm6nyeinek magnetosztratigr~ififija. Annual Report of Hungarian Geological Institute 1988, 109-134. & -1994. High resolution polarity records and the stratigraphic and magnetostratigraphic correlation of late Miocene and Pliocene (Pannonian, s.l.) deposits of Hungary. In: TELEKI, R G., MATTICK, R. E. & KOKAI, J. (eds) Basin Analysis in Petroleum E~ploration. Kluwer Academic Publishers, Netherlands, 111-142. FISCHER, A. G. & BOTTJER, D. J. 1991. Orbital forcing and sedimentary sequences. Journal of Sedimentary Petrology, 61, 1063-1069. GALLOWAY,W. E. & HOBDAY,D. K. 1983. Terrigenous Clastic Depositional Systems. Springer-Verlag. HAMOR, G. (ed.) 1988. Neogene paleogeographic atlas of Central and Eastern Europe. Hungarian Geological Institute, Budapest. HAQ, B. U., HARDENBOL, J. &; VAIL, P. R. 1987. Chronology of fluctuating sea levels since the -
-
Triassic (250 Myr ago to present). Science, 235, 1156-1167. HOMEWOOD, P., GUILLOCHEAU, E, ESCHARD, R. & CROSS,T. A. 1992. Corrdlations haute r6solution et stratigraphie g6ndtique: Une ddmarche int6gr6e. (High resolution correlations and genetic stratigraphy: An integrated approach.) Bulletin du
Centres de Recherche Exploration-Production Elf-Aquitaine, 16, 357-381. HORVATH, E 1984. Neotectonics of the Pannonian basin and the surrounding mountain belts: Alps, Carpathians and Dinarides. Annales Geophysical, 2, 147-154. Hs0, K. J. 1978. When the Black Sea was drained. Scientific American, 238, 5, 53-63. JA.MBOR A. 1980. Pannonian in the Transdanubian
Central Mountains. A Dundntfli - kOz@hegysdg pann6niai kdpz6dmOnyei. A Magyar/i, llami F01dtani Int6zet EvkOnyve, 62. (ed.) 1987. A Magyarorszdgi kunsdgi emeletbeli
--
k~pz6dm~nyek fOldtani jellernz&'e [Geologische Characterisierung der Ablagerung~en der KunsdgStufe in Ungarn]. A Magyar Allami F61dtani ---,
Intdzet I~vk6nyve 69. BAL,a,ZS, E., BALOGH, K. & 24 OTHERS 1987. General characteristics of Pannonian s.1. deposits in Hungary. h~: Proceedings of the 8th Congress,
Regional Committee on Mediterranean Neogene Stratigraphy, Budapest, 15-22 September, 1985. Annals of the Hungarian Geological Institute, 70, 1 5 5 - 1 6 7 .
JUH,a,SZ, E., MtJLLER, P., R1CKETI'S,B., TOTH-MAKK,/~., HAMOR, T., FARKAS-BULLA,J. & SOT0-SZENTAI, M. 1996. High resolution sedimentological and subsidence analysis of the Late Neogene in the Pannonian Basin, Hungary. Acta Geologica Hungarica, 39, 129-152. --, KOVACS, L. O, MI~LLER, R, TOTH-MAKK, /~., PHn.LWS, L. & LANTOS, M. 1997. Climatically driven sedimentary cycles in the late Miocene sediments of the Pannonian Basin, Hungary. Tectonophysics, 282, 257-276. JUHASZ, GY. 1991. Sedimentoiogical and lithostratigraphical framework of the Pannonian (s.1.) sequence in the Hungarian Plain, Eastern Hungary. Acta Geologica Hungarica, 34, 53-72. 1993. Relativ vizszintingadozfisok rdtegtani-szedimentol6giai bizonyftdkai az AlfOid pann6niai s. 1. tileddk6sszlet6ben [Sedimentological and stratigraphical evidences of water-level fluctuations in the Pannonian Lake]. FOldtani KOzlOny, 123, 379-398. K~,ZMER, M. 1990. Birth, life and death of the Pannonian Lake. Paleogeography, Palaeoclimatology, Palaeoecology, 79, 171-188. KOKAY J., HAMOR T., LANTOSM. & MOLLER P. 1991. A Berhida 3. sz. ftirfis paleomfigneses 6s f61dtani vizsgfilata (The paleomagnetic and geological study of borehole section Berhida-3). A Magyar -
-
,4llami FOldtani Int~zet Evi Jelent&e az 1989. Ovr6l. 45-63 [in Hungarian with English summary]. KORPAS-HODI M. 1983. A Dundnt~li-kOzOphegysdg
dszaki el&ere pann6niai mollusca faundjdnak
LATE N E O G E N E SEDIMENTS, PANNONIAN BASIN paleoOkolGgiai ds biosztratigrGfiai vizsgGlata. [Palaeoecology and Biostratigraphy of the Pannonian mollusca fauna in the Northern foreland of the Transdanubian Central range.] A Magyar Allami F/Jldtani Intdzet I~vkGnyve, 66. 1992. A Szombathely II. sz. ftir~is pannGniai (s.1.) molluscai. The Pannonian (s.1.) molluscs of borehole section Szombathely II. Annual Report of Hungarian Geological Institute 1990, 505-525. & POGACSAS,GY. 1992. Paleogeographic outline of the Pannonian s.1. of the southern DanubeTisza interfluve. Acta Geologica Hungarica, 35, 145-163. LANTOS,M. & ELSTON,D. P. 1995. Low- to high-amplitude oscillations and secular variation in a 1.2 km late Miocene inclination record. Physics" of the Earth and Planetary Interiors, 90, 37-53. - - , HAMOR,T. & POGACSAS,GY. 1992. Magneto- and seismostratigraphic correlations of Pannonian s.l. (Late Miocene and Pliocene) deposits in Hungary. Paleontologia i EvoluciG, 24--25, 35-46. MArrvAs J., BURNS,S. J., MULLER R & MAGYARI. 1996. What can stable isotopes say about salinity? An example from the late Miocene Pannonian Lake. Palaios, 96, 31-39. McGOWEN, J. H., GRANATA, G. E. d~ SENI, S. J. 1979. Depositional framework of the lower Dockum Group (Triassic), Texas Panhandle. Bureau of Economic Geology, University of Texas, Austin, Report of Investigation No. 97. MILANKOVITCH,M. 1940. Kanon der Erdbestrahlung und seine Anwendung auf das Eiszeit Problem. Royal Serbian Academy, Beograd, 133. MITCHUM, R. M. & VAN WAGONER, J. C. 1991. Highfrequency sequences and their stacking patterns: sequence-stratigraphic evidence of highfrequency eustatic cycles. Sedimentary Geology, 70, 131-160. MI3LLER, P. d~;MAGYAR,I. 1992a. Continuous record of the evolution of lacustrine cardiid bivalves in the Late Miocene Pannonian Lake. Acta Palaeontologica Polonica, 36, 353-372. & -1992b. A Prosodacnomydk rGtegtani jelentGsGge a K/3tcse kGrnyGki pannGniai s.l. tiledGkekben. Stratigraphic significance of the Upper Miocene lacustrine cardiid Prosodacnomya (KGtcse section, Pannonian Basin, Hungary). FOldtani KOzlOny, 122, 1-38. NAGYMAROSY,A. & MI~LLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian basin. In: ROYDEN,L.H. & HORVATH,E (eds) The Pannonian basin - a study in basin evolution. AAPG Memoirs, 45, 69-77. OLSEN,E E. 1984. Periodicity of lake-level cycles in the Late Triassic Lockatong Formation of the Newark basin (Newark Supergroup, New Jersey and Pennsylvania). In: BERGER, A., IMBRIE, J., HAYS, J., KUKLA, G. & SALTZMAN, B. (eds) Milankovitch and Climate. Understanding the response to Astronomical forcing. D. Reidel Publ. Co., 1, 129-146. PHILLIPS,R. L., REVESZ,I. & BERCZI,I. 1985. Processes and depositional environments' of Neogene deltaiclacustrine sediments, Pannonian basin, Southeast -
-
-
-
355
Hungary: core investigation summary. US Geological Survey Open-file Report, 85-360. & 1994. Lower Pannonian deltaic-lacustrine processes and sedimentation, BGkGs basin. In: TELEKI, E G., MATrICK, R. E. & KOKAI, J. (eds) Basin Analysis' in Petroleum Exploration. Kluwer Academic Publishers, Netherlands, 67-82. PICARD, M. D. &HIGH, L. R., JR. 1972. Criteria for recognising lacustrine rocks. In: RIGBu J. K. & HAMBLIN,W. K. (eds) Recognition of ancient sedimentary environments. SEPM Special Publications, 16, 108-145. & -1981. Physical stratigraphy of ancient lacustrine deposits. In: ETHRIDGE,E G. & ROMEO, I. M. (eds) Recent and ancient nonmarine depositional environments: models for exploration. SEPM Special Publications, 31,233-259. POGACSAS, GY., LAKATOS, L., UJsz/tszI, K., VAKARCS, G., VARKONY1,L., VARNA1,E & REVESZ, I. 1988. Seismic facies, electro facies and Neogene sequence chronology of the Pannonian Basin. Acta Geologica Hungarica, 31, 175-207. --, MA'rrlCK, R. E., ELSTON, D. E, HAMOR, T., JAMBOR, A., LAKATOS,L., LANTOS,M., SIMON,E., VAKARCS, G., VARKONYI, L. & VARNAI, E 1994. Correlation of seismo- and magnetostratigraphy in southeastern Hungary. In: TELEKI, E G., MATTICK R. E. & KOKAI J. (eds) Basin Analysis" in Petroleum Exploration. Kluwer Academic Publishers, Netherlands, 143-160. --, MIJLLER, P. & MAGYAR, I. 1993. The role of seismic stratigraphy in understanding biological evolution in the Pannonian Lake (SE Europe, Late Miocene). Geologica Croatica, 46, 63-69. --, SZAB0, A. & SZALAY, J. 1992. Chronostratigraphic relations of the progradational delta sequence of the Great Hungarian Plain. Acta Geologica Hungarica, 35, 311-327. RIEGEL, W. 1991. Coal cyclothemes and some models for their origin. In: EINSELE, G., RICKEN, W. & SEILACHER,T. (eds) Cycles and events in stratigraphy. Springer, 733-750. ROYDEN, L. H. 1988. Late Cenozoic Tectonics of the Pannonian Basin System. In: ROYDEN, L. H. & HORVA'rH,E (eds) The Pannonian Basin - a study in basin evolution. A A P G Memoirs, 45, 27-48. ROGL, E & STEININGER, E 1983. Vom Zerfall der Paratethys zu Mediterran und Paratethys. Annales der naturhistorisches Museum Wien, 85/A, 135-163. SCLATER, J. G. • CHRISTIE, P. A. E 1980. Continental stretching: an explanation of the post MidCretaceous subsidence of the central North Sea basin. Journal of Geophysical Research, 85, 3711-3739. SEMENENKO,V. N. 1987. Stratigraphycal correlation of the Upper Miocene and Pliocene of the Eastern Parethetys. Ukrainian Academy of Science, Kiev, 'Naukova Dumka' publisher. [in Russian]. STECKLER,M. S. &WATTS,A. B. 1978. Subsidence of the Atlantic-type continental margin of New York. Earth and Planetary Science Letters, 41, 1-13. STEININGER, E E, BERNOR, R. L. d~ FAHLBUSCH, V. -
-
356
E. J U H A S Z ETAL.
1990. European Neogene marine/continental chronologic correlations. In: LINDSAY, E. H., FAHLBUSCH, V. & MEIN, P. (eds) European Neogene Mammal Chronology. Plenum, New York, 15-46. SURLYK, E, NOE-NYGAARD N. & DAM, G. 1993. High and low resolution sequence stratigraphy in lithological prediction - examples from the Mesozoic around the northern North Atlantic. In: PARKER, J. R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. Geological Society, London, 199-214. SWANN, D. H. 1964. Late Mississippian rhythmic sediments of Mississippi Valley. AAPG Bulletin, 48, 637-658. TALBOJ, M. R. & KELTS, K. (eds) 1989. The Phanerozoic record of lacustrine basins and their environmental signals. Palaeogeography Palaeoclimatology Palaeoecology, 701-304. TARI, G.,VARKONu L. & VARNA~,E 1992. Third-order Miocene-Pliocene depositional sequences in eastern Hungary, Pannonian Basin. Stratigraphic framework of the post rift sediments in the
Pannonian Basin based on seismic-reflexion, welllog and detailed paleontologic data. In: Confer-
ence of Sequence Stratigraphy of European Basins, Dijon, France. Abstracts, 260-261. TIEDEMAN, R., SARNTHE[N, M. & SHACKLETON,N. J. 1994. Astronomic timescale for the Pliocene Atlantic ~ 8 0 and dust flux records of Ocean Drilling Program site 659. Paleoceanography, 9, 619-638. VAIL, P. R., AUDEMARD,E, BOWEMAN,S. A., EISNER, P. N. & PEREZ-CRIZ, C. 1991. The Stratigraphic Signitures of Tectonics, Eustacy and Sedimentology - an Overview. In: EINSELE, G., RICKEN, W. & SEILACnER, T. (eds) Cycles and Events in Stratigraphy. Springer-Verlag, Berlin, 617-659. VAKAaCS, G. 1997. Sequence stratigraphy of the Cenozoic Pannonian Basin, Hungary. PhD Thesis, Rice University, Houston, Texas. - - , VALE,R R, TAm, G., PO6ACSAS,GY., MATT1CK,R. E. & SZAB0, A. 1995. Third-order Middle Miocene-Pliocene depositional sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106.
Role of unconformity-bounded units in the stratigraphy of the continental record: a case study from the Late Miocene of the western Pannonian Basin, Hungary MARCO
S A C C H I 1, F R A N K
HORVATH
2 & ORSOLYA
MAGYARI
3
1Research Institute G E O M A R E SUD, CNR, Napoli, Italy 2Department o f Geophysics, EOtvOs University, Budapest, Hungary Abstract: We present an up-to-date stratigraphic framework for the Late Miocene (postrift) non-marine strata of the western Pannonian Basin, based on unconformity-bounded units as they are derived from seismic interpretation. The data set used for this study consisted of some 1700 km of conventional, multi-channel reflection seismic profiles across western Hungary integrated by 190 km of high-resolution, single-channel seismic profiles acquired on Lake Balaton in June of 1993. Seismic stratigraphic analysis has been constrained by selected geological mapping, well-logs and borehole data. A magnetostratigraphic record was also available from a corehole in the study area, together with recent K/Ar dating of basaltic rocks from the Balaton highland. Five third-order (with 106 year periodicities) stratigraphic sequences have been recognized at regional scale in the Late Miocene succession of the western Pannonian Basin. We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-1) and Pannonian-1 (PAN-I) to Pannonian-4 (PAN-4). Reliable time constraints were only available for the two maximum flooding surfaces of sequences PAN-2 and PAN-3, namely mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma), and the boundary of sequence PAN-2 (PAN-2 SB) which is approximately dated at 8.7 Ma. PAN-2 sequence boundary is associated with evidence of relative water-level drop in the Pannonian Lake and significant exposure of lake margins that is widely recorded in the so-called 'marginal facies' of western Hungary. The higher rank unit bounded by PAN-1 SB and PAN-4 SB includes most of the Pannonian s.l. succession of the central Paratethys and approximately correlates with the Tortonian-Messinian of the standard chronostratigraphy. Seemingly, no major palaeoenvironmental impact was perceptible in the western Pannonian Basin during the Messinian salinity crisis of the Mediterranean. However a significant change in the regional stratigraphic patterns may be observed since earliest Pliocene (after PAN-4 SB), possibly associated with the very beginning of a large-scale tectonic inversion within the intraCarpathian area. The case of Late Miocene non-marine strata of Pannonian Basin is a textbook example of how single categories of stratigraphic units do not fit (sometimes do not even approximate) chronostratigraphic correlation. The use of unconformity-bounded units offers new insights into the complex and long debated problem of stratigraphic correlation between Late Neogene deposits of the Pannonian Basin and 'similar' non-marine strata of the Central Paratethys realm. Our study shows that the so-called 'Pontian facies' of western Hungary correspond to an unconformity-bounded unit which is older than the Pontian s. s. facies of the stratotype area (Black Sea basin). Accordingly, we suggest that different stages may be used to discriminate between such similar-in-facies but different-in-age strata. We hence recommend the introduction of a new chronostratigraphic unit ('Danubian' or 'Transdanubian') in the Late Miocene series of Central Paratethys and a three-fold subdivision of the Pannonian (s.1.) strata into Early Pannonian (Pannonian s.s.), 'Middle Pannonian' ('Danubian' or 'Transdanubian') and Late Pannonian (Pontian s.s.) stages.
T h e P a n n o n i a n Basin was part of the Paratethys, a separate b r a n c h of the Tethys o c e a n which d e v e l o p e d d u r i n g t h e O l i g o c e n e to P l i o c e n e f r o m t h e w e s t e r n A l p i n e M o l a s s e B a s i n to central Asia ( L a s k a r e v 1924). T h e gradual separation of this vast epicontinental basin f r o m the o p e n o c e a n was m i r r o r e d by severe provincialism of the aquatic faunas, which has long been a m a j o r source of u n c e r t a i n t y for biostratigraphic
c o r r e l a t i o n b e t w e e n P a r a t e t h y a n and Mediterr a n e a n events. Difficulties resulted mostly f r o m the facies d e p e n d e n c e of faunas (Korp~is-H6di 1983; N a g y m a r o s y & Mt~ller 1988) and facies d i a c h r o n i s m within the various P a r a t e t h y s subbasins (Mtiller & M a g y a r 1992). Biofacies provincialism r e n d e r e d necessary the elaboration of a regional stage system for the P a r a t e t h y s p r o v i n c e ( P a p p et al., 1968, 1985;
SACCHI,M., HORVATH,E & MAGYARI,O. 1999. Role of unconformity-bounded units in the stratigraphy of the continental record: a case study from the Late Miocene of the western Pannonian Basin, Hungary. In:
DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension
within the Alpine Orogen. Geological Society, London, Special Publications, 156, 357-390.
358
M. SACCHI E T A L .
Cicha &Senes 1968; RCMNS 1971; Bfildi 1980; R6gl & Steininger 1983; Nagymarosi & Mt~ller 1988; Steininger et al. 1988,1990; Stevanovid et al. 1990; Miiller & Magyar 1992; R6gl 1996). Paratethys stages were based mainly on mollusc assemblages. Correlation among marine episodes in the Paratethys was established by planktonic microfossils; on the contrary, correlation of nonmarine stages of the Paratethys with standard marine stages still remains substantially obscure (Magyar & Hably 1994; Sacchi et al. 1997). Radiometric age determination, magnetostratigraphy and, more recently, sequence stratigraphy have been used to improve stratigraphic correlation within the Pannonian Basin fill (Tari et al. 1992; Csat6 1993; Ujsz~iszi & Vakarcs 1993; Vakarcs et al. 1994). However, absolute age constraints for the Pannonian strata are still poor and a direct correlation of relative water level changes in the Pannonian Lake with the cycle chart of Haq et al. (1987) has revealed to be driven mostly by conceptual presumption rather than adequate documentation. In this paper we present an up-to-date stratigraphic framework for the Late Miocene (postrift) non-marine strata of western Pannonian Basin (Fig. 1) which relies on unconformitybounded units derived from sequence (or genetic) stratigraphic procedure (Vail et al. 1977, 1990; Galloway 1989; Salvador 1994). Our study was based on the interpretation of about 1700 km of conventional, multi-channel reflection seismic profiles across western Hungary (Fig. 2) complemented by 190 km of high-resolution, single-channel seismic profiles acquired during our expedition on Lake Balaton in June of 1993 (Sacchi et al. 1995, 1998). Seismic stratigraphic analysis has been constrained by geological mapping of selected areas, well-logs and borehole data. The magnetostratigraphic log of Iharosberdny-I well (SW Hungary) was also available, together with recent K/Ar dating of basaltic rocks from the Balaton highland (Lantos et al. 1992; Balogh 1995). We also document that the reconstruction of a reliable stratigraphic framework for the Late Miocene non-marine succession of the Pannonian Basin can be only achieved through an integrated stratigraphic approach. This means full combination of biostratigraphy, magnetostratigraphy, radiometric age determination, sequence (or genetic) stratigraphy and classic field study (Vail 1987; Galloway 1989; Weimer & Posamentier 1993; Shanley & McCabe 1994; Miall 1997). A novelty of our work is the use of highresolution seismic data acquired on Lake Balaton in addition to the existing standard exploration seismics in the western Pannonian
Basin. The acquisition of high-resolution seismics on continental areas is possible when seismic profiling is performed on large rivers or lakes. It is because broad band signals can be generated in the water and the fairly low acoustic impedance contrast between the lacustrine mud and water-saturated deeper layers facilitates the propagation of acoustic waves at depth. Lake Balaton in Hungary is the largest natural water surface in central Europe and it offers ideal conditions to apply high-resolution seismics for the detailed study of the underlying Pannonian strata (Sacchi et al. 1995, 1998).
Geological setting The Pannonian Basin is part of a large back-arc depression superimposed on the Alpine megasuture (Bally & Snelson 1980; Horvfith et al. 1981; Tari 1994) (Fig. 1). Recent studies on the geodynamic evolution of the Pannonian Basin have shown fairly complex mechanisms of basin formation and evolution, which include subduction-related extensional collapse of an overthickened nappe pile and collision-related escape of orogenic terranes (Royden & Horv~th 1988; Tari et al. 1992, 1993; Horvfith 1993, 1995; Tari 1994). Early to Mid-Miocene extensional/strike-slip tectonics caused relatively fast synrift subsidence of narrow basinal areas while a significant part of the Pannonian region still remained intact and elevated. Late Mid-Miocene marked the onset of the post-rift phase when thermal cooling resulted in a generalized subsidence and broadening of the whole Pannonian area. Seismic data coupled with well-log interpretation and core-sample analysis, have shown that the Pannonian depression has been filled up by a fluvial-dominated delta system which prograded into a large lacustrine basin (Pogficsfis 1984; Berczi & Phillips 1985; Mattick et al. 1988; Pogficsfis et al. 1988; Horvfith & Pog~csfis 1988; Juhfisz 1994; Vakarcs et al. 1994). Based on the interpretation of seismic profiles and subsidence analysis of boreholes Horvfith (1993), Tari (1994) and Horvfith & Cloetingh (1996) proposed that, during latest NeogeneQuaternary, the Pannonian Basin underwent significant tectonic inversion. Reverse faulting associated with uplift occurred locally and the Hungarian mountains began to rise, thus causing extensive erosion of large areas covered by Late Neogene deposits. A recent study (Horvfith 1995) suggested that tectonic tranquillity of Pannonian Basin was interrupted even earlier in the synrift phase, at the end of Sarmatian, by compressional (locally transpressional) events.
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
359
Fig. 1. Tectonic map of the intra-Carpathian region and location of the study area.
Neogene geodynamic evolution was accompanied by widespread volcanic activity in the Pannonian Basin. Mid-Miocene to Pliocene calc-alkaline volcanism and Pliocene-Pleistocene alkali-basaltic volcanism resulted in a number of volcanic edifices scattered over the basinal area (Balogh et al. 1986; Szab6 e t al. 1992). During the Late Pliocene-Quaternary, shallow isolated lakes, wetlands, and mostly continental conditions prevailed throughout the basin.
The problem of stratigraphic subdivision of Pannonian s.l. strata. Traditionally, stages and stage systems of the Paratethys have been derived from biostratigraphy complemented with lithologic data. Particularly, the classic subdivisions of Pannonian strata (Roth 1879; L6renthey 1900; Stevanovid 1951) have been mostly based on benthic mollusc assemblages defined within marginal facies of the basin sequence. H6rnes (1851) first described the Pannonian formations of Hungary as Congeria and Paludina beds. The concept of a Pannonian Stage was
introduced by Roth (1879) to designate a relatively monotonous mostly continental sequence, which developed in the Central Paratethys between the Sarmatian (late Mid-Miocene) and the Pleistocene. Since late nineteenth century, the term Pannonian was adopted to include the stages Pontian (Congeria beds), Levantian (Paludina beds) and Thracian (Belvedere beds) when more accurate stratigraphic resolution was impossible. However, both the terms Pannonian and Pontian were used by the majority of stratigraphers as synonyms of Congeria beds. L6renthey (1900) suggested the term Levantian for the upper part of Pannonian s e n s u Roth (1879). This stage was mainly used to designate youngest (late Pliocene to Pleistocene) fluviatile or terrestrial deposits of uncertain stratigraphic position. L6renthey's proposal (1900) included a two-fold subdivision of the Pannonian stage into Lower and Upper Pannonian substages. During the same period, Halavfits (1903) proposed a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) based on Congeria assemblages. However his proposal had little success among Paratethys stratigraphers and was no longer used after that time (Fig. 3).
360
M. SACCHI E T A L .
Fig. 2. Location of seismic profiles used in this study. The detailed seismic grid acquired during our expedition on Lake Balaton in June of 1993 is shown in Fig. 15.
During the first half of the century nomenclature controversies developed among stratigraphers whether the P a n n o n i a n stage or the Pontian stage should be used. SzfideczkyKardoss (1938) suggested a subdivision of Pannonian s e n s u L6renthey (1900) into a lower part (Pannonian s.s.) and an upper part (Pontian and Dacian). Stevanovid (1951) proposed substituting the upper part of the P a n n o n i a n s e n s u L6renthey (1900) with the stage Pontian, based on a presumed coeval appearance of common molluscan species on both sides of the Carpathians. In 1975, a general agreement upon a regional Stage System for the Paratethys was achieved at the VI International Congress on M e d i t e r r a n e a n Neogene Stratigraphy in Bratislava. Accordingly, the Pannonian stage s e n s u Roth (1879) was subdivided into Pannonian s.s., Pontian, Dacian and Rumanian stages and a correlation was proposed with Late Miocene-Pliocene stages of the standard (Mediterranean) chronostratigraphy. Notwithstanding Stevanovid's (1951) redefini-
Fig. 3. Synopsis of Late Neogene chronostratigraphic units for the Central Paratethys according to different authors and correlation with the standard chronostratigraphic scale (after Sacchi et al. 1997). Note the threefold subdivision of Pannonian (sensu L6renthey 1900) strata we adopt in this study (see also Fig. 11 and text for discussion).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN tion of the Pannonian Stage has been officially accepted since 1971 by the International Union of Geological Sciences, practically all the three 'versions' of the term Pannonian (sensu Roth 1879; s e n s u L6renthey 1900 and sensu Stevanovi61951) are still in use in Hungary (Magyar & Hably 1994). In the everyday practice, for instance, many geologists, including those in the oil industry, still use the 'long' Pannonian stage s e n s u L6renthey (1900), and refer to Lower Pannonian instead of Pannonian s.s., and to Upper Pannonian instead of Pontian. The use of the term Pannonian in the 'traditional' Hungarian sense, (Pannonian s e n s u L6renthey, 1900), although in disagreement with the official Paratethys stage system, was somehow dictated by the extreme difficulty of correlating the Pontian facies of the stratotype area (Black Sea Basin) to the timeequivalent Pannonian strata in Hungary. But it also led to the disagreeable practice of assuming an implicit chronostratigraphic correlation between the local Upper Pannonian stage and the Pontian stage of the official stage system without adequate documentation. The Late Miocene chronostratigraphic framework proposed in this study is reported in the last column to the right of the comparative correlation chart of Fig. 3 and illustrated in detail in Figs 11 and 12. Our proposal was based on integrated stratigraphic approach with specific contribution of unconformity-bounded units (Salvador 1994). According to our interpretation none of the stage names in current use in Hungary adequately represents the middle part of Pannonian s.l. stage (c. 9.0-7.4 Ma in the chronology adopted in this study). Based on recent stratigraphic research (Mt~ller & Magyar 1992, 1995; Sacchi et al. 1997) and the results of our work, we adopt and recommend a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) into Lower Pannonian (Pannonian s.s.), Middle Pannonian ('Danubian' or 'Transdanubian' in Sacchi et al. 1997, 1998) and Upper Pannonian (Pontian s.s.). Our 'Middle Pannonian' partly corresponds to C o n g e r i a u n g u l a c a p r a e and C o n g e r i a b a l a t o n i c a beds (Middle Pannonian s e n s u Halavfits 1903).
Lithostratigraphic framework of western Pannonian Basin Pannonian strata unconformably overlie the Sarmatian sequence in the deepest troughs of the basin while they are transgressive on older rocks at the basin margins Szentgy6rgyi & Juhfisz (1988). Sandy turbidite units (Szolnok Sandstone) interbedded with pelagic marl
361
(Endr6d Marl) represent the lowermost part of the Pannonian succession in the deep basins (Danube, Zala and Drava). Towards the top, sandstone (15jfalu Sandstone) and marl (Algy6 formation) follow, indicating delta slope and delta plain settings. The prograding delta complex is overlain by alluvial deposits (Hansfig and Zagyva formations). In the marginal part of the basin, the Pannonian succession commonly overlies a major unconformity. Turbidites and slope deposits are generally missing and basal transgressive sequences consist of nearshore conglomerates. In the Transdanubian Central Range the Pannonian s./. strata have been subdivided into two major groups (Jfimbor 1980, 1987, 1989), the Peremarton and the Dun~nt~l groups. The Peremarton group, represented by the Os Variegated Clay, Kisb6r Gravel, and Cs~kvfir and Sz~k Claymarl formations, has been traditionally considered as corresponding to the Lower Pannonian s e n s u L6renthey (1900). Similarly the Dun~ntt~l group, made up of the Kfilla Gravel, Soml6, Tihany, Torony, Hansfig and Tapolca Basalt formations, has been considered as corresponding to the Upper Pannonian s e n s u L6renthey (1900). The Tapolca Basalt formation includes Neogene basalt and tuff (Balogh et al. 1986; Balogh 1995) from tens of eruptive centres (J~mbor et al. 1981) located in the Bakony mountains and the Balaton highland. Volcanic rocks of Tapolca Basalt formation display pronounced alkaline character (Balogh et al. 1986; Szab6 et al. 1992; Harangi & Harangi 1995). Late Neogene strata of Pannonian Basin are overlain by Quaternary fluvial deposits whose thickness ranges from a few tens to several hundred metres.
Seismic and sequence stratigraphy of western Pannonian Basin Three regional profiles across western Pannonian Basin are presented, which run in a NNW-SSE direction from south of Lake Balaton down to the Drava Basin toward the borders of former Yugoslavia (Figs 2 and 4-6). Based on methods and procedures of sequence (and genetic) stratigraphy (Vail 1987; Galloway 1989), five third-order (with ]06 year periodicities) sequences have been recognized at regional scale in the post-rift succession. We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-I)and Pannonian-1 (PAN-l) to Pannonian-4 (PAN-4). Maximum thickness of the sequence stack is on the order of 3 kin.
362
M. SACCHI E T A L .
Fig. 4. Patterns of third-order stratigraphic sequences across western Hungary (section A). Note the occurrence of the Lower Pannonian-Upper Pannonian boundary (as reported by previous authors) along maximum flooding surfaces (see also Fig. 7 and text for discussion). The major unconformity at the base of the post-rift sequence is often associated with a significant stratigraphic gap whose amplitude increases from WSW (Drava Basin) to E N E (Somogy). This may cause amalgamation of more sequence boundaries (Figs 4-6). A cartoon section, based on the interpreted seismic profiles is illustrated in Fig 7. Reliable time constraints were only available for the stratigraphic interval between 9.0 Ma and 7.4 Ma (Figs 4 and 8) which has been calibrated by the magnetostratigraphic record of Iharosber6ny-I well (Lantos et al. 1992; Ujsz4szi & Vakarcs 1993), revised after Cande & Kent (1992, 1995).
Maximum flooding surface mfs-2 marks the peak of a major flooding event, which occurred in the Pannonian Lake at c. 9.0 Ma. This event was manifested by C o n g e r i a c z j z e k i open lacustrine beds (Sz4k fm) which flooded the basin margins. Mfs-2 represents a quasi-isochronous surface at basin scale that can be strikingly correlated with the top of Pannonian s.s. stage (Lower Pannonian sensu L6renthey 1900). Sequence boundary PAN-2 is associated with a significant drop of base level of erosion within the Pannonian Basin at c. 8.7 Ma, which was accompanied by extensive subaerial exposure of the lake margins. This is documented in the
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
363
9
9 ~,.-4
,-r
O
364
M. SACCHI E T A L .
Fig. 6. Patterns of third-order stratigraphic sequences across western Hungary (section C). Note the amalgamation of SAR-I and PAN-1 sequence boundaries along a major unconformity at the base of Pannonian strata (see also Fig. 7 and text for discussion).
so-called marginal facies of Transdanubia and evidenced by erosional scours due to a several tens of metres lake level fall (Figs 5 & 9). Maximum flooding sequence mfs-3 (c. 7.3 Ma) represents a second important flooding event within the basin which is again characterized by the occurrence of open lake strata, this time associated with a younger Congeria assemblage (Congeria rhomboidea). Mrs-3 represents another useful quasi time line at basin scale and may be considered a good proxy in western Hungary for the base of Pontian as it is defined in the stratotype area (Black Sea Basin). The stratigraphic (genetic) unit b o u n d e d by maximum flooding surfaces mfs-2 and mrs-3 may be correlated with the lower part of the Pontian sensu Stevanovi6 (1951), and regarded as a sort of anticipation of the Pontian s.s. facies in Hungary. As a consequence, it is likely the case that Pontian s. s. (younger than 7 Ma) strata are
practically missing in outcrop in central western Hungary. Sequence boundary PAN-4 (c. 5.0 Ma) is likely to be associated with stratigraphic gap (see also Juh~sz et al. this volume) and significant tectonic overprint as it suggested by the general tectono-stratigraphic patterns within the Neogene Basin fill (Figs 5 & 7). Sequence boundaries SAR-1 SB to PAN-4 SB of this study basically correlate with sequence boundaries III to VII of Ujszfiszi & Vakarcs (1993). The only exception is PAN-2 SB which we find definitely at a higher stratigraphic position with respect to correspondent sequence boundary SB V of Ujsz~szi & Vakarcs (1993). However the absolute ages reported here have been revised in the light of the updated magnetic polarity scale (Cande & Kent 1995; Berggren et al. 1995) and they significantly differ from those reported by the above authors. This would
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
365
Fig. 7. Tectonic and sequence stratigraphic framework of Pannonian strata based on the interpretation of 1700 km of seismic sections across western Pannonian basin (see Figs 2 and 4-6). Pannonian units Q and @ of this figure respectively correspond to the 'Late Miocene sequence' and the 'Pliocene sequence' of Juhfisz et al. (this volume).
suggests that the age dating proposed by gjsz~szi & Vakarcs (1993) on the basis of a direct correlation of P a n n o n i a n third-order sequences with the eustatic curve of Haq et al. (1987) is not applicable, at least for the stratigraphic interval between 9.0 Ma and 7.4 Ma. The higher rank unit bounded by PAN-1 SB and PAN-4 SB approximately correlates with the Tortonian-Messinian stages of the standard time scale. Seemingly, no major palaeoenvironmental impact was perceptible in the western Pannonian Basin during the Messinian salinity crisis of the Mediterranean. However a significant change in the regional stratigraphic patterns is observed since the earliest Pliocene (after PAN-4 SB), which was possibly associated with the very beginning of a large-scale tectonic inversion within the intra-Carpathian area (Fig. 7). The major unconformity at the base of the Pannonian s.l. strata and PAN-4 SB subdivide the Neogene strata of western Pannonian Basin into two main tectono-stratigraphic units (Fig. 7) which namely correspond to the 'Late Miocene sequence' and the 'Pliocene sequence' of Juhfisz et al. (this volume).
The concept of Lower Pannonian-Upper Pannonian boundary from the perspective of unconformity-bounded units The observation that the Lower P a n n o n i a n U p p e r P a n n o n i a n ( s e n s u L 6 r e n t h e y 1900) boundary is basically a time-transgressive facies b o u n d a r y is not new (Pogficsfis et al. 1988; Jfimbor 1989). In particular, Pogficsfis et al. (1988) had pointed out that the Lower PannonJan-Upper Pannonian (LP-UP) boundary gradually rejuvenates from about 9 Ma to about 6 Ma, as depositional sequences of the Pannonian Basin prograde from NNW to SSE at a regional scale. Nevertheless, our study (Figs 4 ~ & 10) shows that a 'virtual' L P - U P lithostratigraphic boundary is likely to occur at each maximum flooding surface, i.e. it is cyclically repeated at the top of each transgressive system tract of consecutive third-order sequences). The patterns of stacking sequences evidence, in other words, that the L P - U P lithostratigraphic boundary does not move in time gradually, but rather rejuvenates with discrete 'jumps' by stepping one maximum flooding surface after another in the direction of prograding sequences. As a
366
M. SACCHI E T A L .
Fig. 8. Magnetostratigraphic calibration of Pannonian strata at Iharosberdny-I well site (after Lantos et al. 1992; Cande & Kent 1995). The adopted three-fold subdivision of Pannonian s e n s u L6renthey (1900) is based on regional maximum flooding surfaces mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma) (see also Figs 11 and 12 and text for discussion).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
367
Fig. 10. Lithostratigraphic versus sequence stratigraphic patterns in the western Pannonian Basin. Note the cyclic 'repetition' of the boundary between 'Lower Pannonian type' and 'Upper Pannonian type' lithofacies along maximum flooding surfaces. Mfs-2 (9.0 Ma) corresponds to the top of Pannonian s.s. (Pannonian sensu L6renthey 1900). Mfs-3 (7.4 Ma) is as a good proxy for the base of Pontian s.s. in western Hungary. The unconformity bounded unit between mfs-2 and mfs-3 represents our 'Mid-Pannonian' stage (after Sacchi et al. 1997). consequence, it is likely the apparent paradox that one could detect a number of 'distinct' L P - U P facies boundaries, which are clearly repeated in time and space, and rejuvenate from NNW to SSE across the same regional stratigraphic section (Fig. 10). The reason that a specific boundary is interpreted as 'the Lower P a n n o n i a n - U p p e r Pannonian boundary' (and then correlated with the Pannonian s.s.-Pontian boundary) seems to lie on the local occurrence of 'proper' conditions in the sedimentary environment, e.g. sharp lithological contrast, associated with transition from 'Pannonian' versus 'Pontian' faunal assemblages. These observations suggest the concept that a boundary between 'facies of Pannonian affinity' and 'facies of Pontian affinity' cannot be correlated unambiguously to a single, distinct lithostratigraphic boundary throughout the Pannonian Basin. In order to avoid confusion and misunderstanding, we recommend that the use of Lower P a n n o n i a n - U p p e r Pannonian boundary in a formal lithostratigraphic sense should be abandoned, or significantly revised.
Difficulties in the application of classic lithostratigraphic and biostratigraphic units for stratigraphic correlation through the Pannonian Basin are mirrored by substantial uncertainty concerning chronostratigraphic position of the boundary between the Pannonian s.s. and the Pontian s.s. stages of the Central Paratethys. Chronostratigraphic miscorrelation due to diachronism of biofacies has already been detected in the Pannonian Basin by Mtiller & Magyar (1992, 1995) who have shown that the 'biostratigraphic Pontian of the P a n n o n i a n Basin' (Pontian sensu Stevanovid 1951), developed nearly 2 million years earlier than the 'biostratigraphic Pontian' of the Black Sea Basin (Pontian s.s. ).
A proposal for a three-fold subdivision of Pannonian
s.1. s t r a t a
Following the results of MUller & Magyar (1992, 1995) we propose that the Pontian of western Hungary (Miiller & Sz6noky 1990), although
Fig. 9. Palaeogeographic sketch-section across western Pannonian Basin at Pan-2 SB (c. 8.7 Ma) showing subaerial exposure at the basin margin due to relative drop of Pannonian lake level. Note the erosional slope at the 'shelf-break' and associated depositional, epigenetic and/or diagenetic features in response to the lowering of base level (after Magyar 1988).
368
M. SACCHI E T A L .
similar to the Pontian facies of the Black Sea, is better understood as a stratigraphic unit sandwiched between the Pannonian s.s. and the Pontian s.s. strata (Sacchi 1997). Such a rock unit is bounded by maximum flooding surfaces mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma) and might be expressed, in terms of chronostratigraphy, by a new stage inbetween the Pannonian s.s. and the Pontian s.s. stages (Fig. 11). Accordingly we recommend a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) into Early Pannonian (Pannonian s.s.), Mid-Pannonian ('Danubian' or 'Transdanubian' in Sacchi et al. 1997, 1998) and Late Pannonian (Pontian s.s.). The introduction of such a 'Mid-Pannonian' may be helpful in enhancing stratigraphic resolution and avoiding prolonged confusion and misunderstanding concerning the apparent diacrhonous nature of the boundary between the Pannonian s.s. and the Pontian s.s. facies of Central Paratethys. It would also fit, as a matter of fact, the present-day 'chronostratigraphic gap' existing between these stages (Figs 11 and 12). A stratotype for our Mid-Pannonian stage might be adequately represented by the Tihany-Fehdrpart section (MOiler & Szdnoky
1990) which is presently used as Pontian faciostratotype in Hungary (Figs 12 & 24). The concept of an intermediate stage between 'Lower Pannonian' and 'Upper Pannonian' is not new. Since early times of stratigraphic study on Pannonian strata, Halavfits (1903) had already defined a 'Middle Pannonian', corresponding to C o n g e r i a u n g u l a c a p r a e and C o n g e r i a balatonica beds, which partly correlates with our Danubian (or Transdanubian) stage.
Acquisition and processing of highresolution seismic data on Lake Balaton During June of 1993, a geophysical survey (cruise GMS93-02) was carried out on Lake Balaton. The aim of the cruise was to acquire high-resolution single-channel seismic reflection data on the Pannonian s.l. succession beneath the Balaton Quaternary deposits. Significant advantage towards a meaningful application of marine-type high-resolution seismics as a tool for the study of Pannonian strata has been offered by the geographic position of
Fig. 11. Late Miocene chronostratrigraphy of Central Paratethys (Pannonian basin). We propose the introduction of a new stage ('Danubian' or 'Transdanubian' in Sacchi et al., 1997, 1998) and a three-fold subdivision of the Pannonian strata sensu L6renthey (1900) into Early Pannonian (Pannonian s.s.), 'MidPannonian' ('Danubian' or 'Transdanubian') and Late Pannonian (Pontian s.s.) stages. Our Mid-Pannonian stage partly corresponds to Congeria ungulacaprae and Congeria balatonica beds (Middle Pannonian sensu Halav~its 1903).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
369
Fig. 12. Sequence stratigraphic concept of the 'Mid-Pannonian' ('Danubian' or 'Transdanubian') stage proposed in this study, and the framework of Late Miocene stratotypes of the Paratethys. The Tihany-Feh6rpart section, which is currently used as Pontian faciostratotype in Hungary (Mfiller and Sz6noky 1990), represents a possible stratotype for our 'Mid-Pannonian' stage (see also Fig. 24 and text for discussion).
the lake itself. The Balaton is located in fact at the southern foot of the Bakony Mountains, where the Palaeozoic-Mesozoic basement crops out and the Pannonian sequence pinches out towards NNW. Consequently, the late Miocene stratigraphic record, which in deepest parts of the Pannonian Basin may reach several thousand metres, is here represented (even though incomplete), within a few hundred metres (Fig. 13).
The survey area Lake Balaton covers an area of c. 600 km 2. It has a S W - N E elongated shape (about 78 km in length and up to 14 km in width) and its surface is at 105 m above sea level. It is the largest lake in central Europe, although very shallow. Water depth of the lake is 3-4 m along its entire axis, with the only exception of the tip of the Tihany peninsula where a local 'hole' of 12 m depth is present. The Balaton is part of a relatively young hydrographic system that has formed in the latest Pleistocene-Holocene during the postWtirm deglaciation of Central Europe. The oldest lacustrine deposits have been proven to be as old as approximately 15 000 years Be. by palynological analysis (Nagy-Bodor 1988) as
well as 14C-dating (Cserny et al. 1995). The lake presumably formed after the joining of individual ponds as climatic conditions changed from cold-arid to warm-humid during the Quercus-Fagus vegetation phase (Cserny & Nagy-Bodor 1996). The present-day lake is quite an artificial feature in the sense that its coastline is mostly a constructed e m b a n k m e n t and the water table is kept at a constant level by regulating the water discharge. The origin of the lake has been long debated among geomorphologists (L6czy 1913; Cholnoky 1918; Erddlyi 1961, 1962; Wein 1967; Mike 1980; Marosi & Szil~rd 1981). Several explanations have been proposed, most of which include various erosional processes superimposed on some tectonic control in the broad shape and location of the lake itself (Fig. 14).
Acquisition and processing The 1993 seismic survey was carried out on board of the RV VizvOdelem.The use of a differential GPS with slave stations located at the lakeshore ensured ship positioning with accuracy on the order of 2-3 m. Ship position was recorded each 30 s; fixes were taken each 10 minutes. The total length of acquired profiles is
370
M. S A C C H I E T A L .
Fig. 13. Sketch-section across western Pannonian Basin based on regional seismic profiles (see Figs 2 and 4-6) and compiled after Tari (1994). The unconformity bounded unit between mfs-2 and mfs-3 represents the 'MidPannonian' stage we introduced in this study. Note the erosional truncation of Pannonian strata towards N W due to the Plio-Quaternary uplift of the Bakony mountains (see also Fig. 7 and text for discussion).
Fig. 14. Cholnoky's (1918) model on the origin of Lake Balaton.
around 190 km. The seismic grid is shown in Fig. 15; details of navigation courses are summarized in Table 2. Seismic acquisition was obtained with a U n i b o o m type source (300 J) operating at a shot rate of I s. The streamer was equipped with a group of eight equally spaced hydrophones within an active length of 4.6 m. Boomer and streamer were towed with a lateral offset of 3 m. The average towing speed was 3.5 nml/h (c. 6.5 kin/h). Seismic signal has been acquired on a two-channel VHS hi-fi analogue tape recorder, channel 1 recording the response of the hydrophone group, channel 2 the trigger signal. A bandpass filter of 100 Hz-5 kHz and Time Varying Gain have been applied before printout on an E P C graphic recorder. The recording window was set on 250 ms. Resolution was on the order of 0.5-1.0 m. A scheme of the acquisition system is shown in Fig. 16; acquisition parameters are listed in Table 1. The shallow-water conditions within Lake Balaton imposed severe operating problems as the length of the hydrophone group was in the same the order of the water depth. Thus, water b o t t o m reflections were not summed in phase in the case of high frequencies; in addition direct arrivals interfered with water bottom reflections. In a second stage, the original analogue data set was converted to digital format. The length of digitized record is 300 ms with a sampling rate of 32.052 kHz.
LATE M I O C E N E S T R A T I G R A P H Y , P A N N O N I A N BASIN
371
Tnble 1. Lake Balaton 1993 seismic survey: parameters o f the acquisition system Boomer Hydrophone
Hydrophone group
Recording system
Power Shot rate Sensitivity Pre-amplification Output resistance Input voltage Sensitivity Bandwidth Plotter Analogue recorder
300 J 1s -103 dB V 1 ~bar -1 40 dB 2 kW 9V -63 dB V -1 p b a r 1 100 Hz-10 kHz EPC-3200 PAL VHS hi-fi
Fig. 15. Sketch map of Lake Balaton area showing the location of high-resolution seismic profiles acquired in June of 1993 and main localities and boreholes cited in the text.
Raw data have been converted to SEG-Y format and processed on P r o M A X . The standard processing sequence started with Trace D C Removal and Bandpass Filtering. Minimum phase Butterwoth filter was chosen with 100-1000 Hz corner frequencies and 12 dB/oct slopes. The effect of spherical divergence was corrected by True Amplitude Recovery, using velocities 1480 m/s, 1600 m/s and 2000 m/s in the time ranges of 0-20 ms, 20-50 ms and 50-300 ms respectively (Fig. 17a, b). Remarkable improvement has been achieved by FK Filtering. Fan filters were used with a spatial extent of 50 traces. Fan filter velocities were + 500 m/s and _+ 5000 m/s, corner frequencies were 50 Hz and 1500 Hz. Primary reflections revealed to be often masked by other types of reflections. The most ubiquitous nonprimary event has a 2-3 ms delay and exhibits reverse
polarity. This was due to boomer plate reverberation. Weaker non-primary events with 8-12 ms delay correspond to base-of-mud multiples. Phase correction was applied first with 2 ms operator length, then predictive deconvolution with the same operator length; 2.5 ms prediction distance suppressed the b o o m e r plate reverberation. Lake Balaton mud deposits show an average thickness of the order of 6-8 m. Base-of-mud multiples were also removed by predictive deconvolution (Fig. 17c, d). Time gates were determined to exclude direct arrivals from both predictions while base-of-mud reflections were excluded from the longer prediction. Bandpass filtering was applied after the deconvolution. Coherency filtering improved the signal/noise ratio but e n h a n c e d the multiples, for this reason c o h e r e n c y filtering was applied before the phase
M. SACCHI E T A L .
372
Table 2. Summary o f navigation courses and seismic profiles acquired.on Lake Balaton in June o f 1993 Date of acquisition 17.06.93 17.06.93 17.06.93 17.06.93 17.06.93 18.06.93 18.06.93 18.06.93 19.06.93 19.06.93 19.06.93 19.06.93 19.06.93 20.06.93 20.06.93 20.06.93 20.06.93 20.06.93 21.06.93 21.06.93
Seismic profile
Heading
Length (km)
L-1 (test) L-2 (test) L-11 (test) L - 12 L - 13 L -4 L-9 L-6 L - 10 L - 10/11 L - 11 L-11/12 L - 3/4 LW - 1 LW - 2 LW - 3 LW - 4 LW - 5 LW - 6 L -7
W-E E-W N-S S-N N-S E-W N-S W-E N-S S-N N-S S-N E-W NE-SW S-N N-S S-N W-E W-E W-E
2.0 2.5 11.5 12.0 9.0 16.0 8.0 19.5 8.0 9.5 11.5 9.0 7.5 11.0 6.5 7.5 5.5 15.0 16.0 11.0
Lake Balaton 1993 Survey Acquisition System
"fzzq
~
vLr~ Channel 1 Channel 2
~.._,V(~H~ ;L Re(~
Fig. 16. Scheme of the acquisition system during the 1993 high-resolution survey on Lake Balaton.
correction. The final step of the processed sections includes Trace Mixing (Fig. 17d).
Seismic and sequence stratigraphy of the Pannonian strata beneath Lake Balaton T h e 1993 survey i n v e s t i g a t e d L a k e B a l a t o n subsurface to a m a x i m u m t i m e - d e p t h of c. 180 ms d o w n to t h e acoustic b a s e m e n t , w h i c h is r e p r e sented by Sarmatian marl and marly limestone
(Sacchi et al. 1995, 1988). C o n s t r a i n t for seismic stratigraphic i n t e r p r e t a t i o n was given by boreh o l e s T i h a n y - 6 2 , Si6fok-3, Balatonf61dv~irMHSz, and fieldwork on selected areas (B~intapuszta, Papv~is~irhegy, Tihany, K6v~ig66rs) in the s u r r o u n d i n g s o f the lake (Fig. 15). Calibration of seismic profiles was o b t a i n e d b y the i n t e r s e c t i o n of seismic profile LW-1 with interpolated stratigraphy between boreholes Tihany-62 and BalatonfOldv~ir-MHSz. S e q u e n c e
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN stratigraphic interpretation of the Balaton seismic profiles (Figs 18-22) was aided by the study of the regional seismic sections across western Hungary and Iharosber6ny-I well-log and magnetostratigraphy (Figs 4-6 & 13). Interpreters should keep in mind that vertical exaggeration for the seismic data set is about ten-fold. Two major unconformities, both associated with significant stratigraphic gap have been
(a)
(b)
373
detected in the subsurface of Lake Balaton: (a) an upper unconformity at the base of the Balaton Quaternary deposits; (b) a lower unconformity at the top of the pre-Pannonian (Middle Miocene) sequence. These two unconformities separate, from bottom to top, three major seismic-stratigraphic units (Fig. 23): (1) PrePannonian strata; (2) Pannonian s.l. (Late Miocene) sequence; (3) Late PleistoceneHolocene deposits of Lake Balaton.
374
M. SACCHI ETAL.
(c)
(d) Fig. 17. Sample of high-resolution seismic profile across Lake Balaton (section L-6, detail): (a) analogue (EPC printout) record; (b) digitized section after trace DC removal, bandpass filtering and TAR; (c) processed section without coherency and F-K filtering; (d) final processed section, including trace mixing.
Pre-Pannonian strata Pre-Pannonian (Sarmatian) strata are dramatically truncated at the top by a mature (polycyclic) erosional surface. Stratigraphy at Tihany-62 site (Fig. 24) indicates that this
surface marks a significant stratigraphic gap (about 12 to 9 Ma) spanning from the Upper Sarmatian to Congeria czjzeki beds (upper part of Pannonian s.s.). The gap includes amalgamation of two third-order sequence boundaries, namely SAR-1 SB (top of Sarmatian sequence)
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
375
9
~E ~13 r
o~9- ~
~
9
~1~+ ...~
r,~
z.-~
"go~ 9
O,.~
9
.o .~ a,
376
M. SACCHI E T A L .
9,,.-i
r~
9
~.~
, ,...~
~~ z..~
::
>~0
,
,x::l " ~~ 9,0~ ~.~ 9 m 0 "
~
~9
~D
~
-~,~.~ r~
~
"~ ~o ,
,.~
E.S : ~
>
"7.
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN
377
O +,,a
9
~Z ..=
e~t)
9
o
Z
u?, 9 ~,,,,,~ t~
o
,,,,~
O ~ O
eq
378
M. SACCHI E T A L .
9
i
r~
2;
O +,,a
O
O
9
+6 Z ,,...z,
.~t-q
E=
, ,,,,~
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
379
-6 O
. ,...,,~ 9
6 9
. ,,,,~
"5 . ,,,,,.
,a=
9
Z~
~Z
. ,,.,~
e4"~ e'.l
380
M. SACCHI ETAL.
Fig. 23. Stratigraphic framework of Neogene-Quaternary strata beneath Lake Balaton and correlation with Neogene sequence stratigraphic units of the western Pannonian Basin (see also Fig. 7 and text for discussion).
and PAN-1 SB (top of first Pannonian thirdorder sequence) (Fig. 23). The duration of this hiatus increases from SW to NE across Lake Balaton area, to include the whole Sarmatian. Good exposures at this stratigraphic level are found at Bfintapuszta, where Pannonian s.s. lacustrine strata directly overlie Badenian marine deposits. Incised valleys and associated fluvial terraces and channel-fills are also recognized from the seismic record (Fig. 25). Channel-fills within valleys are bounded at the top by a transgressive surface and may be interpreted as lowstand systems tract (LST) deposits (Shanley & McCabe 1994).
Pannonian s.l. (Late Miocene) succession beneath Lake Balaton Maximum thickness of this unit is in the order of 200 m. The lowermost part of the Pannonian sequence is represented by the transgressive systems tract (TST) deposits of the Szfik formation (Congeria czjzeki beds). These deposits consist of open lacustrine grey clay-marl and siltstone which directly onlap the pre-Pannonian
basement. The top of Szfik formation is bounded by mrs-2 (c. 9.0 Ma) that correlates with the top of Pannonian s.s. stage. The Soml6 formation and the lower part of Tihany formation follow, which are interpreted, in turn, as early progradational and late progradational-aggradational highstand systems tract (HST) deposits. The Tihany formation developed partly in nearshore areas, mainly lagoons, at shallower depth than the underlying Soml6 formation. Stratigraphic architecture of these deposits is characterized by forestepping strata which downlap above the underlying TST deposits and are accompanied by local development of small coarse-grained prograding deltas (Figs 18, 20, 24). Good exposures of this sequence are found at Tihany-Feh6rpart. Notwithstanding these strata are reported as Pontian faciostratotype in Hungary (Mialler & Szdnoky 1990), it is to be noted they are Danubian (or Transdanubian) in age according to the chronostratigraphy adopted in this study (Fig. 24). A subtle third-order sequence boundary (PAN-2 SB) has been also detected within the Tihany formation, the upper part of which corresponds to the LST of the sequence above.
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN
381
o,..~ ~ ,...--.,
"~cq
r.g/ ,,--,-x
9 ..~
r.~ O
m,..~
9
9
*..o 9
9
O
382
M. SACCHI E T A L .
Fig. 25. Enlarged detail of seismic profile L-11/12 showing evidence of v-shaped valleys incised within the prePannonian (Sarmatian) strata and filled up by lowstand systems tract (LST) deposits of sequence PAN-2 (see Fig. 22). The unconformity at the base of Pannonian strata marks a major stratigraphic gap spanning from c. 12 Ma to c. 9 Ma) and includes amalgamation of SAR-I+ PAN-1 sequence boundaries. Note small mound-shaped features along PAN-2 SB (see also Fig. 27 and text for discussion).
Sequence boundary PAN-2 may be followed as a regional unconformity on seismic profiles all through western Hungary. It is related with a significant lowering of base level in the Pannonian Lake and is possibly associated with tectonic and volcanic activity (Tapolca basalt and Kabhegy formations). PAN-2 SB crops out at the top of Tihany Peninsula. Its occurrence (c. 8.7 Ma) predates the onset of basaltic eruption of the Tihany Volcano (c. 7.8 Ma, Balogh 1995). Stratal patterns towards igneous bodies inferred from seismic profiles suggest that most near-surface magmatic intrusions were coeval and/or slightly postdated the deposition of the Sz~k formation (c. 9.0 Ma). This is also suggested by 'anomalous' truncation of strata at the top of the TST, possibly due to local uplift related to magmatic intrusion (Figs 18, 24 & 26). Scattered over the top of Tihany Peninsula, a number of silicified carbonate mounds are found that are currently believed to be P l i o c e n e Pleistocene in age. These deposits (geyserite of the Hungarian literature) have been so far interpreted as purely chemical deposits related to post-volcanic activity (L6czy 1913). According
to our interpretation they are better interpreted as silicified travertine mounds developed at warm/hot springs. Furthermore, a number of mound shaped features have been detected from the seismic record which can be traced down to the Balaton subsurface along PAN-2 SB (Figs 19, 22, 23, 24, 27). We propose these features may be correlated with the Tihany travertine mounds which, accordingly, would be Late Miocene in age (Sacchi et al. 1995, 1998). According to our sequence stratigraphic framework, we also suggest that the K~lla formation, a coarse-grained foreshore deposit that crops out in the K~I Basin, north of Lake Balaton, is a facies heteropy of the lower part of Tihany fm., being itself part of a HST progradational-aggradational unit that formed along the Pannonian lakeshore between 9.0 and 8.7 Ma. We speculate that significant part of K~lla deposits in this area were possibly removed by erosion during the lake level fall which caused PAN-2 SB to occur. One puzzling feature of the K~lla formation is certainly the k6tenger (sea of stones), a 2 m thick silicified sandstone and conglomerate bed. We
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
383
Fig. 26. Enlarged detail of seismic profile L-4 showing evidence of near-surface magmatic intrusion beneath Lake Balaton. Stratal patterns within the Pannonian succession, suggest the intrusions are coeval and/or slightly postdate mrs-2 (9.0 Ma).
suggest the k6tenger may be interpreted as a groundwater silica-cemented regolith (groundwater silcrete), (A. Mindszenty pers. comm.) and regarded as an epigenetic feature that developed at the phreatic-vadose interface in response to the base-level drop within the Pannonian Lake at Pan 2 SB (8.7 Ma). In other words the k6tenger may be seen as a 'sequence stratigraphic equivalent' of the silicified travertine mounds of Tihany peninsula (Fig. 9). The occurrence of silcrete would also suggest relatively warm (dry/humid seasonal) climatic conditions during that time.
Fig. 27. Enlarged detail of seismic profile L-6 showing mound-shaped features along PAN-2 SB (c. 8.7 Ma). We interpreted these features as travertine build-up, similar to those ~ropping out at the top of Tihany Peninsula (Geyserite of the Hungarian literature).
Late Pleistocene-Holocene deposits of Lake Balaton This unit is mostly represented by the deposits of Lake Balaton; it also includes thin patches of older (Late Quaternary) fluvial clastics at its base that have been documented by drillings
384
M. SACCHI E T A L .
(Cserny & Corrada 1989), but cannot be solved seismically. Lake Balaton deposits display an average thickness of c. 5 m and consist of silt and subordinately clay and fine sand with a carbonate content of 50-70% (Cserny & Corrada 1989). Seismic response of these deposits is characterized by parallel-continuous reflectors. The unconformity at the base of this unit corresponds to a subaerial erosional surface which dramatically truncates the underlying Pannonian strata (Figs 18-22, 24). It marks a major stratigraphic hiatus that practically encompasses the whole Pliocene and most of the Pleistocene (Fig. 23).
Tectonic interpretation Seismic profiles across western Hungary indicate that the Late Miocene strata of western Pannonian Basin underwent significant post-rift tectonic deformation (Figs 4-6, 18-22, 24). Regional scale tectonic deformation is evidenced by severe tilting of the whole Neogene sequence of western Pannonian Basin towards ESE. This is clearly seen on seismic profiles that consistently show E S E dipping strata which are erosionally truncated at the top. In order to explain this
tectono-stratigraphic setting Tari (1994), Horvfith (1995) have suggested that the Pannonian Basin underwent significant late-stage tectonic inversion and uplift with consequent subaerial erosion of Late Neogene strata in Transdanubia and Northern Hungary (Tari 1994; Horvfith 1995). Based on our interpretation we propose that the onset of this tectonic inversion could be placed at the very beginning of the Pliocene (Fig. 7). An older compressive/transpressive tectonic phase can be detected from regional seismic profiles at about the end of Sarmatian (Horvfith 1995). This may account for the significant and almost ubiquitous stratigraphic gap that we have documented at the base of Pannonian strata. Small scale tectonic deformation in the area of Lake Balaton is expressed by S W - N E strike-slip faults and associated folds which postdate the 'Middle Pannonian' succession of Transdanubia (Figs 18-22, 24, 28). Evidence of minor tectonic activity also exists between mrs-2 and PAN-2 SB (9.0-8.7 Ma). This may suggest causal relations between tectonics and coeval volcanic activity in the area of Balaton highland. SW-NE-trending faults had possibly an important role in controlling the recent hydrographic pattern in the Balaton area (even the location and shape of the
Fig. 28. Main faults detected from the interpretation of Lake Balaton high-resolution seismic profiles. Tectonic deformation mostly postdates 'Middle Pannonian' and consists of WSW-ENE strike-slip faults and associated folds. Fault patterns suggest a possible tectonic control on shape and location of the lake itself.
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN lake itself) and the distribution of igneous rocks in the outcrop.
Summary and conclusions Recent stratigraphic research has focused on the sequence architecture in shallow marine deposits of continental margins. Although a consensus seems to be emerging, concerning the application of sequence stratigraphic concepts to the continental record (Shanley & McCabe 1994; Miall 1997), very few well-documented examples of non-marine sequences have yet been described. The case study of the Late Miocene continental record of western Pannonian Basin represents a contribute towards sequence stratigraphy applied to non-marine strata. It also offers new insights into the complex and debated problem of correlation between regional unconformity-bounded units (Salvador 1994) and the more consolidated lithostratigraphic and biostratigraphic units which derive from local or regional use in a given geological province. Furthermore our results suggest that absolute age constraints for the Pannonian strata are still poor and the correlation of stratigraphic sequences of western Pannonian Basin with the cycle chart of Haq et al. (1987) is not documented. The 1993 high-resolution seismic survey on Lake Balaton offered an outstanding complementary data set for our purpose as it served as outcrop-scale link between the regional sequence stratigraphic procedure and the fieldwork. According to the data presented and discussed in this paper, a number of conclusions can be summarized as follows.
Sequence stratigraphic f r a m e w o r k o f western Pannonian Basin Five third-order (with 106 year periodicities) sequences have been recognized at regional scale in the post-rift succession of western Pannonian Basin (Figs 4-8). We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-1) and Pannonian-1 (PAN-l) through Pannonian-4 (PAN-4). Reliable time constraint was only available for the stratigraphic interval between 9.0 Ma and 7.4 Ma which has been calibrated by the Iharosber6ny-I magnetostratigraphic record, revised after Cande & Kent (1995). Maximum flooding surface mrs-2 marks the peak of a major flooding event which occurred in
385
the Pannonian Lake at c. 9.0 Ma. This event was manifested by Congeria czjzeki open lacustrine beds (Szfik fm.) which flooded the basin margins. Mrs-2 represents a quasi time line at basin scale that can be strikingly correlated with the top of Pannonian s.s. stage (Lower Pannonian sensu L6renthey 1900). Sequence boundary PAN-2 is the result of a major water level drop (several tens of metres) in the Pannonian Lake at c. 8.7 Ma. Evidence for subaerial exposure of the lake margins documented by significant erosion along PAN-2 SB, suggests a scenario where a number of depositional, epigenetic and/or diagenetic features developed in response to the lowering of base level. These features are widely represented in the 'marginal facies' of Transdanubia in the area of Lake Balaton. Among them are: continental travertine (geyserite of the Hungarian literature), evaporitic dolomite, swamp deposit, calcrete and palaeosols of Tihany formation, residual rocks associated with epigenesis in the vadose zone, such as silica-cemented regolith (k6tenger silcrete) of the Kfilla formation (Fig. 9). Maximum flooding surface mfs-3 (c. 7.4 Ma) represents a second important flooding event within the Pannonian Basin during Late Miocene. It is again characterized by occurrence of open lake strata, this time associated with a younger Congeria assemblage (Congeria rhomboidea). Mfs-3 represents another useful chronostratigraphic surface at basin scale which can be considered a good proxy in western Hungary for the base of Pontian as it is defined in the stratotype area (Black Sea Basin) (Fig. 12). The higher rank unit bounded by PAN-1 SB and PAN-4 SB approximately correlates with the Tortonian-Messinian of the standard time scale and correspond to the 'Late Miocene sequence' of Juhfisz et al. (this volume).
Chronostratigraphic implication Maximum flooding surfaces mfs-2 (9.0 Ma) and mfs-3 (7.4 Ma) represent two quasi-isochronous surfaces at basin scale. Our study showed (Figs 8, 11, 12 & 13) they individuate a package of strata in between the Pannonian s.s. and the Pontian s.s. stages of the Paratethys (lower part of Pontian sensu Stevanovid 1951). This correlation suggest the opportunity to introduce a three-fold subdivision of the Pannonian stage sensu L6renthey (1900) into Early Pannonian (Pannonian s.s.), Mid-Pannonian ('Danubian' or Transdanubian' in Sacchi et al. 1997, 1998) and Late Pannonian (Pontian s.s.).
386
M. SACCHI ETAL.
The concept of an intermediate stage between Early Pannonian and Late Pannonian is not new. Since an early stratigraphic study on Pannonian strata, Halav~its (1903) had already defined a 'Middle Pannonian' succession, corresponding to Congeria ungulacaprae and Congeria balatonica beds, which partly correlates with our Danubian (or Transdanubian) stage (Fig.
11). High-resolution seismic survey on Lake Balaton and the outcrop-scale stratigraphic f r a m e w o r k o f Pannonian s.1. strata Two major unconformities, both associated with significant stratigraphic gap have been detected in the subsurface of Lake Balaton: (a) an upper unconformity at the base of the Balaton Quaternary deposits; (b) a lower unconformity at the top of the pre-Pannonian (Middle Miocene) strata. These two unconformities separate, from bottom to top, three major seismostratigraphic units (Figs 18-24): (1) pre-Pannonian strata; (2) Pannonian s.l. (Late Miocene) succession; (3) Late Pleistocene-Holocene deposits of Lake Balaton. (1) Pre-Pannonian (Sarmatian) strata are dramatically truncated at the top by a mature (polycyclic) erosional surface that marks a significant stratigraphic gap (about 12-9 Ma) spanning from the Upper Sarmatian to the upper part of Pannonian s.s. The gap includes amalgamation of SAR-1 SB (top of Sarmatian sequence) and PAN-1 SB (top of first Pannonian third-order sequence).V-shaped valleys with associated fluvial terraces and channel-fills are recognized from the seismic record. (2) The Pannonian s.l. strata are essentially represented by part of sequence PAN-2 and include from bottom to top the Sz~ik, Soml6 and Tihany formations. The Sz~ik formation is interpreted as open lacustrine TST deposits which onlap the underlying Sarmatian basement. The top of this formation is represented by mfs-2 (9.0 Ma) and correlates with the top of Pannonian s.s. stage. The Soml6 formation and the lower part of Tihany formation correspond to early progradational and late progradational-aggradational highstand systems tract (HST) deposits. These strata have been designated as Pontian faciostratotype in Hungary. However they are Danubian (or Transdanubian) in age according to the chronostratigraphy adopted in this study. The upper part of Tihany formation is separated from the underlying strata by PAN-2 SB (8.7 Ma) and corresponds to the LST of the
above sequence PAN-3. Unless associated with travertine mounds PAN-2 SB is hardly detectable on seismic profiles. Similarly, this subtle sequence boundary is easily detected in the outcrop only where associated with distinctive depositional/diagenetic features (i.e. paleosols, silcrete). The development of PAN-2 SB slightly predates the onset of basaltic eruption of the Tihany Volcano (c. 7.8 Ma, Balogh 1995). Stratal patterns towards igneous bodies inferred from seismic profiles suggest that most nearsurface magmatic intrusions were coeval and/or slightly postdated the deposition of the Szfik fm (c. 9.0). (3) Lake Balaton is part of a relatively young hydrographic system which evolved in the latest Quaternary, during the post-Wtirm deglaciation of Central Europe. The unconformity at the base of the Pleistocene-Holocene deposits of Lake Balaton, corresponds to a subaerial erosional surface which dramatically truncates the underlying Pannonian strata (Figs 18-22, 24). This unconformity marks a major stratigraphic hiatus that practically encompasses the whole Pliocene and most of the Pleistocene.
Tectonic interpretation The Neogene succession of the western Pannonian Basin is affected by significant post-rift tectonic deformation. A regional tilting of the western Pannonian Basin fill towards ESE is evidenced by seismic profiles, consistently showing ESE-dipping strata which are erosionally truncated at the top. Previous studies (Tari 1994; Horv~ith 1995) have proposed this regional tectono-stratigraphic setting was induced by a late-stage tectonic inversion and uplift with consequent subaerial erosion of Late Neogene strata in Transdanubia and Northern Hungary. Based on our interpretation we suggest this tectonic inversion may have started at the very beginning of Pliocene (after PAN-4 SB). An older compressive/transpressive tectonic phase can be also detected from regional seismic profiles at about the end of Sarmatian (Horv~ith 1995). This may account for the significant and almost ubiquitous stratigraphic gap that we have documented at the base of the post-rift strata of western Pannonian Basin. The tectonic pattern in the area of Lake Balaton is expressed by SW-NE strike-slip faults and associated folds which postdate the 'Middle Pannonian' succession of Transdanubia (Figs 18-22, 24, 28). We kindly acknowledge the officers and the crew of the RV VizvOdelem (Central Transdanubia Water
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN Management Office) for their skilled help and assistance during the seismic survey on Lake Balaton, L. Mirabile and his staff (Oceanography Institute, Istituto Universitario Navale, Naples) for their expertise in the data acquisition, T. McGee (Thalassic Data Limited, Vancouver) for his help during the acquisition and t h e preliminary data processing and T. Cserny and A. Jfimbor (Geological Institute of Hungary, Budapest) for allowing us access to existent seismic and borehole data in the study area. We sincerely thank I. Magyar, A. Mindszenty, P. Mtiller and O. Sztan6 for their help during the field work, together with M. B. Cita, B. D'Argenio, P. D6v6nyi, A. Gal~cz, M. K~izm6r, M. Lantos, F. Molisso, A. Nagymarosy, P. Sclafani and G. Tari for their support at various stages of this work. Thanks are also due to A. Bally for his precious suggestions on the interpretation of Balaton seismic profiles and T. Jacquin who revised the manuscript. The research work has been developed as part of a PhD project that the first author is carrying on at the Department of Geophysics of EOtv6s Lorfind University, Budapest. The digital conversion and processing of the original analogue seismics acquired on Lake Balaton were edited by O. Magyari. Financial support was provided by the IBS Project (Contract JOU2-CT92-0110), the Italian-Hungarian Cooperation Agreement (CNR-MTA) for the period 1995-1997, the Geomare Sud Institute, CNR, Naples, Italy, and the National Science Foundation of Hungary (OTKA 4181).
References BALDI, T. 1980. The early history of the Paratethys. FOldtani K6zIOny, 110, 456472. BALLY, A. W. & SNELSON, S. 1980. Realms of subsidence. In: MIALL, A. D. (ed.) Facts and principles of world petroleoum occurrence. Memoirs of the Canadian Society of Petroleum Geologists, 6, 9-94. BALOGH, K. 1995. K/Ar study of the Tihany .Volcano, Balaton Highland, Hungary. Report on the work supported by the European Community in the flame of program 'IBS'. Institute of Nuclear Research. Hungarian Academy of Sciences. Debrecen. , fikRVA-SOs,E., PI~CSKAY,Z. & RAVASZ-BARANYAI, L. 1986. K/Ar dating of Post-Sarmatian alkali basaltic rocks in Hungary. Acta Mineralogica Petrographica, 28, 75-93. BgRCZI, I. & PHILLIPS,R. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, Southeast Hungary. Geophysical Transactions Special Edition, 31, 55-75. BERGGREN, W. A., HILGEN, E J., LANGEREIS, C. G., KENT, D. V., OBRADOVICH,J. D., RAFFI, I., RAYMO, M. E. & SHACKLETON,N. J. 1995. Late Neogene chronology: New perspectives in high-resolution stratigraphy. Geological Society of America Bulletin, 107, 1272-1287. CANDE, S. C. & KENT,D. V. 1992. A new Geomagnetic Polarity Time Scale for the Late Cretaceous and
387
Cenozoic. Journal of Geophysical Research, 97, 13917-13951. & -1995. Revised calibration of the geomagnetic polarity time scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 100, 6093-6095. CHOLNOKY, J. 1918. A Balaton hidrogr6fi6ja. A Balaton tudom~inyos tanulm~inyoz~is~inak eredm6nyei I/2. CICHA, I. & SENES,J. 1968. Sur la position du Mioc6ne de la Parat6thys Centrale dans le cadre du Tertiaire de l'Europe. Geologica Carpathica, 19, 95-116, Bratislava. COMMITTEE OF HUNGARIAN STRATIGRAPHY 1983. Lithostratigraphic Formations of Hungary. In: FOLOP, J. (ed.) Hungarian Geological Survey Formation Chart. CSAT0, I. 1993. Neogene sequences in the Pannonian Basin, Hungary. Tectonophysics, 226, 377-400. CSERNY,Y. & CORRADA,R. 1989. Complex geological investigation of Lake Balaton (Hungary) and its results. Acta Geologica Hungarica, 32, 117-130. -& NAGY-BODOR, E. 1996. The Pre-Quaternary Morphology and Quaternary Geohistory of Lake Balaton. In: JUHASZ, A. & PECSI, M. (eds) lAG European Regional Geomorphological Conference, Hungary, April 9-12, 1996. Geomorphological studies on the Transdanubian Mountains, 70-73. - - , HERTELENDI,E. & TARJAN, S. 1995. Results of isotope-geochemical studies in the sedimentological and environmental geologic investigations of Lake Balaton. Acta Geologica Hungarica, 38, 355-376. ERDELYI, M. 1961. Ktils6-Somogy vizf61dtana. Hidrol6giai k6zl6ny, 41, 445-528. 1962. Ktils6-Somogy vizf61dtana. Hidrol6giai kOzlOny, 42, 56-65. FEJFAR, 0. 1988. Tableau stratigraphique de l'Eocene superieure au Pleistocene. In: Symposium de Gtinzburg (unpublished). GALLOWAY, W. E. 1989. Genetic stratigraphic sequences in basin analysis I: architecture and genesis of flooding surface-bounded depositional units. AAPG Bulletin, 73, 125-142. GYALOG, L. (ed.) 1996. A fOldtani tdrk~pek jelkulcsa Os a rOtegtani egys~gek rOvid leirdsa [Key of geological maps and short description of the lithostratigraphical units] MAFI alk. kiadv. [Hungarian Geological Survey], Budapest. HALAVATS, J. 1903, Die Fauna der pontischen Schichten der Umgebung des Balatonsees. In: Resultate der wissenschaftlichen Erforschung des Balatonsees 1/1 Palaontologischen Anhag 4/2, Holzel, Wien, 1-80. HAQ, B. U., HARDENBOL,J. & VAIL P. R. 1987. Chronology of fluctuating sea level since the Triassic (250 million years ago to present). Science, 235, 1156-1167. HARANGI, Sz. & HARANGI, R. 1995. Late Neogene magmatism in the NW Pannonian Basin. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of
388
M. SACCHI E T A L .
the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to field trip No. 6, Hungary, 61-74. HORNES, M. 1851. Die fossilen Mollusken des Terti~irBeckens von Wien. Jahrbuch Geol. Reichsanst., 2. HORVATH, E 1993. Towards a mechanical model for the formation of Pannonian Basin. Tectonophysics, 226, 333-357. 1995. Phases of compression during the evolution of the Pannonian Basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology, 12, 837-844. -& CLOETINGH, S. 1996. Stress-induced late-stage subsidence anomalies in the Pannonian Basin. Tectonophysics, 266, 287-300. & POGACSAS,Gv. 1988. Contribution and Seismic Reflection Data to Chronostratigrphy of the Pannonian Basin. In: ROYDEN, L. H. & HORVATH, F. (eds) The Pannonian Basin. A Study in Basin Evolution. AAPG. Memoirs, 45, 97-105. , BERCKHEMER,H. & STEGENA,L. 1981. Models of Mediterranean back-arc basin formation. Phil. Trans. R. Soc. London, s. A, 300, pp. 383M02. JAMBOR, A. 1980. A Dun~inttili-k/3z6phegysdg pann6niai k6pz6dm6nyei. Magyar dllami FOldtani IntOzet EvkOnyve, 62, 1-259. -(ed.) 1987. Kuns6gi Emelet - A Magyarorsz6gi Kunsdgi emeletbeli k4pz6dmOnyek fOldtani jellemzOse Annals of the Hungarian Geological Institute, 69. 1989. Review of the geology of the s. 1. Pannonian formations of Hungary. Acta Geologica Hungarica, 32, 269-324. , PARTI2NYI,Z. & SOLTI, G. 1981. Geological features of basalt volcanics in Hungary. Geological Institute of Hungary, 1979, 225-239. JtJHAsz, G. 1994. Comparison of the sedimentary sequences in Late Neogene subbasins in the Pannonian Basin, Hungary / Magyarorsz~igi neogdn medencerdszek pannGniai s. 1. tiledGksorfinak 6sszehasonlitGelemzdse. FOldtani KOzlOny, 124, 341-365. , PHILLIPS, L., MOLLER, P., RICKETTS, B., TOTHMAKK, A., LANTOS,M. & KovAcs, L.O. 1999. Late Neogene sedimentary facies and sequences of the Pannonian Basin, Hungary. This volume. KORPAS-HODI, M. (ed.) 1983 DunGntfdi KOz~phegys~g -Paleoecology and biostratigraphy of the pannonian mollusca fauna in the northern foreland of the Transdanubian Central Range. Annals of the Institute of Geology, Hungary. 66. KRIJGSMAN,W., GARCI2S,M., LANGEREIS,C. G., DAAMS, R., VANDER MEULEN,A. J., AGUSTi, J. & CABRERA, L. 1996. A new chronology for the middle to late Miocene continental record in Spain. Earth and Planetary Science Letters. 142, 367-380. LANTOS, M., HAMOR, Y. & POGACSAS, G. 1992. Magneto - and seismostratigraphic correlations of Pannonian s. 1. (late Miocene and Pliocene) deposits in Hungary. Paleontologia i EvoluciG, 24-25, 35-46. LASKAREV,V. 1924. Sur les equivalents du Sarmatien superieur en Serbie. Recueil de travaux offert a M. Jovan Cvijic, Beograd. -
-
-
-
-
-
LoczY, L. 1913. A Balaton tudomdnyos tanulmfnyozdsdnak eredm~nyei. I. A Balaton kOrnyGkGnek geolGgiai kGpzGdmGnyei 6s ezeknek vidGkek szerinti telepedGse. LGREYrHEY, I. 1900. Foraminifera der pannonischen Stufe Ungarns. Neues Jarbuch fiir Mineralogie, Geologie und Paliiontologie, 2, 99-107. MAGYAR, I. 1988. Mollusc fauna and flora of the Pannonian quartz sandstone at Mindszentk~illa, Hungary. Annales Universitatis Scientiarun Budapestiensis de R. EOtvOs Nominatae, Sectio Geologia - 28, 209-222. 1995. Late Miocene mollusc biostratigraphy in the eastern part of the Pannonian Basin (Tisz~intul, Hungary). Geologica Carpathica, 46, 29-36. & HABLY, L. 1994. Stratigraphic position of late Neogene paleobotanical sites in Hungary: Miocene or Pliocene? Acta Paleobotantica Polonica, 34, 195-203. MAGYAR,I., GEARY,D., SI)T(3-SZENTAI,M., LANTOS,M. & MOLLER, P. in press. Integrated biG-, magnetoand chronostratigraphic correlations of the Late Miocene Pannonian Lake Deposits. Acta Geological Hungarica. MAROSI, S. & SZIL,/~RD,J. 1981. A Balaton kialakulfisa. FOldrajzi KOzlem~nyek, 1-30. MATrlCK, R., PHILLIPS,R. & RUMPLER,J. 1988. Seismic stratigraphy and depositional framework of sedimentary rocks in the Pannonian Basin in southeastern Hungary. In: ROYDEN, L.H. & HORVATH, E (eds), The Pannonian Basin: a study in basin evolution. A A P G Memoirs, 45, 117-145. MIALL, A. D. 1997. The Geology of Stratigraphic Sequences. Springer. MIKE, K. 1980. Osmedernyomok a Balaton k6rnydk6n. FOldrajzi Ortesit6, 29, 313-334. MULLER, P. & MAGYAR, I. 1992. Stratigraphic significance of the Upper Miocene lacustrine cardiid Prosodacnomya (K6tcse section, Pannonian Basin, Hungary) / A Prosodacnomy~ik r6tegtani jelent6sdge a K6tcse k6rny6ki panndniai s.1. tileddkekben. FOldtani KOzlOny, 112, 1-38, Budapest. & -1995. Designation of the base of Pontian stage in the Pannonian Basin. Romanian Journal of Stratigraphy, 76. Abstracts of the 10th Congress of RCMNS, 1, 83-84. --& SzONOKu M. 1990. Faciostratotype TihanyFehdrpart (Hungary), ("Balatonica Beds" by L6renthey, 1905). In: STEVANOVId, P., NEVESSKAYA, L., MARINESCU, E, SOKAd, A. & JAMBOR, fi~. (eds) Chronostratigrapie und Neostratotypen, Neogen der Westliche ("Zentrale") Paratethys, Pontien, Jazu and Sanu, Zagreb-Beograd, 427-436. NAGY-BODOR, E. 1988. A Balaton pannGniai 6s holocGn kGpzGdmGnyeinek palynolGgiai vizsgfilata. FOldt. int. Evi Jel. 1986-r61., 568-580. NAGYMAROSY,A. & MOLLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian Basin. In: ROYDEN, L. H. & HORVATH,F. (eds) The Pannonian Basin. A Study in Basin Evolution. AAPG. Memoirs, 45, 69-78. PAPP, A. , GRILL, R., JANOSCHEK, R., KAPOUNEK, J., -
-
-
-
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN KOLLMAN, K. & TURNOVSKY,K. 1968. Nomenclature of the Neogene of Austria. Verh. Geol. Bundesanst., Jg. 1968, 1%27. --, JAMBOR, A. & STEININGER, E E (eds) 1985. Chronostratigraphie und Neostratotypen, Mioziin der Zentralen Paratethys 7, Pannoniaen. AkadGmiai KiadG, Budapest. POGACSAS,GY. 1984. Seismic stratigraphic features of Neogene sediments in the Pannonian Basin. Geophysical Transactions, 30, 373-410. , LAKATOS, L., UJSZASZI, K., VAKARCS, G., VARKONYI, L., VARNAI, P. & REVESZ, I. 1988. Seismic facies, electro facies and Neogene sequence chronology of the Pannonian Basin. Acta Geologica Hungarica, 31, 175-207. REGIONAL COMMITTEE ON MEDITERRANEAN NEOGENE STRATIGRAPHY 1971. Stratotypes of Mediterranean Neogene Stages. Giornale di Geologia, 37, 1-266. ROTH, L. 1879. Geologische Skizze des KroisbachRuster Bergzuges und des sudlichen Teiles des Leita-Gerbiges. FOldtani KOzlOny, 9, 139-150, Budapest. ROYDEN, L. H. & HORVATH, E (eds) 1988. The Pannonian basin. A study in basin evolution. A A P G Memoirs, 45. ROGL, E 1996. Stratigraphic correlation of the Paratethys Oligocene and Miocene. Mitt. Ges. Geol. Bergbaustud. Osterr. 41, 65-73. & STEININGER,E E 1983. Vom Zerfall de Tethys zu Mediterran und Paratethys. Die neogene Pal~iogeographie und palinspastik des zirkummediterranen Raum. Ann. Naturhis. Museum Wien, 85/A, 135-163. , BERNOR, R. L., DERMITZAKIS,M. D., MOLLER, C. & STANCHEVA,M. 1991. On the Pontian correlation in the Aegean (Aegina Island). Newsletters in Stratigraphy, 24, 137-158. SACCHI, M. 1997. Two-fold versus three-fold subdivision of Pannonian s.l. strata: a proposal for a new stage in the Upper Miocene Series of the Central Paratethys. In: Abstracts of the Interim Colloquium RCMNS, Catania (Italy), 4-9 Nov. 1997, 87-88. , CSERNY,T., DOVI~NYI,P., HORVATH,E, MAGYARI, O., MCGEE, T. M., MIRABILE, L. & TONIELLI, R. 1998. Seismic stratigraphy of the Late Miocene sequence beneath Lake Balaton, Pannonian Basin, Hungary. Acta Geologica Hungarica, 41, 63-88. , HORVATH, E, MAGYAR, I. & MOLLER, P. 1997. Problems and progress in establishing a Late Neogene Chronostratigraphy for the Central Paratethys. Neogene Newsletter, 4, 37-46. , --& MAGYARI, 0. 1995. High resolution seismics in the Lake Balaton: insights into the evolution of western Pannonian Basin. /n: HORVATH,E,TARI, G. & BOKOR, CS. (eds), Extensional collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to field trip No. 6, Hungary, 171-185.
389
SALVADOR,A. (ed.) 1994. International stratigraphic guide. International Subcommission on Stratigraphic classification of International Commission on Stratigraphy. The Geological Society of America. SHANLEY, K. W. & MCCABE, P. J. 1994. Perspectives on the sequence stratigraphy of continental strata. A A P G Bulletin. 78, 544-568. STEININGER, E E, BERNOR, R. L., FAHLBUSCH,g. ~; MEIN, E 1990. European Neogene marine/continental chronologic relations. In: L1NDSAY,E. H., FAHLBUSCH, V. & MEIN, P. (eds) European Neogene Mammal Chronology. Plenum Press, New York, 15-46. , Mt3LLER, C. & ROGL, E 1988. Correlation of Central Paratethys, Eastern Paratethys and Mediterranean stages. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin: a study in basin evolution. A A P G Memoirs, 45, 79-87. Sa'EVANOV~C, R M. 1951. Pontische Stufe im engeren s i n n e - Obere Congerienschichten Serbiens und der angrenzenden Gebiete. Serbische Akademie der Wissenschaften, Sonderausgabe 187, Mathematisch-Naturwissenschaftliche Klasse no 2, Beograd. , NEVESSKAYA,L. A., MARINESCU,E, SOKAC, A. & JAMBOR, A. (eds) 1990. Chronostratigraphie und Neostratotypen, Neogen der Westlichen ("Zentrale") Paratethys 8, Pontien. Jazu and Sanu, Zagreb-Beograd. SZAB0, Cs., HARANGI, Sz. & CSONTOS,L. 1992. Review of Neogene and Quaternary volcanism of the Carpathian-Pannonian region. Tectonophysics, 208, 243-256. SZADECZKY-KARDOSS,E. 1938. Geologie der rumpfungarlandischen kleinen Tiefeben. A soproni Bfinya 6s ErdGmGrnOki FGiskola KGzl. 10(2). SZENTGYORGYI, K. & JuIqASz, GY. K. 1988. Sedimentological charasteristics of the Neogene sequence in SW Transdanubia, Hungary. Acta Geologica Hungarica. 31, 209-225. TARI, G. 1994. Alpine tectonics of Pannonian Basin. PhD thesis, Rice University, Houston, Texas. , BALDI, T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian Basin: a geodynamic model. Tectonophysics, 226, 433-455. - - , HORVATH,E & RUMPLER,J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. UJszAszI, K. & VAKARCS G. 1993. Sequence stratigraphic analysis in the South Transdanubian region, Hungary. Geophysical Transactions, 38, 69-87. VAIL, E R. 1987. Seismic stratigraphy interpretation procedure. A A P G Studies in Geology 27. - - , AUDEMARD,E, BOWMAN,S., EISNER, P. & PEREZCRUZ, C. 1990. The stratigraphic signatures of tectonics, eustasy and sedimentology. An overview. In: EINSELE,E., RICKEN,A. & SEILACHER,A. (eds) Cycles and events in stratigraphy, Springer-Verlag, New York, 617-659. --, MITCHUM, R. JR & THOMPSON, S. 1977. Seismic stratigraphy and global changes of the sea level,
390
M. SACCHI E T A L .
part 3: Relative changes of the sea level from the coastal onlap. In: PAYTON, C. E. (ed.) Seismic Stratigraphy - A p p l i c a t i o n s to Hydrocarbon exploration, AAPG Memoirs, 26, 63-82. VAKARCS, G., VAIL, P. R., TARI, G., POGACSAS, GY., MATrICK, R. E. & SZAB0, A. 1994. Third-order Miocene-Pliocene depositional sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106. VASS, D., KOVAC,,M. & LEXA, J. 1988. Molasse Basins and volcanic activity in west Carpathian Neogene.
Its evolution and geodynamic character. Geologica Carpathica, 39, 539-561. --, REPCOK, I., BALOGH, K & HALMAI, J. 1987. Revised radiometric time-scale for central Paratethyan Neogene. Annales of the Hungarian Geological Institute 70, 423434. WEIMER, P. & POSAMENTIER,H. W. (eds) 1993. Siliciclastic sequence stratigraphy - Recent developments and applications. AAPG Memoirs, 58. WHN, GY. 1967. D61kelet-Dungmttil hegysdgszerkezete. FOldtani K6zlOny, 97, 371-395.
Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin R. T. V A N B A L E N 1, L. L E N K E Y 2, F. H O R V / k T H 2 & S. A. P. L. C L O E T I N G H
1
1Tectonics~Structural Geology Group, Department o f Sedimentarv Geology, Inst. o f Earth Sciences, Vrije Universiteit, De Boelelaan 1081, 1081 H V Amsterdam, The Netherlands 2Geophysical Department, EOtvOs L o r a n d University, L u d o v i k a tOr 2, Budapest, Hungary Abstract: During the Pliocene-Quaternary time interval, the peripheral parts of the Pan-
nonian Basin system have been uplifted and subsidence in the basin centre accelerated, causing a distinctive truncation pattern in the basin stratigraphy. Stress analyses indicate that the Pannonian Basin system, originally formed in an extensional regime, is subjected to a compressive stress since the early Pliocene. Results of forward modelling of basin subsidence and sedimentary filling along a cross-section through the southern part of the Pannonian Basin demonstrate that a change of the basin shape due to the compressive stress can successfully explain the observed pattern of differential uplift and subsidence occurring since the early Pliocene. In addition, the forward modelling of subsidence and fill provides constraints for the depth of lithospheric necking during extension, the palaeowater-depth history and lake-level changes in the southern part of the Pannonian Basin. Compaction-driven fluid flow modelling shows that the first significant overpressures in the southern part of the Pannonian Basin developed during progradation of a large deltaic system, at a time when sedimentation rates increased rapidly. Due to the stress-induced acceleration of subsidence during Pliocene to Quaternary times, sedimentation rates increased again, causing a further increase of overpressure. The Pliocene stress induced uplift of the basin flanks combined with a preceding lake-level fall created a larger gravity potential of the groundwater table, enhancing the influx of meteoric water into the basin. This can explain observed diagenetic patterns in the southern part of the Pannonian Basin. The P a n n o n i a n Basin is a N e o g e n e intram o n t a n e basin system bounded in the west by the Alps, in the north, east and southeast by the Carpathians and in the southwest by the Dinarides. The intra-Carpathian Basin system comprises a central part, the G r e a t H u n g a r i a n Plain, and several peripheral basins: the Styrian Basin, the Vienna Basin, the Little H u n g a r i a n Plain (the D a n u b i a n Basin), the Transcarpathian Basin and the Transsylvanian Basin (Fig. 1). The P a n n o n i a n Basin system originates from contemporaneous tectonic escape in the Alps (Ratschbacher et al. 1991), asthenospheric upwelling (Stegena et al. 1975; Becker 1993) and s u b d u c t i o n roll-back along the Carpathian front in a back-arc setting, comparable to the present-day A e g e a n (Horv~ith & B e r c k h e m e r 1982). A detailed discussion on the origin of the P a n n o n i a n Basin can be found in Horv~ith (1993). A n overview of basement structures resulting from the basin-forming mechanisms is given in Fig. 2. The basement of the Pannonian Basin consists of Palaeogene deposits and Palaeozoic to Cretaceous rocks, stacked on top of each other as imbricate nappes during the Cretaceous Alpine collision (Csontos et al. 1992; Tari et al. 1992; Grow et al. 1994). The Neogene extension began
in the early Mid-Miocene with the opening of rift- and pull-apart basins and the formation of metamorphic core complexes along reactivated Alpine overthrusts (Rumpler & Horvfith 1988; Tari et al. 1992). The style of extension and the direction of normal faulting varies in the basin system. The strike-slip faults, depicted in Fig. 2, act as transfer zones for the different types and amounts of extensional strain. The total amount of extension in an E - W direction exceeds 100 km and is approximately equal to the amount of c o n t e m p o r a n e o u s shortening in the O u t e r Carpathian flysch belt (Csontos et al. 1992; Tari et aL 1992). Due to changes in the motions of lithospheric plates, the stress field in the Pannonian Basin changed to compressive in Pliocene. The maximum horizontal stress in the southern part of the basin has a S W - N E orientation (Csontos et aL 1991; Gregersen 1992; MUller et al. 1992). Changing intraplate stresses can cause substantial differential subsidence in s e d i m e n t a r y basins, which is recorded as a relative sea-level change in the stratigraphy (Cloetingh et al. 1985). We investigate the effect of the compressive stress on the stratigraphy and compaction driven pore fluid overpressures in the sub-basins of the southern part of the Pannonian Basin
VANBALEN,R. T., LENICEY,L., HORV~,T.,E & CLOETINGH,S. A. P. L. 1999. Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin. In: DURAND,B., JOLIVET,L., HORVATH, E & Ss M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 391-414.
392
R . T . VAN B A L E N E T A L .
Fig. 1. (a) The Pannonian Basin and its surroundings. The names of various sub-basins and mountain ranges have been indicated. (b) Isopach map of the Pannonian Basin. A - A ' denotes the modelled cross-section. The contour lines depict the depth to the pre-Neogene basement (modified after Horv~ith 1993).
MODELLING IN THE PANNONIAN BASIN
393
Fig. 2. Fault pattern resulting from the tectonic escape from the Eastern Alps, subduction roll-back along the Carpathian front and mantle upwelling (modified after Horvfith 1993). Dextral and sinistral strike-slip faults accommodate the general eastward movement of the basin from the eastern Alps towards the Carpathian front. The amount, the style and the direction of extension in between the strike-slip faults vary considerably. A-A' denotes the modelled cross-section.
(Drava, Sava and Mako troughs and the B6k6s Basin) using forward modelling techniques. The location of the modelled cross-section is shown in Fig. 1. Below, we first give an introduction to the Neogene sedimentary fill, palaeo-water depths and lake-level variations, as these are important parameters for the tectono-stratigraphic forward modelling and the compaction-driven fluid-flow modelling. Subsequently, the latestage anomalous subsidence and the lithospheric stress field are discussed. Next, we present and discuss results from the tectono-stratigraphic forward models. Finally, we provide an introduction to overpressure mechanisms in the southern part of the Pannonian Basin and show results of forward modelling of compactiondriven overpressure. The results of forward modelling are compared to predictions obtained by a neural network analysis.
Neogene sedimentary fill of the southern part of the Pannonian Basin The Neogene Pannonian Basin was filled by a large deltaic system originating from the rising Carpathians and Alps. The reconstructed positions of the depositional shelf break of the Neogene deltaic system (Ujszaszi & Vakarcs, 1993; Vakarcs et al. 1994) demonstrate the gradual fill of the basin system from the rims towards the central and southern parts. The Neogene stratigraphy in the southern part of the basin can be subdivided into seven depositional units. From bottom to top these are: a basal unit (B), a deep basin unit (DB), a prodelta unit (PD; Szolnok formation), a delta front-delta slope unit (DS; Algy6 formation), a delta plain-delta front unit (DP; T6rtel formation), an alluvial plain unit (AP; Zagyva and Nagyalf61d formations) and Quaternary unit (Q) (B6rczi 1988; Horv~ith et al. 1988; Mattick et al. 1988; K~izmer 1990; Juh~isz 1991; see
394
R.T. VAN BALEN E T A L .
Table 1. Depositional units in the southern part of the Pannonian Basin Palaeo-water depth (m) Unit name Q AP DP DS PD DB B
Age (Ma)
Sedimentary environment
shallow
deep
Effective sand content (vol%)
0.0-2.4 2.4-3.9 3.9-5.5 5.5-6.3 6.3-8.2 8.2-10.5 10.5-17.5
alluvial plain alluvial plain delta plain delta slope pro-delta deep basin basal
0 0 0 200 n.a. n.a. 0
0 0 0 400 800 >1000 0-2000
0.2 0.2 0.8 0.8 0.2 0.0 1.0
Table 1). These depositional units have diachronous ages, as they are related to a deltaic system. The basal unit was deposited during the initial part of the syn-rift stage. The deposits in this unit in the southeastern part of the Pannonian Basin system consist of redeposited conglomerates alternating with marl beds in the deep subbasins and abrasive conglomerates around the basement highs (B6rczi & Phillips 1985; B6rczi 1988; K~zm6r 1990; Juh~sz 1991). In the southeastern part of the Pannonian Basin, the next depositional unit, the DB unit, consists of argillaceous and calcarous marls (B6rczi & Phillips 1985; B6rczi et al. 1988; K~izm6r 1990; Juh~isz 1991). The first deltaic sediments in the deep sub-basins consist of turbidites (the PD unit). This facies lacks in the shallower parts of the basin; it was, for example, not deposited on the highs in between and flanking the sub-basins. The PD unit consists of grey argillaceous marls, siltstones and light-grey sandstone beds. In general, the marl intercalations decrease upwards (K~zm6r 1990). The next depositional unit consists of delta slope deposits (DS unit), mainly containing mudstone (siltstone and argillaceous marl), with a few interbedded sandstone bodies (Juh~sz 1991). These sediments are widespread over the basin (K~izm6r 1990; Juh~sz 1991), indicating that the prograding delta covered both the deep sub-basins and the basin highs, filling the whole basin almost completely by early Pliocene (K~zm6r 1990). The next unit is deposited in a delta plain environment (DP unit). This is a sand rich unit. The sandstone bodies occur as distributary mouth bars and channel fill rhythms of about 20-50 m thickness. The extension and continuity of these bodies are rather restricted, but they can merge laterally with one another (Juh~isz 1991). The last depositional units consist of alluvial plain sediments (AP and Q units). It contains thin bedded siltstone, claystone and sandstone with the dominance of the fine-grained fraction (Juh~isz 1991).
The modelled profile has a SW-NE orientation and transects the deep sub-basins in the southern part of the Pannonian Basin (the Sava, Drava and Mako troughs and the B6k6s Basin), see Fig. 1. Because the modelling transect is parallel to the trend of the depositional shelf break at 5.5 Ma (Vakarcs et al. 1994), we assume that along the profile line the depositional units are also chronostratigraphic units. The ages of the different units are constrained by the sequence stratigraphic analyses of the B6k6s Basin by Vakarcs et al. (1994; Table 1). Palaeo-water depths
The tectono-stratigraphic forward modelling requires an estimation of palaeo-water depths for the depositional units. Due to endemic fauna, it is hard to assess palaeo-water depths using biostratigraphy (Nagymarosy & MUller 1988). Furthermore, for sediments in the deepest parts of the sub-basins knowledge about palaeo-water depths is even more problematic, as these rocks were almost not drilled. Therefore, indirect indications have to be used in order to estimate palaeo-water depths of sediments in the deep sub-basins and in shallower parts of the basin system. During the mid- and late Miocene the B6k6s Basin was starved, as other sub-basins closer to the sediment source area captured the sediment derived from the uplifted and eroding Carpathians and Alps (K~zm6r 1990; Grow et al. 1994). The combination of low sedimentation rate and high subsidence rate must have produced great water depths (1000-1500 m) in the central part of the B6k6s Basin. The difference between present-day burial depths of abrasive conglomerates around basement highs and turbiditic conglomerates in the Mako trough and B6k6s Basin is more then 4 km (Juh~sz 1991); both are syn-riff deposits. After correction for differential subsidence due to thermal subsidence and isostasy, this indicates that the deepest part of
MODELLING IN THE PANNONIAN BASIN the Mako and B6k6s sub-basins may have obtained palaeo-water depths of about 2 km during and immediately after the rifting stage. The calcarous marls of the DB unit have a pale yellow colour on the basement highs and become darker (and eventually black) as the basin gets deeper (B6rczi et al. 1988; Juhfisz 1991). This, combined with the abundant occurrence of pyrite (K~izm6r 1990; Juhfisz 1991), the decreasing content of carbonate towards the deep sub-basins (Csat6 1993) and the lack of bioturbation, indicates a deep, oxygen-deficient, reducing environment with a stratified water column in the deep sub-basins (Kfizm6r 1990). Therefore, also sedimentological evidence supports very deep palaeo-water depths (>1000 m) during the deposition of the marls of the DB unit. Assuming that the alluvial plain sediments were deposited at zero palaeo-water depth, the thickness of the delta slope deposits gives an estimate for the palaeo-water depths of the prodelta turbidites. The difference in burial depth between the base of the alluvial plain deposits and the pro-delta turbidites (after corrections for compaction, lake-level variation and isostasy) gives an estimate of 700 m for the palaeo-water depths during deposition of the pro delta (PD) unit. Likewise, our estimate for the palaeo-water depth of the DS unit is 400 m, for the delta plain sediments and for the A P unit 0 m (see Table 1). Using similar reasoning, the palaeo-water depths for the depositional units at the flanks of the sub-basins are also estimated. The results are depicted in Table 1. Lake-level
variations
During the extension stage, the Pannonian Basin is part of the Paratethys and connected to the world seas. Eustatic sealevel changes (e.g. Haq et al. 1987) must have influenced the stratigraphy during this time interval. Elevation of the Dinarides and the eastern and southern parts of the Alps at the Sarmatian-Pannonian stage boundary causes the isolation of the Pannonian Basin from the world seas, it becomes a lake (Royden etal. 1983; Csontos etal. 1992). The isolation is associated with a drop in salinity (Jfimbor 1989; Kfizm6r 1990; Tari et al. 1992; Vakarcs et al. 1994) and the creation of an endemic fauna (Nagymarosy & Mt~ller 1988; Kgzm6r 1990). According to seismic sequence stratigraphic analyses combined with palaeomagnetic dating, a correlation exists between the eustatic sealevel curve of Haq et al. (1987) and the relative lake-level variations in the Pannonian Basin (Tari et al. 1992; Csat6 1993; Ujszaszi & Vakarcs 1993; Vakarcs et al. 1994). This can
395
be explained by changes in climatic conditions (drainage in the uplifting Carpathians) and by upstream effects of changes in the ultimate base level of the proto-Danube (Tari et al. 1992). The most dramatic lake-level change occurs at 6.3 Ma (Messinian): the lake-level drops and subsequently rises 100 to 200 m (Tari et al. 1992; Vakarcs et al. 1994). However, also based on seismic correlations, Mattick et al. (1994) have argued that most of the identified sequence boundaries are caused by delta-lobe switching. In this study we follow the inferences of Tari et al. (1992) and Vakarcs et al. (1994).
The Pliocene to Recent stress field and anomalous subsidence Due to lithospheric plate motions, the intraplate stress field has changed to a compressive regime since early Pliocene (Csontos et al. 1991; Gregersen 1992; Maller et al. 1992). Focal mechanism solutions show maximum horizontal stress orientations perpendicular to the Carpathian arc, except for the southeastern part of Hungary where the orientation is roughly N E - S W (Gregersen 1992). In addition, along the Mid-Hungarian-Balaton strike-slip zone positive flower structures are evidence for Pliocene-Quaternary wrench tectonics related to a N-S-oriented compressional stress field (L6rincz & Szab6 1993). These inferences are in agreement with bore-hole breakout and in-situ stress measurements (Mt~ller et al. 1992; Grt~nthal & Stromeyer 1992; Becker 1993; Fig. 3). In the major part of the Pannonian Basin the break-out data are scattered, possibly reflecting detachment of the sedimentary cover from the lithospheric basement. However, in the southeastern part of the Pannonian Basin system a general trend of the maximum horizontal stress orientation of SW-NE can be identified, parallel to our modelling profile line. Finite element modelling of the stress field in and around the Pannonian Basin indicates that a combination of North Atlantic ridge push, forces resulting from the northward movement of Africa and Arabia, subduction forces at the eastern border of the Pannonian Basin and a decreased rigidity of basement in the centre of the basin system can explain the observed stress field (Grt~nthal & Stromeyer 1992; Bada et al. 1995). The decreased rigidity is related to thermal weakening due to asthenospheric upwelling. Anomalous
late stage subsidence
During Pliocene to Quaternary times, subsidence decreases or uplift takes place in the
396
R.T. VAN BALEN E T A L .
Fig. 3. The stress map of the Pannonian Basin (after Mtiller et al. 1992; Becker 1993) showing the consistently SW-NE directions of maximum horizontal stress in the southern part of the basin system. A-A' denotes the modelled cross-section.
external parts of the Pannonian Basin system and subsidence in the basin centre accelerates (e.g. Royden et aL 1983; L~z~rescu et al. 1983; Polonic 1985; D e m e t r e s c u & Polonic 1989; Vakarcs et al. 1994; Horvfith & Cloetingh 1996). Studies of the Danube river terraces and travertine horizons show clear evidence for a 200-300 m uplift of the Transdanubian Central Range (Ronai 1974; Gfibris 1994). Furthermore, Quaternary erosion has removed large parts of Late Pliocene sediments in this area (Ronai 1985) with exceptions of parts which were covered by contemporaneous basaltic lava flows (G~bris 1994; Horv~ith & Cloetingh 1996). Geodetic measurements show that the area of the B6k6s Basin is presently subsiding at a rate of 4 mm a -a whereas Transdanubia rises up to 1.3 mm a -I (Joo et al. 1990). A map of compiled anomalous subsidence data by Horvfith & Cloetingh (1996) is shown in Fig. 4. Further insights into the subsidence history of the Pannonian Basin can be obtained by constructing basement subsidence curves. These curves are obtained by subtracting the deposited sediment thickness from the present-day thickness, in order to obtain the depth of the basement
in the past. Corrections are made for compaction of sediment and changes in palaeo-water depth and eustasy. No isostatic correction is made, as isostasy is one the components contributing to basement subsidence. Basement subsidence curves for several wells in the neighbourhood of the modelling profile line clearly show three different subsidence phases: a syn-rift phase, subsidence due to sediment loading during the deposition of delta-slope deposits and an acceleration or deceleration of subsidence since the early Pliocene (Fig. 5). Backstripping data were obtained from various sources: data for the centre of the Drava Trough (Molve field) are after Bari6 et al. (1991), artificial well data for the flank of the Drava trough are after seismic data in Ujszaszi & Vakarcs (1993), the MI-1 and Bes-1 well stratigraphies are after B6rczi (1988), Usz-1, Doboz-1, Sarkad-1 and Tot-1 are after Horvfith et al. (1988) and Hod-1 and Mako-1 are after Mattick et al. (1988). All stratigraphic data are approximate. Ages for the depositional units are after Vakarcs et al. (1994). The positions of the wells in our modelling transect are indicated in Fig. 6. The wells located in and at the flank of the Drava Trough show uplift since early
MODELLING IN THE PANNONIAN BASIN
397
Fig. 4. The pattern of Pliocene-Quaternary anomalous subsidence and uplift in the Pannonian Basin (modified after Horvfith & Cloetingh 1996). In general, the central part of the basin system shows continuous or enhanced subsidence, whereas the flanks of the system are uplifted. Pliocene (Fig. 5). The wells located in and at the flanks of the Mako trough and B6k6s Basin have experienced an acceleration of subsidence starting at early Pliocene.
Tectono-stratigraphic forward modelling The adopted numerical model for basin evolution is based on the pure shear stretching principle, originally proposed by McKenzie (1978). In this model, thinning of the lithosphere results in post-rift subsidence due to thermal contraction. The pure-shear model has been extended in order to account for two-dimensional heat flow, flexural isostasy (including the effects of changing in-plane stresses), and necking of the lithosphere during extension (Kooi & Cloetingh 1992; Van Balen & Cloetingh 1993). The latter process can be described as thinning of the lithosphere around its strongest part(s), commonly termed the level of necking. The main parameters for the tectono-stratigraphic forward modelling are: pre-rift crustal and subcrustal thicknesses, thinning factors, the depth to the level of necking, sea-(lake-)level history, palaeowater depths and intraplate stresses. The forward modelling predits a stratigraphic
cross-section which is compared to data. Before the modelling, the parameters have to be constrained as much as possible. By fitting the predicted profile to observations, an iterative procedure can provide further constraints, i.e. the modelling gives further limits to parameter values. The basement of the Pannonian Basin consists of Cretaceous Alpine nappes (Tari et al. 1992) and, therefore, the crust was overthickened before the Neogene extension took place (Horv~th & Berckhemer 1982; Csontos et al. 1992; Tari et al. 1992). The pre-rift thickness of the crust and subcrustal parts of the lithosphere are assumed to be equal to the current thicknesses in the eastern Alps, giving a crustal thickness of about 42.5 km and a subcrustal thickness of about 82.5 km before extension (Table 2). The current crustal thickness below the Great Hungarian Plain, without the sedimentary cover, is on average 20 km, whereas the subcrustal thickness is 40 km and less (Horvfith 1993). The present-day crustal strength of the Pannonian Basin, as demonstrated by rheological modelling of earthquake depths by Horv~th & Cloetingh (1996), is concentrated around a depth range of 6-10 km (Fig. 7). Assuming that the
398
R.T. VAN BALEN E T A L .
Fig. 5. Basement subsidence curves for several wells along or in the vicinity of our modelling profile line. Three phases of subsidence can be recognized in all wells: a syn-rift phase, subsidence due to increase of sediment loading during the arrival of the Pannonian delta and the stress induced accelaration of subsidence. The wells located in or at the flank of the Drava trough (depicted in upper panel) show Late Pliocene uplift or cessation of subsidence. Basement subsidence curves for wells located in or at the flanks of the Mako trough and B6k6s basin (depicted in the lower panel) show Late Pliocene acceleration of subsidence. See text for further discussion and data sources.
upper crust has strengthened due to cooling during the post-rift phase, the strongest part of the lithosphere is 1-2 km shallower immediately after rifting. Subtracting an additional 2 km for present-day average sedimentary cover thickness, results in a depth of the strongest part of the crust of 3 km at the beginning of the post-rift
phase. Taking into account a crustal extension factor of 2 to 3 (Royden et al. 1983; Horv~ith et al. 1988; Royden & D6v6nyi 1988) and assuming that the strongest part of the crust coincides approximately with the level of necking during extension (Van Balen & Cloetingh 1994), it follows that the pre-rift depth of necking is probably located at a depth between 5 and 20 km (in the absence of isostasy, the level of necking itself is, by definition, not displaced during extension). We have, therefore, modelled the subsidence along the profile using two end-member values of 7.5 and 15 km for the depth of lithospheric necking during extension. Constraints for the stratigraphic modelling are given by well data (B6rczi 1988; B6rczi & Phillips 1985; Mattick et al. 1988; Rumpler & Horv~ith 1988), a Neogene basement depth map (Horv~ith 1985), palaeogeographic maps for the distribution and thickness of the main sedimentary sequences in the southeastern part of the Pannonian Basin (Juh~isz 1991) and stratigraphic cross-sections through the Drava Trough (Bari~ et al. 1991). In addition, a seismic sequence stratigraphy interpretation of the B6k6s Basin (Vakarcs et al. 1994), palaeomagnetic ages of the depositional units (determined by mollusc faunas) in the Danube-Tisza interfluve area (Korp~s-H6di et al. 1992) and a sequence stratigraphic interpretation of the south Transdanubian region (Ujszaszi & Vakarcs 1993) provide further constraints for the thicknesses and ages of the sedimentary sequences, see Table 1. We have converted the two-way travel time scale of the seismic section of Vakarcs et al. (1994) to a depth scale using seismic velocity-depth logs measured in wells in the B6k6s Basin (Doboz-I, B6k6s-2) and the Mako trough (Hod-I) (e.g. Posgay et al. 1996). As our modelling transect is parallel to the trend of the depositional shoreline break at 5.5 Ma in the southeastern part of the Pannonian Basin (Korp~is-H6di et al. 1992; Ujszaszi & Vakarcs 1993; Vakarcs et al. 1994), boundaries between the lithostratigraphic units are assumed to be proxys for chronostratigraphic boundaries. Because the transition from deposition of basal sandstone to deep basin marls reflects the deepening and drowning of the sub-basins during rifting, this stratigraphic boundary is also almost synchronous. The stratigraphic cross-section based on the data sources mentioned above is shown in Fig. 6. Clearly recognizable are the truncations at the basin margins in Pliocene sequences (the DP and A P units). An important objective for the stratigraphic forward modelling is to reproduce the same stratigraphic cross-section, with special
MODELLING IN THE PANNONIAN BASIN
399
Fig. 6. Constructed stratigraphic cross-section along our model profile line. The position of hydrocarbon pools (Dank 1988) and backstripped wells (see text) are also indicated. V indicates the position of the seismic crosssection in Vakarcs et al. (1994) which is almost perpendicular to our section; other symbols are abbreviated well names. See text for further discussion and data sources.
emphasis to the truncations. Knowledge about rheology and amounts of crustal and lithospheric thinning are used to constrain the modelling. As a result of the forward modelling, predictions for lake-level variations, palaeowater depths during deposition, intra-plate stress history and the level of lithospheric necking depth during extension are obtained by matching the predicted with the known stratigraphic cross-section.
Modelling results
Fig. 7. Depths of Recent earthquakes in and around the Pannonian basin (modified after Horv~th & Cloetingh 1996). Earthquakes occurring within and outside the Pannonian Basin system are indicated. The shape of this frequency curve resembles the shape of lithospheric strength profiles.
The results of the stratigraphic forward modelling are depicted in Fig. 8. Also shown in the same figures are the estimates obtained by the modelling for the lake-level variations and palaeo-water depths for the deepest part of the Mako trough (the position of the Hod-I well). For the lake-level variations we have taken into account that they varied in concert with the eustatic sea-level changes, although not necessarily with the same amplitude (see the previous section on lake-level variations). The most dramatic lake level change occurs a r o u n d the
400
R.T. VAN BALEN E T A L .
Messinian (6.3 Ma): lake level drops 100-200 m and subsequently rises by about the same amount (Tari et al. 1992; Vakarcs et al. 1994). Along our modelling transect, the lake level fall occurs during deposition of the PD unit and lake
(a)
(b)
level rises when the delta slope deposits of the DS unit are deposited. At early Pliocene, after deposition of the DS unit, the Pannonian Basin becomes completely filled and the Pannonian lake no longer exists. As discussed before, the
MODELLING IN THE PANNONIAN BASIN
401
Fig. 8. The stratigraphy predicted by our forward modelling. The insets show the adopted lake-level curve, the palaeo-water depth curve for the deepest part of the Mako trough (the position of the Hod-I well) and the applied intraplate stress history. The lower panel depicts the adopted crustal and subcrustal extension factors. (a) Result of the applying only lake level variations, this model can not explain the thickness distribution of the AP unit. (b) Improved model applying an increase in the level of compressive intraplate stress starting at Late Pliocene; this model result better fits the observed stratigraphy. (c) The result of a model applying a deeper level of necking (17.5 km). This model requires higher palaeo-water depths than the previous models.
Table 2. Adopted numerical values for tectono-stratigraphicforward modelling
Initial crustal thickness Initial subcrustal thickness Density crust, density mantle Thermal diffusivity Thermal expansion coefficient Young's modulus lithosphere Poisson ratio lithosphere Isotherm defining effective elastic thickness
palaeo-water depths are less well constrained. Especially the basal sandstone and the d e e p basin marls of the DB unit were deposited at unconstrained water depths (2000-1000 m). The a d o p t e d numerical forward m o d e l l i n g tool applies differential lithospheric extension. Crustal extension factors, although constrained by crustal thickness maps given by Horvfith (1993), cannot be determined on the scale of our modelling. Therefore, they are determined by a modelling iteration on the basement morphology. Subcrustal extension factors are even less
42.5 km 82.5 km 2.9 g cm-3, 3.3 g cm-3 7.5 • 10-7m2 s 1 3.2 • 10-5 ~ 7 • 101~Pa 0.25 350~
constrained, as the distribution of subcrustal thicknesses is uncertain and thermal relaxation causes the subcrust to thicken during the postrift phase. As can be inferred from the subcrustal thickness maps in Horvfith (1993), the distribution of subcrustal extension factors should be smooth. Subcrustal extension is the prime cause for post-rift (thermal) subsidence (McKenzie 1978; R o y d e n & K e e n 1980). Therefore, subcrustal extension factors can be assessed by fitting the thickness of post-rift sediments. Empirically we found that a subcrustal extension
402
R.T. VAN BALEN E T A L .
factor of 5.0 below the Great Hungarian Plain results in the observed amount of post-rift subsidence. This value is in agreement with a coherency analyses for the amount of heat flow, syn- and postrift subsidence (assuming local isostasy) by Royden & DOv6nyi (1988), which resulted in crustal extension factors ranging from 1.6 to 2.9 and subcrustal extension factors ranging from 6.7 to 20 for wells in the Great Hungarian Plain. The difference in the total amount of thermal subsidence and the thermal subsidence rate for subcrustal extension factors in the range of 5 to 20 is minor (McKenzie 1978). We apply, arbitrarily, subcrustal extension factors which are constant below the sub-basins and linearly decay towards the rims of the basin system. As the amount of post-rift subsidence is not a model-fitting parameter, we apply the same amount and distribution of subcrustal extension in all presented models. Adopted values for other modelling parameters are given in Table 2. In the first model a 7.5 km deep level of lithospheric necking during extension is applied. Crustal extension factors and palaeo-water depths are obtained by a fitting iteration. In this first simulation, we determine whether a lakelevel variation can explain the truncations of the DP and A P units. Therefore, lake-level varies in accordance with the eustatic sea-level curve. The result is depicted in Fig. 8a. The modelling requires a palaeo-water depth maximum of 2300 m during the extension stage in order to yield a good fit to the observed stratigraphy. The required crustal and subcrustal extension factors are displayed in the lower panel of the figure. The result shows that the assumed lake-level variations can indeed explain the truncation patterns at the flanks of the Mako and B6k6s subbasins and the thickness of the DP unit. However, the thickness of the AP unit in the deepest parts of these sub-basins does not fit the observed thickness (Fig. 6). The adopted lakelevel fall causes a predicted thickness of about 500 m for this unit which is about half of the observed thickness. This cannot be corrected by adopting a different magnitude of palaeo-water depth, as it is minor during this time interval. Therefore, we conclude that a lake-level fall can not explain the truncation and thickness distribution of the AP unit. The implications of the adopted palaeo-water depths will be discussed below. In the next model we determine whether an increase in the level of compressive intraplate stress can explain the truncation of the AP unit and its thickness distribution. We adopted an increase in the level of compressive intraplate
stress starting at 3.9 Ma, reaching a value of 3 kbar at 2.4 Ma and subsequently decreasing to a present-day value of 1.5 kbar. As shown by an inspection of Fig. 8b the overall stratigraphy and basin shape predicted by the model compares well with the observed large scale pattern in the basin geometry and basin fill. The Pliocene increase in the level of compressive stress can explain the truncation of the A P unit. The predicted thickness of this unit in the deepest parts of the sub-basin better fits the observed thickness than the previous model which applied a lake-level change. The depth of lithospheric necking controls the shape and depth of the syn-rift basin. Shallow levels of necking cause undeep, wide syn-rift basins. Deep levels of necking induce deep synrift basins with a pronounced topography (Kooi & Cloetingh 1992; Van Balen & Cloetingh 1993). Due to the higher amount of syn-rift subsidence, models invoking deeper levels of necking require lower amounts of extension to generate the same total subsidence (syn- and post-rift) and, therefore, require larger palaeo-water depths to fit the observed thicknesses of syn- and first post-rift deposits. The result of a model adopting a 15 km deep level of necking during extension is depicted in Fig. 8c. This model invokes the same stress and lake-level history as the previous model. The adopted crustal extension factors and palaeo-water depths are displayed in the same figure. This model requires indeed a smaller amount of crustal extension and larger palaeo-water depths. Discussion
The modelling results shows that an increase in the level of compressive stress starting at early Pliocene can explain the observed truncation of the AP unit and its thickness in the deep parts of the Mako and B6k6s sub-basins. The compressive intraplate stress causes an uplift of the basement high located between the Drava and Mako sub-basins and the eastern flank of the B6k6s sub-basin (Fig. 9). This uplift and subsidence pattern is in accordance with the general uplift and subsidence occurring in the Pannonian Basin system (Fig. 4) (Horv~th & Cloetingh 1996). The tectono-stratigraphic forward modelling applying a 7.5 km deep level of lithospheric necking requires very deep palaeo-water depths (2.3 km) during the syn-rift and initial part of the post-rift phase for the deepest parts of the Mako and B6k6s sub-basins. This is in accordance with the syn-rift basin morphology, but is higher than previous estimates based on sedimentological evidence (e.g. Kfizm6r 1990; Pog~.cs~,s et al.
MODELLING IN THE PANNONIAN BASIN
403
Fig. 9. The compressive stress-induced subsidence and uplift pattern at the stress maximum (2.4 Ma). The maximum stress-induced subsidence occurs in the Mako trough. The eastern flank of the B6k6s basin and the high in between the Drava and Mako troughs are uplifted, causing a relative lake-level fall and stratigraphic offlap of the AP unit.
1994). This problem cannot further be resolved. Because the palaeo-water depths are already very deep for a lithospheric necking depth of 7.5 kin, a deeper level of necking is unlikely. The presented model applying a 15 km deep level of necking requires a palaeo-water depth of 2.5 km. We, therefore, conclude that the Pannonian Basin extended with a shallow level of necking (5-10 km) during rifting. This implies that the strength of the subcrustal lithosphere was negligible during the rifting process (Van Balen & Cloetingh 1993). This is in accordance with the results of Gaudemer et al. (1988) and Cloetingh et aL (1995), showing that the bulk lithospheric strength of a wide orogenic belt is almost completely determined by the strength of the upper crust, due to the restricted cooling possibilities and thickness of the crust.
Forward modelling of compaction-driven fluid flow and overpressure Using the results of tectono-stratigraphic forward modelling presented in the first part of this chapter, we have analysed the history of compaction-driven fluid flow and overpressures in the southern part of the Pannonian Basin system. Special emphasis is given to the Pliocene-Quaternary evolution. During this time interval the
subsidence in the central part of the basin system accelerates, causing sedimentation rates to increase. Sedimentation rates exert a first-order control on compaction driven overpressures (Bredehoeft & Hanshaw 1968; Bethke & Corbet 1988). Therefore, an increase in the amount of overpressure is expected to have occurred during the late stage evolution of the Pannonian Basin (Van Balen & Cloetingh 1994). Fluid overpressures
High overpressures are known to occur in the deep buried marly sequences in the Mako Trough and B6k6s Basin (Szalay 1982, 1988; Spencer et al. 1994). Generally, hydrostatic fluid pressures are present down to depths of about 1800 m in the Mako and B6k6s sub-basins (Szalay 1982, 1988; Clayton et al. 1990; Spencer et al. 1994). Hydrostatic pressures also occur in sediments overlying the basement highs flanking the sub-basins. Only few pressure measurements were made in the B6k6s Basin (Spencer et al. 1994). The amount of overpressuring measured at the northeastern flank of the B6k6s Basin is up to 15 MPa. At the western flank of the Mako Trough, in the Mako-I well, the maximum measured overpressure is around 45 MPa (Szalay 1982; Spencer et al. 1994). Overpressured oil flowed into this well during drilling
404
R.T. VAN BALEN E T A L .
(Horv~th et al. 1994). Therefore, part of this overpressure might be due to buoyancy forces. The overpressures measured by drill-stem tests in the H6d-I well in the centre of the Mako Trough are 18 MPa at 3.8 km, 25 MPa between 4.7 and 4.8 km and 40 MPa at 5.0 km (Szalay 1982). The inferred overpressures from the depth-porosity trend of pelitic sediments in the same well also shows a steadily increasing overpressure starting at a depth of about 2.5 km to a value of over 30 MPa at 4.8 km (Szalay 1988). Several overpressure mechanisms for the Mako Trough and B6k6s Basin have been proposed: undercompaction, hydrocarbon generation, aquathermal expansion due to the high heatflow and CO2 generation caused by thermal decomposition of carbonatic rocks in the basement (Spencer et al. 1994). Undercompaction is probably the prime mechanism for shallow overpressures. As overpressure increases dramatically at depths with temperatures of about 125~ hydrocarbon generation has been inferred to be a major contributor to overpressure at larger depths (Spencer et al. 1994). However, as shown by Szalay (1982, 1988), an almost perfect equilibrium relationship exists between porosity and effective pressure, implying that overpressure increase was contemporaneous with porosity decrease and denoting mechanical compaction as the prime overpressure causing mechanism. The overpressures have probably contributed considerably to the migration of hydrocarbons. The major Neogene source rock, the Miocene marl of the DB unit, has a very low permeability and, probably, high capillary pressure, which makes buoyancy an unlikely hydrocarbon migration driving force. Instead, the high overpressure occurring in the same beds has probably promoted migration in lateral and vertical directions, including downward (Spencer et al. 1994). The maximum overpressures, occurring at depths in excess of 3 km, measured in the northwestern part of the Drava Trough range from about 15 MPa in Molve field to about 10 MPa in the Stari Gradac field, promoting a secondary migration of hydrocarbons from WNW to ESE direction. Hydrostatic pressures occur down to a depth of 2 km (Barid et al. 1991).
depth versus porosity E
ECU v
"(30O
+ = sandstone z~ = marl J
(a)
.
,
J
L
]
.
,
L
,
,
,
,
,
L
,_J_•
J
9
i
10 20 porosity (%)
permeability versus porosity
-"o
7-
/
Q'u~
o Modelling parameters
For modelling purposes the lithology in the cross-section has been simplified to a mixture of idealized 'marl' and 'sandstone' types of sediment, based on the sedimentological description of the Hod-I well by Mattick et aL (1988; see Table 1). The porosity versus depth trends for
'i
(b)
~,-t- - sandstone - marl 0 2 3 log hor. permeability x 10^-16 m^2
Fig. 10. (a) Porosity versus depth trends for marls and sandstones in the southern part of the Pannonian Basin (after Szalay 1982; Spencer et aL 1994). (b) The relationship between porosity and horizontal permeability (after Szalay 1982).
MODELLING IN THE PANNONIAN BASIN
405
Table 3. Porosity and permeability constants' First estimate
qb0 bp ak bk kx/k z
Calibrated
Marl
Sandstone
Marl
Sandstone
0.65 8.82 • 10-8 16 10.5 11.7
0.475 4.5 • 10~8 16 18 1.62
0.65 8.82 • 10-8 20 8 11.7
0.475 4.5 • 10-8 18 8
these two lithologies (Szalay 1982; Spencer et al. 1994) are shown in Fig. 10a. The porosity-depth data are translated to porosity-effective pressure data and fitted to an exponential function required for the forward modelling (Van Balen & Cloetingh 1993, 1994): qb = qb0 exp-bp Pe with (b = porosity, 60 = surface porosity, Pe = effective pressure, bp = compaction coefficient. The deep porosities for marls (>3 kin) are not used, because these porosities are abnormally high due to fluid overpressures and, possibly, secondary porosity (A. Szalay pers. comm. 1993). The relationship between porosity and permeability depicted in Fig. 10b is reconstructed from permeability-depth data and the porosity-depth data for the two lithologies given by Szalay (1982) and Spencer et aL (1994). These data are fitted to functions given by: logk = -ak + bkqb with ak, bk = empirical constant, + = porosity, k = permeability. The permeability anisotropy is accounted for by adopting a vertical to horizontal permeability ratio. Porosity and permeability vary strongly in the subsurface. Thin interlayered shale beds control to a large extent the vertical permeability in a rock sequence (BjOrlykke 1993). Such beds may not have been sampled for permeability measurements. Furthermore, fractures were probably not taken into account in the laboratory measurements and presumably porosity and permeability were assessed at atmospheric pressure and surface temperature conditions. Therefore, we consider the derived porosity and permeability functions provided by the data as a first estimate. During the modelling process, the functions are further calibrated and the sand to marl ratios are adjusted until the predicted overpressures match the observations. The first estimate and finally obtained values are shown in Table 3. Essentially, the permeability values
1.62
determined by calibration show much lower values than those derived from the data. The minimum permeability for marl, for example, differs by four orders of magnitude (10 -16 v. 10- 2o m2). However, the minimum permeability value for marl and the values determined for sandstone are in keeping with values for shales given by Freeze & Cherry (1979), Bethke (1985) and Harrison & Summa (1991). The porosity-effective pressure functions did not require calibration. Adopted values for viscosity, compressibility and density of pore fluid and densities of sediment grains are given in Table 4. M o d e l l i n g results
The results of modelling of compaction-driven fluid flow and overpressures are shown for five different time slices (Fig. 11). These time slices are related to important changes in the basin evolution in the southern part of the Pannonian Basin system. The overpressures and flow directions at 6.3 Ma are depicted in Fig. l l a . Up to this stage, sedimentation rates in the sub-basins have been relatively low. Therefore, the overpressures, mainly occurring in the marls of the DB units, are subdued. The maximum overpressure is about 9.5 MPa (B6k6s Basin). The predicted compaction-driven fluid velocity field, shown in the same figure, demonstrates that the pore fluids are mainly expelled in lateral directions. The upward flow directions occur mainly in the upper 0.5 km. The maximum flow velocity is about 5.1 mm a -1. In order to show the large lateral component of fluid fluxes, the velocity fields depicted in Fig. 11 are not scaled. Therefore, true velocity directions are shown in these Table 4. Additional fluid modelling parameters" Pore fluid visocisity Pore fluid compressibility Fluid density Grain density
5 • 10-4 Pa s 4.3 • 10-1~Pa -1 1.024 g cm -3 2.7 g cm ~
406
R . T . VAN B A L E N E T A L .
MODELLING IN THE PANNONIAN BASIN
407
Fig. 11. The history of compaction driven overpressures and fluid flows predicted by our forward modelling. Vertical exaggeration of the basin cross-section is about 25. The velocity fields have not been scaled. (a) Overpressures and flows just before the arrival of the Pannonian delta (6.3 Ma). (b) The increase in sedimentation rates during the delta progradation causes overpressure to increase considerably (5.5 Ma). (c) The compressive stress induced acceleration of subsidence in the basin center causes sedimentation rates to increase again, inducing even higher overpressures (3.9 Ma). (d) Overpressure and flows predicted for the present day situation. The predicted overpressures in the centre of the Mako trough match with the inferred overpressures (Fig. 12). See text for a discussion of the results.
figures. The basin cross-section is, however, scaled with a factor of about 25. If the velocity field would also be scaled, it would show only vertical flow directions. During deposition of the delta slope deposits (DS unit), sedimentation rates increase considerably. As a consequence, overpressures increase to a maximum value of 17.8 MPa in the DB unit (Fig. 11b) at the end of deposition of this unit. Compared to the previous time frame, fluid-flow velocities have decreased slightly. After deposition of the A P unit, at the peak of compressional intraplate stress, overpressures have reached a maximum value of 35.7 MPa, due to the stress-induced increase of sedimentation rates (Fig. 11c). Maximum compaction-driven flow velocities are 2.3 mm a -1. Our prediction for the present day compaction driven overpressures and flows is shown in Fig. 11d. Maximum overpressure, occurring in the deep basin marls of the Mako Trough, is around 40.9 MPa. Flow velocities have decreased slightly to a maximum value of 2.2 mm a -1. The predicted overpressures at the centre of the Mako Trough are in reasonable agreement with measured and inferred overpressures in the Hod-I well (Fig. 12). In general, the overpressures are maximal in the DB unit, which is caused by the combination of low permeability and high porosity at deposition. The overpressures
decrease upwards and downwards from the maximum overpressure zone. The overpressure decrease with depth is high enough to promote downward migration of hydrocarbons into the permeable basal unit and basement (Mako Trough: 11 MPa; B6k6s Basin, 8.5 MPa). Additionally, our modelling predicts slightly less present-day overpressuring in the B6k6s Basin than in the Mako Trough.
Discussion on forward overpressure modelling results During the increase of the level of compressive intraplate stress the southwestern margin of the M a k o Trough was uplifted. This may have caused degassing of pore fluids due to pressure release. Furthermore, the uplift increased the gravity potential of meteoric waters which could, therefore, p e n e t r a t e deeper into the basin. Isotopical, geochemical and diagenetic data show that meteoric water has penetrated into the PD unit around the basement highs of the Derecske Basin (M~ity~s & Matter 1998), which was also uplifted during the P l i o c e n e - Q u a t e r n a r y (Vakarcs et al. 1994; Horv~th & Cloetingh 1996). The tectonic uplift of these highs combined with a preceding lakelevel fall caused an increase of the gravity potential enhancing the influx of these waters.
408
R.T. VAN BALEN E T A L . overpressure ( MPa ) 0
10
20
30
40
0
mud weight data
Trough and B6k6s Basin (PusztafOldvfir field; Dank 1988; Fig. 6). Therefore, the pore fluids flow towards the hydrocarbon pools. Thus, the compaction-driven fluid flow may have contributed to the secondary migration of hydrocarbons from the deeper parts of the sub-basins towards the highs. Our modelling results suggest therefore that, in addition to known hydrocarbon occurrences, more fields can be found on the eastern flank of the B6k6s Basin.
t-,-
Artificial neural n e t w o r k analyses o f overpressure data porosity
estimated X ~ ~ , , ~ , ~ . ~
Fig. 12. Overpressure profile in the centre of the Mako trough (Hod-I well) inferred from porosity trends and drill-stem tests (crosses) (after Szalay 1982) and the predicted present-day overpressure at the same position by our forward modelling. The predicted overpressure is in reasonable agreement with the inferred and measured overpressure.
The hydrocarbon generation started between 9 and 6 Ma (Horv~ith et al. 1987; Szalay 1988), implying that during the stress-induced change of the fluid regime the hydrocarbons are mature and actively expelled in the southern part of the Pannonian Basin. As the marl sequences are the major source rock for hydrocarbons in the Neogene sediments and the overpressure maximum occurs exactly in this unit, the increase of overpressure has probably enhanced the primary migration considerably. Our modelling also shows that the overpressures decrease with depth in the deepest part of the basin, promoting a downward migration of hydrocarbons, enhancing the prospectivity of the deepest unit. The predicted compaction-driven fluid velocity field shows that the pore fluids are expelled from the deep sub-basins mainly in lateral directions towards the basement highs. The largest velocities occur in the delta plain unit at the southwestern margin of the Mako Trough and the northeastern margin of the Bdkds Basin. The major hydrocarbon occurrences in the southeastern part of the Pannonian Basin system are located at the southwestern margin of the Mako Trough (AlgyO field) and the high inbetween the Mako
In order to compare the predicted present day overpressure to known overpressures in the area of interest, we have performed an analyses of overpressure data, based on neural network technology (Van Balen & Cloetingh 1995). Artificial neural networks are software tools inspired by brain models, which are used for pattern recognition purposes (Sejnowski et al. 1988). They are capable of learning, i.e. they can find relationships in data. Artificial neural networks learn by 'investigating' a set of training data. Once the relationship in these data has been found they can interpolate and extrapolate to new values. For data analyses purposes, artificial neural networks provide an alternative to statistical regression methods because they can find functional relationships in data sets without any a priori knowledge about their form (linear, exponential, etc.). We applied a neural network to mapping, interpolation and extrapolation of fluid overpressures in the southeastern part of the Pannonian Basin. The trained networks are used to assess overpressures along the crosssection depicted in Fig. 13a. The locations of the wells from which the overpressure data are used are shown in Fig. 13b. The number of input variables for the analysis is determined by parameters influencing the fluid overpressure in sedimentary basins in general and the Pannonian Basin in particular. We have defined five such parameters: burial depth, composition, thickness of overlying Quaternary deposits and geographical distances (2 parameters, plane coordinates). In a separate test, all five parameters were found to be significantly influencing the results. The data set for the network analysis is compiled from different sources. Fluid overpressure data for several wells are reported in Szalay (1982, 1988) and Clayton et al. (1990). The wells used in this study are indicated in Fig. 13b. In this part of the Pannonian Basin there is only one well which reaches a depth of 5.4 km: Hod-I
MODELLING IN THE PANNONIAN BASIN
409
Fig. 13. (a) The Pannonian basin and its surroundings (after Horvfith 1993). The inset shows the position of (b). A-A' denotes the modelled cross-section. (b) The locations of the wells used in this analyses plotted on the isopach map. X and Y are arbitrary geographical coordinate axes. For scale (b) located in the centre of the Mako trough. All the other wells are much shallower, typical depths are 2-3 km. As can be seen in Fig. 13b, the deep part of the B6k6s Basin is not penetrated by a well. For this sub-basin overpressure information is only available for the outer, shallower part. As will be shown below, neural network
analysis enables extrapolation down to a depth of 6 km in this sub-basin. Sediment compositions (the percentage of sandstone in the deposits) for the lithostratigraphic units in the southeastern part of the Pannonian Basin are based on Mattick et al. (1988). The distribution, thickness and depth of these
410
R.T. VAN BALEN ETAL.
lithostratigraphic units are given by Juhfisz (1991). The thickness and distribution of the Quaternary deposits are presented by R6nai (1974). Finally, the geographical coordinate axes we have defined are indicated in Fig. 13b. Using these sources, we have assembled a database with 193 records. During training of the network 150 patterns (data records) are used, while the remaining 43 patterns are used to compare the predicted and known overpressures during the testing phase. A neural network with a topology consisting of one hidden layer with three neurons is found to be the optimal configuration (Van Balen & Cloetingh 1995). A comparison between overpressure profile predicted by the neural network analyses and the forward modelling is shown in Fig. 14. The results are in fairly good agreement. We, therefore, conclude that results from both methods confirm each other. Discussion
As shown in both numerical forward modelling and neural network analyses, the overpressure decreases at the deepest parts of the sub-basins, in the basal sandstone unit. The amount of overpressure is, however, not equal in the sub-basins. Both methods predict an overpressure in the B6k6s Basin which is slightly less than the overpressure in the Mako Trough. This can be explained by spatial variations in the thickness of the Quaternary deposits. As the Quaternary is thicker in the Mako Trough, sedimentation rates during Quaternary above this sub-basin were higher than in the B6k6s Basin, inducing faster compaction. The anomalous thickness of the Quaternary deposits can be explained by compression induced lithospheric deflection since early Pliocene. Due to the distribution of vertical lithospheric loads, the maximum of stress induced subsidence is located exactly above the Mako Trough.
Conclusions On basis of this modelling study several important conclusions can be made regarding palaeowater depths, lake-level fluctuation, lithospheric necking depth during extension, intraplate stress history, extension factors, timing of overpressures and the directions of fluid flow. These are presented below. The tectono-stratigraphic forward model invoking a lithospheric necking depth of 7.5 km is best in accordance with earthquake, gravity and lithospheric data, and requires reasonable
lake-level variations, palaeo-water depths and stretching values. The relatively shallow necking depth in the Pannonian Basin probably reflects the pre-rift weak bulk rheology of the lithosphere due to Cretaceous and Palaeogene crustal thickening. The stratigraphic forward modelling requires a palaeo-water depth of 2300 m during deposition of the deep basin marls in the centre of the Mako Trough. Although this value is large, it is in keeping with observations. A lake-level fall can explain the truncation and thickness distribution of the delta plain (DP) depositional unit. The truncation of Pliocene alluvial plain sediments (AP unit) can not be explained in terms of lake-level changes This agrees with the fact that during this time interval the Pannonian lake became extinct. Therefore, lake-level changes could not affect the stratigraphy anymore. Palaeo-stress analyses, borehole break-out data, focal mechanism solutions and in-situ stress measurements indicate that the Pannonian Basin is in a compressive state of stress since early Pliocene. Our modelling results show that an increase in the level of compressive stress during Pliocene followed by a slight decrease during Quaternary can successfully explain the observed truncation patterns and thickness distribution of the alluvial plain (AP) and Quaternary (Q) depositional units. The compressive stress induces differential movements across the basin system, consisting of uplift at the basin system flanks and subsidence in the sub-basins. Forward modelling of compaction-driven fluid flow and overpressures in the Pannonian Basin shows that overpressures have increased twice due to accelerations of sedimentation rates. The first overpressure increase event occurs during deposition of the delta slope deposits (DS unit) between 6.3 and 5.5 Ma. Starting in the early Pliocene (3.9 Ma), an increase in the level of compressive stress causes subsidence and sedimentation rates to increase again, which enhances overpressuring substantially. This has several implications for the diagenesis of sediments and migration of hydrocarbons. G. Spadini and P. Szafian are thanked for stimulating discussions. G. Vakarcs and an anonymous reviewer provided useful reviews of the manuscript. This research was funded by the Hungarian Research Fund (OTKA) and IBS (Integrated Basin Studies) project, part of the JOULE II research programme funded by the Commission of European Communities (contract no. JOU2-CT 92-0110). Publication No. 970110 of the Netherlands Research School of Sedimentary Geology.
M O D E L L I N G IN T H E P A N N O N I A N BASIN
411
Fig. 14. (a) Overpressure through the Mako Trough and B6kds Basin produced by the 5 • 3 • 1 neural network. Thick lines are contour lines for the overpressures (MPa), thin lines represent the stratigraphy. A n overpressure reversal occurs in the deepest part of the sub-basins. Generally, the overpressures in the Mako Trough are less than in the B6k6s Basin. (b) Overpressure profile predicted by numerical forward modelling (close up of Fig. 1 ld), showing similarities to the overpressure profiles obtained by the neural network analyses. Thick lines are contourlines for the overpressure (MPa), thin lines represent the modelled stratigraphy. The modelled overpressure profile also shows the overpressure reversal and a slightly less overpressure in the B6k6s Basin inferred from the neural network analyses.
References BADA, G., GERNER, P., SZAFIAN, P., CLOETINGH, S. & HORVATH, E 1995. Sources of recent tectonic stress in the Pannonian region derived from finite element modelling. Terra Abstracts (EUG 8), supplement to Terra Nova, 7, 173. BARIC, G., MESIC, M., JUNGWIRTH, I., & SPANIC, D. 1991. Gas and condensate-fields in the north-west of the Drava depression, Yugoslavia. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons I. European Association of Petroleum Geologists, Special Publications, 1. Oxford University Press, Oxford, 323-340. BECKER, A. 1993. Contemporary state of stress and neotectonic deformation in the Carpathian-Pannonian region. Terra Nova, 5, 375-388.
BI~RCZI, I. 1988. Preliminary sedimentological investigation of a N e o g e n e depression in the Great Hungarian Plain. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin. American Association of Petroleum Geologists Memoir, 45, 107-116. -& PrtILLIPS, R. L. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, southeast Hungary. Geophysical Transactions, 31, 55-74. - - - , HAMOR, H., JAMBOR, A. & SZENTGYORGYI, K. 1988. Neogene sedimentation in Hungary. In: ROYDEN, L. H. & HORVATH,E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoir, 45, 27-48. BETHKE, C. M. 1985. A numerical model of compaction-driven groundwater flow and heat transfer and its application to the paleohydrology of
R.T. VAN BALEN ET AL.
412
intracratonic sedimentary basins. Journal of Geo90, 6817-6828. - & SORBET, T. E 1988. Linear and nonlinear solutions for one-dimensional compaction flow in sedimentary basins. Water Resources Research, 24, 461-467. BJORLYKKE, K. 1993. Fluid flow in sedimentary basins. Sedimentary Geology, 26, 137-158. BREDEHOEFr, J. D. & HANSHAW, B. B. 1968. On the maintenance of anomalous fluid pressures I, Thick sedimentary sequences. Geological Society of America Bulletin, 79, 1097-1106. CLAYTON,J. L., SPENCER, C. W., KONCS, I. & SZALAY,A. 1990. Origin migration of hydrocarbon gases and carbon dioxide, B6k6s Basin, southeastern Hungary. Organic Geochemistry, 15, 233-247. CLOETINGH, S., MCQUEEN, H. & LAMBECK, K. 1985. On a tectonic mechanism for regional sea level variations. Earth and Planetary Science Letters, 75, 157-166. , VAN WEES, J. D.,VAN DER BEEK, R A. & SPADINI, G. 1995. Role of pre-rift Theology in kinematics of extensional basin formation: constraints from thermomechanical models of Mediterranean and intracratonic basins. Marine and Petroleum Geology, 14, 793-807. CSAT0, I. 1993. Neogene sequences in the Pannonian Basin, Hungary. Tectonophysics, 226, 377-400. Csoyros, L., NAGYMAROSY,A., HORVATH,E & KOVAC, M. 1992. Tertiary evolution of the IntraCarpathian area: a model. Tectonophysics, 208, 221-241. , TARI, G., BERGERAT,E & FODOR, L. 1991. Evolution of the stress fields in the Carpatho-Pannonian area during the Neogene. Tectonophysics, 199, 73-91. DANK, V. 1988. Petroleum geology of the Pannonian Basin, Hungary: an overview. In: ROYDEN,L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoir, 45, 319-331. FREEZE, J. A. & CHERRY, J. A. 1979. Groundwater. Prentice Hall, Englewood Cliff, New Jersey. GABRIS, G. 1994. Pleistocene evolution of the Danube in the Carpathian Basin. Terra Nova, 6, 495-501. GAUDEMER, Y., JAUPART, C. 8,: TAPPONNIER, P. 1988. Thermal control on post-orogenic extension in collision belts. Earth and Planetary Science Letters, 89, 48-62, GREGERSON. S. 1992. Crustal stress regime in Fennoscandia from focal mechanism. Journal of Geophysical Research, 97, 11 821-11 827. GROW, J. A., MATTICK,R. E., BERCZI-MAKK, A., PERO, C., HAJD(J, D., POGASCAS, G., V,~,RNAI, P. & VARGA, g. 1994. Structure of the B6k6s Basin inferred from seismic reflection, well and gravity data. In: TELEKI, P. G., MATTICK, R. E. & KOKAI, J. (eds) Basin analysis is in petroleum explor-
physical Research,
ation, a case study from the Bikes Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 1-38. GR13NTHAL, G. & STROMEYER, D. 1982. The recent crustal stress field in Central Europe: trajectories
and finite element modeling. Journal of Geophysical Research, 97, 11 805-11 820. HARRISON,W. J. & SUMMA,L. L. 1991. Paleohydrology of the Gulf of Mexico Basin. American Journal of Science, 291, 109-176. HAQ, B. U., HARDENBOL, J. & VAIL, P. 1987. Chronology of fluctuating sea levels since the Triassic. Science, 235, 1156-1167. HORVATn, E 1985. Thickness of Neogene-Quaternary basin fill (map). Hungarian Geological Institute, Budapest, Hungary. - 1993. Towards a kinematic model for the Pannonian Basin. Tectonophysics, 226, 333-357. & BERCKHEMER, H. 1982. Mediterranean backarc basins. In: BERCKnEMER, H. & HSO, K. (eds) Alpine-Mediterranean geodynamics. Geodynamic Series 7. American Geophysical Union and Geological Society of America, 141-173. -& CLOETINGH,S. 1996. Stress-induced late-stage subsidence anomalies in the PannonianBasin. Tectonophysics, 266, 287-300. - - , DOVENYI, P., PAPP,S., SZALAY,A. & HARGITAY,M. 1987. Tectonic control on migration and trapping of hydrocarbons in the Pannonian Basin. In: DOLIOEZ, B. (ed.) Migration of hydrocarbons in sedimentary basins. Editions Technip, Paris, 667-681. --, SZALAY, A. & ROYDEN, L. 1988. Subsidence, thermal, and maturation history of the Great Hungarian Plain. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 355-372. JAMBOR,/~. 1989. Review of the geology of the s.1. Pannonian formations of Hungary. Acta Geologica Hungarica, 32, 269-324. Joo, I., CZOBOR,A., GAZSO, M. & NEMETH,Z. 1990. On recent crustal movements in the Pannonian Basin.
Acta Geodynamica, Geophysical et Montanista Hungarica, 25, 231-242. JUnASZ, G. 1991. Lithostratigraphical and sedimentological framework of the Pannonian (s.l.) sedimentary sequence in the Hungarian Plain (Alf61d), Eastern Hungary. Acta Geologica Hungarica, 34, 53-72. KAZMER, M. 199(I. Birth, life and death of the Pannonian Lake. Palaeogeography, Palaeoclimatology, Palaeoecology, 79, 171-188. KOOl, H. & CLOETINGH,S. 1992. Lithospheric necking and regional isostasy at extensional basins 2. Stress induced vertical motions and relative sea level changes. Journal of Geophysical Research, 97, 17 573-17 591. KORPAs-HODI, M., POGA('SAS, G. & SIMON, E. 1992. Paleogeographic outlines of the Pannonian s.1. of the southern Danube-Tisza interfluve. Acta Geological Hungarica, 35, 145-163. LAZARESCU, V., CORNEA, I., RAOULESCU, E & POPESCU, M. 1983. Moho surface and recent crustal movements in Romania: geodynamic connections. Proceedings of the Institute of Geology and Geophysics in Bucarest, 63, 83-91. LORINCZ, K. & SZAB0, P. 1993. Seismic analysis of
MODELLING IN THE PANNONIAN BASIN multi-phase tectonism in the central part of the Pannonian Basin in Hungary. In: SPENCER,A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons IlL European Association of Petroleum Geologists, Special Publications, 1. Oxford University Press, Oxford, 311-323. MATYAS, J. & MATTER, A. 1998. Diagenetic indicators of meteoric flow in the Pannonian Basin, SE Hungary. Marine and Petroleum Geology, in press. MATTICK, R. E., PHILIPS, R. L. & RUMPLER, J. 1988. Seismic stratigraphy and depositional framework of sedimentary rocks in the Pannonian Basin, southeastern Hungary. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 117-146. - - , RUMPLER,J., UJFALUSY,a . , SZANYI,B. & NAGY,I. 1994. Sequence stratigraphy of the B6k6s Basion. In: TELEKI,P. G., MATTICK,R. E. & KOKAt, J. (eds) Basin analysis in petroleum exploration, a case study from the B~k~s Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 39~55. MCKENZIE, D. R 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters, 40, 25-31. MOLLER, B., ZOBACK, M. L., FUCHS, K., MASTIN, L., GREGERSEN, S., PAVONI, N., STEPHANSON, 0. & LJUNGGREN, C. 1992. Regional patterns of tectonic stress in Europe. Journal of Geophysical Research, 97, 11 783-11 803. NAGYMAROSY,A. & Mf)LLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian Basin. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists, Memoirs, 45, 58-68. POGACSAS, G., MATTICK,R. E., ELSTON,D. P., HAMOR, T., JAMBOR, A., LAKATOS,L., LANTOS,M., SIMON, E.,VAKARCS,G.,VARKONYI,L. & VARNAI,R 1994. Correlation of seismo- and magnetostratigraphy in southeastern Hungary. In: TELEKI, P. G., MATTICK,R. E. & KOKAI,J. (eds) Basin analysis" in petroleum exploration, a case study from the BOkOs Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 143-160. POLOMC, G. 1985. Neotectonic activity at the eastern border of the Pannonian depression and its seismic implications. Tectonophysics, 117, 109-115. POSGAY,K. T., BODOKY,T., HEGEDOS, E., KOVACSVOLGYI, S., LENKEY,L., SZAFIAN,P., TAKACS,E., IMAR, Z. & VARGA, G. 1996. Asthenospheric structure beneath a Neogene basin in SE Hungary. Tectonophysics, in press. RATSCHBACHER,L., FRISCH,W., LINZER, H. & MERLE, O. 1991. Lateral extrusion in the Eastern Alps, part 2: structural analyses. Tectonics, 10, 257-271. RONAI, A. 1974. Size of Quaternary movements in Hungary's area. Acta Geologica Academiae Scientiarum Hungaricae, 18, 39-44. 1985. Limnic and terrestrial sedimentation and the N/Q boundary in the Pannonian Basin. In:
-
-
413
KRETZOI, M. & PECSJ, M. (eds) Problems of the Neogene and Quaternary. Akad6miai Kiad6, Budapest, 21-49. ROYDEN, L. H. & DOVI~NYI, P. 1988. Variations in extensional styles at depth across the Pannonian Basin. In: ROYDEN,L. H. & HORVATH,E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 235-255. & KEEN, C. E. 1980. Rifting processes and thermal evolution of the continental margin of eastern Canada determined from subsidence curves. Earth and Planetary Science Letters, 51, 343-361. --, HORVATH, E • NAGYMAROSY,A. 1983. Evolution of the Pannonian Basin system 2: subsidence and thermal history. Tectonics, 2, 91-137. RUMPLER,J. & HORVATH,E 1988. Some representative seismic reflection lines from the Pannonian Basin and their structural interpretation. In: ROYDEN,L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 153-171. SEJNOWSKI,T. J., KOCH, C. & CHURCHLAND,P. S. 1988. Computational neuroscience. Science, 241, 1299-1306. SPENCER, C. W., SZALAY,a . & TATAR, E. 1994. Abnormal pressure and hydrocarbon migration in the Bdk6s Basin. In: TELEKI, P. G., MATTICK, R. E. & K(SKAI,J. (eds) Basin analysis in petroleum exploration, a case study from the BdkOs Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 201-219. STEGENA, L., GECZu B. & HORVATH, E 1975. Late Cenozoic evolution of the Pannonian Basin. Tectonophysics, 26, 71-90. SZALAY,A. 1982. A rekonstrukci6s szemldletii fOldtani kutatds lehetOsOgei a szOnhidrog~n-perspektiv6k elOrejelzds~ben [Possibilities of the reconstruction of basin evolution in the prediction of hydrocarbon prospects]. PhD thesis, Hungarian Academic of Sciences, Budapest. 1988. Maturation and migration of hydrocarbons, SE PannonianBasin. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 347-354. TARI, G., HORVATH, E & RUMPLER, J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. UJSZASZI, K. & VAKARCS, G. 1993. Sequence stratigraphic analysis in the south Transdanubian region, Hungary. Geophysical Transactions, 38, 69-87. VAKARCS, G., VAIL, P. R., TARI, G., POGACSAS, GY., MATTICK, R. E., SZABO, A. & BOWMAN, S. 1994. Third-order sequence Miocene-Pliocene sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106. VAN BALEN,R. & CLOETINGH,S. 1993. Stress-induced fluid flow in rifted basins. In: HORBURY,A. D. & ROBINSON, A. G. (eds) Diagenesis and Basin Development. A A P G Studies in Geology series, 36, 87-98. -
-
414 -
-
R.T. VAN B A L E N ETAL. & 1994. Tectonic control of the sedimentary record and stress induced fluid flow: constraints from basin modelling. In: PARNELL, J. (ed.)
Geofluids: origin, migration and evolution of fluids in sedimentary basins. Geological Society, London, Special Publications, 78, 9-26.
-
-
& -1995. Neural network analyses of stressinduced overpressures in the Pannonian Basin. Geophysical Journal International, 121, 532-544.
The present-day tectonics of the Aegean as deduced from seismicity DENIS HATZFELD L a b o r a t o i r e de G ~ o p h y s i q u e Interne et Tectonophysique, I R I G M , U J F - C N R S , B P 53, 38041 G r e n o b l e C e d e x 9, France (e-mail: D e n i s . H a t z f e l d @ o b s . ujf-grenoble.fr)
Abstract: We present a review of the seismicity and focal mechanisms for the Aegean (historical, teleseismically and microseismically computed), which we compare with the recent active faults. A very consistent trend in the T-axes within the Aegean across blocks of different size and orientation suggests that the surface tectonics of the Aegean is controlled by the deformation of the lithosphere as a whole. Some recently active faults (mostly the NW-SE-striking faults that are located west of the Aegean sea) are not associated with seismicity, some seismicity is associated with new E-W-striking faults of modest displacement that are mostly seen in continental Greece. We suggest that, due to the important amount of internal deformation of the Aegean lithosphere and the inferred rotations, the localization of the deformation changes with time and a precise timing of the recent faults activity is necessary to correlate with present-day deformation. We also suggest that, for continental crust, the work to create new faults that strike perpendicular to the main strain direction (as deduced from focal mechanisms and GPS measurements) is less than the work to reactivate faults that strike obliquely to the strain direction. We propose that a diffuse system of normal faults or grabens starts to develop, parallel to the Gulf of Corinth, and links the western termination of the North Aegean Trough to the northern end of the Kefallinia fault. In continental regions, deformation is distributed over wide areas and, if the amount of deformation is significant, the location of the active surface features changes with time. We observe block rotations and creation or death of active faults while there is no change in the boundary conditions (McKenzie & Jackson 1983; Roberts & Jackson 1991). The picture of the active faults is, therefore, strongly dependent upon the considered age of the structure (Mercier et al. 1987). Some faults seem to be 'locked', others seem to slip aseismically, whereas earthquakes occur in places that do not seem related to active faults. Because tectonics, palaeomagnetism and seismology may be related to different periods of time, contradictions occur that may not be due to the different techniques considered but to the dating of the event. For these rapidly deforming regions, it is necessary to ensure that the presentday tectonics is described by observations related to present-day active structures. The A e g e a n is a region of fast and intense deformation whose dimensions are only several h u n d r e d kilometres wide (Jackson 1994). Apparent contradictions are observed between active faults (Jackson 1994; Jolivet et al. 1994; Armijo et al. 1996) and seismicity (Jackson & McKenzie 1988). A n d different geodynamic models have been suggested (e.g. McKenzie 1978; Taymaz et al. 1991; Le Pichon et al. 1995; Armijo et al. 1996) that do not agree with each other. In this paper, we summarize some of the seismological observations that have been
already published elsewhere (e.g. Hatzfeld et al. 1997) and assume they constrain both the present active tectonics and the actual strain pattern. We compare the seismicity with recent faults inferred from surface observations and propose a rough sketch of the principal presentday active faults.
Geodynamic background T h e A e g e a n region is located b e t w e e n the African and European lithospheric plates that converge at a rate of about 1.5 cm a -1 (Fig. 1). T h e motion across the Hellenic trench is, however, of about 5 cm a -1, probably due to the westward motion of Turkey and the southwestward motion of the Aegean (McKenzie 1978; Jackson 1994; Le Pichon et al. 1995). This motion is related to the active lithospheric subduction beneath the Hellenic Trench (Papazachos 1973), but it does not describe completely the tectonics of the Aegean, which experiences also internal deformation as evidenced by geological observations (Angelier 1979), geodetic displacements (Veis et al. 1992), focal mechanisms (e.g. Hatzfeld et al. t993) and paleomagnetic rotations (Kissel & Laj 1988). Shortening is mostly observed along the external part of the Hellenic arc from Epirus to Rhodos, generally oriented perpendicular to the active boundary. Extension, trending N-S, affects the internal part of the Aegean, but trends differently along the Hellenic arc. This
HATZFELD,D. 1999. The present-day tectonics of the Aegean as deduced from seismicity. In: DURAND,B., JOLIVET,L., HORVATH,E & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 415-426.
416
D. HATZFELD
Fig. 1. Sketch of the main active boundaries and NEIC-seismicity (mb > 4.5) until 1996. Principal places described in the text: A, Amvrahikos; C, Corinth; E, Evia; Ka, Kavala; Kr, Kremasta; M, Mygdonian; S, Saros; Te, Thessaly; Tr, Trikhonis; V, Volos; The seismicity defines four branches for the North Aegean Trough that continue toward the Kefallinia zone. The Saros, Thessaly and Evia basins have low seismicity.
extension affecting the Aegean probably started 15 Ma ago (Mercier 1977; Jolivet et al. 1994) or even 22 Ma ago (Gautier & Brun 1994), but the relation with the westward motion of Turkey is unclear because the motion along the western termination of the North Anatolian fault started probably only 5 Ma ago (Jolivet et al. 1994; Le Pichon et al. 1995; Armijo et al. 1996). The most prominent faults, more recent than Miocene time, have been summarized by Mercier et al. (1976), Angelier (1979), Lyberis (1984), Pavlides & Mountrakis (1987), Mascle & Martin (1990), Roberts & Jackson (1991), Jolivet et al. (1994), Armijo et al. (1996). Faults are mostly normal within Central Greece, Pelo-
ponnese and on islands of the Aegean sea, and dextral strike slip in the North Aegean sea. But the details in the sketch of the active faults differ from one author to another, especially for Central Greece and the North Aegean sea. In this paper we try to identify active faults by comparing tectonic information with seismicity and infer a present-day tectonic map.
Data Seismicity
Shallow earthquakes are related to fractures that occur within the brittle part of the crust. In
TECTONICS & SEISMICITY IN THE AEGEAN order to bring relevant information to geodynamical problems, however, the seismicity maps should follow some basic rules. The location of the earthquakes should be mapped with an accuracy that is comparable to the scale of the considered tectonic problem and seismicity should be representative of long term deformation at a scale that is comparable to dimensions of large tectonic faults. Earthquakes of magnitude greater than 5.5 are usually recorded at teleseismic distances (greater than 30~ in a spherical Earth with an accuracy strongly dependent on the velocity structure at the focus of the earthquake and on the number and the azimuthal coverage of the recording stations. We used the NEIC catalogue that is probably complete for magnitude greater than 3.5 since 1963. Earthquakes of moderate magnitude (above 3.0) are recorded in regional seismological networks. The epicentral distances are usually smaller than 30 ~ which is the range where refracted waves are observed with weak onsets. They can provide good locations but generally a poor control in the depth of the earthquakes. Temporary experiments conducted with dense networks usually provide reliable information but for small magnitude earthquakes that occur during a short period of time and which might be not representative of the long term deformation. During the years 1984-1993, we installed several times, in different regions, temporary networks of portable seismological stations to locate the seismicity more precisely (especially the depth of the earthquakes) and compute more focal mechanisms. The duration of the experiments was generally 6-8 weeks and we installed 30-100 stations over regions that were approximately 150 • 150 km 2 in area. Because the instrumental seismicity covers a period of time that is short compared to loading processes of faults, it could be not representative of tectonic activity. We therefore also examined the historical record of seismicity of the Aegean (Papazachos & Papazachou 1997), which is incomplete, but provides complementary information to the recent measurements. Focal mechanisms
Fault plane solutions are also computed with different methods and different datasets. Mechanisms computed at teleseismic distances are generally good for events of magnitude larger than 5.5, when the focal sphere is well sampled. Mechanisms computed with first motion polarities at regional distances are questionable because, due to refracted waves,
417
they sample only a limited part of the focal sphere. Mechanisms computed for microearthquakes are well controlled when enough polarities are available (more than 8-12 readings sampling a minimum of three quadrants). If the earthquake is well recorded in several long period stations, it is possible to compare the observed seismograms with synthetics that are computed for body waves at teleseismic distances. This method, which gives a good control on the parameters of the source properties (depth, focal mechanism, source function), is possible only for magnitudes greater than 5.5 and therefore does not provide many solutions in regions of moderate seismicity. Centroid moment tensor solutions are routinely performed automatically by Harvard (Dziewonski et al. 1988) for earthquakes of magnitude above 5.0. As in any automatic procedure, some unreliable solutions can be computed accidentally.
Shallow seismicity In this section we will examine three different data sets that are complementary. (1) The teleseismically located seismicity by NEIC shows a complex pattern (Fig. 1). Most of the seismicity is observed along the active boundary of western Greece and the Hellenic Trench but does not extend further south. Important seismicity is also observed in the North Aegean sea and is clearly associated with the different branches of the North Aegean trough. The sea of Crete is almost aseismic. In continental Greece, beside the active region located around the Gulf of Corinth, seismicity is spread over wide areas and not clearly limited to obvious surface faults. Most of the seismicity is located south of a line which joins Volos to Kefallinia. (2) The seismicity that we recorded during our temporary experiments (Fig. 2) is not a homogeneous picture of the seismic activity of the Aegean, but it is the juxtaposition of several detailed studies. However, it confirms some of the aspects of the teleseismically located seismicity. We confirm the concentration associated with the Hellenic Trench and the continental collision. The sea of Crete is almost free of earthquakes and we located only a few events around the Cyclades islands. In the Peloponnese (Hatzfeld et al. 1990) and in western Greece (Hatzfeld et al. 1995), we cannot easily associate earthquakes with individual surface faults. The seismicity is diffuse with depth ranging between 5 and 20 km. In a few places in Central Greece, however, as around the Gulf of Corinth
418
D. HATZFELD
Fig. 2. Seismicity map recorded during the microearthquake experiments that we conducted from 1984 to 1993. This is not an homogeneous picture of seismicity but complements Fig. 1.
(Hatzfeld et al. 1990, Rigo et al. 1996), around the Gulf of Volos and in Thessaly (Hatzfeld et al. 1997), or around the Mygdonian graben (Hatzfeld et al. 1987) seismicity seems to underline recent, sometimes not very well developed, normal fault systems that are striking E-W. The very i m p o r t a n t N W - S E normal faults that bound the Gulf of Saros, the Gulf of Evia, the basin of Thessaly, are not related to microseismic activity. (3) These main features (low seismicity in the Sea of Crete, around the Cyclades, the Gulf of Soras, the Gulf of Evia and Thessaly) are confirmed over a longer period of time by the historical seismicity (Papazachos & Papazachou 1997). Particularly, we do not observe earth-
quakes related to the Saros basin and Evia basin, whereas we observe earthquakes related to Volos or to the Kavalla fault (Fig. 3). It seems to us, therefore, that there is not a unique relationship between the faults that are generally mapped as recently active in Greece and seismicity. Some faults are seismically active and some are not. Some seismicity is located in regions where no recent important surface deformation is mapped. Focal mechanisms
The first integrated study of focal mechanisms in the Mediterranean was the pioneering work of McKenzie (1972, 1978), which established the
TECTONICS & SEISMICITY IN THE AEGEAN
419
Fig. 3. Historical seismicity from 550 Bc until AD 1900 (Papazachos & Papazachou 1997). Geographical names are as Fig. 1. Little seismicity is related to Evia, Thessaly and the Saros basin. Kavalla experienced strong events.
global kinematic framework of the Mediterranean. At that time, most of the fault plane solutions were computed using only first motion polarities read on long period records. In this paper, we will use three different sets of data that will help to indicate the strain pattern of the area. (1) We present 61 solutions computed using the body waves modelling technique from three papers (Taymaz et al. 1990, 1991; Baker et al. 1997) which cover the Aegean. They represent the most reliable mechanisms in this area. (2) We compare these solutions with 183 CMT solutions automatically computed by Harvard (Dziewonsky et al. 1988) since 1977.
(3) We use the 746 microearthquake mechanisms that we computed during our different experiments from 1984 to 1993 (Fig. 5). Again, comparing the source properties of magnitude 6 and magnitude 2 e a r t h q u a k e s needs some caution, and we will never consider a single microearthquake mechanism as representative of the strain pattern if it is not consistent with a nearby large event or with other mechanisms that belong to the same area. In order to visualize better the strain pattern as defined by the mechanisms, we display also the T-axes that represent the direction of lengthening and the P-axes that represent the direction of shortening.
420
D. HATZFELD
Fig. 4. Map of the 61 reliable mechanisms computed by body waves modelling (Baker et al. 1997; Taymaz et al. 1990, 1991) in black, and 183 CMT solutions published by Harvard in grey. Reverse faulting is restricted to a very narrow zone along the active boundary. Strike slip is associated with all the different branches of the North Aegean Trough and the Kefallinia fault. Normal faulting is present in most of the Aegean.
A first examination of the mechanisms, both for strong (Fig. 4) or small (Fig. 5) earthquakes clearly show that reverse faulting is observed along the Hellenic Trench, normal faulting within the A e g e a n , and dextral strike slip mechanisms are associated with the different branches of North Aegean Trough and with the Kefallinia Fault. Within the internal Aegean, the T-axes (Fig. 6) trend consistently N-S from Macedonia to the Gulf of Corinth, and from western Greece to Turkey. But, conversely, toward the active
boundary, from Albania to Rhodes, there is a progressive change in the direction of the Taxes. Actually, the T-axes trend parallel to the Hellenic arc, and are therefore different in orientation from those within the interior Aegean. This change in trend suggests that internal deformation affects the A e g e a n (Hatzfeld et al. 1993, 1997). The P-axes (Fig. 7) that are associated with reverse faulting along the Hellenic Trench trend consistently from Albania to South Peloponnese. There is no noticeable rotation of the
TECTONICS & SEISMICITY IN THE AEGEAN
421
Fig. 5. Map of the 746 mechanisms computed during the different microearthquake experiments between 1984 and 1993.
P-axes (or of the slip vectors) across the Kefallinia fault which represents the transition between continental collision and oceanic subduction (Kahle et al. 1993; Hatzfeld et al. 1995). It is therefore unlikely that western Greece has rotated for a long time around a pole that is located near the active boundary (Baker et al. 1997). Further east, between the Kythira strait and eastern Crete, there is a significant change in the trend of the P-axes to a more N-S direction. If we assume a rigid African plate, this change in the orientation of the P-axes for earthquakes located along the boundary implies either a pole of rotation located close to the active boundary (Le Pichon & Angelier 1979) or more likely, as
inferred from the T-axes trend, that the Aegean region is deforming. Discussion T h e strain p a t t e r n
The strain pattern deduced from focal mechanism resembles the strain pattern deduced from geodetic measurements (Veis et al. 1992) or deduced from microtectonic observations for the present time (Mercier etal. 1987). It is homogeneous over all the Aegean, across blocks of various orientation and dimensions and supports the idea that the forces that move the
422
D. HATZFELD
Fig. 6. Map of the T-axes that dip shallower than 45~. Thick arrows are for large earthquakes, thin arrow are for microearthquakes. The direction of lengthening is homogeneous in the North Aegean and rotates slightly clockwise from west to east within the Aegean. The T-axes trend along strike for the Hellenic arc.
crustal blocks are applied to their base by a lithosphere that deforms as a continuum ( Le Pichon & Angelier 1979; England et al. 1985; Jackson et al. 1992; Hatzfeld et al. 1997). This strain pattern is slightly different to the strain p a t t e r n deduced from microtectonic observations for the Pliocene time (Mercier et al. 1987), which implies either a rotation of the Aegean within a constant strain field or a rotation of the strain field. The tectonics
It seems that there is not a unique relationship between the seismicity and the most important
faults displayed by Jackson (1994), Jolivet et al. (1994) or Armijo etal. (1996). Some of the faults are indeed seismically active as the North A e g e a n Trough, the Gulf of Corinth, the Kefallinia fault, the southern Peloponnese and western Crete. These faults are either NE-SWstriking strike-faults, or normal faults of various direction. The Kavalla fault, which is also a E N E - W S W striking strike-slip fault (Lyberis 1984), experienced several very strong earthquakes during historical time, while instrumental seismicity is rather low. On the other hand, some of the faults recently active are not seismic, such as those which bound the western Saros basin, the North Evia basin,
TECTONICS & SEISMICITY IN THE AEGEAN
423
Fig, 7, Map of P-axes that dip shallower than 45~ Same symbols as Fig. 6. We note a constant trend along western Greece across the Kefallinia fault, and a progressive anticlockwise rotation toward Rhodos.
the Argos basin, or Thessaly. This is true for the historical seismicity (Fig. 3), for the NEIC-bulletins seismicity (Fig. 1) and for the microseismicity (Fig. 2). Finally, we note seismic activity in regions where no important active faults are mapped. This is true for western Greece, from Albania to Peloponnese, where little is known about active faults. This is also true for young faults of modest displacement in the Mygdonian graben (Mercier et al. 1983), in the Gulf of Volos (Caputo & Pavlides 1993), in Central Attiki (Armijo et al. 1996) or in Trikhonis lake and in the Amvrahikos Gulf (Brooks et al. 1988; Doutsos et al. 1987). These young faults are all pure normal faults striking roughly E - W and therefore
perpendicular to the strain pattern and they experienced strong e a r t h q u a k e s during the present century or in historical time.
Strength o f the continental crust and o f the
faults We therefore propose that the present-day tectonics is probably different than the Pliocene tectonics. Most of the existing very large N W - S E - s t r i k i n g normal faults which were active during the Pliocene are not seismically active at the present time (Fig. 8), and large earthquakes occur only exceptionally (as the 1894 Atalanti Martinon earthquake). On the other hand, new E-W-striking faults
424
D. H A T Z F E L D 42 ~
22 ~
20 ~
24 ~
26 ~
o
28 ~
42 ~
EUROPE
Albania ~
o
%~ <M
Ka %
.
9 //
40 ~
~
~,~
40 ~
SVRKEV-.
A
9
~
.dz
38 ~
38 ~
:e,oo.nes
....
r;6,"
~t
"~hodos
/
36 ~ Crete
36 ~
/ 34 ~
34 ~ Trertctt
20 ~
22 ~
24 ~
26 ~
28 ~
Fig. 8. Rough interpretative tectonic sketch deduced from tectonic faults (after Caputo & Pavlides 1993; Mercier et al. 1987; Jackson, 1994; Jolivet et al. 1994; Armijo et al, 1996) that experience present-day seismicity. We plot, as presently active, only the faults that are associated with historical or instrumental seismicity. Thin lines are aseismic faults, thick lines are seismic faults. Most of the NW-SE-trending surface faults seem to be inactive at the present time. We note that the western termination of the North Anatolian Fault is a complex system that is connected to the Northern termination of the Kefallinia fault similarly as originally proposed by McKenzie (1972).
of small d i m e n s i o n s start to d e v e l o p in the Mygd o n i a n g r a b e n , in t h e V o l o s basin, in A t t i k i , a r o u n d t h e T r i k h o n i s lake a n d t h e A m v r a h i k o s Gulf. In s o m e places t h e y c r o s s - c u t t h e existing N W - S E - s t r i k i n g faults as in T h e s s a l y ( C a p u t o & P a v l i d e s 1993) or t h e M y g d o n i a n g r a b e n ( M e r c i e r et al. 1983). T h e s e faults strike p e r p e n d i c u l a r to t h e m a i n N - S - t r e n d i n g e x t e n s i o n t h a t affects t h e A e g e a n a n d are l o c a t e d o n l y in c o n t i n e n t a l G r e e c e . M o r e o v e r , m o s t of t h e s e r e c e n t E - W - s t r i k i n g faults are l o c a t e d at t h e
t e r m i n a t i o n of dextral strike-slip faults such as t h e d i f f e r e n t b r a n c h e s of the N o r t h A e g e a n T r o u g h a n d t h e Kavalla fault.
Conclusion A careful e x a m i n a t i o n of reliable seismicity a n d focal m e c h a n i s m s clearly shows t h a t t h e r e is n o t a p e r f e c t fit w i t h t h e active faults as t h e y are g e n e r a l l y d i s p l a y e d . M o s t of t h e N W - S E striking faults are n o t seismically active a n d n e w
TECTONICS & SEISMICITY IN THE A E G E A N E - W - s t r i k i n g faults start to d e v e l o p in continental G r e e c e . We have g o o d e v i d e n c e that important changes in the tectonics o c c u r r e d since late Pliocene ( M e r c i e r et al. 1987; A n g e l i e r et al. 1982; A r m i j o et al. 1996) w h i c h could be d u e either to a c h a n g e in the b o u n d a r y conditions, or m o r e likely to the i m p o r t a n t a m o u n t of internal d e f o r m a t i o n (and r e l a t e d rotation) within the Aegean. T h e r e is, t h e r e f o r e , a possibility that s o m e of the late Pliocene or e a r l y P l e i s t o c e n e faults, that are oblique to the strain field, are no l o n g e r active. N e w faults of m o d e s t dimensions, that m o r e easily a c c o m m o d a t e t h e d e f o r m a t i o n , d e v e l o p at the p r e s e n t time. T h e global picture of p r e s e n t active faults is t h e r e f o r e different to that for the late Pliocene. T h e s e n e w E - W - s t r i k i n g faults are m o s t l y l o c a t e d in c o n t i n e n t a l G r e e c e a n d T u r k e y , w h o s e crust is certainly c o n t i n e n t a l and probably w e a k , r a t h e r t h a n in the n o r t h e r n A e g e a n sea w h i c h has b e e n s t r e t c h e d and t h i n n e d by a factor of 2, and p r o b a b l y o c e a n i z e d (Le P i c h o n & A n g e l i e r 1979), and w h i c h m i g h t be stronger. As a c o n s e q u e n c e , the w e s t e r n t e r m i n a t i o n of the N o r t h A e g e a n t r o u g h is c o n n e c t e d to the K e f a l l i n i a fault by a n e w diffuse s y s t e m of n o r m a l faults or grabens that start to d e c o u p l e P e l o p o n n e s e f r o m central G r e e c e , as p r o p o s e d by M c K e n z i e (1972). The microearthquake seismicity is the result of numerous seismological experiments that were conducted in collaboration with the Universities of Athens (K. Makropoulos) and Thessaloniki (P. Hatzidimitriou, D. Panagiotopoulos, V. Karakostas) and supported by the EEC, contracts Simulation-121 and -353, contract EPOCH CT-91-0043. We benefitted from interesting discussions with J. Jackson, H. Lyon-Caen, J. Martinod, P. Molnar and B. Parsons.
References
ANGELIER,J. 1979. Recent Quaternary tectonics in the Hellenic arc: examples of geological observations on land. Tectonophysics, 52, 267-275. --, LYBERIS, N., LE PICHON, X., BARRIER, E. & HUCHON, P. 1982. The neotectonic development of the Hellenic Arc and the Sea of Crete: a synthesis. Tectonophysics, 86, 159-196. ARMIJO, R., MEYER, B., KING, G. C. R, RIGO, a . & PAPANASTASSIOU,D. 1996. Quaternary evolution of the Corinth Rift and its implications for the late Cenozoic evolution of the Aegean. Geophysical Journal International, 126, 11-53. BAKER, C., HATZFELD, D., LYON-CAEN,H., PAPADIMITRIOU,E. & RXGO,A. 1997. Earthquake mechanisms of the Adriatic sea and western Greece. Geophysical Journal International, 131, 559-594.
425
BROOKS, M., CLEWS,J. E., MELIS,N. S. & UNDERHILL,J. R. 1988. Structural development of Neogene basins in western Greece. Basin Research, l, 129-138. CAPUTO, R. & PAVLIDES,S. 1993. Late Cainozoic geodynamic evolution of Thessaly and surroundings (central-northern Greece). Tectonophysics, 223, 339-362. DOUTSOS, T., KONTOPOULOS, N. & FRYDAS, D. 1987. Neotectonic evolution of northwestern Greece, Geologisches Rundschau, 76, 433-450. DZIEWONSKI,A., FRIEDMAN,A. & WOODHOUSE, J. H. 1983. Centroid-moment tensor solutions for January-March, 1983. Physics of the Earth and Planetary Interiors, 33, 71-75. ENGLAND, P., HOUSEMAN, G. & SONDER, L. 1985. Length scales for continental deformation in convergent, divergent, and strike slip environments: analytical and approximate solutions for a thin viscous sheet model. Journal of Geophysical Research, 90, 3551-3557. GAUTIER, P. & BRUN,J.-P. 1994. Ductile crust exhumation and extensional detachments in the central Aegean (Cyclades and Evvia island). Geodinamica Acta, 7, 57-85. HATZFELD, D., BESNARD, M. MAKROPOULOS, K. & HATZIDIMITRIOU,P. 1993. Microearthquake seismicity and fault plane solutions in the southern Aegean and its tectonic implications. Geophysical Journal International, 115, 799-818. - - , CHRISTODOULOUA. A., SCORDILISE. M., PANAGIOTOPOULOS D. G. c~; HATZIDIMITRIOU, P. M. 1987. A microearthquake study of the Mygdonian graben (Northern Greece). Earth and Planetary Science Letters, 81, 379-396. - - , KASSARAS,I., PANAGIOTOPOULOS,D., AMORESE, D., MAKROPOULOS, K., KARAKAISIS, a. & COUTANT, O. 1995. Microseismicity and Strain Pattern in Northwestern Greece. Tectonics, 14, 773-785. - - , MARTINOD, J., BASTET, G. & GAUTIER, P. 1997. An analog experiment for the Aegean to describe the contribution of gravitational potential energy. Journal of Geophysical Research, 102, 649-659. , PEDOTI'I, G., HATZID1MITR1OU,P. & MAKROPOULOS, K. 1990. The strain pattern in the western Hellenic arc deduced from a microearthqauke survey, Geophysical Journal International, 101, 181-202. JACKSON,J. A. 1994. The Aegean deformation. Annual Review of Geophysics, 22, 239-272. & MCKENZm, D. 1988. The relationship between plate motions and seismic moment tensors, and the rates of active deformation in the Mediterranean and the Middle East. Geophysical Journal International, 93, 45-73. --, HAINES, E & HOLT, W. 1992. The horizontal velocity field in the deforming Aegean Sea region determined from the moment tensor of earthquakes. Journal of Geophysical Research, 97, 17 657-17 684 JOLIVET, L., BRUN, J-P., GAUTIER,P., LALLEMAND,S. & PATRIAT,M. 1994.3-D Kinematics of extension in
426
D. HATZFELD
the Aegean from early Miocene to the present: insight from ductile crust. Bulletin de la Soci~t~ GOologique de France, 165, 195-209. KAHLE, H-G, MOLLER, M. V., MUELLER, S. & VEIS, G. 1993. The Kephalonia transform fault and the rotation of the Apulian plateform: evidence from satellite geodesy. Geophysics Research Letters, 20, 651-654. KISSEL, C. & LAJ, C. 1988. The tertiary geodynamical evolution of the Aegean arc; a paleomagnetic reconstruction. Tectonophysics, 146, 183-201. LE PICHON, X. & ANGELIER, J. 1979. The Hellenic arc and trench system: a key to the neotectonic evolution of the Eastern Mediterranean region. Tectonophysics, 60, 1-42. , CHAMOT-ROOKE,N., LALLEMANT,S., NOOMEN,R. & VEIS, G. 1995. Geodetic determination of the kinematics of Central Greece with respect to Europe: Implications for Eastern Mediterranean Tectonics. Journal of Geophysical Research, 100, 12675-12690. LYBERIS, N. 1984. Tectonic evolution of the North Aegean Trough. In: ROBERTSON, A. H. E & DIXON, J. E. (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 708-725. MASCLE, J. & MARTIN, L. 1990. Shallow structure and recent evolution of the Aegean Sea: A synthesis based on continuous reflection profiles. Marine Geology, 94, 271-299. MCKENZIE, D. P. 1972. Active Tectonics of the Mediterranean Region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. - 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions.
Geophysical Journal of the Royal Astronomical Society, 55, 217-254. JACKSON, J. 1983. The relationship between strain rates, crustal thickening, palaeomagnetism, finite strain and fault movements within a deforming zone. Earth and Planetary Science Letters, 65, 182-202. MERCIER, J-L. 1977. L'arc 6gden, une bordure ddform6e de la plaque eurasiatique; rdflexions sur un exemple d'dtude n6otectonique. Bldletin de la SociOt~ G~ologique de France, 19, 663-672. , CAREY E., PmL1e H. & SOREL D. 1976. La ndotectonique plio-quaternaire de l'arc 6gden externe et de l a m e r Egde et ses relations avec la sismicitd, Bulletin de la SociOtO G~ologique de France, 7, 355-372. --, CAREY-GAILHARDIS, E., MOtIYARIS, N., -
-
&
SIMEAKIS, C., ROUDOYANNI,T. & ANGELIDHIS, C. 1983. Structural analysis of recent and active faults and regional state in the epicentral area of the 1978 Thessaloniki earthquake (Northern Greece). Tectonics, 2, 577-600. --, DELIBASIS, N., GAUTIER, A., JARRIGE, J. J., LEMEILLE,E, PHILIP,H., SI~BRIER,M. & SOREL,D. 1979. La n6otectonique de l'Arc Eg6en. Revue de
G~ographie Physique et de G~ologie Dynamique, 21, 67-92. - - , SOREL, D. • SIMEAKIS,K. 1987. Changes in the state of stress in the overrinding plate of a subduction zone: the Aegean arc from the Pliocene to the Present. Annales Tectonicae, 1, 20-39. PAPAZACHOS,B. C. 1973. Distribution of seismic foci in the Mediterranean and surrounding area and its tectonic implication, Geophysical Journal of the Royal Astronomical Society, 33, 419-428. PAPAZACHOS B. & PAPAZACHOUK. 1997. The Earthquakes of Greece. Ekdoseis Ziti, Thessaloniki. PAVLIDES, S. B. & MOUNTRAKIS, D. M. 1987. Extensional tectonics of northwestern Macedonia, Greece, since the late Miocene. Journal of Structural Geology, 9, 385-392. RIGO, A., LYON-CAEN,H., ARMIJO, R., DESCHAMPS,A., HATZFELD, D., MAKROPOULOS, K., PAPADIMITRIOU, P. & KASSARAS I. 1996. A microseismic study in the western part of the Gulf of Corinth (Greece): implications for large-scale normal faulting mechanisms. Geophysical Journal International, 126, 6634588. ROBERTS, S. & JACKSON,J. A. 1991. Active normal faulting in Central Greece: an overview. In: ROBERTS, A. M., YIELDING, G. & FREEMAN, B. (eds) The Geometry of Normal Faults. Geological Society, London, Special Publications, 56, 125-142. SPAKMAN,W., WORTEL, M. J. R. & VLAAR, N. S. 1988. The Hellenic subduction zone: a tomographic image and its geodynamical implications. Geophysical Research Letters, 15, 60453. TAYMAZ, T., JACKSON, J. A. & WESTAWAY, R. 1990, Earthquake Mechanisms in the Hellenic Trench near Crete, Geophysical Journal International, 102, 695-732. - - , JACKSON,J. A. & MCKENZlE, D. 1991. Active Tectonics of the North and Central Aegean Sea, Geophysical Journal International, 106, 433-490. VEIS, G., BILLIRIS, H., NAKOS,B. & PARADISSIS,D. 1992, Tectonic strain in Greece from geodetic measurements, Academy of Sciences of Athens, 67, 129-166.
Ductile extension and the formation of the Aegean Sea L. J O L I V E T
t & M. PATRIAT 2
1DOpartement des Sciences de la Terre, Universit~ de Cergy-Pontoise, 8 le Campus, A v e n u e du Parc, 95011 Cergy-Pontoise cedex, France, U R A C N R S 1759 eLaboratoire de GOologie, Ecole Normale Sup&ieure, 24 rue L h o m o n d , 75231 Paris cedex 05, France, U R A C N R S 1316 Abstract: The Aegean Sea and Tyrrhenian sea areas offer almost unique examples of
actively collapsing orogens where the internal velocity field, kinematic boundary conditions and seismic activity can be compared and where older structures formed during the early stages of the same process are exhumed. We study these processes along two major transects. The first runs from Mt Olympos to the centre of the Cyclades, it is perpendicular to the direction of extension and offers the opportunity to study the brittle-ductile transition by direct observation of rocks and deformation. A gradient of finite exhumation along the transect has brought various crustal levels to the surface, and the localization of deformation during exhumation along major detachments has preserved older penetrative structures formed below. As these outcrops are distributed along the strike of a single crustal-scale tilted block the observed structures probably corresponds to a single tectonic event, and we can sample the associated deformation at increasing depth along the transect. We propose a stratification of deformation regimes associated to extension: localized normal faults in the brittle crust, distributed extensional shear bands with a partition between domains of noncoaxial and coaxial flow below 8-10 km, and distributed coaxial flow below 20 km. Flat-lying extensional shear bands extend brittle normal faults within the brittle-ductile transition. The second transect runs from Crete to Naxos and Mykonos and is parallel to the main direction of extension. It shows the transition from tectonic accretion and syn-orogenic extension near the subduction front to post-orogenic extension in the backarc region. The exhumation of metamorphic rocks proceeds in two stages: synorogenic 'cold' exhumation along detachments in the upper part of the accretionary complex, with a good preservation of HP-LT parageneses, and post-orogenic exhumation in a warmer environment in the backarc region with a complete retrogression of HP-LT parageneses. The 3D finite strain field is then compared to recent space geodesy data and the mechanism of crustal collapse discussed. E x t e n s i o n has b e e n active in the A e g e a n r e g i o n for a long e n o u g h time (25 Ma) for ductile structures f o r m e d u n d e r the same extensional r e g i m e to crop out at the surface. T h e y are thus accessible to direct o b s e r v a t i o n in various m e t a m o r phic core c o m p l e x e s such as Naxos in the c e n t r e of the Cyclades (Lister et al. 1984) (Figs 1, 2 & 3). T h e y offer t h e o p p o r t u n i t y to study a c o m p l e t e section of an e x t e n d i n g crust f r o m t h e d e e p ductile to the u p p e r brittle crust. In this p a p e r we d e m o n s t r a t e the structural relations b e t w e e n v a r i o u s levels of t h e crust focussing on the b r i t t l e - d u c t i l e transition and discuss the kinematic b o u n d a r y conditions and processes which allowed the f o r m a t i o n of those structures. E x t e n s i o n in t h e b a c k - a r c r e g i o n (Cyclades) o c c u r r e d c o n c u r r e n t l y with the f o r m a t i o n of an a c c r e t i o n a r y c o m p l e x n e a r the s u b d u c t i o n front (Le P i c h o n et al. 1994; T a y m a z et al. 1991) (Fig. 4). D u r i n g t h e e x t e n s i o n a l process frontal accretion and e x t e n s i o n both m i g r a t e d s o u t h w a r d as the s u b d u c t i o n slab r e t r e a t e d (Le P i c h o n & A n g e l i e r 1981). This implies that the m a t e r i a l
first a c c r e t e d in the w e d g e was t h e n t r a n s f e r r e d to the b a c k - a r c region. D u r i n g this m i g r a t i o n d e e p l y b u r i e d m a t e r i a l was e x h u m e d in two d i f f e r e n t settings, t h e a c c r e t i o n a r y c o m p l e x itself and the w a r m b a c k - a r c d o m a i n (Jolivet et al. 1994b). T h e m e t a m o r p h i c history of rocks is different in b o t h cases a n d we describe it along a N - S t r a n s e c t f r o m C r e t e to t h e c e n t r a l Cyclades. K i n e m a t i c b o u n d a r y conditions are well constrained in the e a s t e r n M e d i t e r r a n e a n and the active velocity field inside the A e g e a n d o m a i n precisely k n o w n f r o m space g e o d e s y (Fig. 5). It is thus possible to c o m p a r e the p r e s e n t velocity field to the finite strain field o b t a i n e d after 25 M a of c o n t i n u o u s d e f o r m a t i o n (Jolivet et al. 1994a). We show that b o t h fields are c o m p a t i b l e and that the s a m e k i n e m a t i c s has p r e s i d e d to the deform a t i o n during the last 25 Ma. D i r e c t o b s e r v a t i o n of t h e in situ b r i t t l e - d u c t i l e transition d u r i n g extension in the c o n t i n e n t a l crust is impossible with available t e c h n i q u e s and no consensus exists as to h o w the n o r m a l faults
JOLIVET,L. & PATRIAT,M. 1999. Ductile extension and the formation of the Aegean Sea. In: DURAND,B., JOLlVET, L., HORVATH,F. & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 427--456.
428
L. JOLIVET & M. PATRIAT
Fig. 1. Major active structures in the Aegean domain and P - T paths in the main metamorphic core complexes (Avigad et al. 1992; Jolivet et al. 1996). Insert: present-day kinematics of extrusion of the Anatolian block along the North Anatolian Fault (thick dotted lines for small circles and large black dot for rotation pole) (Le Pichon et al. 1994). Thin lines represent the direction of finite stretching (Jolivet et al. 1994a).
DUCTILE EXTENSION, AEGEAN SEA
429
Fig. 2. Bathymetric and topographic map of the Aegean region (source etopo 5) and sense of shear in exhumed ductile metamorphic core complexes and sense of shear along the brittle-ductile transition deduced from the attitude of normal faults.
seen in the upper crust root in the middle or lower crust. The succession of deformational events shown by exhumed metamorphic rocks gives a rather precise idea of the vertical superposition of deformation regimes (Sibson 1983) but does not constrain the nature of the brittle-ductile transition at a larger scale. One of the pending problems is how large normal faults and tilted blocks are relayed at depth by distributed ductile flow, and whether or not the simple shear model of extension is applicable to the whole crust (Wernicke 1981) or limited to its middle and upper part (Andersen & Jamtveit 1990; Jackson 1987; Lister & Davis 1989). Large-scale detachments seen above extensional metamorphic core complexes separate domains with respectively brittle and ductile deformation but they do not represent the brittle-ductile transition (Lister & Davis 1989). They are the ultimate evolution of a more diffuse shear zone which encompasses a large
crustal thickness. A large part of the finite deformation seen along those detachments is the result of exhumation of deep crustal rocks and thus it cannot give a precise idea of the in situ brittle-ductile transition. Seismological studies in regions of active extension show that seismogenic normal faults are planar down to the base of the brittle crust (Jackson & White 1989; Braunmiller & Nabelek 1996). Decoupling from the lower ductile crust is probably achieved through aseismic slip (Braunmiller & Nabelek 1996). Other more detailed studies in areas of active extension such as the Gulf of Corinth suggest that steep normal faults in the upper crust root in shallow dipping (e.g. 10 ~ extensional shear zones at depth around 10 km (Eyidogan & Jackson 1985; King et al. 1985; Rigo et al. 1996). Those d6collements are seismically active and the sense of motion along them is synthetic to that along major normal faults (Rigo et al. 1996). The active seismic zone
430
L. JOLIVET & M. PATRIAT
Fig. 3. Bathymetry and topography along the Olympos-Naxos transect, and isocontours of crustal thickness (Le Pichon & Angelier 1981). Fault pattern is also shown (Gautier & Brun 1994a; Masclc & Martin 1990; Rigo et al. 1996). Arrows represent the sense of shear observed in exhumed metamorphic rocks (Godfriaux 1965; Schermer 1990, 1993; Schermer et al. 1990; Urai et al. 1990; Buick 1991; Faure, et al. 1991; Godfriaux and Ricou 1991 ; Lee & Lister 1992; Gautier, et al. 1993; Gautier 1994; Gautier & Brun 1994a; Jolivet et al. 1996 and this work).
marks the lower limit of crustal seismicity and likely represents the transition from brittle to ductile behaviour. The depth of the transition between the seismic and aseismic crust varies between 10 and 20 km (Molnar & Chen 1983) and is most likely controlled by the rheology of
quartz and feldspar which have their brittle-plastic transition at respectively 300-350~ and 450-500~ (Voll 1976; Kerrich et al. 1977; Sibson 1983). Independantly of seismological studies, field observations of Mediterranean core complexes
DUCTILE EXTENSION~ A E G E A N SEA
Fig. 4. Synthetic cross sections through the Aegean domain and the two types of retrograde P - T paths, (Jolivet et al. 1994b).
Fig. 5. Displacement field recorded from space geodesy, (Le Pichon et al. 1994). Dotted lines represent the small circles about the rotation pole Anatolia/Eurasia.
431
432
L. JOLIVET & M. PATRIAT
(Alpine Corsica and the Aegean Sea) shows the existence of similar flat-lying ductile shear zones active within the pressure and temperature domain of the greenschist facies which is in the range of the above temperatures (Daniel et al. 1996; Gautier & Brun 1994a,b; Jolivet et al. 1991, 1994a,b). The geological history of these shear zones makes them good candidates for comparison with the seismically active detachments seen in regions of active extension (Jolivet et al. 1994a). The question is however the following: can we go back to the initial geometry while they were active at depth if their present attitude is the result of a succession of deformations during exhumation? or, is it possible to find regions were deep deformations have been preserved (frozen) without much reworking during the process of extension? Generally core complexes show an internal part where such deep deformations were preserved because deformation has a tendancy to localize along the detachment, and the degree of localization increases during exhumation, while rocks pass through the brittle-ductile transition. Observing various core complexes with various finite rates of extension (and exhumation) might allow reconstruction of the vertical stratification of structures in an extending crust. The Aegean domain offers this possibility along the Olympos-Naxos transect (Fig. 3), where a succession of metamorphic core complexes exhumed during the same progressive process of extension crop out, and where, next to the transect, the same phenomenon is active at depth below the Gulf of Corinth.
Regional geology Aegean
Sea
Crustal thinning has been active in the Aegean Sea in the back-arc region of the Hellenic trench since the Early Miocene (McKenzie 1972, 1978; Dewey & Seng6r 1979; Le Pichon & Angelier 1981; Taymaz et al. 1991). Present-day kinematics is governed by rigid extrusion of the Anatolian block along the North Anatolian Fault toward the subduction zone free edge (Fig. 5), and internal deformation of the same block leading to extension in the backarc domain (McKenzie 1972, 1978; Le Pichon 1981; Le Pichon et al. 1994; Armijo et al. 1996). Incipient collision of the African plate and the Anatolian block south of Crete is responsible for arc-parallel extension close to the trench while N-S extension prevails in the back-arc domain (Angelier et al. 1982; Armijo et al. 1992). Before 3 Ma N-S extension was prevailing in the entire
Aegean Sea (Lyberis et al. 1982; Mercier et al. 1979, 1987). Metamorphic rocks were exhumed during the process of extension and the geometry of ductile extension has been studied by several authors (Faure & Bonneau 1988; Buick 1991; Faure et al. 1991; Lee & Lister 1992; Gautier et al. 1993; Sokoutis et al. 1993; Gautier & Brun 1994a; Jolivet et al. 1994a, 1994b, 1996). From Crete to the Cyclades islands and to the northern Aegean Sea the direction and sense of motion along major detachments were elucidated. The results are shown on Fig. 2. Extension was superimposed on an earlier episode of crustal thickening which led to the formation of the Hellenides culminating 45 Ma ago (Avigad & Garfunkel 1991; Avigad et al. 1992; Bonneau & Kienast 1982; Jacobshagen et al. 1978; Wijbrans & McDougall 1988; Wijbrans et al. 1993). The products of this early episode are a stack of nappes of alpine type and highpressure and low-temperature metamorphic rocks known as the Cycladic Blueschists (Blake et al. 1981; Bonneau & Kienast 1982; Okrusch & Br6cker 1990; Avigad & Garfunkel 1991). Crustal collapse started in the Late Oligocene-Early Miocene probably as a consequence of the initiation of the Hellenic trench 40 Ma ago (Berckhemer 1977; Gautier & Brun 1994a; Spakman et al. 1993). Syn- versus post-orogenic
extension
Post-orogenic extension affected most of the Aegean domain while crustal thickening was still active in Crete and Peloponnese (Bassias & Triboulet 1994; Bonneau 1984; Fassoulas et al. 1994a; Jolivet et al. 1994b, 1996; Seidel et al. 1982; Theye & Seidel 1991). Extension is recorded in those regions as soon as the Early and MidMiocene and was responsible for the exhumation of younger high pressure and low temperature metamorphic rocks in the upper part of a convergent accretionnary wedge. It is thus important to distinguish between syn - and post-orogenic extension. Along the Olympos-Naxos transect a similar process of syn-orogenic extension has probably been active and responsible for the exhumation of HP-LT metamorphics during the Eocene crustal stacking event (Avigad & Garfunkel 1991; Wijbrans et al. 1993) as will be discussed later. Crustal thickness
Crustal thickness in the Aegean domain has been estimated from gravity data (Makris 1978; Le Pichon & Angelier 1981) and published as
DUCTILE EXTENSION, AEGEAN SEA maps of Moho depth and crustal thickness which show that the North Aegean Trough and Sea of Crete are the thinnest regions. Crustal thickness decreases gradually from the Hellenic arc to the central Aegean Sea along strike (Fig. 3). Along the Olympos-Naxos transect it varies from 38 km below Mt Olympos to 30 km below the island of Naxos. North and south of Naxos the crust is considerably thinner especially in the Sea of Crete where it can be as thin as 20 km or less. 45-50 km is a fair minimum estimation of the pre-extension thickness by comparison to the Hellenic Chain further west. Metamorphic
core complexes
The first description (Lister et al. 1984) of a cordilleran-type metamorphic core complex in the Aegean Sea was made in Naxos (Fig. 2). Later the islands of Naxos, Paros, Evia, Andros, Tinos, Mykonos, Ikaria were studied by various teams (Urai et al. 1990; Buick 1991; Faure et al. 1991; Lee & Lister 1992; Gautier et al. 1993; Gautier & Brun 1994b; Vandenberg & Lister 1996) and a complete description of the finite strain field during the Late Oligocene Early Miocene high-temperature overprint is available in the form of stretching lineation maps in all these islands (Fig. 2). The island of Thassos and the southern Rhodopian massif were also studied and metamorphic core complexes revealed there also (Dinter & Royden 1993; Sokoutis et al. 1993). Except for the two latter regions and the island of Ios, kinematic indicators related to extension are top-to-the-north. A recent study of palaeomagnetic directions in the Miocene intrusions of Mykonos and Naxos (Morris & Anderson 1996), suggests postductile deformation rotations about vertical axes that have to be considered when discussing the Miocene evolution. Ductile deformation was also studied in the region of Mt. Olympos (Schermer 1990, 1993; Schermer et al. 1990; Godfriaux & Ricou 1991). The most obvious deformation in this region is contemporaneous with the Eocene HP-LT event and dated around 50-60 Ma. Most kinematic indicators are top-to-the-west consistent with the geometry of the crustal stacking episode.
The Olympos-Naxos transect and the brittle-ductile transition Geological
context
The studied transect runs N W - S E parallel to the western branch of the Hellenic arc. Major normal faults, active or recent, separates the arc
433
from the deep basins of the Aegean Sea to the northeast (Fig. 3). The whole region is tectonically divided by a series of normal faults of similar trend (Taymaz et al. 1991; Jackson 1994). Several crustal scale tilted blocks are readily recognized, separated by sedimentary basins (the Gulf of Evvia or the Gulf of Corinth) (Jolivet et al. 1994a; Papanikolaou et al. 1988) (Fig. 4). Studies of the subsurface structure between the blocks (Papanikolaou et al. 1988), as well as studies of the active deformation of the whole area show that major normal faults dip toward the northeast and that blocks are tilted toward the southwest (Taymaz et al. 1991; Jackson 1994). The case is particularly clear for the Gulf of Corinth which is clearly asymmetric with north-dipping normal faults which root in the seismically active decollement (King et al. 1985; Rigo et al. 1996). The metamorphic core complexes belonging to the transect studied here all belong to the same tilted block starting in the Mt Olympos region, running through Evia, Andros, Tinos and Mykonos. At the latitude of Mykonos the morphological expression of the block is less clear and the islands of Mykonos, Naxos and Paros are located on the same wide topographic high (Figs i and 2). P-T-t evolution The Late Oligocene-Early Miocene deformation reworks the Eocene high-pressure-lowtemperature event (Gautier & Brun 1994a) (Figs 1 & 6). A gradient of retrogression toward high temperature parageneses is observed from NW to SE. Little retrogression in the greenschist facies is observed in the Mt Olympos region (Schermer 1990; Schermer et al. 1990), while pervasive retrogression in the greenschist facies characterizes southern Evvia, Andros and part of Tinos, and amphibolite facies overprint is observed from southern Tinos to Mykonos and Naxos where partial melting and formation of migmatites is observed (Alther et al. 1982; Buick & Holland 1989; Jansen 1977; Wijbrans & McDougall 1988). Retrogression P - T - t paths show these different thermal histories (Figs 1 & 6). For the three examples of Mt Olympos, Tinos and Naxos, published P - T - t paths are available (Buick & Holland 1989; Okrusch & Br6cker 1990; Schermet 1990; Schermer et al. 1990). In Mt Olympos the P - T conditions of peak metamorphism are around 8 kbar-300~ and radiometric dates for this event cluster between 53 and 61 Ma (Schermer 1990, 1993; Schermer et al. 1990). The last event recorded in the radiometric data is a fast cooling around 100-150~ at
434
L. JOLIVET & M. PATRIAT
Fig. 6. Retrograde P-T-t paths from Mt Olympos, Tinos and Naxos (Buick & Holland 1989; Okrusch & Br6cker 1990; Schermer 1990; Schermer et al. 1990) and in Crete (Jolivet et ah 1996).
16-23 Ma associated with normal faulting. This shows undoubtedly that the HP-LT metamorphic rocks had already been exhumed to shallow levels when the Aegean extension started in the early Miocene. Peak P - T conditions in Tinos fall in the eclogite facies (15-18 kbar, 450-500~ and ages around 45 Ma were obtained (Okrusch & Br6cker 1990; Br/3cker et al. 1993). A subsequent greenschist-amphibolite overprint is recorded around 21-23 Ma. As will be discussed later this late overprint is associated to ductile reworking of the earlier HP foliation, motion along the detachement and intrusion of a granite at 19 Ma. The deformation contemporaneous with the greenschist reworking has been interpreted to be (Gautier & Brun 1994a; Jolivet et al. 1994a) the result of crustal scale extension. The Aegean extension is thus recorded here as ductile to brittle structures, the deepest structures being well within the greenschist facies. The shape of the P - T - t path shows that a fast exhumation occurred before 21-23 Ma; whether
this is due to extension or to some other process removing overburden is not clear. Peak P - T conditions are less well constrained in Naxos because of a pervasive overprint in the amphibolite facies and partial melting in the core of the dome. P - T conditions around 10 kbar and at least 550-600~ were attained 45 Ma ago (Jansen 1977; Buick & Holland 1989; Buick 1991). The retrogression path involves a peak toward high temperatures (700~ 5 kbar) and is also associated with extensional deformation and formation of the metamorphic core complex. In a first approach these data show that when the Aegean extension started the metamorphic rocks of Naxos had already been equilibrated in high-temperature-low-pressure conditions while those of Tinos or Mt Olympos had escaped from the deep and hot environment. In the Early Miocene the blueschist of Mt Olympos were already in the surficial parts of the crust, while those in Tinos were at intermediate depth (greenschist facies) and those of Naxos deeper still and affected by amphibolite facies and anatexis. The observed pattern is consistent with thinning of the crust from NW to SE. The conclusion of this preliminary analysis is that from Mt Olympos to Naxos one can observe rocks exhumed by extension from a rather superficial setting in the NW to a deep one in the SE. In the following a short summary of the deformation observed in each region is presented. For more detailed studies the reader is referred to Shermer (Schermer 1990, 1993; Schermer et al. 1990) and Godfriaux and Ricou (Godfriaux 1965; Godfriaux & Ricou 1991) for Mt Olympos, to Melidonis (1980), Bavay & Romain-Bavay (1980), Gautier and Brun (Gautier 1994; Gautier & Brun 1994a) Patriat (1996) for the islands of Evvia, Andros and Tinos, to Faure et al. (1991) and Lee & Lister (1992) for Mykonos, and to Gautier et al. (1993) and Buick (1991) or Urai, et al. (1990) for the islands of Naxos and Paros. Mt Olyrnpos
The region is characterized by the Mt Olympos tectonic window which shows the contact between the Pelagonian units (sensu lato) above the Cycladic blueschist (Ambelakia unit) and the parautochtonous Olympos-Ossa unit (Figs 3 & 7). HP-LT parageneses are observed both above and below the tectonic contact which is described as a syn-HP thrust. Ductile deformation localized in a narrow zone along the contact and most syn-HP kinematic indicators indicate top-to-the-southwest motion (Fig. 3).
DUCTILE EXTENSION, AEGEAN SEA
Fig. 7. Cross-sections through Mt Olympos, Tinos and Paros-Naxos (Schermer 1990; Gautier et Gautier 1994; Gautier and Brun 1994a, b). Rare kinematic indicators corresponding mostly to semi-brittle structures on the eastern side of Mt Olympos indicate top-to-the-northeast motion but it cannot be excluded that they represent simply layer-parallel extension due to a late bending of the foliation during normal faulting. The early nappe stack is reworked by extensional tectonics best expressed by the large recent northeast-dipping normal fault which cuts the topography to the northeast of Mt
435
al.
1993;
Olympos and Mt Ossa (Fig. 8a). An older extensional event is recorded along a flat-lying detachment (Schermer 1993) (Fig. 7). No major ductile deformation is associated to this detachment. It can be concluded that the Aegean extension in Mt Olympos is expressed mostly by brittle structures, high-level detachments and normal faults indicating a top-to-the-northeast motion of the hanging wall.
436
L. JOLIVET & M. PATRIAT
9
,....
E.-.
5~ ,,~
~.,,,,.
~,,..~
r.~
.<~,
9~
o
i
~r.~
.~'7,
~2
DUCTILE EXTENSION, A E G E A N SEA
437
t"q
O 9
. ,,...~
. ,,,-t
9
.,,.~
"7,
9
r~
,3
9 9
438
L. JOLIVET & M. PATRIAT
Evia
Tinos
Ductile structures which undoubtedly can be associated with extension are observed in s o u t h e r n Evia where large outcrops of the Cycladic blueschist are found (Fig. 3). The greenschist retrogression is localized to numerous shear bands at various scales (Figs 8b, c & 9). Figure 9 shows metre-scale extensional shear zones in the central part of southern Evia and this structural pattern is characteristic of the whole region. The fabric between the shear zones comprise greenschist facies minerals assemblages. Figures 8b and c show smaller scale shear zones which cut through the greenschist foliation between the larger shear zones described above. Most shear zones dip toward the east or northeast and indicate a regional sense of shear in the direction of the northern A e g e a n Sea. The southwestern part of the island shows conjugate shear zones implying a finite deformation closer to coaxial flow but the direction of stretching is still along the same E N E - W S W direction.
Ductile extension in Tinos has been studied in details by several workers (Avigad 1990; Gautier 1994; Gautier & Brun 1994b). The geology of the island is divided into two main units (Figs 12 & 13): the upper unit is essentially made of serpentinites and metabasites metamorphosed in the grenschist facies (Melidonis 1980). The lower unit is a complex alternation of metabasites and metapelites displaying blueschist and eclogitic parageneses belonging to the Cycladic blueschists widely distributed in the Aegean Sea (Br0cker 1990). Both units are later intruded by a granodiorite dated at 19 Ma (Br0cker et al. 1993). As opposed to Naxos where a younger granodiorite only intrudes the lower unit below the detachment, its seals the extensional detachment in Tinos (Avigad & Garfunkel 1991). Radiometric ages in Tinos are consistent around 45 Ma for the high pressure event, 19-14 for the high-temperature event and the intrusion, and 70 Ma for the greenschist event in the upper unit (Br6cker et al. 1993). Here only the Miocene high temperature event is of concern. The distribution of metamorphism in the lower unit can be summarized as follows (Fig. 12): well preserved eclogites and blueschists are found only in the southwest half of the island, while the northeast one only shows the retrograde event in the greenschist facies except for a few highly retrograded eclogites. The late event is locally recorded in the southwest part, restricted to narrow shear zones. Greenschist metamorphism is synkinematic and formed preferentially along shear zones (Gautier & Brun 1994b). The same opposition between the two parts of the island holds for estimates of the finite strain regime (Fig. 12). The domain which has been completely retrogressed in the greenschist facies is characterized by consistent top-to-the-north sense of shear (Gautier & Brun 1994b), and a deformation regime which evolved from ductile to brittle near the detachment. The detachment itself is a sharp discontinuity marked by a shallow NE-dipping fault soled by a reddisch breccia (Figs 11c, l l d , 12, 13 & 14). High shear strains are recorded below it by penetrative NEtrending stretching lineations and sheath folds which appear immediately below the cataclastic breccia. Late steeply NE-dipping mesoscopic faults cut the detachment along the northeastern coast (Fig. 14). The granodiorite clearly intrudes both the lower and upper units and shows little internal strain. A weak NE-trending lineation is however visible near the detachment. The upper unit also
Andros
The Eocene high pressure event is rather poorly preserved on Andros (Papanikolaou 1977; Reinecke 1982). A pervasive reworking in a barrovian environment is observed throughout the island. The general deformation regime is much the same as in Evia with extensional shear bands distributed on the whole island (Figs 3 & 10). A spatial partitionning between a wide zone of predominant noncoaxial strain and a narrow zone of coaxial strain is indicated by the distribution of shear bands. A narrow stripe parallel to the coastline, on the southwestern side of the island, is characterized by conjugate shear bands, a southwest dipping crenulation cleavage and associated folds, and a better preservation of Eocene high pressure parageneses with garnet and glaucophane. Most of the island is characterized by retrogressive northeastdipping shear bands (Figs 9d, l l a & l l b ) , formed after peak pressures within the P - T range of the greenschist facies. The frequency of shear bands increases along a SW-NE crosssection until the basal contact of an upper tectonic unit. This zone and its associated shear-strain gradient has all the characteristics of an extensional detachment. The contact itself is marked by a tectonic breccia of variable thickness (up to several tens of metres). Northeastdipping normal faults cut through the upper unit above the contact.
DUCTILE EXTENSION, AEGEAN SEA
439
Fig. 10. Structural maps of Andros, Tinos and Naxos (Gautier et al. 1993; Gautier 1994; Gautier & Brun 1994a, b) and this work. Light shading represent domains where noncoaxial deformation predominates during extension. Darker shading represent the upper non metamorphic plate. shows evidence of top-to-the-NE shear immediately above the granite (Fig. 15a), contact metamorphic minerals and peripheric dykes being synkinematic. The detachment has apparently
migrated upward within the upper plate, above the granite, during its intrusion. Later steeply NE-dipping normal faults cut both the upper unit and the granite (Figs 14 & 15b).
440
L. JOLIVET
& M.
PATRIAT
. ,...~
~E
~.~
9 ,,-"
~.~ 4...a
'~.=
e=~D ,~.~ i
~9. ~ o
~2 ,_~.~ ~'E o o
g=,
. . ~,--...x
,~.~
~D ~D "--J ,...a r~
m
O
~a N~
Fig. 12. (a) Structural map of the island of Tinos showing the syn-greenschist deformation (Gautier & Brun 1994a; Melidonis 1980, and this work). Directions of finite stretching and sense of shear are shown as arrows. Ambigious sense of shear or clear coaxial stretching are shown by double arrows. Palaeostress tensors in the granite and in the upper plate near the contact with the granite were calculated using Angelier's method (Angelier 1984). The contours of well preserved high pressure rocks are after Melidonis (Melidonis 1980). (b) Enlargement of the central part of the island, measured lineations and foliation are displayed on stereograms. The entire data set shows the smooth antiformal shape of the foliation and the consistent NE-trend of the stretching lineation. Two stereograms then show the difference between the northeast and southwest sides of the island. Only stretching lineations are shown with a color code which differenciate lineations based on the associated shear sense: yellow is or top-to-NE, red for top-to-SW and dark blue for unknown sense of shear of coaxial figures.
442
L. JOLIVET & M. PATRIAT
Fig. 13. Photograph of the detachment on Planitis island (see location on Fig. 12).
Fig. 14. Synthetic diagram illustrating the structural relations between the lower and upper plates and the intrusive granite.
The n o r t h e a s t part of Tinos can thus be described as a major top-to-the-NE shear zone which evolved before, during and after the intrusion of the granodiorite 19 Ma ago. A transition in time from ductile to brittle deformation is observed consistently. The southwest part of the island has a more complex structure. As shown on the structural map inconsistent kinematic indicators are found (Fig. 12). We have distinguished between clear
kinematic indicators which indicate clear sense in either directions along the NE-trending stretching lineation, and indicators which were symmetric and thus did not indicate any preferred sense. These could have been envisaged only as unreliable criteria and rejected. However their distribution at the scale of the island is not random and we think they represent evidence for coaxial deformation. This is strengthened by the observation at all scales of
DUCTILE EXTENSION, A E G E A N SEA
443
~E
.--Z~ g ~-.~ ,-~
,._~ .~
~ ~~ . ,,,,.
o .~ ~
M
~~
*"
~
444
L. JOLIVET & M. PATRIAT
conjugate shear bands, indicated either top-tothe-NE or top-to-the-SW sense of shear. These shear zones transect the high-pressure fabric and is associated with synkinematic greenschist assemblages. Between the shear bands greenschist minerals apparently statically overgrow the high pressure foliation. The distinction between a coaxial regime and a non-coaxial one is thus very dependent upon the scale of the observed outcrop. The systematic variation of observed shear senses in the southwest half of the island c o m p a r e d to the consistency observed on the other side is symptomatic of different strain regimes at the scale of the island. An outcrop observed southeast of the city of Tinos on the southern coast is very characteristic of the coaxial regime. As shown on Fig. 16 the outcrop consists of foliated mafic rocks stretched along the NE direction and giving lensoid boudins of various scales. No clear asymmetry is observed and at the scale of the entire outcrop (hundreds of metres) and the finite bulk deformation is apparently coaxial. A detail of one of those boudins is shown on the same figure. The observation of the boudin itself shows the coexistence of conjugate shear bands where no preferred sense of shear can be determined. If one can use the map of Fig. 12 as a crude statistics of the number of shear bands indicating one sense or the other, the conclusion should be that coaxial stretching predominates the deformation regime. Or, at least, that the strain regime is less non-coaxial than in the northeast half of the island. There is thus a spacial partition between a domain of highly non-coaxial strain (the shear zone on the NE side) and a domain of bulk coaxial stretching. This conclusion has to be paired with the observation that high-pressure metamorphism is preserved only on the southeastern side (Fig. 12). Hence, there is a clear spatial and temporal link between the formation of the large scale north-dipping shear zone and the greenschist retrogression at the scale of the island. One important point of observation is that the shear zone is cut by the granite and had thus ceased to move at the depth of intrusion (Avigad & Garfunkel 1991). The detachment thus locally migrated above the granite between the rigid upper plate and the ductile intrusion. This together with the similar pressure gap below the detachment on both sides of the islands suggest that the shear zone has not accomodated very large displacements, and that most of the pressure gap was already achieved before this deformation occurred.
M y k o n o s , Paros, N a x o s The easternmost end of the transect is characterized by high t e m p e r a t u r e assemblages (amphibolite facies and migmatites) and intrusion of granitoids overprinting the Eocene highpressure event. Noncoaxial ductile deformation with top-to-the-northeast sense of shear characterizes the lower unit in Mykonos (Faure et al. 1991; Lee & Lister 1992). A progressive localization of deformation within the detachment zone below the upper non-metamorphic unit is well established. The ductile deformation occurs mostly within the granitoid (13 Ma) which makes most of the island. In Naxos and Paros two migmatites domes are seen below the contact with the upper, non-metamorphosed unit with a progressive localization of shear strain along the detachment during exhumation (Figs 7 & 11). A granodiorite that intruded the metamorphic dome at 12 Ma in Naxos and was also sheared along the detachment with the same top-to-the-north kinematics. Migmatization in Naxos (Wijbrans & McDougal11988) was contemporaneous with the deposition of the oldest sediments in asymetric grabens above the detachment (Gautier et al. 1993). The cores of the migmatite domes in Naxos, Paros and Mykonos show a clear stretching lineation but a less well defined foliation and no clear asymmetrical structures which could be used as shear sense criteria (Fig. 11). A crosssection through the dome in Naxos shows the shear sense criteria progressively vanishing toward the core while the first evidence of partial melting appear.
Conclusion The first conclusion to be drawn is the evolution seen along the transect in terms of structures related to the Miocene to present extension (Figs 3 & 7). A simple set of east-dipping largescale normal faults, associated with brittle flatlying detachments, is seen in Mt Olympos region. This led to a differential uplift of the hanging wall of the major normal fault. The preextensional Eocene deformation and metamorphic history is well preserved, and the contractional event precisely described. Southern Evvia and Andros display widely distributed shear bands, which were active within the P - T range of the greenschist facies. Most of them dip to the northeast and indicate a regional sense of shear toward the Aegean Sea. Locally on the southwestern side of the islands a domain of coaxial deformation is observed. It is also the region where HP parageneses are best
DUCTILE EXTENSION, A E G E A N SEA
(a)
445
(b)
(c)
(d) Fig. 16. Example of a symmetric stretching in metabasites along the southwestern coast of Tinos (see location on Fig. 12) and detail of one boudin.
446
L. JOLTVET & M. PATRIAT
preserved. Tinos shows a more evolved structure with a localization of strain along one major shear zone which occupies the whole northeast half of the island. The remaining half displays coaxial extensional deformation and wellpreserved HP minerals. In the late stage evolution of the detachment a granite intruded both the lower and upper plates. The southeastern end of the transect shows the ultimate evolution of the same structure with an extreme localization of the shear strain along one major shear zone at the base of the upper unit. The exhumed metamorphic rocks have evidence of very high shear strain within the shear zone and mostly the coaxial deformational regime in the core of the dome was not affected by the detachment. Along the strike of a single crustal scale tilted block from Mt Olympos to Mykonos and Naxos extensional structures evolve from brittle normal faults to distributed shear bands, then to localized large scale shear zones and finally a migmatitic domain characterized mostly by coaxial stretching. The asymmetry of deformation consistently indicates a top-to-the-northeast or top-to-the-north sense of shear at crustal scale. Deeper rocks have been exhumed by the larger extension in the southeastern end of the transect. Similar structures have been described in the Tyrrhenian Sea in Alpine Corsica and Isola d'Elba but with a less gradual transition. Structures seen in Corsica are quite similar to those of Tinos. The geometry of extension in Elba resembles that of Mykonos, especially concerning the relations between detachments and intrusion of granitoids (Daniel & Jolivet 1995; Daniel et al. 1996). Stratification o f d e f o r m a t i o n r e g i m e s
Describing the vertical stratification of deformation regimes within an extending crust based on this lateral evolution requires that the structures that formed as a consequence of exhumation are clearly identified. Obviously the Cordilleran-type metamorphic core complex of Naxos does not represent a good image of this vertical stratification because the intense shear strain localized along the detachment is the product of a progressive evolution. The transition at depth between the brittle and ductile behaviour is certainly not as sharp as seen in Naxos and is rather a thicker zone of mixed rheologies (between the brittle-ductile transitions of quartz and feldspar for instance, the temperature range is 150-200~ It is thus necessary to eliminate from our picture the sharp detachments seen on all those islands. What remains
might represent in-situ structures frozen during exhumation: normal faults in the upper crust, distributed shear bands within the brittleductile transition (10 kin) and symmetrical structures indicating coaxial flow in the deeper crust (20 km and below) (Fig. 17). The flat Moho observed in region of distributed intracontinental extension such as the Basin and Range requires that the localized detachments seen in the superstructures do not reach the lower crust (Gans 1987; Wernicke 1992). Lateral spreading of a low resistance lower crust, possibly molten to a large extent (the fluid layer; Wernicke 1992) explains this geometry. Coaxial flow is indeed what is observed in regions where deep crustal rocks have been exhumed through an extensional process such as the Western Gneiss Region in the Norwegian Caledonides (Andersen & Jamtveit 1990; Dewey et al. 1993; Andersen et al. 1994). This coaxial flow in the deep and weak crust is not recorded in the seismic activity. Seismic activity is usually restricted to the upper 10--20 km in the continental crust (Molnar & Chen 1983; Hatzfeld et al. 1993; Rigo et al. 1996). The normal fault seen at the surface and the main shocks of earthquakes are usually contained in a single plan which dips around 50 ~ (Jackson & White 1989). The main shocks and aftershocks of extensional earthquakes in the Aegean region s.l. are distributed near the base of the seismogenic layer. Major normal faults root within a flat-lying fault near the base of the seismogenic layer (Eyidogan & Jackson 1985; King et al. 1985). The motion along the flat portion of the fault is synthetic to that along the major steep fault. In the Corinth Gulf the major faults dip toward the north and the displacement along the flat portion is expected to be top-tothe-north as well (King et al. 1985). This flatlying decollement has been recently imaged below the Corinth Gulf (Rigo et al. 1996) using micro-earthquakes recorded during one month by a local network. A clear concentration of earthquakes along a thin zone gently dipping toward the north is seen near the brittle-ductile transition. Focal mechanisms of earthquakes along the ddcollement are compatible with a flat-lying normal fault. A comparison with the domain of non-coaxial strain observed along the Olympos-Naxos transect, in the islands of Evvia, Andros and Tinos can be made. A similar comparison with early works on the Corinth Gulf (King et al. 1985) was made on the examples of Alpine Corsica (Jolivet et al. 1991; Daniel et al. 1996), and was already suggested for the Aegean domain (Jolivet et al.
DUCTILE EXTENSION, AEGEAN SEA
447
Fig. 17. Idealized cross-section through an extending crust showing the succession of structures seen along the OlympoNaxos transect. Black dots represent hypocenters of earthquakes with a geometry similar to that described (Rigo et al. 1996) in the Corinth Gulf.
1994a). The P - T conditions of the greenschist facies are compatible with those of the brittle-ductile transition and the geometries seen in the exhumed metamorphic rocks and in the actively deforming crust are fully compatible. The stratification of deformation regimes proposed above is thus reasonable. The brittle~tuctile transition in extensional setting would thus be characterized by widely distributed extensional shear bands. The partitioning between noncoaxial and coaxial flow within the domain of the greenschist facies seen in Alpine Corsica or in the Aegean, suggests that conjugate
shear bands and overall coaxial flow characterizes most of the brittle-ductile transition zone and that a single set of shear bands and local noncoaxial flow exists below major normal faults. This makes of the brittle-ductile transition a zone of decoupling between the localized deformation of the upper crust and the distributed flow in the lower crust. Figure 18 shows a cartoon of the evolution of such a geometry during progressive extension. A t the onset of extension shear bands are widely distributed within the brittle-ductile transition zone. Further extension sees the formation of a set of major normal faults in the upper crust. The
448
L. JOLIVET & M. PATRIAT
Fig. 18. Cartoon showing the progressive evolution of a continental crust subjected to extension with flat-lying shear zone in the brittle-ductile transition zone.
normal faults all dip in the same direction in the case of the Aegean region. The localization of deformation in the upper crust is accompanied by a similar phenomenon in the middle crust where a few major shear bands are selected, those which extend normal faults at depth. Between the shear zones coaxial vertical shortening occurs. As extension proceeds the upper crust thins through normal faulting and a rebound of the lower crust follows. Cooling accompanies exhumation and thus the brittleductile material migrates downward in the exhumed material. Once the rocks are above the brittle-ductile transition they are 'frozen' and the deformation they experienced before preserved. Only along major shear zones are rocks deformed until they reached the surface.
Conclusion A gradient of finite exhumation along the transect has brought to the surface various crustal
levels, and the localization of d e f o r m a t i o n during exhumation along the major detachment has preserved deep structures until they reached the surface. As outcrops are distributed along the strike of single crustal-scale tilted blocks we can assume that the observed deformation corresponds to a single tectonic event, and that we sample the resulting deformation at increasing depth along the transect. We then propose a stratification of deformation regimes on the basis of observations: localized normal faults in the brittle crust, distributed e x t e n s i o n a l shear bands with a partition between domains of non-coaxial and coaxial flow below 10 kin, and distributed coaxial flow below 20 km. Flat-lying extensional shear bands e x t e n d n o r m a l faults within the brittle-ductile transition. This transect should attract detailed studies of the relations between d e f o r m a t i o n and m e t a m o r p h i c recristallization, as well as studies of the regime of fluid transfer.
DUCTILE EXTENSION, AEGEAN SEA
Migration of extension and compression, exhumation of metamorphic rocks The high pressure-low temperature parageneses described above were partly exhumed during the Miocene extensional episode but only from the depth of the greenschist facies (Tinos) or amphibolite facies (Naxos-Paros). Some of them were clearly already in the upper crust when extension started (Olympos). The excellent preservation and the radiometric ages obtained in Syros and Siphnos suggest that they had been already largely exhumed in the Eocene. This is confirmed by a reevaluation of A r - A r ages in Sifnos (Lister & Raouzaios 1996) showing that a large part of the exhumation took place as early as 40 Ma and that rocks were already in the greenschist facies P - T field 32 Ma ago. A large part of their exhumation is thus older than the formation of the Aegean Sea. In Syros the sense of shear recorded in the blueschist facies reworking of eclogites are similar to those observed associated with greenschist parageneses (top-to-the-eastnortheast) (Gautier 1994; Gautier & Brun 1994b; Patriat 1996). There thus seem to exist a continuum in the kinematics of exhumation. We shall examine this problem along a transect running from Crete to Naxos. The southward migration of thrust fronts from a Cretaceous one in the Rhodope massif, to an Eocene one in the Cycladic blueschists (Bonneau 1982; Bonneau & Kienast 1982; Wijbrans et al. 1993), a Miocene one in Crete and Peloponnese (Seidel, et al. 1982; Fassoulas et al. 1994b; Jolivet et al. 1994b, 1996; Kilias et al. 1994), and finally the active thrust front south of the Mediterranean Ridge (Le Pichon & Angelier 1981; Lallemant et al. 1994) was probably not entirely continuous and it might instead have jumped during the progressive incorporation of new crustal domains in the accretionary complex, but we shall in a first approach consider it as a continuous process of southward migration. Crete is then in a particularly favorable position. Our observations in Crete are published elsewhere (Jolivet et al. 1996) and we present here only a short summary useful for the final discussion. High-pressure-low-temperature rocks of Early Miocene age crop out in the core of a metamorphic core complex (Seidel et al. 1982; Theye & Seidel 1991; Fassoulas et al. 1994b; Jolivet et al. 1994b, 1996; Kilias et al. 1994) (Fig. 19). The lower plate (Phyllite-Quartzite and Plattenkalk units) shows HP-LT parageneses characterized by Fe-Mg carpholite, chloritoid,
449
lawsonite, aragonite, phengites in metapelites and glaucophane, lawsonite, Na-pyroxene in metabasites. The upper plate, made of several stacked units, shows no significant alpine metamorphism of Tertiary age. The extensional deformation is characterized mostly by coaxial N-S stretching (Fassoulas et al. 1994b; Kilias et al. 1994) or non-coaxial with a top-to-the-north predominant sense of shear during exhumation and extension (Jolivet et al. 1994b, 1996). In any case a strong N-S stretching associated with vertical shortening is observed during exhumation. Retrograde P - T paths are compatible with the existence of a detachment between the two plates and a fast exhumation (Fig. 6). All units were exhumed along a HP-LT gradient around 10~ -1 and no temperature increase is observed. Moreover, rocks situated immediately below the detachment cooled faster than those situated deeper in the nappe stack because they record the cooling effect of the upper plate as has been observed in the Himalayas (Hodges et al. 1993) and numerically modelled (Ruppel et al. 1988). Radiometric dates recently obtained on single grains of phengite by the A r - A r method (Jolivet et al. 1996) confirm and further precise previous findings (Seidel et al. 1982): the peak of pressure and the subsequent exhumation occurred in a short time range between 24 and 15 Ma. The HP-LT gradient maintained throughout the exhumation process requires continuous underthrusting. The fast exhumation observed in Crete thus occurred within an accretionnary complex in formation ('syn-orogenic exhumation' hereafter). This makes it very different from what is observed in Naxos ('post-orogenic exhumation' hereafter) further north during the same time interval (Fig. 4). While compression in a cold environment and formation of an accretionary complex were active in Crete with formation of detachments in the upper parts, crustal extension was active in the centre of the Cyclades in a warmer P - T environment. It should be noted at this stage that the sense of shear observed in Crete (Jolivet et al. 1994b, 1996) is similar to that seen in Naxos at the same period. The transition from syn- to postorogenic exhumation in space occurs with the same general kinematics. It can then be proposed that it has always been the case and that the transition from syn- to post-orogenic exhumation t h r o u g h time had the same kinematics. During the southward migration of the thrust front units which were first incorporated in the
450
L. JOLIVET & M. PATRIAT
Fig. 19. Simplified cross-section through western Crete showing the detachment and the retrograde P - T paths calculated from parageneses in the metapelites (Jolivet et al. 1996).
nappe stack near the thrust front were then transferred to the backarc domain. They could be exhumed early within the accretionary complex and in this case they followed HP-LT gradients much as the Phyllite-Quartzite unit of Crete. If exhumation had been entirely posttectonic it would have followed a warmer geotherin or later once they had reached the warmer back-arc region and they would have had time to equilibrate thermally. Their retrograde P - T paths is illustrated by a loop toward high temperatures as in Naxos. Those which had escaped reheating because they were exhumed syn-orogenically instead show HP-LT retrograde paths (Figs 1, 4 & 6).
Finite strain and instantaneous kinematics Figure 2 shows the distribution of the Miocene core complexes analysed so far by various authors (Faure et al. 1991; Lee & Lister 1992; Sokoutis et al. 1993; Gautier & Brun 1994a; Jolivet et al. 1994b). Except for the Thassos and Kavala examples in the northern Aegean region and Ios in the central Aegean (Vandenberg & Lister 1996) all detachments (syn-orogenic and post-orogenic) show northerly sense of shear (see above). The strike of stretching lineations progressively changes from E N E - W S W in the north to N-S in the south. The transition is progressive and the entire strain field has been described by
DUCTILE EXTENSION, AEGEAN SEA small circles centered on a pole located near the eastern end of the Hellenic trenches (Jolivet et al. 1994a) (Fig. 20b). This distribution could suggest that (1) either the observed deformation records the ductile flow below a rigidly rotating plate or (2) that the entire region is collapsing toward the free boundary (the subduction zone) and that particles move around a fixed point giving roughly circular trajectories. The present day directions of stretching lineations should be discussed in the light of palaeomagnetic data (Kissel & Laj 1988; Morris & Anderson 1996). Clockwise rotations in the Hellenic arc and counterclockwise rotations on the Turkish side have been documented since the last 5 Ma while no rotation is observed in Crete. The question is the following: have the ductile stretching direction been rotated rigidly or not? The lineations in the northern Aegean domain from the western boundary of the Rhodope to the northern Cyclades point northeastward which is compatible with the clockwise rotation evidenced by palaeomagnetic data. However had they been rotated their exact parallelism with the trend of the North Anatolian Fault and the slip lines of the rigid geodetic rotation would then be coincidental. Directions observed further south do not show any rotations. Lineations recorded in the southern Menderes Massif trend NNE in a domain which has suffered counterclockwise rotation. This last observations preclude the idea that all those lineations were originally N-S and
451
have then been rigidly rotated. In fact no data really constrained the paleomagnetic rotation of the ductilely deformed domain since the Miocene until a recent analysis performed in Naxos and Mykonos (Morris & A n d e r s o n 1996). These authors suggest differential post-ductile deformation in Naxos and Mykonos. If the direction of the stretching lineation is restore to its pre-rotation position, both are parallel and trend NE approximately parallel to the trend of the North Anatolian Fault. The direction of stretching then swings to more northerly in the southern Cyclades and Crete where no rotation have been detected. It is noticeable that the trend of the North Anatolian Fault is in good agreement with the small circles which describe the finite strain field, suggesting that the same kinematics is at work at present (Jolivet et al. 1994a). The present-day velocity field can be calculated by space geodetic measurements (Le Pichon et al. 1994). The first order of the velocity field is compatible with a rigid extrusion of the Anatolian block about a pole located in the southeastern corner of the Mediterranean (Fig. 5), i.e. south of the pole which describes the finite ductile strain in the Aegean. Though the general picture is similar the curvature of the ductile flow lines is stronger. Recent geodetic data clearly show a change in horizontal velocity across the Aegean Sea with the outer arc moving faster toward the southwest than the bulk of the Anatolian block. This
Fig. 20. Possible kinematic evolution of the Aegean domain and its relation to deformation. Dashed lines show kinematic flow lines (Le Pichon et al. 1994) and continuous lines show finite strain trajectories for the Miocene period (Jolivet et al. 1994a).
452
L. JOLIVET & M. PATRIAT
acceleration is mostly due to the extensional deformation now active in the Corinth Gulf region (Le Pichon et al. 1994) which is also the place where the maximum of seismicity is recorded. GPS measurements across the Corinth Gulf yield values of about 1 cm a -a (Rigo et al. 1996). The present-day direction of extension across the Corinth Gulf is N-S, i.e. parallel to what is fossilized in Crete. Directions of stretching are also recorded in the Menderes massif (Hetzel et al. 1995a,b), As shown on Fig. 2 a NNE trend is observed similar to that recorded in the central Cyclades. Their strike clearly departs from that expected from the small circles geometry and departs even more from the rigid rotation shown by space geodesy. As shown on Fig. 20 the strike of the North Anatolian Fault fits equally well the rigid rotation small circles and those of the finite strain. This suggests that the observed ductile deformation observed in the north (Thassos, Kavala, Tinos, Andros, Evia) is the direct image of the rigid rotation of the Anatolian Block. A two-component-process can be suggested: (1) rigid rotation of the Anatolian Block guided by the North Anatolian pole which probably started only in the Mid- or Late Miocene and (2) crustal spreading of the entire Aegean thickened crust toward the free boundary which probably started at least 25 Ma ago. If one considers the rate of extension observed in the Corinth Gulf region, 1 cm a -1 is probably a maximum for crustal spreading. The 3 cm a -1 accomodated by the rigid rotation dominate the geodetic data. An additional crustal spreading component should make the flow lines diverge toward the Hellenic trenches if it is still active at present. The quasi-absence of seismicity in the central Aegean domain suggests that extensional deformation is not presently active. Extensional deformation has apparently been recently localized along the northwest border of the Aegean domain by the propagation within the back-arc region of the North Anatolian Fault (Armijo et al. 1996). Given the data set presently available one can propose the following scheme to explain the observed finite strain as shown in the core complexes. (1) Crustal spreading started some 25 Ma ago (or earlier) allowed by the free edge along the subduction zone. This collapse was recorded in the whole Aegean domain while nappe stacking was still active along the convergence front. No available data can constrain the velocity of the trench retreat, but a value of I cm a -1 is reasonable by comparison with the present rate of
extension across the Corinth Gulf. On Fig. 20 the directions of stretching lineations were restored to their pre-rotation position (Morris & Anderson 1996): the general scheme still shows a change in the direction of stretching from N E - S W in the north to N-S in the south which is well describe by the small circles proposed earlier (Jolivet et al. 1994a). (2) The inception of the fast westward displacement of the A n a t o l i a n block some 12-15 Ma ago dramatically increased the rate of subduction south of the Aegean domain by a factor of 3. However, the internal deformation of the Aegean domain remained very similar to that of the Early Miocene because stress conditions along the subduction zone did not change with a free edge. Deformation progressively localized along the northwest boundary of the Aegean domain during the propagation of the North Anatolian Fault in the backarc region (Armijo et al. 1996). (3) More recently during the Pliocene the stress boundary conditions changed drastically along the subduction zone as attested by the change of the strike of 0-3 recorded in palaeostress tensors (Mercier et al. 1979) from N E - S W to N-S in the Aegean domain, and from N-S to E - W along the external arc (Crete) (Angelier et al. 1982; Armijo et al. 1992). Armijo et al. interpreted this change as evidence for the inception of collision between the African plate and the Aegean domain. This starting collision might have reduced the extensional component along the subduction zone. The authors express their thanks to their colleagues B. Goff6, J. P. Brun, P. Gautier, X. Le Pichon and T. Andersen for stimulating discussions during field work and in the lab before and during the production of the manuscript.
References ALTHERR, R., KREUZER, H., WENDT,I., LENZ, H., WAGNER,G. A., KELLER,J., HARRE,W. r162HOHNDORF,A. 1982. A Late Oligocene/Early Miocene high temperature belt in the anti-cycladic crystalline complex (SE Pelagonian, Greece). Geologisch Jahrbuch, 23, 97-164. ANDERSEN,T. B. & JAMI'VEI'[;B. 1990. Uplift of deep crust during orogenic extensional collapse: a model based on field studies in the Sogn-Sunnfjord region of Western Norway. Tectonics, 9, 1097-11t 1. --, OSMUNDSEN, P. T. & JOLIVET, L. 1994. Deep crustal fabric and a model for the extensional collapse of the southwest Norwegian Caledonides. Journal of Structural Geology, 16, 1191-1203. ANGELIER,J. 1984. Tectonic analysis of faultslip data sets. Journal of Geophysical Research, 89, 5835-5848.
D U C T I L E EXTENSION, A E G E A N SEA --,
LYBERIS, N., LE PICHON, X. BARRIER, E. &. HUCHON. P. 1982. The tectonic development of the Hellenic arc and the sea of Crete, a synthesis. Tectonophysics, 86, 159-196. ARMIJO, R., LYON-CAEN,U. & PAPANIKOLAOU,D. 1992. East-West extension and Holocene normal fault scarps in the Hellenic arc. Geology 20, 491-494. - - , MEYER, B., KING, G. C. R, RIGO, A. & PAPANASTASSIOU, D. 1996. Quaternary evolution of the Corinth Rift and its implications for the Late Cenozoic evolution of the Aegean. Geophysical Journal International 126, 11-53. AVIGAD, D. 1990. The geodynamic evolution of the Cycladic massif (Aegean Sea, Greece) - a contribution to the study of continental collision PhD thesis, Hebrew University. & GARFUNKEL, Z. 1991. Uplift and exhumation of high-pressure metamorphic terranes: the example of the Cycladic blueschists belt (Aegean Sea). Tectonophysics, 188, 357-372. - - , MATTHEWS,A. EVANS, B. W. & GARFUNKEL, Z. 1992. Cooling during the exhumation of a blueschist terrane: Sifnos (Cyclades, Greece). European Journal of Mineralogy, 4, 619-634. BASSIAS,Y. & TRIBOULET,C. 1994. Tectono-metamorphic evolution of bleschist formations in the Peloponnesus (Parnon and Taygetos Massifs, Greece): a model of nappe stacking during Tertiary orogenesis. Journal of Geology, 102, 697-708. BAVAY,P. & ROMAIN-BAVAY,D. 1980. L'unitO de StyraOchi, un ensemble mOtamorphique de type schiste bleu d'~ge alpin dans le massif d'Attique-Cyclades (Eubde du Sud, GrOce). Thbse, Universit6 Paris Sud. BERCKHEMER, H. 1977. Some aspects of the evolution of marginal seas deduced from observations in the Aegean region. In: International Symposium on Structural History of the Mediterranean Basins, Technip, Paris, 303-314. BLAKE, M. C., BONNEAU,M. GEYSSANT,J. KIENAST,J. R. LEPVRIER, C. MALUSKI, H. & PAPANIKOLAOU,D. 1981. A geological reconnaissance of the Cyclacic blueschist belt, Greece. Geological Society of America Bulletin, 92, 247-254. BONNEAU, M. 1982. Evolution gdodynamique de l'arc 6g6en depuis le Jurassique Supdrieur jusqu'au Mioc6ne. Bulletin de la Socidtd GOologique de France 7, 229-242. 1984. Correlation of the Hellenic nappes in the south-east Aegean and their tectonic reconstruction. In: DIXON, J. E. & ROBERTSON,A. H. E (eds) The Geological Evolution of the Eastern Mediterranean, Geological Society, London, Special Publications 17, 517-527. KIENAST, J. R. 1982. Subduction, collision et schistes bleus: exemple de l'Egde, Grace. Bulletin de la SociOt~ Gdologique de France, 7, 785-791. BRAUNMILLER, J. & NABELEK, J. 1996. Geometry of continental normal faults: seismological constraints. Journal of Geophysical Research. 101, 3045-3052. BROCKER, M. 1990. Blueschist-to-greenschist transition in metabasites from Tinos island, Cyclade, -
-
-
-
-
-
&
453
Greece: compositional control or fluid infiltration. Lithos, 25, 25-39. --, KREUZER, H. MATTHEWS, A. & OKRUSCH, M. 1993. 4~ and oxygen isotope studies of polymetamorphism from Tinos island, Cycladic blueschist belt, Greece. Journal of Metamorphic Geology. 11, 223-240. BUICK, I. S. 1991. Mylonite fabric development on Naxos, Greece. Journal of Structural Geology 13, 643-655. & HOLLAND,T, J. B. 1989. The P - T - t path associated with crustal extension, Naxos, Cyclades, Greece, In: DALY, Ji S. (ed.) Evolution of metamorphic belts. Geological Society, London, Special Publications 43, 365-369. DANIEL, J. M. & JOLIVET, L. 1995. Interaction of detachments and granitic plutons during extension in the Tyrrhenian Sea (Elba island). Bulletin de Soci~td Gdologique de France, 166, 341-354. --, JOLIVET, L. GOFFE, B. & POINSSOT, C. 1996. Crustal-scale strain partitionning: footwall deformation below the Alpine Corsica Oligo-Miocene detachment. Journal of Structural Geology, 18, 41-59. DEWEY, J. E & SENGOR,A. M. C. 1979. Aegean and surrounding regions: complex multiplate and continuous tectonics in a convergent zone. Geological Society of America Bulletin. 90, 84-92. --, RYAN, P. D. & ANDERSEN, T. B. 1993. Orogenic uplift and collapse, crustal thickness, fabrics and metamorphic phase changes: the role of eclogites. In: PRICHARD,H. M.,ALABASTER,T., HARRIS,N. B. W. & NEARY, C. R. (eds) Magmatic Processes and Plate Tectonics'. Geological Society, London, Special Publications, 76, 325-345. DINTER, D. A. & ROYDEN, L. 1993. Late Cenozoic extension in northeastern Greece: Strymon valley detachment system and Rhodope metamorphic core complex. Geology, 21, 45-48. EYIDOGAN, H. & JACKSON, J. 1985. A seismological study of normal faulting in the Demirci, Alasehir and Gediz earthquakes of 1969-1970 in western Turkey: implications for the nature and geometry of deformation in the continental crust. Geophysical Journal of the Royal Astronomical Society, 81, 569--607. FASSOULAS, C., KILIAS, A. & MOUNTRAKIS,D. 1994a. Postnappe stacking extension and exhumation of high-pressure/low-temperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. ,- - & 1994b. Postnappe stacking extension and exhumation of high-pressure/lowtemperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. FAURE, M. & BONNEAU, M. 1988. Donndes nouvelles sur l'extension n6og6ne de l'Eg6e: la d6formation ductile du granite mioc6ne de Mykonos (Cyclades, Gr6ce). Comptes Rendue de l'Acadamie des Sciences, Paris, 307, 1553-1559. & PONS, J. 1991. Ductile deformation and syntectonic granite emplacement during the late Miocene extension of the Aegean (Greece). Bulletin de la SocidtO GOologique de France, 162, 3-12. GANS, P. B. 1987. An open system, two layer crustal -
-
454
L. JOLIVET & M. PATRIAT
stretching model for the eastern Great Basin. Tectonics, 6, 1-12. GAUTIER,P. 1994. GOomOtriecrustale et cinOmatique de l'extension tardi-orogOnique dans le domaine centre-OgOen (iles des Cyclades et d'EubOe, GrOce), -
Thesis, Universit6 de Rennes Univ. & BRUN, J. P. 1994a. Crustal-scale geometry and kinematics of late-orogenic extension in the central Aegean (Cyclades and Evvia island). Tectonophysics, 238, 399-424. 1994b. Ductile crust exhumation and extensional detachments in the central Aegean (Cyclades and Evvia islands). Geodinamica Acta, 7, 57-85. & JOLIVET,L. 1993. Structure and kinematics of upper Cenozoic extensional detachement on Naxos and Paros (Cyclades Islands, Greece). Tectonics, 12, 1180-1194. GODFR1AUX,Y. 1965. Etude gOologique de la rOgion de l'Olympe. ThGse d'Etat, Universit6 de Lille Univ. & RIcou, L. E. 1991. Direction et sens de transport associds au charriage synmGtamorphe sur l'Olympe. Bulletin of the Geological Society of Greece, 25, 207-229. HATZFELD, D., BESNARD, M., MAKROPOULOS, K. & HATZIDIMITRIOU,P. 1993. Microearthquake seismicity and fault-plane solutions in the southern Aegean and its geodynamic implications. Geophysical Journal International, 115, 799-818. HETZEL, R., PASSCHIER,C. W., RING, U. & DORA, O. 0. 1995a. Bivergent extension in orogenic belts: the Menderes massif (southwestern Turkey). Geology, 23, 455-458. --, RING, U., AKAL, A. & TROESCH, M. 1995b. Miocene NNE-directed extensional unroofing in the Menderes massif, southwestern Turkey. Journal of the Geological Society, London, 152, 639-654. HODGES,K. V., BURCHFIEL,B. C., ROYDEN,L. H., CHEN, Z. & LIE, Y. 1993. The metamorphic signature of contemporaneous extension and shortening in the central Himalayan orogen: data from the Nyalam transect, southern Tibet. Journal of Metamorphic Geology, 11, 721-737. JACKSON, J. 1994. Active tectonics of the Aegean region. Annual Review of Earth and Planetary Science, 22, 239-271. 1987. Active normal faulting and continental extension. In: ROBERTS, A. M. ET AL. (eds) Geometry of Normal Faults, Geological Society, London, Special Publications. 28, 3-18. & WHITE, N. J. 1989. Normal fauting in the upper continental crust: observations from regions of active extension. Journal of Structural Geolology, 11, 15-36. JACOBSHAGEN,V., DIJRR, S., KOCKEL, E, KoPP, K. O., KOWALCZYK,G., BERCKHEMER,H. & B ~TTNER,D. 1978. Structure and geodynamic evolution of the Aegean region. In: CLOOS, H., ROEDER, D. & SCHMIDT, K. (eds) Alps, Apennines, Hellenides. IUGG, Stuttgart, Reports, 38, 537-564. JANSEN, J. B. H. 1977. Metamorphism on Naxos, Greece. Thesis, Utrecht University. JOLIVET, L., BRUN,J. P., GAUTIER,P., LALLEMANT,S. & -
-
-
&
-
-
-
-
-
-
PATRIAT,M. 1994a. 3-D kinematics of extension in the Aegean from the Early Miocene to the Present, insight from the ductile crust. Bulletin de la SociOtO GOologique de France, 165, 195-209. - - , DANIEL, J. M. & FOURNIER, M. 1991. Geometry and kinematics of ductile extension in alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. -- - , TRUFFERT,C. & GOFF14,B. 1994b. Exhumation of deep crustal metamorphic rocks and crustal extension in back-arc regions. Lithos, 33, 3-30. , GOFFI~, B., MONII~, P., TRUFFERT-LUXEY, C., PATRIAT, M. & BONNEAU, M. 1996. Miocene detachment in Crete and exhumation P-T-t paths of high pressure metamorphic rocks. Tectonics, 15, 112%1153. KERRICH, R., BECKINSALE,R. D. & DURHAM,L. J. 1977. The transition between deformation regimes dominated by intrercristalline diffusion and intracrystalline creep evaluated by oxygene isotope thermometry. Tectonophysics, 38, 241-257. KILIAS, A., FASSOULAS, H. & MOUNTRAKIS, D. 1994. Tertiary extension of continental crust and exhumation of Psiloritis "metamorphic core complex" in the central part of the Hellenic arc (Crete, Greece). Geologisches Rundshau, 83, 417-430. KING, G., OUYANG,Z., PAPADIMITRIOU,P., DESCHAMPS, A., GAGNEPAIN,A., HOUSEMAN, G., JACKSON,J., SOUFLEmS, C. & V m m u x , J. 1985. The evolution of the Gulf of Corinth (Greece) an aftershock study of the 1981 earthquake. Geophysical Journal of the Royal Astronomical Society, 80, 677-693. KISSEL, C. & LAJ, C. 1988. The Tertiary geodynamic evolution of the Aegean arc: a paleomagnetic reconstruction. Tectonophysics, 146, 183-201. LALLEMANT, S., TRUFFERT, C., JOLIVET, L., HENRY,P., CHAMOT-ROOKE,N. ~: VOOGD,B. D. 1994. Spatial transition from compression to extension in the western Mediterranean Ridge accretionnary complex. Tectonophysics, 234, 33-52. LE PICHON, X. & ANGEUER, J. 1981. The Hellenic arc and trench system: a key to the neotectonic evolution of the eastern mediterranean area. Philo-
sophical Transactions of the Royal Society, London, 300, 357-372. --,
CHAMOT-ROOKE,N., LALLEMANT,S. L., NOOMEN, R. & VE~S,G. 1994. Geodetic determination of the kinematics of Central Greece with respect to Europe: implications for eastern Mediterranean tectonics. Journal of Geophysical Research, 100, 12675-12690. LEE, J. & LISTER, G.S. 1992. Late Miocene ductile extension and detachment faulting, Mykonos, Greece. Geology, 20, 121-124. LISTER, G. S. & DAVIS, G. A. 1989. The origin of metamorphic core complexes and detachment faults formed during Tertiary continental extension in the northern Colorado River region, U.S.A. Journal of Structural Geology, 11, 65-94. -& RAOUZAIOS,A. 1996. The tectonic significance
D U C T I L E EXTENSION, A E G E A N SEA
-
of a porphyroblastic blueschist facies overprint during Alpine orogenesis: Sifnos, Aegean Sea, Greece. Journal of Structural Geology, 18, 1417-1436. - - , BANOA, G. & FEENSTRA,A. 1984. Metamorphic core complexes of cordilleran type in the Cyclades, Aegean Sea, Greece. Geology, 12, 221-225. LYBERIS,N., ANGELIER, J., BARRIER, E. ~: LALLEMANT, S. 1982. Active deformation of a segment of arc: the strait of Kythira, Hellenic arc, Greece. Journal of Structural Geology, 4, 299-311. MAKRIS, J. 1978. The crust and upper mantle of the Aegean region from deep seismic sounding. Tectonophysics, 46, 269-284. MASCLE, J. & MARTIN, L. 1990. Shallow structure and recent evolution of the Aegean Sea: a synthesis based on continuous reflection profiles. Marine Geology, 94, 271-299. MCKENZIE, D. 1972. Active tectonics in the Mediterranean region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions. -
Geophysical Journal of the Royal Astronomical Society, 55, 217-254. MELIDOMS, M. G. 1980. The geology of Greece: the geological structure and mineral deposits of Tinos island (Cyclades, Greece). Institute of Geological Mineral Exploration Athens, 13, 1-80. MERCIER, J. L., ARMIJO, R., TAPPONNIER, P., CAREYGAILHARDIS, E. &; LIN, H. T. 1987. Change from Late Tertiary compression to Quaternary extension in southern Tibet during the India-Asia collision. Tectonics, 6, 275-304. --, DELIBASSIS, N., GAUTHIER, A., JARRIGE, J. J., LEMEILLE,E, PHILIP,H., SI2BRIER,M. & SOREL, D. 1979. La ndotectonique de l'arc 6gden. Revues de
GOologie Dynamique et de Geographie Physique, 21, 67-92. MOLNAR, P. & CHEN,W. P. 1983. Focal depths and fault plane solutions of earthquakes under the Tibetan plateau. Journal of Geophysical Research, 88, 1180-1196. MORRIS, A. & ANDERSON, A. 1996. First paleaomagnetic results from the Cyclaic Massif, Greece, and their implications for Miocene extension directions and tectonic models in the Aegean. Earth and Planetary Science Letters, 142, OKRUSCH, M. & BROCKER, M. 1990. Eclogites associated with high-grade blueschists in the Cyclades archipelago, Greece: a review. European Journal of Mineralogy, 2, 451-478. PAPANIKOLAOU,D. J. 1977. Contribution to the geology of the Aegean Sea: the island of Andros. Annale de GeoIogie de Pays Hellenique, 29, 477-553. , LYKOUSIS,V., CHRONIS,G. & PAVLAKIS,P. 1988. A comparative study of neotectonic basins across the Hellenic arc: the Messiniakos, Argolikos, Saronikos and Southern Evoikos gulfs. Basin Research, 1, 167-176. PATRIAT,M. 1996. Etude de la transition cassnat-ductile
en extension, application au transect OlympeNaxos, GrOce. Th6se de Doctorat, Universit6 Pierre et Marie Curie.
455
REINECKE,T. 1982. Cymrite and celsian in manganeserich metamorphic rocks from Andros island / Greece. Contributions to Mineralogy and Petrology, 79, 333-336. RIGO, A., LYON-CAEN,H., ARMIJO, R., DESCHAMPS,A., HATZFELD, D., MAKROPOULOS, K., PAPADIMITRIOU,P. & KASSARAS,I. 1996. A microseismicity study in the western part of the Gulf of Corinth (Greece): implications for large-scale normal faulting mechanisms. Geophysical Journal International, 126, 663-688. RUPPEL, C., ROYDEN, L. & HODGES, K. V. 1988. Thermal modelling of extensional tectonics: application to pressure-temperature-time history of metamorphic rocks. Tectonics', 7, 947-957. SCHERMER, E. R. 1990. Mechanism of blueschist creation and preservation in a A-type subduction zone, Mount Olympos region, Greece. Geology, 18, 1130-1133. 1993. Geometry and kinematics of continental basement deformation during the Alpine orogeny, Mt. Olympos region, Greece. Journal of Structural Geology, 15, 571-591. - - , Lux, D. R. & BURCHFIEL,B. C. 1990. Temperature-time history of subducted continental crust, Mount Olympos region, Greece. Tectonics, 9, 1165-1195. SEIDEL, E., KREUZER, H. & HARRE, W. 1982. The late Oligocene/early Miocene high pressure in the external hellenides. Geologische Jahrbuch, E23, 165-206. SIBSON,R. H. 1983. Continental fault structure and the shallow earthquake source. Journal of the Geological Society, London, 140, 741-767. SOKOUTIS, D., BRUN, J. P., DRIESSCHE, J. V. D. & PAVLIDES, S. 1993. A major Oligo-Miocene detachment in southern Rhodope controlling north Aegean extension. Journal of the Geological Society, London, 15ti, 243-246. SPAKMAN,W., VAN DER LEE, S. & VAN DER HILST, R. 1993. Travel-time tomography of the EuropeanMediterranean mantle. Physics of the Earth and Planetary Interiors, 79, 3-74. TAYMAZ,Y., JACKSON,J. & MCKENZIE, D. 1991. Active tectonics of the north and central Aegean Sea. Geophysical Journal International, 106, 433-490. THEYE, T. & SEIDEL, E. 1991. Petrology of low grade high pressure metapelites from the external hellenides (Crete, Peloponese), a case study with attention to sodic minerals. European Journal of Mineralogy, 3, 343-366. URAI, J. L., SHUILING, R. D. & JANSEN, J. B. H. 1990. Alpine deformation on Naxos (Greece). In: KNIPE, R. J. & RUTTER, E. H. Deformation Mechanisms, Rheology and Tectonics. Geological Society, London, Special Publications, 54, 509-522. VANDENBERG, L. C. & LISTER, G. S. 1996. Structural analysis of basement tectonics from the Aegean metamorphic core complex of Ios, Cyclades, Greece. Journal of Structural Geology, 18, 1437-1454. VOLL, G. 1976. Recristallization of quartz, biotite and feldspar from Erstfeld to the Levantina Nappe, -
-
456
L. JOLIVET & M. PATRIAT Swiss Alps, and
its geological significance.
Schweizerische Mineralogische und graphiche Mitteilungen, 56, 641-647.
Petro-
WERNICKE, B. 1981. Low-angle normal faults in the Basin and Range province: nappe tectonics in an extending orogen. Nature, 291, 645-648. 1992. Cenozoic extensional tectonics of the U.S. cordillera. In: BURCHFIEI.,B. C., LIPMAN, P. W. c~z ZOBACK, M. L. (eds) The Cordilleran Orogen: Conterminous U.S., Geological Society of America, Geology of North America, G3, 553-581.
-
-
WIJBRANS,J. R. & McDOUGALL, I. 1988. Metamorphic evolution of the Attic Cycladic Metamorphic Belt on Naxos (Cyclades, Greece) utilizing 4~ age spectrum measurements. Journal of Metamorphic Geology, 6, 571-594. --, VAN WEES, J. D., STEPHENSON, R. A. & CLOETmNGn, S. A. R L. 1993. Pressure-temperature-time evolution of the high-pressure metamorphic complex of Sifnos, Greece. Geology, 21, 443-446.
New insights from 4~ laserprobe dating of white mica fabrics from the Pelion Massif, Pelagonian Zone, Internal Hellenides, Greece: implications for the timing of metamorphic episodes and tectonic events in the Aegean region A . L. W. L I P S 1'2, J. R. W I J B R A N S 2 & S. H . W H I T E 1
1Geodynamics Research Institute, Faculty o f Earth Sciences, Utrecht University, P O B o x 80.021, 3508 TA, Utrecht, the Netherlands' (e-mail." [email protected]) 2Faculty o f Earth Sciences, Vrije Universiteit, Amsterdam, the Netherlands Abstract: 4~
laserprobe dating of mylonites fabrics from the Pelion Massif in the Pelagonian Zone of mainland Greece has characterized its Mid-Late Alpine deformation history. Following high pressure (HP) metamorphism, ductile deformation occurred under greenschist-facies conditions from c. 54 Ma, and continued to affect the Pelion Massif until c. 15 Ma. The prolonged episode of ductile deformation in the Pelion Massif has resulted in the formation of an Oligocene-Early Miocene ductile domal structure. The new geochronological data obtained for the Pelion contribute to a detailed record of the Alpine kinematic history in the Pelagonian Zone and allow a discussion of P-T-t data from Aegean HP rocks to characterize the regional thermotectonic history. Comparison with the P-T-t data from the Cycladic region reinforces the point that Mid-Eocene phengite ages, commonly taken as the age of peak HP metamorphism in the Cyclades, do not always reflect the metamorphic culmination, but rather the retrograde paths of the HP rocks. It is shown that, on a regional scale, termination of HP metamorphism is a diachronous process in the Aegean region, being c. 54 Ma in the north (Pelagonian Zone) and shifting to younger ages, chiefly c. 40 Ma in the Cyclades, and c. 20 Ma on Crete, as the present-day subduction zone is approached. In contrast to the diachronous exhumation of Aegean HP assemblages, the well documented Miocene phase of ductile regional extension appears to be synchronous across the whole Aegean region and affected basement rocks until c. 15 Ma. The Hellenic Orogen in the Aegean region (Fig. 1) contains a high-pressure, low-temperature, (HP) metamorphic belt which occurs at the interface of the largely u n m e t a m o r p h o s e d External Hellenides in the (south)west and the amphibolite grade metamorphic rocks of the I n t e r n a l Hellenides in the (north)east (e.g. Blake et al. 1981; Seidel et aL 1982). In the northern part of the Aegean region, the H P metamorphic belt crops out in the Pelagonian Zone, where it runs semi-continuously along the eastern coast of the Greek mainland (Fig. la), and has been thrusted over weakly metamorphosed rocks, which are exposed in the Olympos and Ossa tectonic windows (Fig. lb) (e.g. Godfriaux 1970; Derycke & Godfriaux 1976). In the southern part of the A e g e a n region, the belt crops out in the Cyclades Archipelago, where it changes its overall strike from northwest-southeast to roughly east-west and can be traced further to the east into western Turkey. South and southwest of the Cyclades and closest to the present-day subduction zone, high-pressure metamorphic sequences have been documented from the Peloponese and Crete.
The H P metamorphic belt in the Hellenides reflects the temporal and spatial evolution of the convergent tectonics between the African and E u r a s i a n plates. A l t h o u g h characterized by overall convergence, involving the subduction of both oceanic and continental crust, much of the Late Alpine tectonic history of the belt is dominated by regional extension (e.g. Jolivet et al. 1994). Since the exact timing of peak H P metamorphism and the subsequent uplift and exhumation of H P assemblages is an integral part of the understanding of the orogenic history of the Hellenides, a combination of detailed structural/petrological studies and sensitive dating techniques is required to characterize the metamorphic and deformational events that affected the polydeformed H P belt. The application of the 4~ laserprobe technique not only allows dating of individual grains, using the incremental heating technique (e.g. Wijbrans et al. 1990), but also allows the spot dating of individual crystals to characterize the resetting of the argon isotopic system during recrystallization (e.g. Scaillet et al. 1990) and to investigate
Lies, A. L. W., WIJBRANS,J. R. & W~ ITE. S. H. 1999. New insights from 4~ laserprobe dating of white mica fabrics from the Pelion Massif, Pelagonian Zone, Internal Hellenides, Greece: implications for the timing of metamorphic episodes and tectonic events in the Aegean region. In: DURAND,B., JOLIVET,L., HORVATH,E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 457-474.
458
A.L.W. LIPS E T A L .
Fig. 1. (a) Locations of exposed HP assemblages (dots) in the circum-Aegean region. (b) Simplified geological map of the Pelagonian Zone (modified after IGME 1983), showing 4~ sample locations (numbered) and observed directions of tectonic transport.
the argon gradients in single crystals (e.g. Reddy et al. 1996). This paper demonstrates that the high sensitivity and low blank characteristics of the laserprobe dating is a significant advance in the dating of multiple tectonothermal events. Numerous petrological studies in the Cyclades over the past decades provide a good coverage of data constraining the petrological evolution of the HP rocks in the central-southern part of the Aegean region (e.g. van der Maar & Jansen 1983; Okrusch et al. 1984; Ridley 1984; Schliestedt & Matthews 1987; Buick & Holland 1989; Okrusch & Br6cker 1990; Avigad et al. 1992; Franz & Okrusch 1992). Pressure and temperature estimates of the HP metamorphism range from blueschist-facies conditions (7-9 kbar, c. 400-450~ to eclogite-facies conditions (14-17 kbar, c. 450-500~ Tectonic studies in the Cyclades suggest that the HP rocks are part of a composite stack of thrust sheets that have
experienced variable PT histories (e.g. Avigad 1993; Vandenberg & Lister 1996). In the past, several studies have concentrated on the geochronology of the HP rocks in various parts of the Cyclades by the 4~ dating technique (Maluski et al. 1981, 1987; Wijbrans & McDougall 1986, 1988; Wijbrans et al. 1990; Br6cker et al. 1993; Raouzaios et al. 1996). The peak of HP metamorphism is dominantly interpreted as being Mid-Eocene (c. 40-45 Ma) in age, which was established by early K - A r and Rb-Sr studies (e.g. Altherr et al. 1979; Andriessen et al. 1979). Although older ages have been reported (Maluski et al. 1987) and recent work has argued that the culmination of HP metamorphism in the Cyclades is not directly constrained by the published geochronological data (Wijbrans et al. 1990, 1993; Raouzaios et al. 1996), many papers have continued to use the Mid-Eocene ages for the
TIMING OF METAMORPHISM & TECTONISM, AEGEAN peak of HP metamorphism in the Cycladic area (e.g. Avigad et al. 1992; Avigad 1993; Gautier & Brun 1994; Vandenberg & Lister 1996). In contrast to the extensive literature published on the HP rocks of the Cyclades, fewer publications exist on the northern and southern parts of the HP belt. Petrological and geochronological work on Crete, south of the Cycladic region, has resulted in Oligocene to Early Miocene ages (i.e. pre-24 Ma to c. 20 Ma) for the HP metamorphism (Seidel et al. 1982; Triboulet & Bassias 1988; Skarpelis 1989; Theye et al. 1992; Fassoulas et al. 1994; Jolivet et al. 1996). The results of an extensive study around the Olympos tectonic window in the northern Pelagonian Zone (Fig. 1) have been presented by Schermer (1990, 1993) and Schermer et al. (1990), following first recognition of high-pressure assemblages in the 1970s (e.g. Derycke & Godfriaux 1976). Schermer (1993) established that peak metamorphic pressure and temperature conditions for HP metamorphism in the Olympos region occurred at c. 5-8 kbar and c. 300-350~ Conventional 4~ age dating on bulk samples (c. 100 mg) resulted in dominantly Palaeocene (53-61 Ma) ages for the peak of the HP metamorphism in the Olympos Massif. The P - T conditions of HP metamorphism and its timing in the Pelagonian Zone have been further refined after detailed structural, microstructural and petrological studies, combined with 4~ laserprobe experiments on mica separates (c. 0.5 mg) from selected samples of the Ossa Massif (Lips et al. 1998), which had been subject of a limited investigation by Schermer et al. (1990) and is similar in style to the Olympos Massif (Schermer et al. 1990; Schermer 1993). The occurrence of either lawsonite bearing blueschists or epidote bearing blueschists, depending on host lithology, in the Ossa Massif (Lips et al. 1998) further constrain the metamorphic conditions in the area to 7-8 kbar and 320-360~ for the Pelagonian HP event ( P - T estimates following Evans 1990). Laserprobe dating on samples from the Ossa Massif has constrained the episode of HP metamorphism to a period from pre-85 Ma to c. 54 Ma (Lips et al. 1998), following final cooling of the basement rocks to c. 350 _ 30~ (taken as average closure temperature range of argon in white mica) by c. 100 Ma. The local identification of lawsonite porphyroblasts in Tertiary flysch (first reported by Derycke & Godfriaux 1978), indicates that at least parts of the flysch sequence, which forms part of the autochthonous Olympos-Ossa sequence has experienced elevated pressures and low temperatures by c. 45 Ma ( P - T estimates in Lips et al. 1998). It is
459
emphasized that the recognized lawsonite occurrences by Dercyke & Godfriaux (1978) have been reported for flysch thrust slices in the metamorphic sequence, which originally were interpreted to have a slightly older age (c. Palaeocene) than the autochthonous flysch with an estimated Lutetian-Priabonian biostratigraphical age (Derycke & Godfriaux 1978). We present the results of an integrated structural, petrological, and 4~ laserprobe study in the Pelion Massif and adjacent areas (Fig. lb) of the southern Pelagonian Zone. Here, basement rocks have been affected by blueschist-facies metamorphism and have been severely overprinted by greenschist-facies deformational fabrics (Walcott & White 1998). A more intense greenschist facies ductile overprint in the Pelion Massif suggests that ductile deformation continued in the Pelion Massif after it had terminated in the Olympos-Ossa Massifs. This might give insight into the timing of ductile fabric forming events at younger Alpine times. Careful application of 4~ laserprobe dating to structurally well defined white micas will help to constrain the timing of the fabric forming events during the ductile deformation and the possible temporal difference in the observed directions of tectonic transport in the Pelion Massif. Additionally, the integration of the 4~ laserprobe and kinematic data for the Pelion Massif, with the results of Schermer et al. (1990) and Lips et al. (1998) for the Ossa-Olympos region will produce a more complete overview of the thermotectonic evolution of the Pelagonian Zone. This in turn provides a platform to re-evaluate the thermotectonic evolution for the Aegean region as a whole.
Structural setting of the Pelion Massif The basement sequence in the Pelion Massif has been studied in the past (e.g. Ferriere 1976; Jacobshagen & Wallbrecher 1984) and consists of gneisses and schists of felsic to intermediate compositions, tectonically interlayered with marble and serpentinite units (see also Walcott & White 1998). In contrast to a well-defined lithotectonostratigraphy for the Olympos and Ossa Massifs (Schermer 1993, following, e.g., Ferriere et al. 1988; Godfriaux et al. 1988), the sequence of rocks in the Pelion Massif occurs in a regionally mylonitized interval in which the lithologies have been severely intermixed tectonically to produce a mylonitic melange. This mylonitized interval might be correlated with the lower structural levels of the metamorphic sequence in the Ossa Massif (Ambelakia and lower Pierien units, following terminology by
460
A.L.W. LIPS ETAL.
Schermer 1993, after Schmitt 1983), albeit that in the Pelion Massif the sequence is drastically thinned and partly d i s m e m b e r e d and no units at structurally lower positions than the relict blueschists are exposed (e.g. potential existence of an a u t o c h t h o n o u s series as observed in the O s s a - O l y m p o s Massifs). R e m n a n t s of blueschist-facies assemblages have b e e n observed in the lower structural levels and are characterized by d o m i n a n t l y e p i d o t e - a l b i t e blueschists and rare l a w s o n i t e - a l b i t e blueschists, depending on parent lithologies. As concluded for the Ossa Massif, these blueschist rocks constrain the peak metamorphic pressure and temperature conditions to c. 7-8 kbar and 320-360~ respectively (following Evans 1990). The mylonitic melange in the Pelion Massif is characterized by a strong greenschist-facies overprint, which largely obliterates the blueschist-facies fabrics. The tectonic characteristics of the Pelion Massif are basically similar to those of the Ossa and Olympos Massifs (Schermer 1993; Lips et al. 1998) and are extensively discussed by Walcott & White (1998). In the intensely d e f o r m e d sequence of rocks of the Pelion Massif a welldeveloped extension lineation is often observed. It parallels mineral lineations defined by aligned (blue or green) amphiboles and is characterized by dominantly stretched recrystallized quartz aggregates or stretched albite crystals. Where extension lineations are well developed, kinematic indicators such as shear bands or asymmetric strain shadows around rigid clasts have been used to obtain a direction of tectonic transport which indicates both a top-to-the-northeast and a top-to-the-southwest sense of shear (Figs 1 & 5). Samples with well-developed mica bearing mylonitic foliations and clear kinematics were selected for micro-structural analyses and for the characterization of the white mica populations as a prelude to 4~ laserprobe experiments to directly date the fabric forming events, an approach which was successful for a similar setting in the Ossa Massif (Lips et al. 1998). Samples were also collected from the eastern and western margins of the Pelagonian Zone (most likely to be correlated with levels of the Pierien unit); in the Antihassia Mountains, near the village of Longa (Fig.1 loc. 1), and from the island of Skiathos, near Troulos (Fig.1 loc. 11), because the margins appear to have been totally overprinted by greenschist-facies metamorphism or, alternatively, have escaped blueschist-facies m e t a m o r p h i s m . Sample locations and their deduced directions of tectonic transport are indicated in Fig. 1. After 4~ laserprobe experiments, the kinematic data and 4~
fabric ages are combined to characterize the temporal and regional distribution of the directions of tectonic transport.
4~
laserprobe dating
Analytical techniques The selection and separation of white mica single crystals and multigrain populations for 4~ laserprobe analyses have been carried out as described by Lips et al. (1998). In summary, different white mica generations were identified and characterized in the field and, subsequently, in thin section by optical microscopy and SEM studies. The metamorphic assemblage of the mylonitic fabric was determined in thin section and the mica generations were selected. In the present study a maximum of two mica generations were separated for every sample. Large sized muscovite porphyroclasts in mylonites were used to deduce the protolith thermal histories. After establishing these, recrystallized white micas (phengitic in composition), associated with the mylonitic foliation were selected to date the fabric development of the mylonite. The muscovite porphyroclasts have an average diameter of c. 500-800 pm, the largest recrystallized white micas in the mylonitic foliation have average diameters which range from c. 100 pm to c. 500 pm. After selection, the samples were crushed and then sieved firstly into 250 pm size intervals. Subsequently, the micas were separated in these sieve intervals using a Faul vibration table and sieved a second time into narrower 50 pm size intervals. Before final handpicking, some samples were further separated using a Frantz magnetic separator and heavy liquids to minimise any contribution from a multigeneration white mica population in the final mica separate. 4~ laserprobe analyses and data regression techniques were carried out as described by Lips et al. (1998). It is essential that contamination with fragments of micas belonging to other populations is minimized. To achieve this, we decided to work with the smallest possible amount of material (i.e. less than c. 0.5 nag). A side effect is that some low temperature steps of the stepwise heating experiments contained very little argon and hence, show rather large uncertainties.
Results o f 4~ laserprobe dating on samples" f r o m the Pelion Massif The age spectra of the 4~ laserprobe experiments on samples from the locations are presented in Fig. 2 with a two sigma analytical error. Age spectra from muscovite porphyroclasts and from mylonitic fabrics or spectra from the same sampling area have been combined in single diagrams. The relevance of the obtained ages is further outlined in Table 1 and presented in Fig. 4. To arrive at the initial regional thermal conditions, the 4~ laserprobe ages which
TIMING OF METAMORPHISM & TECTONISM, AEGEAN were obtained from the muscovite porphyroclasts have been combined with published data from Schermer et al. (1990) and Lips et aL (1998) (Fig. 3). These results show a regional preservation of pre-Alpine and Early Alpine cooling ages, which cluster around 280-300 Ma and 100-120 Ma, with the local occurrence of some intermediate ages. The regional preservation of a pre-Tertiary argon isotopic signature in the porphyroclasts indicates that subsequent metamorphic temperatures were low enough to preserve the 4~ isotopic signatures in white mica of the Alpine deformational fabrics. The preservation of these signatures permits the direct dating of the mylonite forming events by 4~ analyses of recrystallized white micas defining the mylonitic foliations, as long as the preservation of the argon isotopic signature in white mica is not obscured by excess argon (e.g. due to crystallization of white mica in an oversaturated Ar environment), and/or by argon loss (e.g. due to diffusion loss from the fabric micas with a, dominantly, slightly finer grainsize; chemical alteration, subsequent recrystallization etc.). The 143.1 + 4.2 Ma age from the mylonitic fabric of the Mavrovouni Massif (Fig. 2, loc. 2) is concluded to be a cooling age, as the result is not significantly younger than the age obtained from muscovite porphyroclasts of the same sample (149.9 + 5.4 Ma). Fabric ages from recrystallized micas from Antihassia (Fig. 2, loc. 1) and Skiathos (Fig. 2, loc. 11) both fall in the range around 90 Ma. All ages from the Pelion Massif were obtained from 4~ analyses of mylonitic foliations, which formed under greenschist facies conditions, following older HP metamorphism, and have ages ranging from 54 to 15 Ma (Fig. 2, Table 1). The results tend to rule out the interpretation that the obtained fabric ages can be interpreted as cooling ages related to slightly lower temperatures (e.g.c. 350~ than the closure temperatures for the muscovite porphyroclasts (e.g.c. 400~ Such a scenario would have resulted, dominantly, in a homogeneous spread of data across the region, which is not reflected by the data, as several different fabric ages have been obtained from the same locations or nearby locations. It is shown in several spectra that some loss of argon has occurred in several mica populations (e.g. locations 9 & 10), which resulted in the produced lower ages in the first steps of the spectrum. This partial argon loss might be caused by diffusion of argon when the mica had been positioned for a long time interval at ambient temperatures, which still fall in the lower limits of the argon closure temperature range (e.g. 330 ~ It is more likely however that some fabric micas have
461
been affected by minor recrystallization or chemical alteration at younger times, which caused argon loss at the rims of the mica crystals. In general, the consistency of the fabric age data suggest that a reliable dataset has been obtained, which can be used to further constrain the tectonothermal history of the Pelion Massif. The greenschist mylonite fabric ages imply that blueschist facies metamorphism in the Pelion is older than 54 Ma. The mylonite ages have been combined with the established directions of tectonic transport to summarize the temporal distribution of the tectonic transport directions (Fig. 4, lower left panel). During the 54 Ma to 15 Ma interval existing mylonitic structures have been reactivated. This is best illustrated by the studied samples from location 9 (Fig. 2, Table 1; near the village of Neohori) by the 39 Ma top-tosouthwest directed mylonitic fabric which occurred in the same mylonitic interval as the 54 Ma top-to-northeast directed mylonitic fabric. In the field both mylonitic fabrics in the zone were interpreted to have formed as a result of a single phase of layer parallel stretching, but the 4~ laserprobe results show that they represent two separate events during which tectonic transport was reversed. The final stage of ductile deformation in the Pelion Massif is characterized by a protracted event, with dominantly layer parallel extension, from c. 40 Ma until 15 Ma (Fig. 4, lower left panel), producing a domal structure, after which it was cross-cut by highangle brittle faults.
Thermotectonic evolution of Pelagonian Zone The results of the current study together with those of Schermer et al. (1990) and Lips et al. (1998) provide an improved insight into the thermotectonic evolution of the Pelagonian Zone as a whole. The fabric forming events in the Pelion Massif and Olympos-Ossa Massifs are summarized in Fig. 4, and are discussed below in chronological order (from oldest to youngest). Figure 4 shows that the Pelagonian basement, which is currently exposed at the surface, had regionally cooled to c. 350 ___30~ by 100 Ma. The two dominant clusters of cooling ages, which occur in the Pelion Massif, had also been established in the Ossa Massif (Lips et al. 1998; Fig. 4 right panel) and are related to the relative position of individual thrust sheets in the tectonostratigraphic sequence (Fig. 5). In the Ossa Massif Hercynian porphyroclast cooling ages (c. 300 Ma) were derived from the basal thrust sheet of the allochthonous sequence and Cretaceous cooling ages (c. 110 Ma) were
462
A . L . W . LIPS E T A L .
TIMING OF M E T A M O R P H I S M & TECTONISM, A E G E A N
463
Table 1. Summary of kinematic analyses and 4~
laserprobe results of white mica generations from the Pelion Massif (P), Antihassia Mountains (A), Mavrovouni Massif (M) and Skiathos (S) Sample description
Analysed fabric
Transport direction
Age + 20(Ma)
A:I.
Greenschist mylonite
M:2.
Greenschist mylonite*t
M:3. P:4.
Greenschist* Greenschist mylonite
P:5.
P:7.
Greenschist mylonite Greenschist mylonite Greenschist mylonite Greenschist mylonitet Calc-mylonite Greenschist mylonite
P:8.
Greenschist mylonitew Greenschist mylonite
P:9.
Greenschist mylonite
P:10. S:11.
Greenschist Greenschist Greenschist Greenschist
Porphyroclast Mylonitic foliation Porphyroclast Mylonitic foliation Foliation Porphyroclast Mylonitic foliation Mylonitic foliation Mylonitic foliation Mylonitic foliation Mylonitic foliation Mylonitic foliation Porphyroclast Mylonitic foliation Mylonitic foliation Porphyroclast Mylonitic foliation Porphyroclast Mylonitic foliation Mylonitic foliation Mylonitic foliation Mylonitic foliation Porphyroclast Mylonitic foliation
SW SW Ambiguous Ambiguous NE NE NE NE NE SW NE NE SW SW NE Ambiguous
108.2 _+2.6 89.0 _+1.4 149.9 _+5.4 143.1 _+4.2 121.2 _+4.0 293.6 _+4.7 37.1 _ 1.1 21.5 _ 1.8 15.5 _ 3.2 34.9 _+2.6 15.5 _+2.0 293.1 + 3.5 20.7 + 2.7 26.8 + 4.1 304.8 + 4.9 95.3 + 3.7 53.8 _+5.9 39.1 _+1.6 39.4 _+1.5 32.3 _+1.25 218.2 + 5.8 92.9 _+2.7
Loc no.
P:6.
mylonite mylonite mylonite mylonite*
Comments Cooling age Fabric age Cooling age Cooling age Cooling age Cooling age Too fine grainedt Fabric age Fabric age Fabric age Fabric age Fabric age Cooling age Mixed age Fabric age Cooling age Too fine grained ~t Cooling age Fabric age Fabric age Fabric age Fabric age Cooling age Fabric age
Greenschists main constituents; quartz, albite, epidote, white mica, chlorite + actinolite + calcite; calcmylonite; calcite, quartz, white mica _+chlorite; *Sample contains relict garnet. tSample contains relict biotite. w contains relict glaucophane. *Sample contains blue-green amhibole. ~Fraction size below 63 pro.
d e r i v e d f r o m a t h r u s t sheet, w h i c h was in a relatively h i g h e r t e c t o n o s t r a t i g r a p h i c p o s i t i o n . I n c o n t r a s t to t h e p r o p o s e d t e c t o n o s t r a t i g r a p h i c a l d i s t r i b u t i o n in t h e O l y m p o s a n d O s s a Massifs, w h i c h has b e e n r e l a t e d to a c. 110 M a p h a s e of b a s e m e n t i m b r i c a t i o n , a n d was b a s e d o n t h e distribution of clusters of Hercynian (lower s t r u c t u r a l levels) v e r s u s C r e t a c e o u s ( h i g h e r s t r u c t u r a l levels) m u s c o v i t e c o o l i n g ages, it is concluded that the formation of a mylonitic
m e l a n g e u n d e r post-c. 54 M a g r e e n s c h i s t - f a c i e s conditions has strongly affected the tectonos t r a t i g r a p h i c a l o r d e r of t h e P e l i o n Massif, obliterating the potential regional correlation of t h e s e c h a r a c t e r i s t i c t e c t o n o s t r a t i g r a p h i c intervals (Fig. 5). T h e m y l o n i t e f o r m i n g events, as r e c o r d e d in t h e P e l a g o n i a n Z o n e , illustrate that t h e o l d e s t A l p i n e fabric f o r m i n g e v e n t , w h i c h is directly d a t a b l e by t h e a p p l i e d t e c h n i q u e , o c c u r r e d
Fig. 2. Age spectra obtained by 4~ laserprobe incremental heating experiments. All analytical errors are in a 20- interval. Location numbers as indicated in Fig. 1; significance of selected samples are outlined in Table 1. Spectra which have been combined in one diagram, represent: (i) analyses of two mica generations obtained from one single sample, i.e. porphyroclast cooling age versus mylonitic fabric age (e.g. loc. 1, loc. 2, loc. 11), (ii) analyses of mylonitic fabric ages from different samples from the same sample area (e.g. loc. 5, loc. 6), (iii) a combination of the porphyroclast cooling age versus mylonitic fabric age from one sample plus additional mylonitic fabric ages from samples of the same sample area (loc. 7; porphyroclast cooling age and 20.7 Ma total gas age from the same sample, 26.8 Ma fabric age from additional sample; loc. 9; porphyroclast cooling age and 53.7 Ma fabric age from the same sample, 39.1 and 39.4 Ma fabric ages from additional samples). Inset in age spectrum of location 9 presents close-up of the age spectra in the 15 to 60 Ma age interval over 0% to 100% of the cumulative 39Ar fraction).
464
A.L.W. LIPS E T A L .
Fig. 3. Simplified geological map (modified after IGME 1983) showing the spatial distribution of 4~ muscovite porphyroclast cooling ages.
around 85-90 Ma. It has been recorded in greenschist-facies mylonites (Antihassia and Skiathos) and blueschist facies mylonites (Ossa Massif). The tectonic significance of the conflicting tectonic transport directions, being regionally topto-southwest for the Antihassian sample and locally top-to-northeast for the Ossa sample, would suggest that transport reversal at deeper structural levels might have taken place. The youngest blueschist facies mylonitic fabrics in the Pelagonian Zone were found in the OssaOlympos region and cluster around 54 Ma (Schermer etal. 1990; Lips etaL 1998). The 54 Ma blueschist facies fabrics have formed during the same event as greenschist facies mylonitic fabrics which have been observed in the Ossa and Pelion Massif. The observations are that HP metamorphism in the Pelagonian basement sequence had terminated by 54 Ma. The observation of lawsonite bearing Eocene (allochthonous) flysch intercalations in the Ossa Massif (Derycke &
Godfriaux 1978) indicates that parts of the flysch sequence experienced elevated pressures during southwestward transport of the basement sequence (which already was, or had been, affected by HP metamorphism) across the autochthonous Olympos-Ossa sequence, which resulted in the present-day observed tectonostratigraphical order by c. 40 Ma (Fig. 5). After transport of the basement sequence across platform carbonates and phyllites in the Olympos-Ossa region by c. 45 Ma, relative uplift might have been accelerated by the rebound of the low density carbonates underlying the basement sequence. The result is the positioning of the currently exposed Olympos-Ossa sequence at (semi-)brittle crustal conditions by c. 36 Ma and, consequently, the absence of any ductile deformational events during Oligocene and Miocene times. The youngest recorded white mica ages in the Olympos region have been dated at 36-40 Ma (Schermer et al. 1990) and
TIMING OF METAMORPHISM & TECTONISM, A E G E A N
465
Fig. 4. Summary of the thermotectonic events in the Pelagonian Zone. Data on the left side, Pelion data (this study), have been used in conjunction with data from the Olympos-Ossa area (right side) from Schermer et al. (1990), and Lips et al. (1998). Sample frequency corresponds to ages obtained from individual samples. Thin vertical bars between age-axis and shaded horizontal bars (age domain), represent two sigma individual samples (data from this study, left side, and Lips et al. 1998, right side). Light shaded age domains represent cooling ages, dark shaded age domains represent fabric ages. Tectonic transport directions obtained from individual samples (Fig. 1, this study, and Lips et al. 1998) are presented; size of arrow represents number of samples with comparable kinematics.
466
A.L.W. LIPS E T A L .
Fig. 5. Cartoon illustrating different tectonic scenario within the Ossa Massif compared to the Pelion Massif at Miocene times. In blow-ups; structural position of obtained chronological data (Ossa data, biostratigraphic age ages, this study); after Derycke & Godfriaux 1978; 4~ ages after Lips et al. 1998: Pelion data; 4~ biostratigraphical age (italic); 4~ laserprobe white mica ages; porphyroclast cooling age (squares), blueschist fabrics (bold), greenschist fabrics (underlined). Star symbol in upper blow-up represents structural position of lawsonite bearing Tertairy flysch, after Derycke & Godfriaux (1978).
have been related by Schermer (1993) to a phase of top-to-southwest tectonic transport. In contrast to the termination of ductile deformation in the Ossa-Olympos region by c. 36-40 Ma, the basement sequence of the Pelion Massif experienced continued ductile deformation at greenschist facies conditions from c. 40 Ma until 15 Ma, with younger ductile fabrics located in the lower structural levels (Fig. 5). The continued greenschist facies deformation is reflected by the development of the mylonitic domal structure in the Pelion Massif and the contrasting morphology of the Olympos and Ossa domes (see Fig. 5 and Schermer et al. 1990).
Exhumation histories of Aegean high pressure sequences The P - T - t evolution of the Pelion Massif (star symbols in Fig. 6a) went through four phases as outlined above. They are, in summary: (1) cooling of basement rocks below c. 350 _+ 30~ under dominantly greenschist facies conditions (pressures of c. 4-6 kbar) by c. 100 Ma, (2) peak HP conditions of 7-8 kbar and 320-360~ before c. 54 Ma; with continued greenschist-facies metamorphism at shallower crustal levels (pressures of c. 3-5 kbar), at c. 90 Ma in Skiathos and Antihassia, (3) ductile deformation under, post HP, greenschist facies conditions (pressures of c. 3-6 kbar, temperatures below 350 + 30~
TIMING OF METAMORPHISM & TECTONISM, AEGEAN during which older fabrics were continuously overprinted and reworked, between c. 54 Ma and c. 15 Ma, (4) brittle deformation from 15 Ma to Present, which is not recorded by 4~ white mica dating. The results from the Ossa Massif (dots in Fig. 6a after Lips et al. 1998) further constrain the Early Alpine history by the recording of the onset of HP metamorphism in the Pelagonian region after 110 Ma, the 85 Ma HP mylonitic fabric and the 54" Ma deformational event which is related to the termination of blueschist-facies metamorphism. The Oligocene-Early Miocene ductile deformation in the Pelion Massif (15 and 21 Ma stars in Fig. 6a) shows the contrast to the recorded brittle deformation which affected the Olympos at the same time (squares in Fig. 6a, Schermer et al. 1990), where Early Miocene brittle deformation is illustrated by 16-23 Ma K-feldspar cooling ages, which indicate that temperatures were below c. 200 ___50~ (left column in Fig. 6a indicates the closure temperature range in K-feldspar). The extended P - T - t history of the Pelagonian Zone allows comparison with existing P - T - t data from the Cycladic region (Sifnos and Naxos island) and from Crete to place our results in a wider regional perspective. Before comparison of the retrograde paths from Crete, Sifnos and Naxos with the Alpine P - T - t path from the Pelagonian Zone, two theoretical P - T - t paths are presented (Fig. 6b) which will be discussed in detail. The theoretical P - T - t paths will help to clarify the role of the argon closure temperature range for white mica in the blueschist-greenschist transition domain on the obtained P - T - t information. The theoretical retrograde paths are presented for a HP metamorphic sequence which was affected by two hypothetical fabric forming events during the retrograde history (at temperatures around 350~ and 300~ respectively). The two theoretical retrograde paths start at HP metamorphic conditions of c. 14 kbar and 450~ (Fig. 6b, point I). As the metamorphic temperature at peak HP conditions are too high to preserve the argon isotopic signature in white mica, the oldest recorded 4~ white mica age will be a cooling age (Fig. 6b, points III, IV), obtained when temperatures drop below c. 350 _+ 30~ The P - T conditions which are linked to the cooling ages might be actually located in the blueschist field (Fig. 6b, point III) or in the greenschist field (Fig. 6b, point IV), dependent on the cooling path followed. The ambient metamorphic conditions at the time of closure of the argon system in white mica are, in both cases, at pressures and temperatures much lower than indicated by the observed peak metamorphic
467
assemblage of the selected sample. In addition to the lower cooling path, a blueschist to greenschist transition might be obtained from petrological investigations (Fig. 6b, point II), which also will not be recorded in the 4~ white mica ages. After closure of the argon isotopic system in white mica, newly formed metamorphic fabrics and/or deformational fabrics will be directly recorded in the argon signature of white mica. The two theoretical deformational events might reflect either a fabric-forming event at blueschist-facies conditions (Fig. 6b, point V) overprinted by a greenschist-facies deformational fabric (Fig. 6b, point VII) in the upper cooling path, or the recording of two separate fabric-forming events at greenschist-facies conditions (Fig. 6b, points VI, VIII) in the lower cooling path. In summary, the theoretical retrograde P - T - t paths demonstrate that important differences in the thermotectonic significance of the observed 4~ white mica cooling ages and fabric development ages (i.e. cooling ages in the retrograde path after HP metamorphism or fabric forming ages under greenschist and/or blueschist-facies conditions) are dependent upon subtle differences in the cooling path followed. The investigation of the published P - T - t path from Crete (Fig. 6c) illustrates the relationships between retrograde metamorphic conditions and closure of the argon isotopic system. The path shows that the relevant chronological information is obtained in the temperature range below 350 _+ 30~ As parts of the basement sequence of Crete have not experienced temperatures above 350 + 30~ during HP metamorphism, the age of near peak HP metamorphism of these parts is constrained at c. 20 Ma (Fig. 6c upper retrograde path, Western Crete, Southern Unit; after Jolivet et al. 1996). Other parts of the basement from Crete (Fig. 6c, lower retrograde path, Western Crete, Northern Unit; after Jolivet et al. 1996) have followed retrograde P-Tpaths which are characterized by a phase of isothermal decompression following HP metamorphism. Because ambient metamorphic temperatures might have been too high to preserve the argon isotopic signature in the white mica of the HP assemblage, the 4~ white mica ages at 24 Ma might constrain the cooling of this part of the sequence to c. 350 _+ 30~ Biostratigraphical data from Crete confines the onset of HP metamorphism to be post32 Ma (youngest ages of units involved in HP metamorphism) and the last phase of exhumation to be c. 15 Ma (oldest age of non-metamorphosed units overlying the metamorphosed sequence).
468
A.L.W. LIPS E T A L .
Fig. 6. Summary of retrograde P - T - t paths for basement rocks from different blueschist facies settings in the Aegean region. (a) Pelagonian Zone (Schermer et al. 1990; Lips et al. 1998; this study); (b) theoretical retrograde paths of high pressure metamorphic sequence, positions I to VIII are discussed in text; (c) Crete (Jolivet et al. 1996); (d) Sifnos (Wijbrans et al. 1990; Avigad 1992; Lister & Raouzaios 1996; Raouzaios et al. 1996); (e) Naxos (Wijbrans & McDougall 1988). Open symbols indicate P - T information, not constrained by 4~ dating results, filled symbols represent P - T - t information constrained by 4~ dating results (timing either obtained by correlation or direct analysis). Vertical columns represent average argon closure temperature ranges in white mica (Tc: white mica), hornblende (Tc: hornblende), and K-feldspar (Tc: K-spar) (average closure temperature ranges of white mica (a-e), hornblende (d), and K-feldspar (e) after Berger & York 1981; Harrison 1981; Harrison & McDougall 1982; Sisson & Onstott 1986; McDougall & Harrison 1988; Blanckenburg et al. 1989; Baldwin et al. 1990; Hanes 1991; Hames & Hodges 1993; Hodges et al. 1994). Dark shaded regions represent position of transition reactions of metamorphic stability fields; LBS, lawsoniteblueschist facies; EBS, epidote-blueschist facies; GS, greenschist facies (after Evans 1990). The published retrograde paths from Sifnos in the central Cyclades (Fig. 6d) reflect conflicting P - T - t paths as the absolute ages that have been obtained (Raouzaios et al. 1996; Wijbrans e t al. 1990) have been related to different
metamorphic conditions (e.g. Wijbrans et al. 1990; Avigad et al. 1992). As peak metamorphic temperatures have been too high, 4~ white mica ages will reflect cooling to c. 350 + 30~ The P - T - t information indicates that the
TIMING OF METAMORPHISM & TECTONISM, AEGEAN recording of 4~ white mica ages of c. 42 Ma and younger occurred when the H P sequence of Sifnos had already been exhumed to crustal levels equivalent to pressures of c. 8-10 kbar. It also indicates that the initial retrograde metamorphic history (i.e. directly following peak HP metamorphism) with pressures decreasing from 14-16 kbar to 8-10 kbar is not constrained by 4~ white mica ages. The P - T - t path from Naxos in the southcentral Cyclades (Fig. 6e) is constrained by two dominant metamorphic events. The first event is a HP metamorphic event (Fig. 6e, M1) with temperature conditions which are, like Sifnos, too high to preserve the argon isotopic signature in white mica. Cooling of the HP sequence records c. 50 Ma 4~ white mica ages which are related to pressure and temperature conditions lower than the peak metamorphic conditions, indicating that the timing of peak HP metamorphism (Fig. 6e, M1) is older than the 50 Ma white mica cooling ages. The second event (Fig. 6e, M2) is related to the development of the Naxos migmatitic dome at Early Miocene times and is characterized by H T - L P metamorphism (e.g. Andriessen et al. 1979). Depending upon the location of the retrogressed HP sequence with respect to the migmatite dome, 4~ white mica ages may have been reset and will record cooling of this Miocene thermal event (e.g. summarized in John & Howard 1995). Initial cooling after the thermal event has been constrained by a c. 15 Ma 4~ hornblende cooling age which is also indicated in Fig. 6e (the right column indicates the closure temperature range in hornblende: 525 + 25~ and continued until c. 11 Ma (cooling ages on biotite, Wijbrans & McDougall 1988; average closure temperature c. 300~ In summary, the four P - T - t paths from the Aegean area presented in Fig. 6 show that the ages obtained from HP assemblages have different meanings. They reflect prograde, peak metamorphic, and retrograde episodes in the Pelagonian Zone and on Crete, and only retrograde metamorphic episodes in the Cycladic region. An estimate of timing of peak HP metamorphism in the Cyclades has been obtained by thermal modelling of the existing P - T - t data from Sifnos (Wijbrans et al. 1993) and resulted in a suggested minimum age of peak pressure metamorphism of around 60 Ma. However, as real rocks often show more complex histories than can be accommodated in numerical models, we contend that individual thrust sheets may have experienced locally different prograde HP paths. It is shown that the first stage of exhumation in the Cyclades, with a corresponding
469
pressure decrease of c. 6 kbar is hidden in the preEocene tectonic history.
Summary and conclusions The continental margin of Eurasia, as exposed in the Hellenides, shows evidence for subduction related HP metamorphism and subsequent ductile deformation under retrograde conditions. The regional preservation of relic Hercynian white mica ages reflects the lower maximum temperatures reached in the Pelagonian Zone in the north compared to the Cyclades and Crete in the south. These lower metamorphic temperatures allow direct dating of Alpine deformational fabrics in the Pelagonian Zone. Here the ductile deformation history can be traced back in time from the Miocene into the Cretaceous (Fig. 4). The earliest, directly resolvable, fabric forming event (Fig. 4) at c. 90-85 Ma is characterized by southwestward directed transport under greenschist facies conditions from shallow parts of the Pelagonian basement (which are most likely equivalent to Pierien units and are now exposed on Skiathos and in the Antihassia mountains), while at deeper crustal levels blueschist-facies mylonites developed during a northeastward direction of transport (Ambelakia and lower Pierien units; now exposed in the Ossa Massif). Whereas the oldest blueschistfacies mylonitic fabric-forming event is recorded at 85 Ma, seen in the Ossa Massif, the youngest recorded blueschist-facies mylonitic fabrics are c. 54 Ma. This would imply that conditions favourable for HP metamorphism occurred over a timespan of at least 30 million years. During the event at 54 Ma, the basement sequence of the Pelagonian was partially exhumed and was affected by subsequent ductile deformation at greenschist-facies metamorphic conditions. Following the 54 Ma event, southwestward transport of the metamorphosed sequence to, and across, the autochthonous Olympos-Ossa sequence by c. 45 Ma caused elevated metamorphic pressures in, parts of, the Eocene flysch, which is related to the autochthonous sequence. This would indicate that the emplacement of a metamorphosed basement sequence, with a recorded HP episode of c. 85-54 Ma, would have caused an HP episode with slightly younger ages (as young as c. 45 Ma) at the present-day location. The greenschist-facies deformation in the Pelagonian Zone had terminated by 40-36 Ma in the Ossa-Olympos region and continued until 15 Ma in the Pelion Massif. In the Cyclades, to the south of the Pelagonian Zone, the phengite ages obtained from HP
470
A.L.W. LIPS E T A L .
metamorphic assemblages consistently indicate the timing of cooling after HP metamorphism and cluster around 40 to 45 Ma, with local exceptions to slightly older ages (c. 50 Ma; Maluski et al. 1987; Wijbrans & McDougall 1988). We suggest that the duration of the prograde and peak metamorphic episodes in the Cycladic blueschists might be similar in style as the observed history in the Pelagonian Zone, with
peak HP ages in the Cyclades expected to be slightly younger than observed in the Pelagonian Zone. This hypothesis could potentially be tested by a systematic geochronological study on minerals with high isotopic closure temperatures (e.g. U - P b zircon and monazite studies). The implications that the peak pressure conditions in the Cyclades might have been reached earlier than c. 45-50 Ma are of great importance when
Fig. 7. Temporal and spatial distribution of metamorphic episodes and tectonic events in the Aegean. (a) the timing of the termination of blueschist facies metamorphism, (b) timing of youngest recorded ductile extension, (c) and (d) distribution of metamorphic events at 40 and 60 Ma, respectively (unrestored geographical distribution); P - T - t data of Rhodope region (northern region) extrapolated from data by Dinter et al. (1995), Liati & Seidel (1994, 1996), Wawrzenitz & Mposkos (1997). Metamorphic stability fields: A, amphibolite facies; AEA, albite-epidote-amphibolite facies; E, eclogite facies; EBS, epidote-blueschist facies; GS, greenschist facies (after Evans 1990).
TIMING OF METAMORPHISM & TECTONISM, AEGEAN P - T - t paths for the region form the basis of regional tectonic models (e.g. Wijbrans et aL 1993; Lister & Raouzaios 1996). Comparison of retrograde metamorphic histories from H P localities in the Aegean shows a clear decrease in age at which blueschist-facies m e t a m o r p h i s m terminated, shifting progressively from c. 54 Ma in the Pelagonian Zone, to c. 40 Ma in the Cycladic region, and as young as c. 20 Ma on Crete. That is younging is towards the active subduction zone (Fig. 7a). The age distribution related to recrystallization of white mica under post-peak pressure metamorphic conditions suggests a diachronous exhumation of the high pressure rocks. The ductile deformation had terminated in the Pelion Massif by c. 15 Ma, which is contemporaneous with the regionally recognized phase of Early Miocene extension, which also affected ductile crustal levels in the Cycladic region, Crete (Fig. 7b), the R h o d o p e Massif (Dinter & Royden 1993; Dinter et aL 1995), the Menderes Massif in western Turkey (Hetzel et aL 1995), and may have extended further east into central Anatolia to the Nigde Massif (Whitney & Dilek 1997). The 15 Ma phase of extension affected (semi-)brittle crustal levels in the Olympos Massif, where it has been resolved, albeit less precisely, by Early Miocene a~ Kfeldspar cooling ages (Figs 4 and 6a). In contrast to the diachronous shift of termination of H P m e t a m o r p h i s m in the A e g e a n towards the active subduction zone (Fig. 7a), the regionally observed phase of Early Miocene extension and subsequent termination of ductile deformation appears to be synchronous across the region (Fig. 7b). The shift from a diachronous uplift of H P sequences to a synchronous phase of ductile extension across the region at Early Miocene times suggests a change in dominant exhumation mechanism for the currently exposed A e g e a n basement rocks. In addition to the temporal distribution of the termination of H P metamorphism and ductile deformation in the region, a p r e s e n t a t i o n of synchronous A l p i n e m e t a m o r p h i c episodes have been included to illustrate the spatial distribution of 40 and 60 Ma events (Fig. 7c, d), albeit unrestored for the palaeogeographical distribution. The 40 Ma and 60 Ma episodes illustrate the distribution of H T metamorphism and H P metamorphism in the prolonged history of the A e g e a n region and the Hellenic subduction complex.
The investigations were supported by the Netherlands Geosciences Foundation (GOA) with financial aid from the Netherlands Organization for Scientific
471
Research (NWO), grant no 750-19-411 to Lips. The research of J. R. W is covered under NSG#gg0106 Constructive reviews from D. Avigad, P. Gautier, L. Jolivet and two anonymous reviewers on previous versions of the manuscript have been appreciated and have helped to initiate and to improve the current manuscript.
References ALTHERR, R., KREUZER, H., WENDT, I., LENZ, H., WAGNER, G. A., KELLER, J., HARRE, W. & HOHNDORF,A. 1982. A Late Oligocene/Early Miocene high temperature belt in the Attic-Cycladic crystalline complex (SE Pelagonian, Greece). Geologisches Jahrbuch, 23, 97-164. , SCHLIESTEDT, M., OKRUSCH, M., SEIDEL, E., KREUZER, H., HARRE, W., LENZ, H., WENDT, I. & WAGNER, G. A. 1979. Geochronology of highpressure rocks on Sifnos (Cyclades, Greece). Contributions to Mineralogy and Petrology, 70, 245-255. ANDRIESSEN,P. A. M., BOELRIJK,N. A. I. M., HEBEDA, E. H., PRIEM, H. N. A., VERDURMEN,E. A. TH. & VERSCHURE, R. H. 1979. Dating the events of metamorphism and granitic magmatism in the Alpine orogen of Naxos (Cyclades, Greece). Contributions to Mineralogy and Petrology, 69, 215-255. AVIGAD,O. 1993. Tectonic juxtaposition of blueschists and greenschists in Sifnos Island (Aegean Sea) Implications for the structure of the Cycladic blueschist belt. Journal of Structural Geology, 15, 1459-1469. - - - , MATTHEWS,A., EVANS,B. W. & GARFUNKEL,Z. 1992. Cooling during the exhumation of a blueschist terrane: Sifnos (Cyclades), Greece. European Journal of Mineralogy, 4, 619-634. BALDWIN,S. L., HARRISON,T. M. & FITZ GERALD,J. D. 1990. Diffusion of 4~ in metamorphic hornblende. Contributions to Mineralogy and Petrology, 105, 691-703. BERGER, G. W. & YORK, D. 1981. Geothermometry from 4~ dating experiments. Geochimica et Cosmochimica Acta, 45, 795-811. BLAKE, M. C., BONNEAU,M., GEYSSANT,J., KIENAST,J. R., LEPVRIER, C., MALUSKI,H. & PAPANIKOLAOU, D. 1981. A geologic reconaissance of the Cycladic blueschist belt, Greece. Geological Society of America Bulletin, 92, 247-254. BLANCKENBURG, E V., VILLA, I. M., BAUR, H., MORTEANI, G. & STEIGER, R. H. 1989. Time calibration of a PT-path from the Western Tauern Window, Eastern Alps: the problem of closure temperatures. Contributions to Mineralogy and Petrology, 101, 1-11. BONNEAU,M. 1984. Correlation of the Hellenic nappes in the south-east Aegean and their tectonic reconstruction. In: DIXON,J. E. & ROBERTSON,a . H. E (eds) The Geological Evolution of the Eastern Mediterrenean. Geological Society, London, Special Publications, 17, 517-527. BROCKER, M. 1990. Blueschist-to-greenschist transition in metabasites from Tinos Island, Cyclades,
472
A . L . W . LIPS E T A L .
Greece: Compositional control or fluid infiltration? Lithos, 25, 25-39. , KREUZER, H., MATrHEWS, A. & OKRUSCH, M. 1993. 4~ and oxygen isotope studies of polymetamorphism from Tinos Island, Cycladic blueschist belt, Greece. Journal of Metamorphic Geology, 11, 223-240. BUICK, I. S. & HOLLAND,T. J. B. 1989. The P-T-t path associated with crustal extension, Naxos, Cyclades, Greece. In: DALu J. S., CLIFF, R. A. & YARDLEu B. W. D. (eds) Evolution of Metamorphic belts. Geological Society, London, Special Publications, 43, 365-369. DERYCKE, E & GODFRIAUX,I. 1976. Metamorphismes <schistes bleus et schistes verts> dans l'Ossa et le Bas-Olympe (Thessalie-Grece). Bulletin de la SociOtO GOologique de France, 2, 252. - & -1978. Ddcouverte de microfaunes pal6ogenes dans le flysch m6tamorphique de Spilia (Ossa, Grece). Comptes Rendus de l'Academie des Sciences, Paris, 286, 555-558. DINTER, D. A. & ROYDEN, L. 1993. Late Cenozoic extension in northeastern Greece. Geology, 21, 45-48. --, MACFARLANE, A., HAMES, W., ISACHSEN, C., BOWRING, S. & ROYDEN, L. 1995. U-Pb and 4~ geochronology of the Symvolon granodiorite: Implications for the thermal and structural evolution of the Rhodope metamorphic core complex, northeastern Greece. Tectonics, 14, 886-908. EVANS, B. W. 1990. Phase relations of epidote blueschists. Lithos, 25, 3-23. FASSOULAS, C., KILIAS, A. & MOUNTRAKIS, D. 1994. Postnappe stacking extension and exhumation of high-pressure/low-temperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. FERRIERE, J. 1976. Etude preliminaire d'un secteur des massifs cristallins internes (): la region de Volos, Gr6ce continentaleorientale. Bulletin de la SociOtO GOologique de France, 2, 265-272. , GODFRIAUX, I. & SCHMITF,A. 1988. Les series metamorphiques infrapieriennes et pagasitiques; bordure orientale de la zone (isopique) Pelagonienne (Thessalie, Gr6ce)? Deltio tes Ellenikes Geologikes Etaireias, 20, 193-206. FRANZ, L. & OKRUSCH, M. 1992. Aragonite-bearing blueschists on Arki island, Dodecanese, Greece. European Journal of Mineralogy, 4, 527-537. GAUTIER, P & BRUN, J.-P. 1994. Ductile crust exhumation and extensional detachments in the central Aegean (Cyclades and Evvia Islands). Geodinamica Acta, 7, 57-85. GODFRIAUX, I. 1970. Etude geologique de la region de l'Olympe (Grbce). Annales of Geologie du Pays Hellenique, 19, 1-271. --, FERRIERE, J. & SCHM1TT,A. 1988. Le development en context continental d'un metamorphisme HP/BT; les "schistes bleus" tertiaires Thessaliens. Deltio tes Ellenikes Geologikes Etaireias, 20, 175-192. HAMES, W. E. & HODGES, K. V. 1993. Laser 4~
evaluation of slow cooling and episodic loss of 40Ar from a sample of polymetamorphic muscovite. Science, 261, 1721-1724. HANES, J. A. (1991). K - A r and 4~ geochronology: Methods and Applications. In: HEAMAN,L. & NUDDEN, J. N. (eds) Short course handbook on applications of radiogenic isotope systems to problems in geology. Mineralogical Association of Canada, 27-57. HARRISON, T. M. 1981. Diffusion of 40Ar in Hornblende. Contributions to Mineralogy and Petrology, 78, 324-331. -& McDOUGALL, I. 1982. The thermal significance of potassium feldspar K-At ages inferred from 40Ar/39Ar age spectrum results. Geochimica et Cosmochimica Acta, 46, 1811-1820. HETZEL, R., RING, U., AKAL, C. & TROESCH, M. 1995. Miocene NNE-directed extensional unroofing in the Menderes Massif, southwestern Turkey. Journal of the Geological Society, London, 152, 639-654. HODGES, K. W., HAMES, W. E. & BOWRING, S. A. 1994. 40Ar/39Ar age gradients in micas from a hightemperature-low-pressure metamorphic terrain: Evidence for very slow cooling and implications for the interpretation of age spectra. Geology, 22, 55-58. IGME 1983. Geological map of Greece, scale 1:500,000. Second edition. Institute of Geology and Mineral Exploration, Athens. JACOBSHAGEN, V. & WALLBRECHER, E. 1984. PreNeogene nappe structure and metamorphism of the North Sporades and the southern Pelion peninsula. In: DIXON, J. E. & ROBERTSON,A. H. E (eds) Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 591-602. JOHN, B. E. & HOWARD, K. A. 1995. Rapid extension recorded by cooling-age patterns and brittle deformation, Naxos, Greece. Journal of Geophysical Research, 100 (B7), 9969-9979. JOLIVET, L., BRUN, J.-P, GAUTIER, P., LALLEMANT,S. & PATRIAT, M. 1994. 3D-kinematics of extension in the Aegean region from the early miocene to the Present, insights from the ductile crust. Bulletin de la SociOtO GOologique de France, 165, 195-209. --, GOFFf2, B., MONII~, P., TRUFFERT-LUXEY, C., PATRIAT, M. & BONNEAU, M. 1996. Miocene detachment in Crete and exhumation P - T - t paths of high-pressure metamorphic rocks. Tectonics, 15, 1129-1153. LIATI,A. & SEIDEL,E. 1994. Sapphirine and hOgbomite in overprinted kyanite-eclogites of central Rhodope, N. Greece: first evidence of granulitefacies metamorphism. European Journal of Mineralogy, 6, 733-738. - & -1996. Metamorphic evolution and geochemistry of kyanite eclogites in central Rhodope, northern Greece. Contributions to Mineralogy and Petrology, 123, 293-307. LIPS, A. L. W., WHITE, S. H. & WIJBRANS, J. R. 1998. 4~ laserprobe direct dating of discrete
TIMING OF METAMORPHISM & TECTONISM, A E G E A N deformational events: a continuous record of Early Alpine tectonics in the Pelagonian Zone, NW Aegean area, Greece. Tectonophysics, 298, 133-153. LISTER, G. S. & RAOUZAIOS,A. 1996. The tectonic significance of a porphyroblastic blueschist facies overprint during Alpine orogenesis: Sifnos, Agean Sea, Greece. Journal of Strucural Geology, 18, 1417-1435. MALUSKI, H., BONNEAU, M. & KIENAST, J. R. 1987. Dating the metamorphic events in the Cycladic area: 39Ar/4~ data from metamorphic rocks of the island of Syros (Greece). Bulletin de la Soci~t~ GOologique de France, (8), III, 5, 833-842. - - , VERGELY,P,. BAVAY,O., BAVAY,P. & KATSIKATSOS, G. 1981. 39Ar/40Ar dating of glaucophanes and phengites in Southern Euboa (Greece) geodynamic implications. Bulletin de la Soci(t~ G~ologique de France, (7), XXIII 5, 469-476. McDOUGALL, I. &HARRISON,T. M. (1988). Geochronology and Thermochronolgy by the 4~ Method. Oxford University Press, New York. OKRUSCH, M. & BROCKER,M. 1990. Eclogites associated with high-grade blueschists in the Cycladic archipelago, Greece: a review. European Journal of Mineralogy, 2, 451-478. - - , RICHTER,P. & KATSIKATSOS,G. 1984. High-pressure rocks of Samos, Greece. In: DIXON, J. E. & ROBERTSON, A. H. E (eds) The geological evolution of the Eastern Mediterrenean, 529-536. RAOUZAIOS, A., LISTER, G. S. & FOSTER, D. 1996. Oligocene exhumation and metamorphism of eclogite-blueschists from the island of Sifnos, Cyclades, Greece. Geol. Soc. Australia, Abstracts of 13th AGC, 141, 358. REDDY, S. M., KELLEY, S. P. & WHEELER, J. 1996. A 4~ laser probe study of micas from the Sesia Zone, Italian Alps: implications for metamorphic and deformation histories. Journal of Metamorphic Geology, 14, 493-508. RIDLEY,J. 1984. Evidence of a temperature-dependent 'blueschist' to 'eclogite' transformation in highpressure metamorphism of metabasic rocks. Journal of Petrology, 25, 852-870. SCAILLET,S., FERAUD, G., LAGABRIELLE,Y.,BALLEVRE, M. & RUFFET, G. 1990. 4~ laser-probe dating by step heating and spot fusion of phengites from the Dora MalTa nappe of the western Alps, Italy. Geology, 18, 741-744. SCHERMER, E. R. 1990. Mechanisms of blueschist creation and preservation in an A-type subduction zone, Mount Olympos region, Greece. Geology, 18, 1130-1133. 1993. Geometry and kinematics of continental basement deformation during the Alpine orogeny, Mt. Olympos region, Greece. Journal of Structural Geology, 15, 571-591. , Lux, D. R. & BURCHFIELD,B. C. 1990. Temperature-time history of subducted continental crust, Mount Olympos region, Greece. Tectonics, 9, -
-
1 1 6 5 - 1 1 9 5 .
SCHLIESTEDT, M. &. MATTHEWS,A. 1987. Transformation of blueschists to greenschist facies rocks as
473
a consequence of fluid infiltration, Sifnos (Cyclades), Greece. Contributions to Mineralogy and Petrology, 97, 237-250. SCHMIqq', A. 1983. Nouvelles contributions a l'~tude g~ologique des Pieria, de l'Olympe, et de l'Ossa (Greece du Nord). Thesis at Facult6 Polytechnique de Mons, Belgium. SEIDEL, E., KREUZER, H. & HARRE, W. 1982. A Late Oligocene/Early Miocene high pressure belt in the External Hellenides. Geologisches Jahrbuch, E23, 165-206. SISSON,V. B., & ONSTOTT,T. C. 1986. Dating blueschist metamorphism: A combined 40Ar/39Ar and elctron microprobe approach. Geochimica et Cosmochimica Acta, 50, 2111-2117. SKARPELIS,N. 1989. Metamorphosed ultramafic rocks in the blueschist (arna unit) of the external metamorphic belt of the Hellenides (Peloponnesos and Kythira Island, Greece). Bulletin of the Geological Society of Greece, 23, 345-358. THEYE, T., SEmEL, E. & VIDAL, O. 1992. Carpholite, sudoite and chloritoid in low-grade high-pressure metapelites from Crete and the Peloponnese, Greece. European Journal of Mineralogy, 4, 487-507. TRIBOULET, C. & BASSIAS, Y. 1988. Geothermobarometry on zoned amphiboles and progressive metamorphism in glaucophane-bearing metabasites from the Parnon massif (external Hellenides, Peloponnesus, Greece). Comptes Rendus de l'Academie des Sciences, Paris, 306, Serie II, 991-996. VAN DER MAAR, E A. & JANSEN, J. H. B. 1983. The geology of the polymetamorphic complex of Ios, Cyclades, Greece, and its significance in the Cycladic Massif. Geologisches Rundschau, 72, 283-299. VANDENBERG,L. C. & LISTER, G. S. 1996. Structural analysis of basement tectonites from the Aegean metamorphic core complex of Ios, Cyclades, Greece. Journal of Structural Geology, 18, 1437-1454. WALCOTT, C. R. & WHITE, S. H. 1998. Constraints on the kinematics of post-orogenic extension imposed by stretching lineations in the Aegean region. Tectonophysics, 298, 155-175. WAWRZENITZ, N. ~; MPOSKOS,E. 1997. First evidence for Lower Cretaceous HP/HT-metamorphism in the Eastern Rhodope, North Aegean region, North-East Greece. European Journal of Mineralogy, 9, 659-664. WHITNEY, D. L. ~; DILEK, Y. 1997. Core complex development in central Anatolia, Turkey. Geology, 25, 1023-1026. WIJBRANS, J. R. & McDOUGALL, I. 1986. 4~ dating of white micas from an Alpine highpressure metamorphic belt on Naxos (Greece); the resetting of the argon isotopic system. Contributions to Mineralogy and Petrology, 93, 187-194. & - - 1988. Metamorphic evolution of the Attic Cycladic metamorphic belt on Naxos (Cyclades, Greece) utilizing 4~ age spectrum -
-
A . L . W . LIPS ETAL.
474
measurements. Journal of Metamorphic Geology, 6, 571-594. & YORK, D. 1990. Single grain argon laser probe dating of phengites from the blueschist to greenschist transition on Sifnos (Cyclades, Greece). Contributions to Mineralogy and Petrology, 104, 582-593. ,
-
-
, VAN WEES, J. D., STEPHENSON, R. A. &
CLOETINGH, S. A. R L. 1993. Pressure-temperatute-time evolution of the high-pressure metamorphic complex of Sifnos, Greece. Geology, 21, 443-446.
Tethyan sutures of northern Turkey A R A L I. O K A Y & O K A N TI21YSUZ
I T U Eurasian Institute of Earth Sciences and Maden Faki~ltesi, Jeolofi BOli~mii, Ayaza~,a, 80626, istanbul, Turkey (e-maik [email protected]) Abstract: The two main Tethyan sutures of Turkey, the Izmir-Ankara-Erzincan and the
Intra-Pontide sutures, are reviewed through several well-studied transects crossing the suture regions. Both sutures have formed during the Early Tertiary continental collisions following northward subduction of Tethyan oceanic lithosphere. The Izmir-AnkaraErzincan suture is represented along most of its c. 2000 km length by Paleocene and younger thrust, which emplace the upper crustal rocks of the northern continent over that of the southern continent with an intervening tectonic layer of Cretaceous subductionaccretion complexes. These thrusts constitutes a profound stratigraphic, structural, magmatic and metamorphic break, of at least Carboniferous to Palaeocene age and form the main boundary between Laurasia and Gondwana in the Turkish transect. Voluminous subduction-accretion complexes of Triassic and Cretaceous ages occur respectively to the north and south of the suture giving the antithetic subduction polarities during these two periods. This, and evidence for a major accretionary orogeny of Late Triassic age north of the Izmir-Ankara-Erzincan suture suggest that two separate oceanic lithospheres, of Carboniferous to Triassic (Palaeo-Tethys) and of Triassic to Cretaceous ages (Neo-Tethys) respectively have been consumed along the suture. The final continental collision along the Izmir-Ankara-Erzincan suture was slightly diachronous and occurred in the earliest Palaeocene to the west and in the Late Palaeocene to the east. The c. 800 km long IntraPontide suture is younger in age and have formed during the Early Eocene and younger continental collisions linked to the opening of the Western Black Sea Basin as an oceanic back-arc basin. At present the North Anatolian Fault, which came into existence in the Late Miocene, follows the course of the older Intra-Pontide suture.
Sutures r e p r e s e n t the b o u n d a r i e s of f o r m e r lithospheric plates. In an idealized case of collision of two continental plates, the suture will be manifested as a major fault or fault zone in the brittle seismogenic part of the crust and by a shear zone in the ductile lower crust and lithospheric mantle. The sutures will form profound stratigraphic, palaeogeographic, structural, magmatic and metamorphic breaks and as such will be easy to recognize and distinguish from o t h e r major intraplate faults or fault zones. Here, the term 'suture' is used if the former plate boundary can be defined as a single fault or fault zone less than one kilometre in width, suture zone refers to a belt more than one kilometre in thickness, occupied by former oceanic crustal rocks. In the case of the suture zone, the former boundary between two plates cannot be shown as a single fault line on a map of 1:500 000 scale or smaller. The structures that represent sutures frequently show major along-strike variation in type and age, and most are probably post-collisional. For example, the suture in the Western Alps is represented from west to east, by the Late Cretaceous thrusts between the Sesia Z o n e and the Penninic units (e.g., Compagnoni 1977), by the segment of the post-collisional dextral strike-slip fault (the Insubric line) b e t w e e n the
Sesia Z o n e and the Eastern Alps (e.g., Schmid et al. 1989), and by the Late Cretaceous thrusts b e t w e e n the A u s t r a - A l p i n e and P e n n i n i c nappes along the Arosa Schuppen Z o n e (e.g., Ring et al. 1988). The only c o m m o n link b e t w e e n these three different structures is that they are believed to separate rocks deposited on different margins of an ocean. The delineation of a suture in the field is, hence, partly subjective, as the decision which stratigraphic units belong to which former plate can be controversial especially in metamorphic areas (e.g., see Michard et al. 1996, for a recent discussion on the problems associated with the delineation of the Alpine suture). H e r e , we describe s o m e of the T e t h y a n sutures and neighbouring former continental margins in Turkey. The description is aimed in answering some of the p r o b l e m s associated with sutures, such as can the boundaries of f o r m e r plates be d e s c r i b e d by single m a j o r faults, or by areally extensive suture zones. To what extent do the former continental margins fit into a simple W i l s o n i a n cycle of rifting, passive margin sedimentation, subduction and collision? H o w can variation of deformation, m e t a m o r p h i s m and m a g m a t i s m along the sutures be explained?
OKAY,A. I. & T(;rYSOZ,O. 1999. Tethyan sutures of northern Turkey. In: DURAND.B., JOLIVET,L., HORVATH,E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, J56o 475-515
476
A.I. OKAY & O. TISYS(JZ
Major continental blocks of Turkey The Tethyan Ocean, which existed between Laurasia and Gondwana as a westward narrowing embayment since the Carboniferous, was not a single continuous oceanic plate during its long evolution (S,eng6r 1987; Ricou 1994; Stampfli 1996). Throughout its history, small continental fragments were rifted off from either side of the Tethyan Ocean and moved towards the other side, creating new oceanic lithosphere in their wakes, and eventually colliding with the opposite continental margin. Therefore in most regions, the former plate boundary between the two megacontinents, Laurasia and Gondwana, cannot be represented by a single suture. This is most obvious in Turkey, where several sutures isolate drifting continental fragments between Laurasia and Gondwana (S,eng6r & Yllmaz 1981). The final amalgamation of these fragments into a single continental mass occurred in Turkey as late as Late Tertiary, when the Arabian plate collided with the Anatolian plate. Figure 1 shows the sutures and major continental fragments in Turkey and the surrounding regions. There are six major lithospheric fragments in Turkey: the Strandja, the istanbul and the Sakarya Zones, the Anatolide-Tauride Block, the Karsehir Massif and the Arabian Platform (Fig. 1, ~engOr & Yllmaz 1981; ~engOr et al. 1982; Okay 1989a; Okay et al. 1994). The first three zones, which show Laurasian affinities, are classically referred to as the Pontides. They are separated by the |zmir-Ankara-Erzincan suture from the Klr~ehir Massif and the Anatolide-Tauride Block, the latter is in contact with the Arabian Platform along the Assyrian-Zagros suture (Fig. 1). Although separated by the Assyrian suture, the Anatolide-Tauride Block shows a similar Palaeozoic stratigraphy to that of the Arabian Platform, and hence to the northern margin of Gondwana. The Klrsehir Massif, which consists mainly of metamorphic and granitic rocks with Cretaceous isotopic ages, is in contact along the controversial Inner Tauride suture with the Anatolide-Tauride Block, while the Intra-Pontide suture constitutes the former plate boundary between the Sakarya and istanbul zones. Here, we describe two of the major sutures of northern Turkey, large sections
of which we have mapped or seen in the field. These are the |zmir-Ankara-Erzincan suture, which extends for c. 2000 km from the Aegean Sea to Georgia, and the c. 800 km long IntraPontide suture. First we briefly give the relevant geological features of the continental blocks, which are separated by these sutures. The Istanbul Zone
The Istanbul Zone is a small continental fragment, about 400 km long and 70 km wide, located in the southwestern margin of the Black Sea (Fig. 1). It is made up of a Precambrian crystalline basement overlain by a continuous, welldeveloped transgressive sedimentary sequence extending from Ordovician to Carboniferous, which was deformed during the Carboniferous Hercynian Orogeny (Aydln et al. 1986; Dean et al. 1997; G6r0r et al. 1997). The stratigraphy of the |stanbul Zone is shown in Fig. 2 on two sections, one from the western part near istanbul, and the other from the eastern part around Zonguldak (Fig. 1). These sections illustrate the notable facies differences along the length of the istanbul Zone, especially marked during the Carboniferous. In the Istanbul region the Carboniferous deposits consists of Vis6an pelagic limestones and radiolarian cherts, which pass up into thick turbidites probable of Namurian age, while in the east, around Zonguldak, Visean is represented by shallow marine carbonates overlain by Namurian to Westphalian paralic coal series (Dill 1976; Kerey et al. 1986). Westphalian megafloras show close affinities to those of western Europe and Donetz basins (Charles 1933). The Upper Carboniferous siliciclastic rocks in the Zonguldak area were derived from the north (Kerey et al. 1986) and were deposited in a large deltaic basin located on the southern margin of the Laurasian plate. These features, as well as palaeomagnetic results from the Palaeozoic rocks of the Istanbul Zone (Evans et al. 1991) underscore its Laurasian affinity. In the |stanbul Zone Hercynian deformation started earlier and is stronger to the west (MidCarboniferous) than to the east (latest Carboniferous). In the |stanbul region it is characterized by minor folds, relatively rare and discontinuous
Fig. 1. Tectonic map of northeastern Mediterranean region showing the major sutures and continental blocks. Sutures are shown by heavy lines with the polarity of former subduction zones indicated by filled triangles. Heavy lines with open triangles represent active subduction zones. The Late Cretaceous oceanic crust in the Black Sea is shown by grey tones. Small open triangles indicate the vergence of the major fold and thrust belts. BFZ denotes the Bornova Flysch Zone. Modified after SengOr (1984), Okay (1989a), and Okay et al. (1994, 1996).
T E T H Y A N S U T U R E S OF N O R T H E R N T U R K E Y
477
478
A.I. OKAY & O. TOYSOZ ISTANBUL istanbul
ZONE Zonguldak .-.-.-.-.-.-.-.-.-.-.-
-.-.-.-.-.-
i~_--~_--~_--~i~-~i~:~i ~i-:~6
Eocene Paleocene
a,.ede,o.ma,,on
!1--1-1--1--1"-1--'1--1--14 I-I-I-I-I-I-I-II'l--l--l--l--l--l---
-/ --I-I
Cretaceous
Jurassic
. -. .-.-.-. -_-.-. -. . . . . . _ . - - s _ _ _ _ _ _ _
9
Triassic
--,--,--,--,-5-5--F-F 51 Cimmeride deformation _;_;_;_; ..... .;_
~_;_%.%o_;822m_';~
.~.~.~.~o.o.o.o.o
Permian -o-o-o-o-o-o-o-o-.
Nm1r o-o-o-c ~,,,~,~, > 1000 m~o ;%?;?$.%.~ ;?$?$?;?;?;%?;?,
Carboniferous :::::::::::::::>3000
m: I
~ : l l t : [ : [ i l : : ] : : :
I
I Devonian
t
7-F-f~prTv7=7=:=7 -
Silurian
~
Ill
I [ I i
I t
I I
I
I
t
t
I 350 m t
I
I
I
I
~
-r op ra
============================ :iiiiiiiiiiiiiii250m[;!il; --.'O'.'O'_'O_O'_'d,'O.'O.d_',.
.-o-o~o-o-o-o~o-o-~ o-o-o-o-o-o-o-o-o.
Ordovician
Hercynian deformation
i:::::I
Flysch
[:.iiii;iiiii] Sandstone
Ophiolite
~
Marl Shale
Metamorphic units
-o-o-o-o-o-o-o-o-<
Coal
,o-o-o-o-o-o-o-o-o. -o-o-o-o-o-o-o-o-~
Neriticlimestone
f@iii-::i I:::::'l
o-o-o-o-o-o-o-o-o,
Pelagiclimestone
IV - I
o-o-o-o-o-o-o-o-o. -o-o-o-o-o-o-o-o-,
o-o-o-o-o-o-o-o-o. -o-o-o-o-o-o~o-o-~
._o-o-o-o_~-~-~-~-,
2;2;2;2:>1500m: ....
Conglomerate
Granitoid
F_2::t
Basaltic, andesitic lava Pyroclastite Dacitic, rhyodacitic lava
;?;% %%%
Conglomerate, sandstone Conglomerate, sandstone, mudstone Subduction-accretion complex Kuruko-type massive sulphide ore deposits
Fig. 2. S.y.nthetic stratigraphic sections of the istanbul Zone from the istanbul and Zonguldak regions based on Yurtta~-Ozdemir (1973), Dean et al. (1996) and G6rtir et al. (1997).
TETHYAN
SUTURES
OF NORTHERN
thrust faults, and minor but widespread strikeslip faults and shear zones. The deformed Palaeozoic sequence is unconformably overlain by the earliest Triassic sedimentary rocks (Fig. 2). The Triassic sequence is well developed in the Istanbul region and comprises an approximately 800 m thick transgressive sequence of earliest Scythian to Norian age (Fig. 2, Assereto 1972; Yurttas-(3zdemir 1973; Gedik 1975). It ends with a flysch-like sequence of sandstone and shale with Halobia sp. marking the onset of Late Triassic Cimmeride deformation, which is particularly strong in the Sakarya Zone to the south. In marked contrast with the |stanbul region, Triassic sequence in the eastern part of the Istanbul Zone is represented by terrigenous, variegated sandstones and conglomerates with a total thickness of 800 m (Fig. 2). The upper parts of these continental deposits locally comprise lacustrine marls and limestones, which yield latest Triassic ages (Akyol et al. 1974; Alisan & Derman 1995). The palaeomagnetic results from the Triassic rocks of the Istanbul Zone indicate its continued affinity with Laurasia (Sarlbudak et al. 1989; Theveniaut 1993). In the western part of the lstanbul Zone, the Jurassic and Lower Cretaceous sequence is absent, and the Palaeozoic and Triassic rocks are unconformably overlain by Upper Cretaceous-Palaeocene clastic, carbonate and andesitic volcanic rocks (Dizer & Meriq 1983; Tansel 1989). In contrast, in the eastern part of the Istanbul Zone there is a thick Middle Jurassic to Eocene succession marked by small unconformities (Fig. 2, Akyol et al. 1974). Senonian andesitic lavas, dykes and small acidic intrusions, which are widespread in the northern part of the istanbul Zone, were produced during the northward subduction of the Intra-Pontide Ocean. The lstanbul Zone shows a similar Palaeozoic-Mesozoic stratigraphy to that of Moesian Platform, and prior to the late Cretaceous opening of the West Black Sea Basin it was situated south of the Odessa shelf (Fig. 1, Okay et al. 1994). Together with very similar Palaeozoic sequences farther west, including the Montagne Noire in France, Carnic Alps in Austria, Krajstides in Bulgaria, the istanbul Zone and its continuation in the Scythian platform formed part of the passive continental margin of Laurasia. With the inception of back-arc spreading, the Istanbul Zone was rifted off from the Odessa shelf and was translated southward, bounded by two transform faults, the West Black Sea fault in the west and the West Crimean fault in the east (Fig. 1, Okay et al. 1994). Both of these transform faults are mapped in the offshore seismic
TURKEY
479 Strandja Massif
Maastrich._ i~-')~i~,~-')~,~-'),~il Campanian i-~ -4~;-i~'--S~"->5()bv Sant.-Con. TuronianCenoman.
i I i [ i [25)' .n-~ ~ 4. ~ 4. ~ 4.~.ff
Albian Aptian Barremian Hauteriv. Valangin. Berriasian l'ithonian ._ Kimm.-Oxf Callovian Bathonian_ Bajocian _ Aalenian _ roarcian
}
Pliensb.
[
Sinemurian Hettangian Nor.-Rhae. _
I I I 1300 m_ [ ~!:! ..... ~176176176176 m ~ § 4.+4-~ ..... - ~-- 4. + 4. +:II;22Z-_---~ 2L
C3~ ~
+ 4, + 4. 4,...... .12_~2 + + 4. 4- 4_--Jill ++++4.*--4. 4. 4. 4. 4. .ta~..... -++++++~-LZ5 4. 4. 4. 4. 4. 4. 4. 4:'~.... . .L ..i..L
.I. a-
.4..,i..4.'--
Fig. 3. Synthetic stratigraphic section of the Strandja Zone based on Aydln (1974). Symbols as in Fig. 2.
sections (Finetti et al. 1988). The southward translation of the Istanbul Zone led to the gradual closure of the Intra-Pontide Ocean by north-dipping subduction. During the Early Eocene the Istanbul Zone collided with the Sakarya Zone leading to the formation of the eastern part of the Intra-Pontide suture. The Strandja Zone
The Strandja Zone constitutes the easternmost part of the little understood crystalline region of the Rhodope-Strandja Massif in northeastern Greece, southern Bulgaria and Turkish Thrace (Fig. 1). The stratigraphy of the Strandja Zone is strikingly different from that of the istanbulZone. In the Turkish Thrace it consists of a
480
A.I. OKAY & O. TOYSUZ associated granodiorites (Moore et al. 1980), formed during the n o r t h w a r d subduction of the I n t r a - P o n t i d e O c e a n (Boccaletti et al. 1974; S eng/Sr & Yflmaz 1981).
b a s e m e n t of m e t a m o r p h i c rocks intruded by P e r m i a n granites (244 Ma, A y d m 1974), which are u n c o n f o r m a b l y overlain by a Triassic sedim e n t a r y sequence (Fig. 3; Aydln 1974). The Triassic sequence shows a d e v e l o p m e n t similar to the G e r m a n i c Triassic facies a n d has clear Laurasian affinity. In the Bulgarian part of the Strandja Z o n e the Mesozoic sequence extends up to the mid-Jurassic (Chatalov 1988). The Triassic to Middle Jurassic sequence as well as the crystalline b a s e m e n t was folded, thrust-faulted and regionally m e t a m o r p h o s e d during the midJurassic, and were u n c o n f o r m a b l y overlain by C e n o m a n i a n conglomerates and shallow marine limestones (Fig. 3). These pass up to Senonian andesitic volcanic and volcanoclastic rocks and
The Sakarya Zone T h e Sakarya Z o n e is an east-west-oriented continental fragment, about 1500 km long and 120 k m wide, between the A n a t o l i d e - T a u r i d e Block to the south and the istanbul and Strandja zones and the eastern Black Sea to the north (Fig. 1). It includes the Sakarya C o n t i n e n t of ~eng6r & Yflmaz (1981) as well as the Central and Eastern Pontides, which show a similar stratigraphic and tectonic development ( O k a y 1989a). The most
SAKARYA ZONE NWTurkey
Central Pontides
Priabonia] Bartonian Lutetian Ypresian ThanetianDanian : o ~ o ~ 0 0 ~ ~~ - - :.-
:
- : - :
Alpide deformation 15-5--(5- 5- -~ 150 m~ ,.
- - - 2 - - : : - :
Maastrich ::-:212:2:215:::52:2:5:2:5::i:1:151:: Campani~ :ii?!i~!:!:!~:i!i!:!i{:~! ~:!:!:!: Sant.-Con . . . . . . . . . . . . . . . . . . Cenoman -~:-'5"~ 77-'.'-7-.'5--.-.'5--3--/-" -3 2.52:-_:2--.722
-
-
- - -
.
.
.
.
.
.
.
.
.
.
.
.
i! !i!:!i! ! ! !:!ii 2000-~! : : 2 : 7 2
U
. : 4 : 4 : 4 : 4 : - 5 : 4 : 4 •
A]bi~,
.
5 : 5 2
_ _._
_._
_:._2
v v v v ~ v v ;~
- 2
!i]!i!i:i!i!iiii!iiiiigb-dd-m~i 9_ _ _ _ . _ _ _ . _ . _ . . . . . . - _ . . . . _._-_._._._ _ _ _._._ _. . . . . . . . . . .
Aptian
9: . _ ' : : - . : - . - . 7 : :
:
-
-
- :-
2
::5:::::::::::::::::::::::::::::::: i
I
I
I
I
Barremial I I I I I I I Hauteriv !111 I I Valangin I I I I I Berriasiar I I I I Tithonian 1 I I 1800 m Kimm.-O I I I I I Callovian l l l l l l l l Bathoniar - - - -,-,-,-,-,-,-,-,-,-,-,-,Bajocian ' ~i:!:~:!:!:!:!:!:!:!:i:!:i Aalenian <2!~i00-5-00-mToarcian ~:::::::::::::::::::::::: . . . . . . . . . . . . . . . . . . .
Pliensb.
:
-
-.2
:
- :._-.:.:.-.:
: :
:
-
o2o oio:o%O . . . . . .
---
-:-:-,----,-,-,-,-,-,-,-,-,-,-,-,_ _ _ _ _ _ - - - - - - - . . . . . . . .
Sinemuri~ ~ Hettangia
. . . . .
.
~ Cimmeride deformation
r,,
~[Nor.-Rha~
~.'~'.2%.~~ ............
Karakaya Complex ,'5
Y5
~
:
[
Hercynian basement
Fig. 4. Synthetic stratigraphic sections of the western and central parts of the Sakarya Zone (Saner 1978, 1980; Altlner et al. 1991; T~Jys~z 1993). See Fig. 18 for a stratigraphic section of the eastern part of the Sakarya Zone. Symbols as in Fig. 2.
TETHYAN SUTURES OF NORTHERN TURKEY striking geological feature of the Sakarya Zone is the widespread presence of Triassic subduction-accretion complexes, rare in the rest of the Alpide-Himalayan chain, which form a strongly deformed and partly metamorphosed basement to the overlying Lower Jurassic-Eocene sequence (Fig. 4, Tekeli 1981). The Triassic subduction-accretion units, called the Karakaya Complex in the western part of the Sakarya Zone, comprise a lower section of PermoTriassic metabasite-marble-phyllite series, over three kilometres in thickness, with exotic Triassic eclogite (Okay & Monie 1997) and blueschist lenses (Monod et al. 1996). This sequence, named as the Nilufer Unit, occurs throughout the Sakarya Zone. In northwest Turkey Mid-Triassic conodonts have been described from the upper parts of the Niltffer Unit (Kaya & Mostler 1992), while the age of the regional metamorphism is latest Triassic (Monod et al. 1996; Okay & Monie 1997). An oceanic seamount (Pickett & Robertson 1996) or alternatively an oceanic intra-arc to fore-arc tectonic setting (Okay et al. 1996) have been proposed for the Nilufer Unit. The Nilufer Unit is tectonically overlain by chaotically deformed but unmetamorphosed clastic and basic volcanic rocks of Triassic age with exotic blocks of Carboniferous and Permian neritic limestone, basalt, Carboniferous and Permian radiolarian chert (Bing61 et al. 1975; Okay & Mostler 1994; Kozur & Kaya 1994; Leven & Okay 1996; Okay et al. 1996; Tt~yst~z 1996). Most of these clastic rocks probably represent trench deposits (Okay et al. 1996; Pickett & Robertson 1996). The basement of the Nilufer Unit, as observed in a few tectonic windows, consists of high-grade metamorphic rocks with Late Carboniferous zircon ages (Okay et al. 1996). The final phase of deformation and metamorphism of the subduction-accretion complexes occurred during the latest Triassic, and the various units of the Karakaya Complex are unconformably overlain by Jurassic terrigenous to shallow marine clastic sedimentary rocks. The Jurassic-Eocene stratigraphy of the Sakarya Zone is summarized in Fig. 4 in two sections from the western and central part of the Sakarya Zone. The succession to the west consists of Lower to Middle Jurassic continental to shallow marine clastic rocks with ammonitico rosso horizons, U p p e r Jurassic-Lower Cretaceous carbonates, Upper Cretaceous-Palaeocene volcanic and sedimentary rocks (Fig. 4). The Senonian andesitic volcanism observed in the northern parts of the Sakarya Zone is related to the northward subduction of the I z m i r - A n k a r a - E r z i n c a n Ocean. Palaeomagnetic data from the Sakarya Zone indicate that it was close to the Laurasian margin during the
481
Liassic and Late Cretaceous time (Channel et al. 1996). The Anatolide-Tauride
Block
The Anatolide-Tauride Block forms the bulk of the southern Turkey and in contrast to the Pontide continental fragments shows a similar Palaeozoic stratigraphy to the Arabian Platform, and hence to that of Gondwana. During the obduction, subduction and continental collision episodes in the Late Cretaceous-Palaeocene, the Anatolide-Tauride Block was in the footwall position and therefore underwent much stronger Alpide deformation and regional metamorphism than that observed in the Pontide zones. During the Senonian a massive body of ophiolite and accretionary complex was emplaced over the Anatolide-Tauride Block. The northern margin of the Anatolide-Tauride Block underwent HP-LT (high pressure-low temperature) metamorphism at depths of over 50 km under this oceanic thrust sheet. Erosional remnants of this thrust sheet of ophiolite and accretionary complex occur throughout the AnatolideTauride Block. In the Turkish geological literature the accretionary complex is often referred to as ophiolitic melange; however, it generally lacks a matrix and is structurally more similar to an imbricate thrust stack. With the inception of continental collision in the Palaeocene the Anatolide-Tauride Block was internally sliced and formed a south to southeast-vergent thrust pile. The compression continued until the Early to Mid-Miocene in the western Turkey and is still continuing in the eastern Anatolia. The lower parts of the thrust pile to the north were regionally metamorphosed, while the upper parts to the south form large cover nappes. This leads to subdivision of the Anatolide-Tauride Block into zones with different metamorphic and structural features, in a similar manner to the subdivision of the Western Alps into Helvetics and Penninics albeit with a different polarity. There are three main regional metamorphic complexes: a Cretaceous blueschist belt, the Tavsanh Zone, to the north and two Barrovian-type metamorphic belts, the Afyon Zone of Palaeocene age and the Menderes massif of Eocene age farther south (Fig. 1). To the northwest of Menderes Massif there is a belt of chaotically deformed uppermost Cretaceous-Paleocene flysch with Triassic to Cretaceous limestone blocks. This Bornova Flysch Zone has an anomalous position between the Izmir-Ankara suture and the Menderes Massif (Fig. 1). The Taurides, which lie south of the metamorphic regions, consist of a stack of thrust sheets of Palaeozoic and Mesozoic sedimentary rocks (e.g., Gutnic et al. 1979; Ozgfil
482
A.I. OKAY & O. TUYSOZ
1984). Although the Anatolide-Tauride Block shows a variety of metamorphic, structural and stratigraphic features, there are some elements of stratigraphy common to all of these zones and which distinguish the Anatolide-Tauride Block as a single tectonic entity. These are a Precambrian crystalline basement, a discontinuous Cambrian to Devonian succession dominated by siliciclastic rocks, a Permo-Carboniferous sequence of intercalated limestone, shale and quartzite, and a thick Upper Triassic to Upper Cretaceous carbonate sequence. On the other hand, Hercynian deformation or metamorphism, and Triassic subduction-accretion units, characteristic features of the Sakarya Zone, are not observed in the Anatolide-Tauride Block. The Ktrs.ehir M a s s i f
The Klr~ehir Massif is a region of metamorphic and voluminous granitic rocks with Cretaceous isotopic ages. It is generally regarded as separated by the Inner Tauride Suture from the Anatolide-Tauride Block (Fig. 1, ~eng6r et al. 1982), however, some workers consider the K~rsehir Massif as the northern metamorphosed margin of the Anatolide-Tauride Block (e.g., Poisson et al. 1996; Yahmz et al. 1996). The metamorphic rocks of the Klr~ehir Massif constitute a coherent metasedimentary sequence of granulite, gneiss, micaschist, metaquarzite, marble and calcsilicate rock, which are isoclinally folded and multiply deformed (Seymen 1983). The regional metamorphism varies from greenschist to granulite facies and is of high-temperaturemedium/low-pressure type. The age of regional metamorphism, although widely accepted as Cretaceous, is poorly constrained. The few K/Ar isotopic data on biotite and muscovite from the micaschists yield Late Cretaceous ages (75-80 Ma, GOnctio~,lu 1986). The metamorphic rocks are tectonically overlain by an unmetamorphosed Late Cretaceous accretionary complex of basalt, radiolarian chert, pelagic limestone, sandstone and serpentinite. The accretionary complex as well as the metamorphic rocks are intruded by granitic rocks, which cover large areas in the Klrsehir Massif. They are mainly monzogranites, quartz-monzonites with a calcalkaline character, and with a trace element geochemistry suggesting a syn- to post-collisional tectonic setting (Aklman et al. 1993; Erler & G6nctiofglu 1996). The K/Ar biotite and hornblende, and Rb/Sr hornblende and whole rock ages from the granitic rocks fall in the range of 110 Ma to 75 Ma (Ataman 1972; GOnctio~lu 1986; Gialeq 1994). An upper age limit on the granitic magmatism as well as on regional
metamorphism is provided by the Upper Maastrichtian terrigenous to shallow-marine clastic and carbonate rocks, which lie unconformably on the metamorphic as well as on the granitic rocks (Seymen 1981).
The lzmir-Ankara-Erzincan suture This is the major suture in Turkey and represents the scar of the main Tethys ocean between Laurasia and Gondwana between the late Palaeozoic and Early Tertiary. The izmirAnkara-Erzincan suture extends from north of izmir eastwards to the border with Georgia from where it continues as the Sevan-Akera suture in the Lesser Caucasus (Fig. 1, Khain 1975; Adamia et al. 1977; Knipper 1980). In the west, the lzmir-Ankara-Erzincan suture probably connects across the Aegean Sea to the Vardar suture. The various structures, which define the izmir-Ankara-Erzincan suture in the field, form a profound stratigraphic, structural, metamorphic and magmatic break. It separates the Sakarya Zone to the north from the Klrsehir Massif and the various units of the Anatolide-Tauride Block to the south (Fig. 1). The izmir-Ankara-Erzincan suture is not rectilinear but is strongly segmented forming several major loops along its length (Fig. 1). The suture will be described on the basis of these segments starting in the west. The i z m i r - A n k a r a - E r z i n c a n suture between izmir and Bahkesir
Along this northeast-trending, 180 km long segment of the izmir-Ankara-Erzincan suture between the Aegean Sea and the region of Bahkesir, the Sakarya Zone is in contact with the Bornova Flysch Zone of the AnatolideTauride Block (Fig. 1). The Sakarya Zone north of the suture consists of the Karakaya Complex unconformably overlain by a Jurassic to Lower Cretaceous succession (Fig. 5). In this region the Karakaya Complex is represented by the Niltifer Unit, tectonically overlain by the Upper Triassic arkosic sandstones (Okay et al. 1996). The arkosic clastic rocks pass upwards and eastwards to extensive olistostromes, which form a 15 km wide and 150 km long belt immediately east of the suture (Fig. 5). The olistostromes are dominated by exotic Carboniferous and Permian neritic limestone blocks; there are also rare blocks of basalt, Carboniferous pelagic limestone and radiolarite (Okay et al. 1991, 1996; Okay & Mostler 1994; Leven & Okay 1996). The age of the exotic
TETHYAN SUTURES OF NORTHERN TURKEY ~
Neogene ~
Oligo-Miocene
l
Peridotite
483
~
SAKARYA ZONE F - ~ ~ ~ 4/+ + ~_ Jurassic-Cretaceoussediments ~ Eocene sediments ~ ~ ~ , ~F~F~i~.6. ~ , ~ . . . ~ j ) ~ ao zx~ ~ UpperTriassic olistostromes" [~176~ Triassicelastic rocks t>Karakaya Complex I / ~ ~ ~ 3 . . ~ 1 1 ~ Metabasite ohyllite marble J I / ~ r ...........@ ~ 1 1 1 ~ Iv v (Niltiferun]t3 ' ; I~, v~//, ~ t MENDERES MASSIF TAV$ ANLI ZONE I~} ~-~ ~1~ ~ ,, Eocene micaschists ~ [ ~ ~] C,retac,e9us [--''~A@d~ ~ ~...~ ~ 3930~ 26 " @~
.~
~
esoo
Aegean
.....
~
Sea
Chios ------~'%
~
" " . ~ x ~
Karaburun Pe~nsula :~. '~--~
~ ' :~ :.~.~ :~
grgl
~ ~ Manisa %
~'~"~'~":~
L.~~
"
I
G6rdes ~,-.~
~'t ?,,.~5 ^5 ,-J ] ~.,~~ ~ 4 : ~ ,.~~ ~t~,~ "-'-'~ '-'-~ ' Turgutlu .
:y
~ / ;
.:
:~?~;,~.." ' "
:
." . f~
:~"~'.~..~~~Se~fe4:~
. ~.. . . . . . . ~....~.~
.i~!7 : ~
~2~30'~--~U~J~'0,~'~.~'~ ~0~ 7 4~,km
BORNOVA FLYSCH ZONE [=--=---qPre-Lower Triassic ~U. CretaceousBasalt ~radiolarite, serpentinite Paleocene flysch v/i/I/. Accretlonarycomplex I i I Triassic-Cretaceous limestone I------4 rocks
Fig. 5. Simplified geological map of the region around the [zmir-Ankara-Erzincan suture between |zmir and Bahkesir (modified from Okay & Siyako 1993; Akyt~z & Okay 1996).
neritic limestone blocks ranges from MidCarboniferous (Bashkirian) to the latest Permian (Dorashamian) with a general trend of younging in the age of blocks eastwards towards the suture (Leven & Okay 1996). The Jurassic-Cretaceous sequence, largely eroded during the Early Tertiary, is preserved 15 km west of suture in the Balya region (Fig. 5). In this region Liassic sequence consists of continental to shallow-marine sandstones and shales, which are overlain disconformably by the Upper Jurassic-Lower Cretaceous neritic limestones, 800 m thick. These are unconformably overlain by Albian-Cenomanian pelagic limestones (Okay et al. 1991). The Bornova Flysch Zone to the east of the suture consists of chaotically deformed late
Maastrichtian to Palaeocene greywackes with Triassic to Cretaceous limestone and rarer peridotite, radiolarian chert, and basalt blocks (Okay & Siyako 1993). The limestone blocks, which reach a few kilometres in size, range in age from Middle Triassic to Maastrichtian ($ahinci 1976; Poisson & Sahinci 1988; Erdo~an 1990; Okay & Siyako 1993). Within the blocks the Triassic to Lower Cretaceous is represented by neritic platform carbonates, while Senonian is represented by red, pelagic, cherty limestones. Towards the east the limestone-flysch of the Bornova Flysch Zone is associated with a Cretaceous oceanic accretionary complex of radiolarian chert, basalt, pelagic shale, pelagic limestone, sandstone and serpentinite. Isolated bodies of peridotite rest on this accretionary complex (Fig. 5).
484
A . I. O K A Y
& O. TUYSOZ
l
00'
28 20'
S6ve
+
+\"vt..
+
+
4-5,,~,.,,\ - + ++ + 4v
'.._,v'"b
("v,,., <..v ,, '~
+$v~
"
+ ~ y/+
t , , ~ ,, v ,,
- ++ 4 ++ + /v,v ,)v v •
-
'vv v ,, ,,
~ # t
' V"),,,,~.y V v ,,~,fv
'
+
Susurluk 9
~,,~ v~d
I t %
~
+
+
+ +
+ -+
Alluvium, Neogene sedimentar) and volcanic rocks ]
Oligo-Miocene granodiorite
. o' ,,."
+
+ ~
+
+
.t.k~t, SiJnntkk + +
+
+ +
+ +
l - - 39~4()'
E~eler<
o,
9 9"
Kepsut 9
10 km I
~ Ma']']'~.
9
9 Tepek6y
Oligo-Miocene granodiorite
NW
SE
Blueschist, marble
o" I' o.
Blueschist
A
Metabasite, marble
~~
152 4
gkm
9 ~ v-
~ +
+
" ..
~
-t-
+
,r a,s c
A'
+
"-~,
~" -
BORNOVA FLYSCH and TAVSANLI ZONES
Greywackewith Permo-Carboniferous] efi l i m e s t o n e
metapelite
I
SAKARYA ZONE
Triassic
.~ -7--
v
9 .
+
+ + , , , q ~
v v -7-'-'J--"
Detachment fault
0
-I-
+ ~atalda~ Granod~orite 4t + -b + + + + + ~ +
,
Suture thrust fault
5 I
~
t
Foliation
0 !
+
+- " " ~ +
~,)
OlOCKS
m..b
~
Greywacke, shale
~
Basalt, radiolarite~erpentinite accretionary complex -[~ Blueschist metapelite, marble
Fig. 6. Geological map and cross-section of the suture region north of Bahkesir in northwest Turkey. For location see Fig. 5 (simplified from Akytiz & Okay 1996).
The Kepsut transect The suture trace between izmir and Bahkesir is almost completely covered by Neogene sedimentary and volcanic rocks, and only in the Kepsut region northeast of Bahkesir the pre-Neogene rocks from both sides of the suture are seen in contact (Fig. 5). A map and cross-section of this region based on Akyiiz & Okay (1996) are shown in Fig. 6. This region is characterized by a dome structure centred on the Oligo-Miocene ~atalda~ granodiorite, which has exposed the tectonic relation between the Sakarya and Bornova Flysch zones. The Sakarya Zone in the Kepsut region is made up of the
metabasite-marble-phyilite sequence of the Niliifer Unit tectonically overlain by Triassic greywackes with exotic Permian and Carboniferous limestone blocks (Akyiiz & Okay 1996). The Bornova Flysch Zone is represented by strongly sheared turbidites with Triassic and Cretaceous limestone as well as blueschist blocks. The chaotically deformed turbidites, which in many areas form a tectonic mdlange, are gradually replaced eastwards by a Cretaceous accretionary complex of basalt, radiolarian chert, pelagic limestone and shale. Blueschist-facies marbles and metapelites are exposed under the accretionary complex
TETHYAN SUTURES OF NORTHERN TURKEY I
:..
, ~
9
~t,"
t
J x_/ ~ - +2 +
rJ
485 I;~'i
'
l
_, ~
~
/
~
_
_
] _ I 2I -
++
.-.~.~,m,,,,,,.v~.-~_.. ~, ~2\ ~ ~ ~ ~
+
~
,
~
I~ h a r ~ B e y p a z a
__
~ ~
I
r U+.... ?
oEski~ehir ~ ~ ~ + + . F i g . i O
7 _ ~ Neogenedeposits ~ O l i g o - M i ..... g.... di~ "~2"-TT.'.""Esediments -~. ..... SAKARYAZONE AFYONZONE ~ J..... ic-E....... di.... ts ~-~--~ Mi .... hist,phyllite, ma.vble of Palaeocene metamorphic age Pre-Jurassic basement ~mainlyTriassicsubduction-accretionunits, r-.arakayaComplex) 0! 20! 401km
TAVSANLIZONE ~ Eocenegranodiorite ~ ~ ~
Peridotite Basalt,radiolarite~ serpentinite Cretaceous accretlonary complex isotopicBlueschistSagesWi Cretaceous th
Fig. 7. Geological map of the region around the lzmir-Ankara-Erzincan suture between Bursa and Nalhhan (modified from Okay 1984b).
(Fig. 6). In the Kepsut region the sheared turbidites of the Bornova Flysch Zone lie tectonically over the Niltifer Unit and the Triassic greywackes. This north-vergent thrust fault, named as the Btikdere Fault, represents the suture fault in the Kepsut region (Akytiz & Okay 1996). The Btikdere Fault is an upper crustal structure with a zone of cataclasis several metres thick. Prior to the Early Miocene updoming, related to the intrusion of the ~atalda~ granodiorite, the fault plane must have had a shallow southward dip. The klippe of greywackes and glaucophane-lawsonite blueschists west of SOve (Fig. 6) indicate a minimum northward translation of 40 km. The movement on the Btikdere Fault is constrained to the latest Cretaceous to Late Oligocene interval.
The Izrnir-Ankara-Erzincan Suture between Bahkesir and Beypazart Along this 280 km long east-west-trending segment of the izmir-Ankara-Erzincan suture between Bahkesir and Beypazan, the Sakarya Zone is in contact with a regional blueschist belt (the Tavsanh Zone) of the Anatolide-Tauride Block (Fig. 7). To the north of the suture there
are units of the Karakaya Complex and the overlying Jurassic-Cretaceous succession of the Sakarya Zone, whereas to the south there are regional blueschists with Cretaceous isotopic ages, a Cretaceous accretionary complex, ophiolite and Eocene granodiorites of the Tavsanh Zone (Okay 1984b). In this segment of the suture, continuous transects across the suture exists south of Bursa and in the geologically poorly known region northeast of Eskisehir (Fig. 7). In addition, in the area north of Sivrihisar, a sequence belonging to the Sakarya Zone outcrops in a fault-bounded block surrounded by the blueschists and peridotites of the Tavsanla Zone (Fig. 7).
The Orhaneli transect A well-exposed continuous transect across the suture between Bursa and Orhaneli based on the works of Lisenbee (1971), Okay & Kelley (1994) and Okay et al. (1998) is shown in Figs 8 and 9. Like in the Kepsut region, the Sakarya Zone in the Orhaneli region consists at the base of intercalated metabasite, marble and phyllite of Permo-Triassic age, over 3 km thick, strongly deformed and metamorphosed in greenschist facies during the latest Triassic. These are tectonically overlain by strongly deformed but unmetamorphosed Late
486
A.I. OKAY & O. TfSYSI3Z
'4N q-'3
v
v
~plnar - - --~..~4~
v
~"~v
v ./ \V
F"
v
v ~.. v
l
Topuk Granodiorite 5+ +_
Cebelgfine
0
2
I
4
6 km I
-4- + + 5- - + h+\ i 4 4+ + + +c/~ + § Orhaneli Granodiorite + / +
%
I
--] Alluvium, Neogene sedimentary and volcanic rocks SAKARYA ZONE U. J u r a s s i c ~ L Jurassic J - . ' - - . j Triassic
~
FaultTriassic (?) ~ - - v - ~ j
Bedding
~
t
I
I
1
Siltstone-shale m61ange ($ehriman m61ange)
TAVSANLI ZONE
Platform limestone (Bilecik Lst) Eocene ~ Granodiorite Sandstone, shale, conglomerate ~ Gabbro ] ~ (Baylrk6y Fm) Harzburgite l "-~ Cretaceous (?) Dunite ] oE"~ Greywacke,siltstone ] Garnet-amphibolitej N o (Orhanlar Greywacke) Fault Basalt, radiolarite, serpentinit~ Cretaceous Metabasite, phyllite, ] ~ W/I/IA accretionary complex marble (Nilfifer Unit) Fault ;~ Cretaceous 11 ] ] Blueschist metapelite, marble I'1'11
Foliation Mineral lineation"~"-'-'-'a~
High angle fault
Low angle basal fault
Fig. 8. Geological map of the suture region south of Bursa in northwest Turkey. For location see Fig. 7 (simplified from Okay et al. 1998).
Triassic greywackes, over I km in thickness, with exotic Permo-Carboniferous limestone olistoliths (Okay et al. 1998). The greywackes are unconformably overlain by 50-200 m thick continental to shallow-marine Liassic clastic rocks. Neritic Jurassic (Callovian to Tithonian) limestones, c. 800 m in thickness, lie
disconformably on the Liassic sandstones and congomerates (Altmer et al. 1991). The Tavsanh Z o n e to the south is made up at the base of a c o h e r e n t , clastic-limestone sequence, over 2 k m in thickness, metamorphosed in blueschist facies. The blueschists are strongly deformed and show a well-developed
TETHYAN SUTURES OF NORTHERN TURKEY ZONE Cretaceous blueschist Triassicmetabasite Suture fault metapelite,marble_ phylhte,marble ' Early Eocene bE granodiorite h Triassicgreywacke h Peridotite , h Af SAKARYA ZONE
NW lkm
A
0i 0
487
TAVSANLI
j. 2
4 km
Fig. 9. Geological cross-section of the suture region south of Bursa. For location of the section see Fig. 8.
foliation, isoclinal folding and an east-westtrending mineral lineation (Okay et al. 1998). The metapelites contain the typical blueschistfacies minerals jadeite, glaucophane and lawsonite. Geothermobarometry indicates HP/LT metamorphism at 20 + 2 kbar and 430 _+ 30~ (Okay & Kelley 1994). A r / A r isotopic data indicate a Late Cretaceous age (80-90 Ma, Santonian/Campanian) for the blueschist metamorphism (Okay & Kelley 1994). The blueschists are overlain by a Cretaceous accretionary complex of basalt, radiolarian chert, pelagic limestone, shale and serpentinite (Fig. 8). A Senonian age has been obtained from the pelagic limestones in the accretionary complex. A large slab of ultramafic rocks with minor gabbro and diabase dykes lies with a subhorizontal contact over the accretionary complex and the blueschists (Figs 8 and 9). The mineralogy of the diabase dykes in the peridotite indicates that the peridotite has not undergone the HP/LT metamorphism observed in the underlying blueschists. At the base of the ultramafic slab there are isolated slices of garnet-amphibolite, similar to the metamorphic soles described from the base of other ophiolites (e.g., Williams & Smyth 1973; Spray et al. 1984). The garnetamphibolite metamorphic soles give an A r - A r isotopic age of 101 _+4 Ma (Harris et al. 1994). Two granodiorites of Early E o c e n e age (52-48 Ma, Harris et al. 1994) intrude the blueschists and the overlying peridotite, and seal their contacts (Figs 8 and 9). The mineral assemblages in the contact metamorphic aureole of the granodiorites as well as Al-in-hornblende geobarometry indicates that the granodiorites were emplaced at a depth of c. 10 km (Harris et al. 1994). This indicates that the blueschists were exhumed from 60 km depth to 10 km in the Santonian-Early Eocene interval. In the Orhaneli area the suture trace is delineated as a narrow well-defined discontinuity ranging from a single strike-slip fault to a maximum 3.5 km wide discontinuous fault zone
bounding a tectonic m61ange (Fig. 8). This ,Sehriman m61ange consists of recrystallized limestone, greywacke, mafic volcanic rock and radiolarite blocks, ranging in scale from centimetre to kilometre, e m b e d d e d in a highly sheared siltstone-shale matrix. Ozko~ak (1969) reports poorly preserved G l o b o t r u n c a n a sp. tests in the fine clastics suggesting a post-Late Cretaceous age for the f o r m a t i o n of the m61ange. The suture fault system can be followed for 60 km along strike in the area studied (Fig. 7). Individual faults dip steeply (70-80 ~) towards the north or northwest. They are upper-crustal, brittle faults and the main fault branch cuts the Miocene conglomerates of the Erenler Basin northwest of G6ktepe. The Erenler Basin is a small Miocene pull-apart basin, bounded to the north and to the south by two fault strands of the suture fault (Fig. 8). This suggests that the suture zone acted as a sinistral strike-slip fault during and after the Miocene. The zone was probably also active during the Eocene, as the Topuk granodiorite intrudes very close and parallel to an off-shoot of the main suture fault (Figs 7 and 8). The Sivrihisar transect In the Sivrihisar region, 50 km south of the main suture trace, rocks of the Sakarya Z o n e occur as an east-westtrending fault block, 30 km long and 3 km wide (Fig. 7). A geological map and cross-section of this region, based on the works of Kulakslz (1981), Gautier (1984) and Monod et al. (1991) are given in Fig. 10. In the Sivrihisar region the Sakarya Z o n e is represented at the base by strongly sheared greywackes and shales with exotic blocks of basalt and Carboniferous, Permian and Middle Triassic limestone, forming part of the Karakaya Complex (Monod et al. 1991). These are unconformably overlain by basal clastic rocks of Jurassic age, which pass up to Middle Jurassic (Bathonian) to Lower Cretaceous (Upper Berriasian) limestones, 800 m in thickness (Gautier 1984). Senonian sandstone
488
A.I. OKAY & O. TUYSOZ 30' -- 39"40'
V V
V
Haliba~
v
V V
V V
K~zflca6ren v V
V V
V V
V V
V V
V V
V
V V
V
V V V
v
V V
"~ v
_ V
v
v
I
IIII I I~- I I
1I 111.1 SW A [ lkm
Dtimrek
11111 iiiii NE
Blueschist marble, Eocene metapelite Peridotite granodiorite k ~' . . . . . . . . i ~ . ~ ~
Eocene conglomerate . 9 . "~ Jurass,c isI. ,, Peridotite
A'
_------
0
[Alluvium,Neoeene ~ sedimentary an'd volcanic rocks'
Detachment fault
~
SAKARYA ZONE Eocene
~
L. C r e t a c e o u s ~ Jurassic ~ Triassic ~ I
0 I
2 I
[
4 ~
Thrust fault TAVSANL1 ZONE
Conglomerate.marl,sandstone Eocene
Granodiorite
Limestone Conglomerate, sandstone Grevwacke, shale, basalt, exotic Permo-Carb. 1st blocks (Karakaya Complex) 6 km
Peridotite Amphibolite
|
Cretaceous (?) ~ .,...... Cretaceous
v/.f/.f/.~ ~1#1#~
Cretaceous
~
~
Basalt, radiolarite, serpentinite complex Blueschist metabasite Blueschist marble, metapelite
accretionary
Fig, 10, G e o l o g i c a l m a p a n d c r o s s - s e c t i o n of the Sivrihisar r e g i o n . F o r l o c a t i o n o f t h e s e c t i o n see Fig. 8 ( s i m p l i f i e d a n d m o d i f i e d f r o m M o n o d e t a l . 1991).
and shales with G l o b o t r u n c a n a sp., probably representing the uppermost part of this cover sequence, occur as a fault-bounded block north of Dtimrek (Fig. 10, Gautier 1984). The Tavsanh Zone in the Sivrihisar region consists of a marble-schist sequence showing a well-developed foliation and isoclinal folding. This sequence contains generally greenschistfacies mineral assemblages, however, relicts of lawsonite and sodic amphibole in the schists indicate an earlier period of blueschists facies metamorphism (Monod et al. 1991). In an area
south of Haliba~ the metamorphic sequence is rich in metabasic rocks, and the blueschist minerals, including glaucophane, lawsonite, sodic pyroxene and garnet, are well preserved. K - A r and A r - A r isotopic ages from this Haliba~l region indicate a Senonian age for the HP/LT metamorphism (Sherlock et al. 1997). The blueschists are tectonically overlain by ultramafic rocks and are intruded by Mid-Eocene granodiorites (Fig. 10). Tectonic slices of accretionary complex and sub-ophiolitic garnetamphibolites occur under the peridotites. A
TETHYAN SUTURES OF NORTHERN TURKEY Campanian age has been obtained from a pelagic limestone block in the accretionary complex west of the village of Karakaya (Fig. 10, Gautier 1984). In the Sivrihisar region the Sakarya sequence lies tectonically over the blueschists with an intervening tectonic slice of ultramafic rocks (Fig. 10). An upper age limit for the thrusting of the Sakarya Zone over the Tavsanh Zone is given by the Lower to Middle Eocene terrigenous to shallow marine sedimentary rocks, which lie unconformably over both the Sakarya sequence and the ultramafic rocks (Fig. 10). A lower age limit for the thrusting is provided by the Campanian pelagic limestones in the accretionary complex. A second phase of thrusting occurred after the Mid-Eocene, when the initial thrust stack was imbricated and the ultramafic rocks were emplaced over the Eocene clastic rocks (Fig. 10). The Sakarya sequence in the Sivrihisar region, which is stratigraphically similar to that observed in the Central Sakarya Basin north of Bilecik (Figs 4 and 7), represents a klippe of the Sakarya Zone, comparable with the Dent Blanche Nappe in the Western Alps (e.g., Trtimpy 1980). This Sivrihisar klippe indicates that during the Eocene, the front of the suture thrust was at least 50 km farther south than the present suture.
Evolution o f the i z m i r - A n k a r a - E r z i n c a n suture between [zmir and Beypazart Age of the izmir-Ankara-Erzincan ocean. The best evidence for the age span of the izmir-
Ankara-Erzincan ocean would be the ages recorded in the accretionary complex, which represent the cover sediments of the subducted Tethyan oceanic crust. However, at present there are very few documented radiolarian ages from the widespread accretionary complexes in the western Anatolia. Servais (1982) mentions Upper Jurassic-Lower Cretaceous radiolaria from cherts from an accretionary complex north of Kiitahya, while Okay & Kelley (1994) describe Pithonella ovalis, a foraminifera characteristic for Cenomanian-Maastrichtian from a limestone-radiolarian chert slice from southwest of Bursa (Fig. 7). A more precise Campanian age, based on species of Globotruncana, has been obtained from Couches Rouges limestone blocks in the accretionary complex in the Sivrihisar region (Fig. 10, Gautier 1984). The ages of the ophiolites overlying the accretionary complex are not known directly, as only the lower plutonic parts of the ophiolite sequence are preserved. However, isotopic studies in the metamorphic soles of several ophiolite bodies, e.g., in the Semail ophiolite in Oman (Lanphere 1981), in Greek and Yugoslavian ophiolites (Spray et al. 1984), have shown that the age of the metamorphic sole is very similar to that of the overlying ophiolite. The isotopic ages from the subophiolite metamorphic rocks from northwest Turkey (c. 101 Ma, Onen & Hall 1993; Harris et al. 1994) suggest an Albian-Cenomanian age at least for some of the ophiolites. These data give a minimum age span of Late Jurassic to Campanian for the Izmir-Ankara-Erzincan ocean.
Senonian H P - L T metamorphism and deformation. The first Alpide orogenic event south of
Obducting oceanic litosphere Blueschist metamorphismof the accrctionary prism Anatolide-TaurideBlock at 60 km depth (Tav~anh ZOne) Disruption of the Anatolide-TaurideBlock
~.. . . . . Fl,schzooo,, Translormfault
/
/
/
]
489
, /~,~\/
7-J-~'~-~/
----<-~'~/ .~r
--
~/
__I"/
-------"-~
-d . ~.D
// / ~
1
%
"?"
Fig. 11. Block diagram illustrating the Senonian evolution of the northwestern margin of the Anatolide-Tauride Block.
490
A.I. OKAY & O. TuYSOZ
the suture is the blueschist metamorphism and related deformation of the Anatolide-Tauride Block, which took place during the early Senonian (Okay et al. 1998). During the Senonian there was continuous marine sedimentation in the Sakarya Zone north of the suture (Fig. 4), which shows that the HP-LT metamorphism and related deformation were not related to the continental collision between the Anatolide-Tauride Block and the Sakarya Zone. The HP-LT metamorphism was caused by the deep burial of the northern margin of the Anatolide-Tauride Block under a southward obducting oceanic lithosphere (Fig. 11) in a similar manner to that inferred in Oman (e.g., Goffe et al. 1988; Michard et al. 1991). The Senonian corresponds to a major change in sedimentation in the westernmost part of the Anatolide-Tauride carbonate platform as observed in the limestone blocks in the Bornova Flysch Zone. In different limestone blocks in the Bornova Flysch Zone Senonian red pelagic limestones rest unconformably over Triassic or Jurassic neritic limestones indicating a period of erosion prior to the deposition of the pelagic carbonates (Poisson & Sahinci 1988; Okay & Siyako 1993). This phase of uplift and erosion of the Triassic-Lower Cretaceous carbonate
platform must have been related to the Late Cretaceous loading of the Anatolide-Tauride Block by ophiolite, accretionary prisms, and continental margin sediments, widespread relicts of which are preserved as erosional remnants. The progressive southwestward migration of this disruption of the carbonate platform starting from the beginning of Santonian in the Savastepe-Soma region through end of Santonian in the area between [zmir and Seferihisar and finally to Campanian in the Karaburun peninsula indicates a northeasterly origin for the allochthons, which supports a temporal and causal relation between the blueschist metamorphism and disruption of the Mesozoic carbonate platform (Okay & Siyako 1993; Okay et al. 1996). However, the absence of regional metamorphism in the Bornova Flysch Zone indicates that it was never deeply buried under the southward obducting ophiolite and accretionary complexes, which therefore must have by-passed this westernmost margin of the Anatolide-Tauride Block (Fig. 11). This also provides an explanation for the scarcity of oceanic accretion complexes and ophiolite along the suture between Izmir and Bahkesir, while they are common to the east farther away from the suture (Fig. 5). The lzmir-Ankara-Erzincan suture between
Fig. 12. Stratigraphic sections of the northern (Mudurnu-GOyntik area, after Meriq & Sengiiler 1986 and Altlner et al. 1991) and southern margins (Nalhhan area, after Tansel 1980, and Altlner et al. 1991) of the Central Sakarya Basin, and the Haymana Basin (after f2nalan et al. 1976). For location of the sections in the Central Sakarya Basin see Fig. 7. All the stages and substages shown have been determined palaeontologically.
TETHYAN SUTURES OF NORTHERN TURKEY [zmir and Bahkesir forms a discordant segment between the northwest-trending Vardar suture and east trending i z m i r - A n k a r a - E r z i n c a n suture east of Bahkesir (Fig. 1). A number of observations suggest that this is an original feature reflecting a transform fault margin connecting two passive continental margin segments (Okay et al. 1996). These observations are: (a) The Triassic to Lower Cretaceous limestone blocks in the Bornova Flysch Zone are neritic and betray no evidence of a proximity to a passive continental margin; (b) the Senonian obduction, which led to the 60 km deep burial and H P - L T metamorphism of the northern margin of the Anatolide-Tauride Block, did not result in the regional metamorphism in the Bornova Flysch Zone, which would have been highly unlikely if the northern margin of the Anatolide-Tauride Block was rectilinear; (c) arc-magmatism of Late Cretaceous age, common farther east, is not found northwest of the suture between lzmir and Bahkesir; (d) an originally northeast-trending passive continental margin would be unlikely in the Tethyan ocean, which was an east-west-trending seaway characterized by similar trending passive continental margins and subduction zones (e.g., Seng6r et al. 1988; Ricou 1994); (e) in Greece, the ophiolites were obducted during the earliest Cretaceous southwestward over the Vardar and Pelagonian zones (e.g., Burchfiel 1980; Jacobshagen 1986). This earliest Cretaceous ophiolite obduction can be traced as far east as the Sporades islands in the Aegean Sea (Jacobshagen & Wallbrecher 1984), while in the Karaburun peninsula the carbonate sedimentation continues without a break from Ladinian to the Albian (Erdo~an et al., 1990) with no evidence of an Early Cretaceous deformation. All of these features suggest that a transform fault formed the northwestern margin of the Anatolide-Tauride Block. During the Early Cretaceous the transform fault relayed the movement associated with the ophiolite obduction in Greece to an intra-oceanic subduction zone north of the Anatolide-Tauride platform (Okay et al. 1996), while during the Senonian it initiated and controlled the western limit of the obducting ophiolite (Fig. 11). S e n o n i a n s u b d u c t i o n and earliest Palaeocene continental collision. The onset and duration of
subduction is probably best dated by the arc magmatism. In the Central Sakarya Basin, north of the suture between Ballkesir and Beypazari (Fig. 7), the Valanginian to Aptian sequence is represented by pelagic micrites deposited in a south-facing passive continental margin (Fig. 12,
491
Saner 1980; Altlner et al. 1991). These pass up into a heterogeneous and lateraly variable sequence of siliciclastic turbidites, pelagic limestones, olistostromes, tufts and lava flows of Cenomanian to Maastrictian age, which were deposited in a fore-arc basin above the northward subducting izmir-Ankara-Erzincan ocean ( ,Seng6r & Yllmaz 1981). Blocks of serpentinite, radiolarian chert and basalt, which occur in the Maastrichtian turbidites (Saner 1980) are probably derived from the emergent segments of the accretionary complex in the south. Collision between an arc and passive margin would be reflected in a change of subsidence and sedimentation pattern in the opposing continental margin. The onset of collision would produce a regressive siliciclastic sequence and will be followed by regional folding and thrusting, and finally by uplift. Therefore, high-resolution stratigraphic data would be able to date the onset of collision fairly accurately (e.g., Rowley 1996). However, as shown in this study, the region south of the suture bears the strong imprint of the obduction-related, pre-collisional deformation and metamorphism, and most of the Mesozoic and Lower Tertiary cover has been stripped off from a several-kilometres-wide belt north of the suture between [zmir and Beypazan (Figs 5 and 7). The Cretaceous-Lower Tertiary sequence nearest to the suture is found in region of Nalhhan (Fig. 7). Here, the Tithonian to Early Campanian is represented by pelagic micrites, the first sandstone interbeds begin in the Early Campanian and by the Early Maastrichtian the sequence has been transformed into siliciclastic turbidites with rare marl intervals with planktonic microfossils probably deposited in a forearc basin (Fig. 12, Tansel 1980). The turbidites show a regressive development so that by the Late Maastrichtian the sequence consists of conglomerate and sandstones with benthic foraminifera. These pass up without a break into terrigeneous red conglomerates and sandstones presumably Early Palaeocene in age (Tansel 1980). This is followed by south-vergent thrust imbrication whereby the Cretaceous sequence is emplaced over the Palaeocene red beds. The stratigraphic data from the Nalhhan region indicates that the arc-continent collision occurred at around the Maastrictian-Palaeocene boundary at about 65 Ma. Geological relations in the Sivrihisar region indicate that during the Palaeocene collision the Sakarya sequence was sheared off from a level below the Karakaya Complex and was thrust south for at least 50 km over the Anatolide-Tauride Block. The major southward translation and subsequent erosion of the Sakarya Zone also provide an explanation for
492
A. I. OKAY & O. TUYSOZ
BLACK
l 34' ~ ............ ~
T
33 _ , ~
32
'
:
'
"
r 36'
~incm
SEA
Bolu
~ankm Fig. 9Beypazarl
"- - ,a~ l V d ~ , ~ ~ Sivrihisar " ~ - ~ ~2~i~) Hayma
~K ama ~ . ,%,_, z~ 9 ....
~! [[~I~
i~ ~
Fig.16
I La~'i1 Salt~
t;
~
~
~
~
- 7r `
~'~
~
~
316~39
-
~
N,eoge.nedeppsits.andTertiary ~ Accretionarycomplex,op,hiolite, ueposlts soum ol the suture intra-oceanicfore-arc anti mtra-arcdeposits ISTANBUL ZONE SAKARYA ZONE TAURIDES ~ - ~ Senonianisland-arc ~ Senonianisland-arcvolcanics~'~ sedimentary Palaeozoic-Mesozoic sequence volcanics ~ Jurassic-Eocene sedimentarysequence ~ otCampanian-E~ raaymana t~asm'dep~ KIRSEH[R MASSIF Ordovician-Triassic ~-~_~1~ Jurassic-EocenesedimentaryF~T~ ~ Cretaceousgranites sedimentarysequence sequence Cretaceous metamorphic basementPrecambriacrystalline n +l+l+l+l~+~-q ~ rocks 1++,++] Mid-Jurassicgranites k--"~ TAVSANLI ZONE rx-x-x] Pre-Jurassicbasement 9 ~<>
i
i
i
I
Crctaccous blueschists Fig. 13. Geological map of the region around the izmir-Ankara-Erzincan suture between Nalhhan and Tokat.
the apparent anomaly that the Jurassic sandstones and limestones, which south of Bursa approach as near as 1.5 km to the suture (Fig. 8), still retain their shallow-marine character and show no evidence of a proximity to a continental margin. The collision-related southward thrusting was followed by the intrusion of Early Eocene granodiorites. During the Miocene the suture was locally used by the strike-slip faults related to the North Anatolian Fault.
The Haymana loop of the lzmir-Ankara-Erzincan suture between Beypazart and Ankara The width of the Tavsanh Zone increases towards the east and shows its widest extent of about 150 km south of Beypazan and then is abruptly terminated along a poorly defined line trending south from Beypazan (Fig. 13). To the east of this line there is a major turbidite-dominated Late
TETHYAN SUTURES OF NORTHERN TURKEY Cretaceous to Eocene basin, called the Haymana Basin. The contact between the Haymana Basin and blueschists are covered by extensive, posttectonic Neogene sedimentary rocks but is likely to be tectonic. The basement of the Haymana Basin consists of an accretionary complex and the structurally underlying Jurassic-Lower Cretaceous carbonates of the Sakarya Zone (Fig. 12, Onalan et al. 1976; Batman 1978; G6rt~r et al. 1984; Koqyi~,it, 1991). The sequence in the Haymana Basin starts with a few tens of metres-thick basal conglomerates, which pass up into a 1800 m thick distal turbidites of Upper Campanian to Maastrichtian age. The basin shallows in the latest Maastrichtian and the turbidites are overlain by Palaeocene terrigeneous conglomerate and sandstones up to 1300 m in thickness (Unalan et al. 1976; G6rtir et al. 1984). Ophiolite and blueschist detritus, including pebbles of glaucophane-lawsonite schists, have been described in the Upper Campanian-Maastrichtian turbidites (Batman 1978) and Palaeocene clastic rocks (Norman & Rad 1971). The Palaeocene terrigenous siliciclastic rocks are overlain by a second cycle of regressive turbidite sequence, up to 2000 m thick, of Early to Mid-Eocene age (Fig. 12, Unalan et al. 1976, G6riJr et al. 1984, Koqyi~it 1991). The stratigraphic position of the Haymana Basin over an accretionary complex, its tectonic setting along a suture and the clastic dominated Upper Cretaceous-Eocene sequence, over 5 km in thickness, led to its interpretation as a fore-arc basin on the southern edge of the Sakarya Zone (G6rtir et al. 1984; Koqyi~it 1991). The shallowing of the basin during the latest Maastrichtian, and evidence for earliest Palaeocene continental collision from farther west, indicate that the Haymana Basin is a composite basin; the lower section of Campanian-Maastrichtian age represents a fore-arc basin, whereas the overlying Paleocene-Middle Eocene sequence has been deposited in a foreland basin formed during and after the continental collision. The lower part of the Haymana sequence can be compared with the turbidites of the same age in the Central Sakarya Basin, which had a similar tectonic setting (Fig. 12). However, the accretionary complex which occurs below the Upper Campanian turbidites of the Haymana region is not observed to the west. This enigmatic mid-Cretaceous event of northward obduction of the accretionary complex over the Sakarya Zone, just before the onset of northward subduction, is a feature observed throughout the rest of the IzmirAnkara-Erzincan suture.
493
T h e ~ a n k t r t l o o p o f the lzmir-Ankara-Erzincan
suture
Between the H a y m a n a Basin and Sivas the i z m i r - A n k a r a - E r z i n c a n suture defines the contact between the Sakarya Zone and the Klrsehir Massif. In this large region the suture is generally represented by post-Eocene thrusts which emplace rocks of the Sakarya Zone over the Tethyan subduction-accretion complexes. The Tethyan subduction-accretion complexes form a 5-10 km wide tectonic belt, which circles and radially thrusts the Eocene-Miocene sedimentary rocks of the (~anklrl Basin resulting in a large loop of the suture (Fig. 13). The A n k a r a transect. A geological map and cross-section of a relatively well-studied part of the suture east of Ankara is given in Figs 14 and 15 (Akytirek et al. 1988). This region forms part of the classical area of Ankara M61ange studied by Bailey & McCallien (1953). The Sakarya Zone in this region shows a similar stratigraphy as that in the previous traverses (cf. Figs 8 and 10). At the base there is a metabasite-phyllite-calc-schist sequence with rare ultramafic lenses metamorphosed in greenschist facies (Koqyi~it 1987; Akytirek et al. 1988). This is overlain tectonically, by a chaotically deformed Late Triassic series of greywacke, siltstone, basalt and exotic blocks of Carboniferous and Permian neritic limestone. Little deformed Liassic conglomerate, sandstone and siltstone with ammonitico rosso horizons lie with a pronounced unconformity over the deformed Triassic series (Fig. 14). The Liassic clastic rocks pass up into hemipelagic limestones of Upper Jurassic-Aptian age, which are preserved as small erosional outliers (Fig. 14; Ko~yi~it 1987; Akytirek et al. 1988). This Sakarya sequence is thrust eastward over the Cretaceous subduction-accretion units. A minimum eastward translation of 30 km can be estimated from the small tectonic windows in the Sakarya basement exposing ophiolitic units (Akytirek et al. 1988). The Cretaceous subduction-accretion complex in the Ankara region can be divided into two composite thrust sheets (Akytirek et al. 1988; T/Jystiz et al. 1995). The upper sheet consists essentially of ophiolite imbricates including peridotite, gabbro, basalt, Cenomanian to Maastrichtian pelagic limestone and radiolarian chert, a few metres to a few kilometres in size (Fig. 14, Ttiyst~z et al. 1995). The lower thrust sheet, termed the Kalecik Unit, has a more ordered stratigraphy and consists of a Cenomanian to Maastrichtian volcano-sedimentary sequence, c. 3500 m thick, deposited on an ophiolitic
494
A.I. OKAY & O. TUYSUZ 33'010,
~
N
dP-(.~5~ -7 . . . . . . .
33' 10' ] Neogenesedimentaryand volcanicrocks
SAKARYAZONE ~
40'20'
-
~
I'INa','T1 l I I1"1.%~ 4o~oo, 33'30,] "
]i~ }~] Paleoceneturbidites
UpperJurassic limestone
~ 1
Maastrichtianturbidites
Liassic sandstone,conglomerate
~
Cenomanian-Campanianvolcanogenic sandstone, shale, Basalt Olistostromeswith ophioliteblocks Pelagic limestone,radiolarite Basaltic pillow lava
Metabasite,phyllite,calcschist I
; - *i
33'20' SUBDUCTION-ACCRETIONUNITS
.......... Unconformity ll~ ~ Triassicsandst~176 with Permianand Carboniferous1st,blocks
0 I
| I" '
~
~ ~ ' [
A~ 33!20,
4 I
I
8 km I
~
Gabbro Peridotite
~ ~ T
hrust fault
Fig. 14. Geological map of the suture region northeast of Ankara (simplified from AkyOrek et al. 1988).
substratum (Fig. 14, TOysOz et al. 1995). The sequence consists essentially of pelagic limestones intercalated with basalts and basaltic andesites with geochemical signatures suggestive
of island-arc and/or hot-spot magmatism (Floyd 1993; TOys0z et al. 1995). Debris flow and olistostrome horizons, up to several hundred metres in thickness, with blocks of peridotite, gabbro,
TETHYAN SUTURES OF NORTHERN TURKEY
495
NW Triassicvolcanosedimentary SE A Permo-Triassic sequencewith olistoliths Dismembered Cenomanian-Palaeocene A' phyllite j ~h~ m~~ 2 km metabasite, v -. ~ " ^ . _ ~ . 7 "v'~ . . . . . . intra-arcto forearc sequence[_
I ~
2
~
~
_
_ 9
~
F
4 km
Fig. 15. Geological cross-section of the suture region northeast of Ankara. For location of the section see Fig. 14.
basalt, occur in the pelagic limestone or pyroclastic rocks. This volcanic rock-pelagic limestone sequence is overlain by an over 1000 m thick regressive turbidite sequence of sandstone, siltstone and shale of Maastrichtian age. Ttiystiz et al. (1995) suggested an ensimatic intra-arc for the depositional setting of the lower part of the Kalecik Unit. This is based on its ophiolitic basement, the close intercalation of basalt and pelagic limestone, absence of continent-derived detritus, and on the geochemistry of the magmatic rocks. The ophiolite olistoliths in the Kalecik sequence were probably derived from emerged segments of the oceanic accretionary complex. The turbididic upper part of the Kalecik Unit was probably deposited in a fore-arc basin following a possible landward migration of the magmatic arc. T h e K a m a n transect. A r o u n d the ~ a n k l n loop the Cretaceous subduction-accretion units are thrust over the Eocene-Miocene sedimentary rocks. Their relation with the Klr,sehir Massif is best observed in the region of Kaman northeast of the Salt Lake (Fig. 16, Seymen 1981, 1983, 1984). In the Kaman region the lowest exposed unit is an isoclinally folded sequence of gneiss, migmatite, micaschist, metaquarzite, calcsilicate rock and marble m e t a m o r p h o s e d in amphibolite to granulite facies. The clinopyroxene + orthopyroxene + hornblende + plagioclase + quartz paragenesis in the gneisses, and the presence of sillimanite, cordierite and garnet in the micaschists indicate P - T conditions of 700 to 800~ and 3-5 kbar (Seymen 1981). The age of this HT/LP metamorphism is believed to be Cretaceous based on K - A r cooling ages from farther southeast (G6nc~o~lu 1986). The metamorphic rocks in the Kaman region are tectonically overlain by an u n m e t a m o r p h o s e d Cretaceous accretionary complex of gabbro, dolerite, basalt, radiolarian chert, pelagic limestone, greywacke, shale, and serpentinite. Cam-
panian ages have been obtained from the pelagic limestones from the accretionary complex (Seymen 1983). A large ultramafic slab lies structurally above the accretionary complex (Fig. 16, Seymen 1981). Although the contact between the metamorphic rocks and the overlying accretionary complex has been interpreted as a thrust (Seymen 1981, 1983), it could be a detachment fault with a normal sense of displacement, which allowed the exhumation of the m e t a m o r p h i c rocks. Both the m e t a m o r p h i c rocks and the overlying accretionary complex are cut by a granodiorite with a Rb/Sr age of 71 _+1 Ma (Ataman 1972). Terrigenous to shallowmarine clastic rocks and limestones of Maastrichtian to Lower P a l a e o c e n e age lie unconformably over the accretionary complex (Fig. 16; G6rtir et al. 1984; Seymen 1984). Evolution suture Age
of the Izmir-Ankara-Erzincan
region of
the
between
Beypazart
lzmir-Ankara-Erzincan
and Tokat ocean.
Recent determination of radiolaria from different chert blocks from a single locality of Cretaceous accretionary complex northwest of A n k a r a has yielded Upper Norian, Lower Jurassic, K i m m e r i d g i a n - T i t h o n i a n and A l b i a n Turonian ages (Bragin & Tekin 1996). These radiolarian chert blocks occur in an ophiolitic m61ange and are associated with blocks and thrust slices of gabbro, basalt, pelagic limestone and volcano-clastic rocks. Red radiolarian chert blocks with T i t h o n i a n radiolaria are also recently described from a slice of accretionary complex, several metres thick, imbricated in the basement of the Sakarya Zone north of Tokat (Bozkurt et al. 1997). These data indicate that the age of the l z m i r - A n k a r a - E r z i n c a n ocean spans at least from Late Norian to Albian, while the age of the Kalecik Unit further extends the age of the ocean to the end of Maastrichtian.
496
A.I. OKAY & O. TfJYSI]Z
Dan6zti
++++ +++~--~
% + +++ -t-+++ 4 + + +x..-~, + "++ +++c + + + +~--
lake ~
9v
v 5
10km
i
i "~x-.
I
NW 2 km.[ Paleocene granitoid |
High grade metamorphic rocks _,
Accretionary complex
,J I]
Paleocene
SE ]-"
O
A [,
-
2
4
6 km
7 Eocene and younger deposits
~
~
Paleocene granitoid ~
Peridotite
~
mi~matite, micaschist~ Gneiss, marble witti Cretaceous isotopic ages
Maastrichtian-Paleocene deposits ~i~
A'
Basalt, radi~ serpentinite Cretaceous accretionary complex
Detachment fault ~__..__....~ Thrust fault
Fig. 16. Geological map and cross-section of the western part of the Klr~ehir Massif illustrating the relation between Cretaceous metamorphic rocks, accretionary prism and granitoids (simplified from Seymen 1984). Senonian subduction and latest Maastrichtian collision. The age of the Kalecik Unit, interpreted as intra-arc to fore-arc deposits, indicates that the subduction in the izmir-Ankara-Erzincan Ocean in this segment started during the Cenomanian-Turonian, which ties well with the inception of arc volcanism in the Central Pontides during the Coniacian (Fig. 4, Ttiystiz 1993). The Senonian subduction must have built up a very wide accretionary complex south of the Sakarya Zone; even after the shortening follow-
ing the collision, the width of the accretionary complex, estimated from erosional outliers, is in excess of 80 kin. This wide accretionary complex was thrust over the Klrsehir Massif during the early Maastrichtian. The relation between the subduction-collision processes in the |zmir-Ankara-Erzincan, and granitic magmatism and associated regional metamorphism in the Klrsehir Massif is difficult to ascertain in the absence of reliable and precise geochronological data, especially on the
TETHYAN SUTURES OF NORTHERN TURKEY
[ , ~ i ~-f BaZaar._..
B L A C K Or.du
S E A ..-. Trabzon
497
Paz.~ - v ~ . . . Y ")~x~v ZX " ~ V ~
~/
~--Q412
Oires n
0"
w.)_'_ '
-_I~-
_ _ _
~
v
.~ _ ) I .
-
'_5.%_
-
-~
~ J ~
----ta.of-u.~o,10rtumf
Horasan 940' grzurum
I
Post-Eocene terrigenousrocks
Middl.eEocene I" 9 1 deposits
~Basalt radiolarite serpentinite~ accreti~narycomi~lex
SAKARYAZONE
TAURIDES
~
~
U. Cretaceous andesite,dacite,tuff1| EOUtterrn Cretaceous granitoid
I Pontides
Kuruko-type sulphideore deposits} -- Jurassic-Paleocene Pre-Jurassicbasement
~ Inner Eastern Pontides
Peridotite
Triassic-Cretaceouslimestone - - ~ Cretaceousgranitoid
KIRSEHiR MASSIF Micaschist,marblewith Cretaceous isotopic ages 0 40 80 km I
i
i
i
I
Fig. 17. Geological map of the Eastern Pontides (modified from Okay & Sahintiark 1998). age of regional metamorphism. If the regional metamorphism in the Klrsehir Massif is Aptian-Cenomanian in age, as seems likely from Rb/Sr isotopic data from the intrusive granitoids (e.g., Gtileq 1994), then it can not be genetically related to the subduction or collision along the lzmir-Ankara-Erzincan suture. A possible cause for the HT/LP regional metamorphism and granitic magmatism in the Klrsehir Massif is mid-Cretaceous northward subduction of the Inner Tauride Ocean under the Klrsehir Massif (Fig. 1). Collision between the Klrsehir Massif and the Sakarya Zone is poorly constrained, especially since the Cretaceous sequences have been eroded in the Sakarya Zone up to 60 km north of the suture (Fig. 13). However, the regressive development of the Maastrichtian turbidites of the Kalecik Unit, interpreted to have been deposited in a fore-arc, suggests uplift of the forearc during the latest Maastrichtian due to underthrusting of the Ktrsehir Massif. A collision of
latest Maastrichtian age would also explain the upper Maastrichtian and Palaeocene terrigeneous to shallow marine clastic and carbonate rocks, which lie unconformably over the accretionary complex in the Kaman region (Fig. 16). Like the case farther west, the collision in the Ankara-Tokat segment involved the delaminat ion of the Sakarya Zone and the southward thrusting of the upper part of the sequence over the ophiolitic units and accretionary complex (Fig. 15). The steep belt, marking the subduction of the lower continental crust must lie farther north than the present suture trace.
The Izmir-Ankara-Erzincan suture south of the Eastern Pontides The Klrsehir Massif tapers out towards the east, and the Sakarya Zone comes again in direct contact with the Anatolide-Tauride Block in the region south of the Eastern Pontides (Fig. 1). In
498
A.I. OKAY & O. TUYSIJZ
this 320 km long east-west-trending segment the Alpide orogen shows an organization as expected from a classical collisional orogeny (Fig. 17). To the north in the Outer Eastern Pontides there is a well-developed major Upper Cretaceous magmatic island arc, which borders to the south a Senonian fore-arc region represented by the Inner Eastern Pontides. Farther south there is a wide suture zone marked by large bodies of peridotite and accretionary complex. To the south of the suture is a thick Mesozoic carbonate platform overlain by a thrust sheet of ophiolite and accretionary complex. Below we first outline the main geological features of the Eastern Pontide magmatic arc, which forms one of the largest Tethyan magmatic arcs, then describe a wide transect from Gt~mt~shane to the Munzur mountains. The Eastern Pontide Magmatic Arc. The Eastern Pontide Magmatic Arc is a 40 km wide zone bordering the Black Sea characterized by an over 2 km thick sequence of Senonian volcanic and intercalated sedimentary rocks and intrusive granodiorite plutons (Boccalatti et al. 1974; Akin 1978; ~eng6r & Yflmaz 1981; Okay & S ahinttirk 1998). The magmatic arc is interpreted to have formed during the northward subduction of the | z m i r - A n k a r a - E r z i n c a n ocean under the Sakarya Zone (Akin 1978; Seng6r & Yflmaz 1981). Small inliers of Upper Jurassic-Lower Cretaceous limestones in the outer Eastern Pontides indicate that the magmatic arc has been constructed on the Lower Cretaceous and older rocks of the Sakarya Zone. The Senonian volcanic cycle starts with basalts and andesites and passes to dacitic and rhyolitic lavas and breccias and tufts with limestone intercalations and ends with a variegated series of mudstone, marl, tuff, radiolarian chert and limestone (SchultzeWestrum 1962; Zankl 1962; Hirst & E~in 1979; Akmcl 1984). The volcanic sequence, over 2000 m in thickness, is well dated through the intercalated limestones and extends from Turonian (Taner & Zaninetti 1978) to Late Maastrichtian and Danian (Hirst & E~in 1979), which places tight constraints on the duration of subduction. The Senonian volcanic rocks are generally subalkaline and give typical island arc geochemical signatures (Peccerillo & Taylor 1975; Akin 1978; E~in et al. 1979; Manetti et al. 1983; Aklncl 1984). Large number of granodiorite intrusions are associated with the Cretaceous volcanic rocks (Fig. 17). Their isotopic ages range from about 95 to 65 Ma (Taner 1977; Moore et al. 1980). Although they show a wide modal scatter, they are dominantly hornblende-biotite
granodiorite and quartz-diorite; geochemically the granitoids show a typical calc-alkaline trend (Taner 1977; Moore et al. 1980). In the Eastern Pontide magmatic arc there are also numerous massive and stockwork-type polymetallic (Fe, Cu, Pb, Zn) sulphide ores, which occur within the Maastrichtian dacitic-rhyolitic lavas, breccias and tufts (Akin, 1978; E~in et al., 1979; Hirst and E~in, 1979; Aklncl 1984). These are very similar to the Japanese Kuroko sulphide ores of Miocene age formed during arc volcanism around the volcanic centres (Mitchel & Garson 1976). The Cretaceous magmatic arc of the outer Eastern Pontides is little deformed. The structure of the Mesozoic series is characterized by block-faulting and gentle seaward dips. To date no major folds or thrusts have been mapped. The steep-dipping faults defining complex horst and grabens are generally conjugate and follow NE and NW directions (Schultze-Westrum 1962; Zankl 1962; Kronberg 1970; Akin 1978). The Senonian volcanic arc of the Eastern Pontides was an extensional arc as shown by the submarine nature of the volcanism. The marginal basins behind extensional arcs usually develop by the splitting of the volcanic arc axis (e.g., Karig 1971). The distribution of the Kuroko type ore deposits in the Eastern Pontides indicates that during the Maastrichtian the volcanic arc axis was located close to the present day Black Sea coast (Fig. 17, Okay & Sahinttkk, 1998). Thus, the East Black Sea Basin probably started to open during the Maastrichtian by the splitting of this arc axis. The Gami~'hane-Munzur transect. The Gtimt~shane-Munzur transect is the best exposed and continuous transect across the izmir-Ankara-Erzincan suture. In addition, the preservation of Upper Cretaceous and Lower Tertiary sequences in the suture region allow a precise dating of the subduction and collision events in the |zmir-Ankara-Erzincan ocean (Fig. 18). A geological map and cross-section of a wide region from Gfimfishane to the Munzur mountains is given in Figs 19 and 20. This region will be described in three parts: (1) the Inner Eastern Pontides, which constitute the southern margin of the Sakarya Zone, (2) the Munzur sequence representing the northern margin of the Anatolide-Tauride Block and (3) the suture zone itself. (1) Inner Eastern Pontides south o f Gami~shane. The Inner Eastern Pontides south of Gtimtishane show the typical stratigraphic features of the Sakarya Zone including a complex pre-Jurassic basement, a major Liassic
TETHYAN SUTURES OF NORTHERN TURKEY
499
SAKARYA ZONE Taurides
Inner Pontides
Munzur Mountains
Maden S
Outer Pontides Alucra N Extension Continental collision
Mre-arc a g mevomtlon a arc t i ca.nd Initiation of subduction
Passive mar~in developrnent
Rifling
Fig. 18. Stratigraphic sections of the Sakarya Zone and the Anatolide-Tauride Block in the Eastern Pontides (modified from Okay & Sahinttirk 1998). Symbols as in Fig. 2.
transgression, and the ubiquitous Upper Jurassic-Lower Cretaceous carbonates overlain by accretionary prism obducted northward during Cenomanian-Turonian times. During the Senonian the Inner Eastern Pontides were in a fore-arc position and received volcano-clastic sediments. Structurally the Inner Eastern Pontides is a north-vergent fold and thrust belt formed during the Palaeocene continental collision (Okay et al. 1997). The pre-Jurassic basement of the Inner Eastern Pontides is exposed in a number of north-vergent thrust sheets. In contrast to the regions farther west, the pre-Jurassic basement is heterogeneous and consists of Hercynian and Cimmeride sequences. The Hercynian crystalline units comprise sillimanite-cordieritegarnet-bearing high-grade metamorphic rocks (Okay 1996) with mid-Carboniferous Sm-Nd
garnet ages (322-312 Ma, Topuz et al. 1997), as exposed in the Pulur Massif (Fig. 19) and granites with Early Carboniferous Rb/Sr isochron ages (360 Ma, Bergougnan 1987), which cover large areas south of Giimtishane (Fig. 19). Similar Hercynian metamorphic and granitic rocks have been described from the Lesser Caucasus (Adamia et al. 1983). An uppermost Carboniferous sequence of conglomerate, sandstone, neritic limestone and quartzites, which crop out east of Kelkit at the base of the thrust stack (Figs 19 and 20, Ketin 1951; Okay & Leven 1996), probably lies unconformably over the Hercynian crystalline basement. This Upper Carboniferous sequence of shallow marine to terrigenous origin, can be compared with the Rotliegende formation of Central Europe. The Cimmeride units, which are rare in the Eastern Pontides, include the m e t a b a s i t e - m a r b l e -
500
A.I. OKAY & O. TUYSUZ
Eocene and younger rocks
Ultramafic rocks
Basalt, radiolarian chert, serpentinite (accretionary complex) SAKARYA ZONE
Paleocene- L. Eocene sandstone, conglomerate
Senonian volcanic rocks
Parautocthonous Jurassic-Cretaceous sediments Allochtonous JurassicCretaceous sediments
Triassic('?) Metabasite phyl ite, marble
U. Carboniferous shale, sandstone, limestone
Carboniferous granite ANATOLIDE-TAURIDE BLOCK Munzur Unit /~
Triassic-Cretaceous carbonates
Keban Unit ~[-~l Permian-Cretaceous metasedimentary rocks
Carboniferous gneiss, migmatite, micaschist
Fig. 19. Geological map of the suture region between GtimiJ~hane and Munzur mountains (modified from C)zgi]l & Tur~ucu 1984 and Okay & Sahinttirk 1998).
phyllite sequence in the A~vanis Massif (Okay 1984a). The Mesozoic sedimentary sequence in the Inner Eastern Pontides starts, like elsewhere in the Sakarya Zone, with a major Liassic transgression possibly coming from the south (Fig. 18, Akin 1978). The Lower-Middle Jurassic throughout the Eastern Pontides is represented by a characteristic volcanosedimentary formation, up to 2000 m thick, of basaltic and andesitic lithic tufts and lavas, volcanogenic sandstone, shale, conglomerate. Within this Kelkit Formation there are also thin, discontinuous coal and ammonitico rosso horizons (Pelin
1977; Bergougnan 1987). Ammonites from the ammonitico rosso horizons (Bassoullet et al. 1975; Bergougnan 1987), palynology of the coal horizons (A~rah et al. 1965) and dinoflagellate and palynomorph assemblages (Robinson et al. 1995) indicate an age span for the Kelkit Formation from Early Pliensbachien up to at least the end of the Bathonian (Fig. 18). The Kelkit Formation shows rapid lateral facies and thickness changes and in general exhibits a deeper marine character towards the south. In the Gtimtishane region, the Upper Jurassic-Lower Cretaceous carbonates lie conformably over the Kelkit Formation. They occur
TETHYAN SUTURES OF NORTHERN TURKEY A
RELATIVE AUTOCHTHON
FIRST THRUST SHEET
501
SECOND TRUST SHEET
NW 2J
I
SE
Carboniferous sequence Jurassic1st. I
1
9
i
.;:-..'.:.
....
KelkitFin (Jurass]cj
"\
Pulurmetamorphicrocks ~~____.___._..~:~
. ,. "l,
~
:'.'.'.
~.. ~
., .:.~
o
THIRD THRUST SHEET
NW ~ ~
Neocomianpelagiclimestone. i o n a r y
t
,
t1-'
'
~
Eocene
A'
Paleoceneforeland basinsediments
co
~ i i i i i i -
2
I
FOURTH THRUST SHEET
i
I
~
Jurassic ~ U.Carboniferous
km
....................... *.
9
"~,
. . . . . . . .
Andesite,dacite
~
Basalt,radiolarite,limestone ~
Neriticlimestone L . C r e t a c e o u s ~ Pelagiclimestone Jurassic~ Sandstone
U. Carboniferous~
Sandstone,shale, Carboniferous[ ~ limestone
i
i i i i i
', ', I ', I "4.'
Peridotite_ _ I ~k
i i i i i i
ill
I',I',', I II I "
Tuff, shale,sandstone Gneiss,micaschist, anaphibolite
Fig. 20. Geological cross-sections of the Gtimiishane-Munzur area. For location of the sections see Fig. 19 (modified from Okay et al. 1997).
in all the thrust sheets as well as in the relative autochthon to the north. When the thrust sheets are palinspastically restored, the Upper Jurassic-Lower Cretaceous carbonate show a transition from c. 900 m thick neritic limestones and dolomites to the north to c. 1200 m thick pelagic radiolarian biomicrites and calciturbidites to the south (Fig. 18, Pelin 1977; Robinson et al. 1995; Okay et al. 1997; Okay & S,ahinttirk 1998). The carbonates range in age from Oxfordian to Barremian/Aptian and are bounded above by a major Cenomanian unconformity. A major episode of uplift and erosion occurred throughout the Eastern Pontides during the Cenomanian and Turonian. In many localities Campanian limestones lie unconformably over the Jurassic sediments (Pelin 1977; Bergougnan 1987; Robinson et al. 1995) and even on Carboniferous granites (Yllmaz 1972). This phase of uplift and erosion was caused by the northward obduction of ophiolitic accretionary complex over the south-facing Pontide continental margin, an episode observed both farther west and farther east in the Lesser Caucasus (Okay &
1-
Peridotite
,Sahinttirk 1998). The accretionary complex consists of tectonic slices of basalt, radiolarian chert, pelagic shale, Lower Cretaceous pelagic and neritic limestone, sandstone and serpentinite. It includes lithologies from the Tethyan oceanic crust as well as from the Pontide continental margin. The accretionary complex rests on the Lower Cretaceous pelagic carbonates and is unconformably overlain by Maastrichtian limestones (Ketin 1951; Fenerci 1994). During the Senonian the Inner Eastern Pontides were in a fore-arc position and a tuffaceous flyschoid sequence with pelagic limestone intercalations, 500-900 m thick, were deposited. Major crustal shortening in the Inner Eastern Pontides occurred during the Palaeocene-Early Eocene and the continental margin was telescoped into a series of stacked north-vergent thrust slices. A major foreland basin, filled with c. 1200 m thick conglomerates and turbidites of Late Palaeocene-Early Eocene age formed in front of these thrust sheets (Figs 19 and 20, Norman 1976; Bergougnan 1976; Elmas 1996; Okay et al. 1997). In the outer Eastern Pontides
502
A.I. OKAY & O. TOYSUZ
this period was marked by a phase of folding, uplift and erosion, so that Lutetian sedimentary rocks lie with an angular unconformity on folded Jurassic and Cretaceous sediments as well as on the pre-Liassic basement (Nebert 1961; Pelin 1977; Okay 1984a). (2) Munzur mountains: the other side o f the ocean. To the south of the Eastern Pontides the Mesozoic lithologies of the Anatolide-Tauride Block are best exposed in the Munzur Mountains southwest of Erzincan, where late Triassic-Cenomanian is represented by platform carbonates, 1200 m thick, which are overlain by Turonian-upper Campanian pelagic micrites (Fig. 18, Ozgtil & Tursucu 1984). In the early Maastrichtian an accretionary complex with large slices of peridotite were emplaced southward over the carbonates, and the carbonate platform was internally sliced into southvergent thrust sheets. The Mesozoic Munzur limestones were thrust southward over a similar platform carbonate sequence of Permian to Senonian age (Keban Unit), which, however, shows low-grade regional metamorphism (Fig. 19). Following this major compressive event of Early Maastrichtian age, Upper Maastrichtian-Palaeocene shallow marine limestones covered the thrust contacts as well as the accretionary complex over a 350 km long region between south of Sivas and Horasan (Yflmaz et al. 1988). The second period of carbonate deposition ended with uplift and erosion during the Late Palaeocene-Early Eocene, so that Lutetian sedimentary and volcanic rocks unconformably overlie all the older units. (3) Suture zone. South of the Eastern Pontides the suture is characterized by a several ten kilometres wide belt of accretionary complex and peridotite (Figs 17 and 19), and unlike the regions farther west, here it is not possible to draw a single tectonic line to represent the Tethyan suture. The reason is the mushroomtype structure of the accretionary complex and peridotite, which lie both on the Pontide and Anatolide-Tauride units. E v o l u t i o n o f the I z m i r - A n k a r a - E r z i n c a n ocean south o f the Eastern Pontides Significance o f the Liassic rifting. Sedimentological features of the Lower to Middle Jurassic Kelkit Formation led to its interpretation as a rift facies related to the opening of the izmir-Ankara-Erzincan ocean (G0rOr et al. 1983). These include the rapid lateral facies and thickness changes, the presence of basic volcanic and volcanoclastic rocks, and the deepening of the basin towards the south as inferred from the
sediments of the Kelkit Formation. However, palaeontological data from the accretionary complexes from farther west indicate that the [zmir-Ankara-Erzincan ocean was in existence at least since the Late Triassic. Furthermore,the pre-Jurassic stratigraphy of the Eastern Pontides is very different from that of the Taurides. In the Munzur and Keban units of the Taurides there is a well developed Triassic and Permian carbonate succession not seen anywhere in the Pontides (Fig. 18; Ozgtil & Tursucu 1984). During the Liassic there were also faunal differences between the two sides of the suture (Bassoullet et al. 1975; Enay 1976); Liassic ammonites from the Pontides are similar to those from the southern Laurasian margin, whereas Liassic fauna from the Munzur mountains of the Taurides resemble those from the southern margin of the Tethys. Thus, all the data indicate that the Anatolide-Tauride Block was separated from the Sakarya Zone since at least from Late Triassic onwards. The rifting inferred from the Kelkit Formation might be related to the splitting of a different continental fragment from the eastern part of the Sakarya Zone. One possibility is that the Sanandaj-Sirjan Zone in southwest Iran, which shows evidence for Palaeozoic and Triassic orogenic activity and has an uppermost Triassic to Middle Jurassic unconformable cover of thick coal-bearing clastic rocks (Shemshak Formation), similar to the Kelkit Formation of the Eastern Pontides (e.g., St0cklin 1968; S,eng6r 1990), could represent the continental fragment that rifted off from the Eastern Pontides during the Jurassic. During the Late Jurassic-Early Cretaceous a passive continental margin was established in the Inner Eastern Pontides. This is shown by the Upper Jurassic-Lower Cretaceous carbonates in the thrust sheets, which when palinspastically restored, indicate a carbonate platform in the north, which passed into a carbonate ramp representing the northern passive continental margin of the izmir-Ankara-Erzincan ocean (GOrtir 1988). Late Cretaceous obduction and Late Palaeocene collision. In the suture region south of Giamtishane there is well documented case of obduction of ophiolite and accretionary complex both northward during the Cenomanian-Turonian and southward during the early Maastrichtian. The northward obduction occurred just before the initiation of arc magmatism in the Outer Eastern Pontides. It is possibly related to the partial subduction of the Eastern Pontide passive continental margin into a south-dipping juvenile intra-oceanic subduction zone (Okay & Sahintiirk 1998). This was
TETHYAN SUTURES OF NORTHERN TURKEY 27'
9~
28'
*Ed,me
~.
29'
503 30
~ :5
%
B 1a c k
31'
Se a
41'
~...... . ............
9'/
:
~
' Istanbul ' ~ .~.~.
U , - - > ". . . .
Marmara
~
~'x, G G ? 5 ~ ~
[ - - 1 T e r t i a r y deposits ~ - ~ T e r t i a r y STRANDJA MASSIF ~]U. Cretaceous andesite ~
granitoid ~ - x ] U .
~
~
t
t
J
b
~.-"
,
J -- i
Cretaceous granitoid ~
Triassic sedimentary rock s
.~; ; ; ,...~ . ~
~
i
ISTANBUL ZONE ~ U. Cretaceous limestone, marl v- volcm~ic facies
Triassic_Jurassic metasediments ~
~-~-]Permian metagranites
S
~
~ - -
~-- i ~,o),nuK,.-7 t - i
I 7- I
B asalt,radiolarite,serpentinite Cret.-Palaeocene accretionary complex SAKARYA ZONE ~ Jurassic-Eocene sequence ~
~
Ordovician-Carboniferous sedimentary rocks
~
Metamorphic rocks and metaophiolite of unknown age ano ongln
Pre-Jurassic basement (mainly Triassic subductionaccretion units) 0 I
20 t
40 ~
60 I
80 kin I
Fig. 21. Geological map of the Marmara Sea region.
followed by the flip of the subduction zone from a south- to north-dipping subduction, which is required to explain the start of the subductionrelated volcanism in the Outer Eastern Pontides during the Turonian. The collision between the Eastern Pontide magmatic arc and the Anatolide-Tauride Block resulted in thrust imbrication of the active margin. The collision is well dated as Late Palaeocene from the age of the foreland basin sequence in front of the north-vergent thrust sheets in the Inner Eastern Pontides. This is corroborated by the end of the arc magmatism in the outer Eastern Pontides during the Late Paleocene (Elmas 1995; Okay & Sahinttirk 1998). Data from the Tauride margin, where the strongest deformation occurred during the Early Maastrichtian, are more ambiguous but does not exclude a Late Palaeocene collision.
The Intra-Pontide suture The Intra-Pontide suture forms the c. 400 km long boundary between the istanbul Zone and the Sakarya Zone. It also extends for approximately another 400 km farther west through the
Sea of Marmara defining the contact between the Rhodope-Strandja Massif and the Sakarya Zone, and bending south may join the izmir-Ankara-Erzincan suture in the central Aegean Sea (Fig. 1). The Intra-Pontide suture can, thus, be divided into two segments, the eastern segment between the istanbul and Sakarya zones and the western segment between R h o d o p e - S t r a n d j a Massif and the Sakarya Zone.
Intra-Pontide suture between the Istanbul and Sakarya zones The Intra-Pontide suture in this segment can be subdivided into a 400 km long east-west-trending collisional suture, and the two limiting, north-trending transform faults, which during the Cretaceous also formed a plate boundary (Figs 13 and 21). Although the suture constitutes a profound stratigraphic, metamorphic, magmatic and structural boundary, the lithological units and structures along the suture are poorly known. Along most of its length the suture is defined by the North Anatolian Fault, a major post-Miocene strike-slip fault with a cumulate
504
A.I. OKAY & O. TOYSLIZ
r-xJ
r-.-/ /--,-/
r-~
/
/'xj
'
rx../
1
,
r,~
,'-'x-/ r-x../
r".../ f'x..r..._/r~. ~ X X
x
--xJx x
X
X
X
X
X
X
X
~-'-zs
x
38x
x
k
X
x
x
%/
•
x
•
X
A-
X
•
X X /~
X
•
X
>32x
X
x
X
•
--
•
--
30
2
x
o Mudurnu .
.
.
.
....)J_
-0 [
2
4
6
I
I
I
8 km'-----
I
--
--
--
,
--
SAKARYAZONE Paleocene ~ Paleocene [~'.~-:~ U. Cret. ~_----~ L.Cret. ~_ ~ U.Jurassicl i--r ~ m
M. JurassicI x
m
I ] Lacustrinelimestone Sandstone,shale, marl ~ U . Shale,sandstone,tuff, marl Pelagic limestone F-~ Limestone,tuff
~--j
xj Basic tuft',basalt
Md~K~ )cene and younger ~
Metabasite,marble
Cretaceousturbidites ~ Basalt, radiolarite accretionarycomplex ~ Slightlycrystallized clastic rocks with Dewmianfossils
Gabbro,amphibolite Ultramaficrock Marble, phyllite
Fig. 22. Geological map of the region around the Intra-Pontide suture south of Almaclk mountains (modified from Abdiisselamo~lu 1959; Yflmaz et al. 1982).
dextral offset of 25-40 km ($eng6r 1979; Barka 1992). The North Anatolian Fault disguises the pre-Miocene relations between the ]stanbul and Sakarya zones. Furthermore, there are large metamorphic areas along the suture, whose provenance and age are unknown. A good example of the problems associated with the Intra-Pontide suture is provided by the geology of the Almaclk mountains east of Adapazarl studied by Abdtisselamo~lu (1959), GOztibol (1980), Yllmaz et al. (1982) and Greber (1996). A geological map based on these works is given in Fig. 22. In the Almaclk region the North Anatolian
Fault defines the suture and divides the region into two parts (Fig. 22). To the south of the North Anatolian Fault there is a well-developed Jurassic to Eocene sequence of the Sakarya Zone (Fig. 12, Saner 1980; G6ziibol 1980; Altlner et al. 1991), which indicates continuous marine deposition from Mid-Jurassic to Early Eocene. To the north the Intra-Pontide suture is defined by a few kilometre wide North Anatolian Fault zone, where elongate tectonic slivers of serpentinite, pre-Jurassic basement of the Sakarya Zone and terrigeneous Neogene deposits outcrop (Fig. 22). In the Almaclk mountains north of the North Anatolian Fault there is a thrust stack separated
TETHYAN SUTURES OF NORTHERN TURKEY by north-south-trending thrust traces (Fig. 22). Sanbudak et al. (1990) have explained the discordance between the general east-west trend of the regional structures and the north-south thrust traces by the Miocene flake rotation of the Almaclk thrust stack in the North Anatolian Fault Zone. At the base of the Almaclk thrust stack there is a metamorphic sequence of quartzite, phyllite, micaschist and marble, over 1000 m in thickness. The metasedimentary sequence is overthrust by a dismembered and metamorphosed ophiolite of peridotite, pyroxenite, gabbro and basalt (Fig. 22, Abdtisselamo~lu 1959; Ydmaz et al. 1982; GOziibol 1980). The common greenschist to amphibolite facies metamorphic grade shown by the dismembered ophiolite and metasedimentary sequence indicates that the ophiolite was tectonically emplaced over the sedimentary sequence prior to the regional metamorphism. A slightly metamorphic clastic-limestone sequence with scarce Devonian fossils (Abdtisselamo~lu 1959) lies with a problematic contact on the metamorphosed ophiolite (Fig. 2). Abdtisselamo~lu (1959) and GOztibol (1980) regard the metasedimentary sequence and the meta-ophiolite as of pre-Devonian age. On the other hand, Ydmaz et al. (1982) consider the meta-ophiolite of
ISTANBUL ZONE Ulus Basin Cretaceousturbidites Carboniferousand Olistostromes Triassicolistoliths
NNW A
lokmt
. . . . . ~ _ ~ ~ 9 ..,_,_,_ N
~
~
Cretaceous age and regard the contact with the overlying Devonian rocks as tectonic. The only consensus on the age of the metamorphic rocks is the poorly preserved unconformable cover of Maastrichtian to Palaeocene neritic limestones (Greber 1996). A similar problem exists in the Armutlu peninsula farther west along the suture, where ages ranging from Precambrian to Devonian have been suggested for a similar pre-Upper Cretaceous metamorphic sequence (Akartuna 1968; GOnctio~lu & Erendil 1990; Ydmaz et al. 1995). As there are no data on the depositional or isotopic ages of these metamorphic complexes, the tectonics and geological evolution of the Intra-Pontide suture remain highly uncertain. O n s h o r e c o n t i n u a t i o n s o f the West B l a c k Sea a n d West C r i m e a n faults
These fossil transform faults form the north-south-trending boundaries of the istanbul Zone with the Strandja Massif to the west and the Sakarya Zone to the east (Fig. 1). Although they do not represent collision-related sutures, nevertheless they are former plate boundaries and form distinct stratigraphic, metamorphic, magmatic and structural discontinuities.
SAKARYA ZONE Triassic-Liassicshale ESE andsiltstone WestCrimean Mid-Jurassic granite ~ c o n t . Qfaultl !
, " -_"~~<' 5~- - -jC . . - ~~ ; ~'.~ld~ <~ . / "" II',A-/ ~
505
t ~
----
7~
~ ~ ~ ~ ~t
/ - -
wsw
ESE A' ~/
U. Jurassiclimestone LowerCretaceous Pre-Jurassicmetabasite 1km ~ - - . ~ . . . . . . ~.~turbidite + ~ ~ Serpentinite 0 ~ -"--%<_ "_-~7-.. . . . . . . .
v
v
0
1
1
1
2 km
v
v
~
v I
Fig. 23. East-west geological cross-section in the Central Pontides showing the onshore continuation of the West Crimean Fault. The section is continuous from top to bottom. For location of the section see Fig. 13.
506
A.I. OKAY & O. TOYSOZ
Onshore continuation o f the West Black Sea Fault. The West Black Sea Fault forms the boundary between the unmetamorphosed Palaeozoic sedimentary rocks of the Istanbul Zone and the regionally metamorphosed Lower Mesozoic sediments of the Strandja Zone. West of the city of istanbul the Carboniferous ~reywacke, siltstone and shale sequence of the Istanbul Zone approach within 20 km of the Triassic metaquarzite, marble, phyllite sequence of the Strandja Zone (Fig. 21). The region between is occupied by undeformed Middle Eocene marl, limestone and shale, which places an upper age limit for the activity along the West Black Sea Fault. Onshore continuation of the West Crimean Fault. The existence of a north-south-trending sinistral fault in the Central Pontides, forming the continuation of the West Crimean fault, was postulated by Okay et al. (1994). Recent fieldwork in this area has shown the presence of such a fault in the Central Pontides (Tiiystiz in prep.). The |stanbul Zone within a few tens of kilometres west of the West Crimean Fault consists of A p t i a n - A l b i a n turbidites, 2000 m thick, deposited in the Ulus basin, which opened during the late Barremian on the Upper Jurassic limestones (Figs 13 and 23). The Jurassic limestones and conglomerates rest unconformably on Triassic and Carboniferous clastic rocks (Akyol et al. 1974). The deformation which affected the fill of the Ulus Basin increases toward the east. In the western and central parts of the basin deformation style is clearly compressional while it is transpressional in the east. Close to the shear zone along the West Crimean Fault the rocks belonging to the Ulus Basin are penetratively deformed and generally foliated (Fig. 23). In this area all micro and mesostructures within the Ulus turbidites clearly indicate a left-lateral transpressional deformation. The Sakarya Zone within a few tens of kilometres to the east of the West Crimean Fault consists mainly of Triassic to Early Liassic black shale and siltstone, comparable to the Tauridian Flysch in the Yayla Range in Crimea and to the upper parts of the Karakaya Complex ($engOr et al. 1980; Ttiystiz 1990). However, unlike the other regions of the Sakarya Zone, in the Central Pontides post-tectonic Mid-Jurassic (165 Ma) granitoids intrude these fine-grained clastics (Boztu~ et al. 1984; Ydmaz & Boztu~ 1986), and both are unconformably overlain by the Upper Jurassic limestones and A p t i a n Albian turbidites (Fig. 23, Ttiystiz 1993). The Triassic to Early Liassic sequence is penetratively deformed with the deformation intensity
increasing westward towards the West Crimean Fault. Thus, new field evidence support a transform-type suture between the Istanbul and Sakarya zones in the Central Pontides. Westward continuation o f the Intra-Pontide suture The Intra-Pontide suture west of the West Black Sea Fault is disguised under the Marmara Sea and comes again onshore in the region of SarkOy in Thrace (Fig. 21). The Thrace Basin, filled with up to 8 km thick siliciclastics of Mid-Eocene to Oligocene age (GOrtir & Okay 1996), hides much of the pre-Eocene geology. However, an accretionary complex of serpentinite, blueschist, radiolarian chert, Upper Cretaceous and Lower to Middle Palaeocene pelagic limestone and basalt outcrops in basement uplifts, and is also found as olistoliths in the Middle Eocene turbidites north of S,arkOy (Fig. 21, Senttirk & Okay 1984; GOrtir & Okay 1996). This region is taken as the suture trace of the Intra-Pontide Ocean (Seng6r & Yflmaz 1981). West of Sark6y the Intra-Pontide suture is covered again by the Neogene sediments and the waters of the Aegean (Fig. 21). Sequences equivalent to the Sakarya Zone, in terms of stratigraphy and tectonic development, are apparently not present in the mainland Greece. Therefore, the IntraPontide suture joins the |zmir-Ankara-Erzincan suture in the northern Aegean Sea (Fig. 1). Evolution o f the lntra-Pontide ocean The Intra-Pontide ocean formed an embayment of the izmir-Ankara-Erzincan ocean, as indicated by the suture triple junction in the northern Aegean Sea (Fig. 1, Okay et al. 1996). There are no data on the age of opening of this embayment. The ocean was in existence by the Late Cretaceous, as indicated by the ages of pelagic limestone blocks, and started to close through northward subduction during the Senonian, as attested by the Senonian arc magmatism in the lstanbul and Strandja zones. The northward subduction gave rise to the opening of the West Black Sea Basin as a back arc basin. The collision between |stanbul and Sakarya zones in the eastern part of the Intra-Pontide Ocean is constrained as Early Eocene based on the stratigraphy of the Sakarya Zone. In the Central Sakarya Basin north of G6yntik, 12 km south of the suture (Fig. 22), there is a continuous deep marine transition from Upper Maastrichtian marls and pelagic micrites to Lower to Middle Paleocene (Danian-Montian) turbidites of sandstone and mudstone with a
TETHYAN SUTURES OF NORTHERN TURKEY 26
\28
~. STRANDJA ~ .
30
BLAC K
32
..
34
~ < : ~ Zonguld~ [ ~
~-~ ~i" ~
~ A
"
Ank~ OLIDE-TAURIDE BLOCK ~
..~:
507 36
"'~-..=>> }':
SEA
~'8
Amasya~~,~
KIREHII~
~ ~'-----'---~ :" " ANATOLILDoEcTAURIDE
Fig. 24. Distribution of the Triassic subduction-accretion units in the northern Turkey. pelagic microfauna (Fig. 12, Meri~ & S,engtiler 1986). Turbidites show a regressive development and the region became a land area during the Early Eocene (Saner 1978). An Early Eocene collision is also in agreement with the stratigraphic data from the northern and central parts of the Istanbul Zone, where continuous marine carbonate deposition has been documented throughout Late Cretaceous and Palaeocene (Dizer & Meri~ 1983). Blocks of pelagic radiolarian biomicrite of Early to Mid-Paleocene age, associated with serpentinite blocks north of Sark6y (Okay & Tansel 1994), suggests that continental collision in the western segment of the Intra-Pontide suture occurred later. G6rtir & Okay (1996) suggested that the continental collision in this segment occurred during the Oligocene. According to their model, the Thrace Basin developed as a fore-arc basin above the northward subducting Intra-Pontide ocean, which was locked in the east by the Early Eocene collision and could not drift northward. The tectonic setting must have been comparable to that of the present Hellenic subduction zone and the overlying Cretan Basin.
The question of the Palaeo-Tethyan suture Palaeogeographic maps show the existence of a Tethyan ocean as a westward narrowing oceanic embayment between Gondwana and Laurasia since at least the Carboniferous (e.g., Smith et al. 1981; Scotese & Golonka 1992). However, almost all the subduction-accretion complexes and ophiolite preserved in the western and central Mediterranean, bear evidence only of Jurassic and younger oceans. This dilemma, pointed out by Smith (1973), led to the subdivision of the Tethys into Palaeo- and NeoTethys. In several papers Seng6r (1979, 1984,
1987) argued that during the Triassic a continental fragment was rifted off from the northern margin of the Gondwana, and drifted north closing Palaeo-Tethys in its front and opening the Neo-Tethys at its back. This continental fragment, called the Cimmerian continent, finally collided with the Laurasian margin during the Triassic-Jurassic thereby ending the PalaeoTethyan evolution. According to this model a Palaeo-Tethyan suture of Triassic-Jurassic age should separate a Cimmerian continental fragment from the Laurasian margin. Triassic subduction-accretion complexes occur widely in the Sakarya Zone (Fig. 24), where they form a separate and easly distinguishable tectonostratigraphic unit from the Neo-Tethyan subduction-accretion complexes. In the last ten years extensive geological data have been collected on these Triassic subduction-accretion complexes in northern Turkey (Koqyigit 1987; Tt~ystiz 1990; Okay et al. 1991, 1996; Usta6mer & Robertson 1994; Genq & Yflmaz 1995; Pickett & Robertson 1996). Although the details are still obscure, these data suggest Triassic southward subduction of an oceanic lithosphere probably as old as Carboniferous and substantiate the Palaeo- and NeoTethyan subdivision. However, although there are scattered, small outcrops of pre-Jurassic granitic and metamorphic rocks of continental origin, it has not been possible to identify a coherent Cimmerian continent in the Sakarya Zone. Similarly a Karakaya suture, representing the closure of a Permo-Triassic marginal basin in the Sakarya Zone ($eng6r & Yllmaz 1981) could not be mapped in the field. There is no real distinction between the various units of the Karakaya Complex in northwestern Turkey and the pre-Jurassic orogenic sequences in the Central Pontides; both represent Triassic
508
A.I. OKAY & O. TOYSOZ
subduction-accretion complexes and occur very close to the I z m i r - A n k a r a - E r z i n c a n suture. Thus, a Palaeo-Tethyan suture as distinct from the 'Neo-Tethyan' | z m i r - A n k a r a - E r z i n c a n suture appears not to exist. This suggests that the latest Triassic orogeny recorded in the Sakarya Zone resulted largely from the obduction of an ensimatic subduction-accretion complex over the Laurasian margin. In such a case, the I z m i r - A n k a r a - E r z i n c a n suture represents a lithospheric plate margin for at least the Carboniferous-Palaeocene period.
Conclusions The 2000 km long Izmir-Ankara-Erzincan suture forms the main geological boundary between Laurasia and Gondwana in the Turkish transect for the period from at least Carboniferous to Palaeocene. However, the intervening Tethys ocean was not a single lithospheric plate for the whole duration of this period but consisted of at least two plates of Carboniferous-Late Triassic and TriassicPalaeocene ages respectively. The final continental collision, which led to the formation of the lzmir-Ankara-Erzincan suture was slightly diachronous and occurred in the earliest
Palaeocene to the west and in the Late Palaeocene to the east. This period corresponds to a major slow down of the motion of Africa relative to Europa in the vicinity of Turkey (Livermore & Smith 1984; Patriat et al. 1982), which might be related to this major continental collision. Throughout most of its length the izmirAnkara-Erzincan suture is represented by an easily recognizable and mappable fault or fault zone of thrust or strike-slip nature, which constitutes a profound stratigraphic, structural, magmatic and metamorphic divide. However, many of these upper crustal brittle structures are generally of Eocene and younger ages, and do not represent structures formed during the Paleocene collision. During the continental collision the upper plate, represented by the Sakarya Zone, was delaminated and the upper crustal part was thrust for at least 50 km, and probably for much larger distance, over the lower plate with the Cretaceous subduction-accretion complexes forming an intermediate composite thrust sheet (Fig. 25). Along most of its length the i z m i r - A n k a r a - E r z i n c a n suture marks the erosional limit of this major southward translation. This provides an explanation for the frequent presence of terrigenous to shallow marine Jurassic rocks of the Sakarya Zone within a few
Fig. 25. Block diagram showing the schematic geometry of the izmir-Ankara-Erzincan suture.
TETHYAN SUTURES OF NORTHERN TURKEY kilometres of the i z m i r - A n k a r a - E r z i n c a n suture (Fig. 25). The lower crust and lithospheric mantle of the Sakarya Zone must have been subducted along a line, which lies north of the i z m i r - A n k a r a - E r z i n c a n suture. The izmir-Ankara-Erzincan suture is not rectilinear but is strongly segmented (Fig.. 1). Some of these segments, like that between Izmir and Bahkesir, probably reflect an originally irregular plate boundary. Others, like the major southward concave loop in Central Anatolia is probably an effect of the extensive Late Tertiary cover. To the south of the i z m i r - A n k a r a - E r z i n c a n suture there are extensive Cretaceous subduction-accretion complexes, comprising mainly basalt, radiolarian chert, pelagic limestone, pelagic shale and serpentinite. Palaeontological data from the subduction-accretion complexes indicate that they have formed during the Late Cretaceous subduction of an oceanic lithosphere of Late Triassic to Cretaceous age. The presence of a major Late Cretaceous magmatic arc and fore-arc to the north of the I z m i r - A n k a r a Erzincan suture, as well as the south-vergent structures to the south of the suture indicate unambiguously north-directed subduction starting in the Senonian. In this context an enigmatic feature observed in the eastern part of the I z m i r - A n k a r a - E r z i n c a n suture is n o r t h w a r d thrusting of an accretionary complex during the Cenomanian-Turonian just before the onset of the northward subduction. The active margin to the north of the i z m i r - A n k a r a - E r z i n c a n suture shows much less lateral geological variation than the passive margin to the south, which in some segments was m e t a m o r p h o s e d during the pre-collisional ophiolite obduction. In fact, in the region south of the I z m i r - A n k a r a - E r z i n c a n suture obduction-related deformation is much stronger than collision-related deformation. S u b d u c t i o n - a c c r e t i o n complexes formed during the Triassic subduction of a Carboniferous to Triassic oceanic lithosphere occur immediately north of the Izmir-Ankara-Erzincan suture suggesting southward polarity of subduction for the Palaeo-Tethys. T h e y are distinguished from Cretaceous s u b d u c t i o n accretion complexes by the dominance of siliciclastic sedimentary rocks, and scarcity of radiolarian chert and pelagic limestone. The c. 800 km long Intra-Pontide suture formed as a result of closure of a major embayment of the Izmir-Ankara-Erzincan ocean. The final continental collision occurred during the Early Eocene to the east but may have been as young as Oligocene to the west. The
509
Intra-Pontide suture consists of an east-westtrending segment, used later in some parts by the North Anatolian Fault, and two north-southtrending limiting branches. Even on a very high structural level in the orogen, the faults, which form major geological discontinuities and thus represent sutures can be mapped and shown precisely on detailed geological maps. This is because the subduction-accretion complexes and ophiolite of the intervening oceanic lithosphere occur as flatlying allochthonous sheets, thrust over the passive continental margin rather than as steeply dipping wide belts (Fig. 25). Thus, suture zones made up of subduction-accretion complexes several tens of kilometres wide, as shown for example in Koqyigit (1991), are not warranted. A second general conclusion is the subdivision of the sutures into collisional and the much rarer transform-fault-type sutures. The latter are easy to distinguish from the collisional sutures, as they are not associated with ophiolite and s u b d u c t i o n - a c c r e t i o n complexes. Two examples for the transform-fault-type sutures are the offshore continuations of the West Black Sea and West Crimean faults. Supported by TUBITAK-Glotek and TOBITAK419/G grants. We thank Turkish Petroleum Company (TPAO) and Directorate of Mineral Research and Exploration (MTA) for supporting field work in Anatolia since 1970s, A.M.C. ,Seng/~rfor discussions on the tectonics of Turkey and surrounding regions, and A. Poisson and an anonymous reviewer for critical and helpful reviews.
References ABDOSSELAMOGLU, ,S. 1959. Geology of the Almactkdag~l and the region between Mudurnu and GOyni&. Istanbul Universitesi Fen Fakt~ltesi Monografileri, 14, [in Turkish]. ADAM1A, S., KEKELIA, M. & TSIMAKURIDZE,G. 1983. Pre-Variscan and Variscan granitoids of the Caucasus. IGCP Project No. 5, Newsletter, 5, 5-12. , LORDKIPANIDZE,M. B. & ZAKARIADZE, G. S. 1977. Evolution of an active continental margin as exemplified by the Alpine history of the Caucasus. Tectonophysics, 40, 183-199. AGRALI, B., AKYOL, E. & KONYALIY. 1965. Preuves palynologiques de l'existence du Dogger dans la r6gion de Bayburt. Bulletin of the Mineral Research and Exploration Institute of Turkey, 65, 45-57. AKARTUNA,M. 1968. Geology of the Armutlu Peninsula. istanbul Oniversitesi Fen Faktiltesi Monografileri, 20, [in Turkish] AKIMAN,O., ERLER,A., GONCOO~LU,M. C., GOLE~,N., GEVEN, A., TORELI, T. K. AND KADIO~LU,Y. K. 1993. Geochemical characteristics of granitoids
510
A.I. OKAY & O. TOYSOZ
along the western margin of the Central Anatolian Crystalline Complex and their tectonic implications. Geological Journal, 28, 371-382. AKIN, H. 1978. Geologie, Magmatismus und Lagerstaettenbildung im ostpontischen Gebirge-Ttirkei aus der Sicht der Plattentektonik. Geologisches Rundschau, 68, 253-283. AKINCI, O. T. 1984. The Eastern Pontide volcano-sedimentary belt and associated massive sulphide deposits. In: DIXON, J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the Eastern Mediterranean, Geological Society, London, Special Publications, 17, 415-428 AKYOL, Z., ARPAT, E., ERDOffJAN, B., GOffJER, E., ~ARO~LU, E, ~ENTORK,I., TOTUNCO,K. & UYSAL, $. 1974. Geological map and explanatory text of the region Cide-Kurucasile. Mineral Research and Exploration Institute, Turkey, Ankara [in Turkish]. AKYUREK,B., AKBA~;,B. & DA~ER, Z. 1988. ( a n k m El6 Quadrangle, 1.100 000 scale geological map and explanatory text. Maden Tetkik ve Arama Genel Miidtirl~i~ti, Ankara. AKVt)z, S. & OKAY, A. I. 1996. A section across a Tethyan suture in northwest Turkey. International Geological Reviews, 38, 405-418. ALI~AN,C. & DERMAN,A. S. 1995. The first palynological age, sedimentological and stratigraphic data for (~akraz Group (Triassic), Western Black Sea. In: ERLER, A., ERCAN,T., BINGOL, E. & OR(~EN,S. (eds) Geology of the Black Sea Region. Directorate of the Mineral Research and Exploration, Ankara, 93-98. AI;FINER, D., KOq'YI~IT,A., FARINACCI,A., NICOSIA,U. & CONTI, M. A. 1991. Jurassic, Lower Cretaceous stratigraphy and paleogeographic evolution of the southern part of north-western Anatolia. Geologica Romana, 28, 13-80. ASSERETO, R. 1972. Notes on the Anisian biostratigraphy of the Gebze area (Kocaeli Peninsula, Turkei). Zeitschrift ftir Deutschen Geologischen Geselleshaft, 123, 435-444. ATAMAN, G. 1972. Preliminary study on the radiometric age of the Cefahk da~l granite-granodiorite southeast of Ankara. Hacettepe Universitesi Fen ve Mtthendislik Bilimleri DergisL 2, 44-49 [in Turkish]. AYDIN,M., SAHINTORK,0., SERDAR,H.S., OZ(ELIK, Y., AKARSU, I., UNGOR, A., ~OKUGRA~, R. & KASAR, S. 1986. Geology of the region between Balhda~~angalda~ (Kastamonu). Ti;trkiyeJeoloji Kurumu Biilteni, 29, 1-16. AYD1N,Y. 1974. Etude pdtrographique et g~ochirnique de la partie centrale du Massif d'lstranca (Turquie). PhD Thesis, Universit6 de Nancy. BAILEY, E. B. & MCCALLIEN,W. J. 1953. Serpentinite lavas, the Ankara melange and the Anatolian thrust. Transactions of the Royal Society of Edinburgh, 62, 403-442. BARKA, A. A. 1992. The North Anatolian Fault. Anneles Tectonicae, 6, 164-195. BASSOULLET,J. P., BERGOUGNAN,H. & ENAY,R. 1975. R6partition des faunes et facihs liasiques dans l'Est de la Turquie, r6gion du Haut-Euphrate.
Comptes Rendus de l'Academie des Sciences, Paris, 280, 583-586. BATMAN, B. 1978. Geological evolution of the region north of Haymana and the investigation of the m61ange. Yerbilimleri, 4, 95-124 [in Turkish]. BERGOUGNAN, H. 1976. Structure de la Chaine pontique dans le Haut-Kelkit (Nord-Est de l'Anatolie). Bulletin de la Soci~t~ G~ologique de France, 13, 675-686. 1987. Etudes gOologiques dans l'est-Anatolien. PhD thesis, University Pierre et Marie Curie, Paris, France. BINGOL, E., AKYOREK, B. & KORKMAZER, B. 1975. Geology of the Biga peninsula and some characteristics of the Karakaya blocky series. Proceedings of the Congress of Earth Sciences for the 50th Anniversary of the Republic Turkey. Mineral Research and Exploration Institute, Ankara, 70-77 [in Turkish]. BOCCALETTI,M., GOCEV, P. 84 MANETI"I,P. 1974. Mesozoic isopic zones in the Black Sea region. Bolletino della Societ~ Italiana, 93, 547-565. BOZKURT, E., HOLDSWORTH, B. K. & KO~YI~IT, A. 1997. Implications of Jurassic chert identified in the Tokat Complex, northern Turkey. Geological Magazine, 134, 91-97. BOZTU~3,D., DEBON,E, LE FORT,E & Y1LMAZ,0. 1984. Geochemical characteristics of some plutons from the Kastamonu granitoid belt (Northern Anatolia, Turkey). Schweizer Mineralogisch Petrologisches Mitteileingen, 64, 389404. BRAGIN, N. Y. • TEKIN, U. K. 1996. Age of radiolarian chert blocks from the Senonian ophiolitic m61ange (Ankara, Turkey). The Island Arc, 5, 114-122. BURCHFIEL, B. C. 1980. Eastern European Alpine system and the Carpathian orocline as an example of collision tectonics. Tectonophysics, 63, 31-61. CHANNEL,J. E. T., TfJYSOZ, 0., BEKTA$, 0. & ,SENGOR, A. M. C. 1996. Jurassic-Cretaceous palaeomagnetism and paleogeography of the Pontides (Turkey). Tectonics, 15, 201-212. CHARLES, F. 1933. Contribution a l'6tude des terrains pal6ozoiques de l'Anatolie du Nord-Ouest (Asie Mineure). M4moires de la Soci(t~ G4ologique de Belgique 1933, 54-152. CHATALOV, G. A. 1988. Recent developments in the geology of the Strandzha Zone in Bulgaria. Bulletin of the Technical University of istanbul, 41, 433-465. COMPAGNONI,R. 1977. The sesia-Lanzo zone: high pressure-low temperature metamorphisn in the Austra-Alpine continental margin. Rendiconti di Societdt Italiana Mineralogia Petrologia, 33, 335-374. DEAN, W. Y., MARTIN, E, MONOD, 0., DEMIR, O., RICKARDS,A. B., BULTYNCK,P. ~: BOZDO(~AN,N. 1997. Lower Paleozoic stratigraphy, KaradereZirze area, central Pontides, northern Turkey. In: GONCOO~LU, M. C. & DERMAN,A. S. (eds) Early Paleozoic Evolution in NW Gondwana. Turkish Association of Petroleum Geologists Special Publications, 3, 32-38. DILL, N. 1976. Etude micropaleontologique du
TETHYAN SUTURES OF NORTHERN TURKEY Dinantian de G/Skg61 de Kokaksu (Turquie). Annales de la Socidtd Gdologique de Belgique, 98, 213-228. DIZER, A. &MERIr E. 1983. Late Cretaceous-Paleocene stratigraphy in northwest Anatolia Maden Tetkik ve Arama Enstitasii Dergisi, 95196,149-163 [in Turkish]. EG1N, D., HIRST, D. M. & PHILLIPS,R. 1979. The petrology and geochemistry of volcanic rocks from the northern Harsit river area, Pontid volcanic province, northeast Turkey. Journal of Volcanology and Geothermal Research, 6, 105-123. ELMAS, A. 1995. Geology of the Kop Da~l area (Bayburt, Erzurum): Evolution of a fore-arc basin. Tiirkiye Petrol Jeologlarz Derne~,i Biilteni, 6, 19-37 [in Turkish]. 1996. Geological evolution of northeastern Anatolia. International Geological Review, 38, 884--900. ENAY, R. 1976. Faunes anatoliennes (Ammonitina, Jurassique) et domaines biog6ographiques nord et sud t6thysiens. Bulletin de la Socidtd Gdologique de France, 18, 533-541. ERDOdAN, B. 1990. Stratigraphic features and tectonic evolution of the Izmir-Ankara Zone in the region between izmir and Seferihisar (in Turkish). Ti~rkiye Petrol Jeologlari Dernegi Biilteni, 2, 1-20. , ALTINER, D., GUNGOR, T. & SACIT, O. 1990. Stratigraphy of the Karaburun Peninsula. Bulletin of Mineral Research Exploration Turkey, 111, 1-24 [in Turkish]. ERLER, A. & GONC00~LU, M. C. 1996. Geologic and tectonic setting of Yozgat batholith, northern central Anatolian crystalline complex, Turkey. International Geological Review, 38, 714-726. EVANS,I., HALL, S. A., SARIBUDAK,M. A. & AYKOL,A. 1991. Preliminary palaeomagnetic results from Palaeozoic rocks of the istanbul-Zonguldak region, N.W. Turkey. Bulletin of the Technical University of lstanbul, 44, 165-190. FENERCI, M. 1994. Rudists from Maden (Bayburt) area (NE Turkey). Turkish Journal of Earth Sciences, 3, 1-12. FINETTI, I., BRICCHI, G., DEL BEN, A., PIPAN, M. & XUAN, Z. 1988. Geophysical study of the Black Sea. Bollettino di Geofisica Teorica ed Applicata, 30, 197-324. FLOYD, P. A. 1993. Geochemical discrimination and petrogenesis of alkalic basalt sequences in part of the Ankara melange, central Turkey. Journal of the Geolological Society, London, 150, 541-550. GAUTIER, Y. 1984. Ddformations et mdtamorphismes associds d la fermeture tdthysienne en Anatolie Centrale (Rdgion de Sivrihisar, Turquie). PhD thesis, Universit6 de Paris-Sud, Centre d'Orsay. GEDIK, 1. 1975. Die Conodonten der Trias auf der Kocaeli-Halbinsel (Tiirkei). Palaeontographica, 1511, 99-160. GENr ,S. C. & YILMAZ,Y. 1995. Evolution of the Triassic continental margin, northwest Anatolia. Tectonophysics, 243, 193-207. GOFFE, B., MICHARD,A., KIENAST,J. R. & LE MER, O. 1988. A case of obduction-related high-pressure, low-temperature metamorphism in upper crustal -
-
511
nappes, Arabian continental margin, Oman: P-T paths and kinematic interpretation. Tectonophysics, 151, 363-386. GONCOO~ LU, M. C. 1986. Geochronological data from the southern part (Ni~de area) of the Central Anatolian Massif. Bulletin of Mineral Resarch and Exploration, 105/106, 83-96. & ERENDIL, M. 1990. Pre-Late Cretaceous tectonic units of the Armutlu Peninsula. In: Proceedings of the 8th Petroleum Congress of Turkey, 161-168. GOR~)R, N. 1988. Timing of opening of the Black Sea basin. Tectonophysics, 147, 247-262. & OKAY,A. I. 1996. Fore-arc origin of the Thrace Basin, northwest Turkey. Geologisches Rundschau, 85, 662-668. --, MONOD, 0., OKAY, A. I., S,ENGOR, A. M. C., TOYSOZ, O., YIGITBA$, E., SAKIN(~,M. & AKKOK, R. 1997. Palaeogeographic and tectonic position of the Carboniferous rocks of the western Pontides (Turkey) in the frame of the Variscan belt. Bulletin de la Socidtd Gdologique de France, 168, 197-205. - - , OKTAY,E g., SEYMEN,I. & ,SENG()R,A. M. C. 1984. Paleotectonic evolution of the Tuzg61ti basin complex, Central Anatolia: sedimentary record of a Neo-Tethyan closure. In: DIXON, J. E. & ROBERTSON, A. H. E (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 455-466. - - , S,ENGOR,A. M. C., AKKt)K, R. & Y1LMAZ,Y. 1983. Sedimentological data regarding the opening of the northern branch of the Neo-Tethys in the Pontides. T~irkiye Jeoloji Kurumu Biilteni, 26, 11-20 [in Turkish]. GOZOBOL, A. M. 1980. Geological investigation of the Mudurnu-Dokurcun-Abant area (Bolu province) and the structural behaviour of the North Anatolian Transform Fault. istanbul (]niversitesi Fen Fakiiltesi Mecmuast, Series B, 45, 9-34. G~3LEr N. 1994. Rb-Sr isotope data from the A~aq6ren granitoid (east of Tuz G61ia): geochronological and genetical implications. Turkish Journal of Earth Sciences, 3, 39-43. GREBER, E. 1996. Stratigraphic evolution and tectonics in an area of high seismicity: Akyazl/Adapazan (Pontides, northwestern Turkey). Turkish Journal of Earth Sciences, 5, 63-80. GUTNIC, M., MONOD, O., POISSON,A. & DUMONT,J.-F. 1979. Gdologie des Taurides Occidentales (Turquie). Mdmoire de la SociOtd GOologique de France, 137. HARRIS,N. B.W., KELLEY,S. R &OKAY, A. I. 1994. Postcollision magmatism and tectonics in northwest Turkey. Contributions to Mineralogy and Petrology, 117, 241-252. HIRST, D. M. & EdIN, D. 1979. Localisation of massive, polymetallic sulphide ores in the northern Har~it river area, Pontid volcanic belt, northeast Turkey. Annales de la Socidtd Gdologique de Belgique, 102, 465-484. JACOBSHAGEN,V. (ed.) 1986. Geologie von Griechenland. Gebrtider Borntraeger, Berlin.
512 --&
A.I. OKAY & O. TOYSUZ
WALLBRECIqEa, E. 1984. Pre-Neogene nappe structure and metamorphism of the North Sporades and the southern Pelion peninsula. In: D~XON,J. E. & ROBE~TSON,A. H. E (eds) The Geological Evolution of the Eastern Mediterranean, Geological Society, London, Special Publications, 17, 591-602. KARIG, D. E. 1971. Origin and development of marginal basins in the western Pacific. Journal of Geophysical Research, 76, 2542-2561. KAYA, O. & MOSTLER, H. 1992. A Middle Triassic age for low-grade greenschist facies metamorphic sequence in Bergama (Izmir), western Turkey: the first paleontological age assignment and structural-stratigraphic implications. Newsletters on Stratigraphy, 26, 1-17. KEFtE',', I. E., KEELING,G. "& WAGNER, R. H. 1986. An outline stratigraphy and palaeobotanical records from the middle Carboniferous rocks of northwestern Turkey. Annales de la SociOtOGOologique du Nord, 55, 203-216. KETIN, I. 1951. Ober die Geologie der Gegend von Bayburt in Nordost-Anatolien. lstanbul Oniversitesi Fen Fakiiltesi Mecmuasz, Series B, 16, 113-127. KHAIN, V. E. 1975. Structure and main stages in the tectono-magmatic development of the Caucasus: An attempt at geodynamic interpretation. American Journal of Science, 275-A, 131-156. KNIPPER, A. L. 1980. The tectonic position of ophiolites of the Lesser Caucasus. In: PANAYIOTOU,A. (eds) Proceedings of the International Ophiolite Symposium Cyprus. Geological Survey Department, Nicosia, 372-376. KOCYI~IT, A. 1987. Tectono-stratigraphy of the Hasano~lan (Ankara) region: evolution of the Karakaya orogen. Yerbilimleri, 14, 269-293. 1991. An example of an accretionary forearc basin from northern central Anatolia and its implications for the history of subduction of NeoTethys in Turkey. Geological Society of America Bulletin, 103, 22-36. - - , OZKAN,S. & ROJAY,B. 1988. Examples from the fore-arc basin remnants at the active margin of northern Neo-Tethys: Development and emplacement ages of the Anatolian Nappe, Turkey. Middle East Technical University, Journal of Pure and Applied Science, Series A, 21, 183-210. KOZUR, H. & KAYA,O. 1994. First evidence of pelagic Late Permian conodonts from NW Turkey. Neues Jahrbuch fiir Geologie und Paliiontologie Monatshefte, 1994(6), 339-347. KRONBERG,R 1970. Photogeologische Daten zur Tektonik im ostpontischen Gebirge (NE-Ttirkei). Bulletin of the Mineral Research and Exploration Institute of Turkey, 74, 24-33. KULAKSIZ,S. 1981. Geology of the region of northwest of Sivrihisar. Yerbilimleri, 8:103-124 [in Turkish]. LANPHERE, M. A. 1981. K-Ar ages of metamorphic rocks at the base of the Semail ophiolite, Oman. Journal of Geophysical Research, 86, 2777-2782. LEVEN,E. J. & OKAY,A. I. 1996. Foraminifera from the exotic Permo-Carboniferous limestone blocks in
the Karakaya Complex, northwest Turkey. Rivista Italiana Paleontologia e Stratigrafia, 102, 139-174. LISENBEE, A. 1971. The Orhaneli ultramafic-gabbro thrust sheet and its surroundings. In: CAMPBELL, A. S. (ed.) Geology and History of Turkey. Petroleum Exploration Society of Libya, Tripoli, 349-360. LIVERMORE, R. A. & SMITH, A. G. 1984. Relative motions of Africa and Europa in the vicinity of Turkey. In: TEKELI,0. • GONCIJO~LU,M. C. (eds) Geology of the Taurus Belt. Mineral Research and Exploration Institute, Ankara, 1-10. MANETrI, P., PECCERILLO,A., POLl, G. & CORSINI, E 1983. Petrochemical constraints on the models of Cretaceous-Eocene tectonic evolution of the Eastern Pontic chain (Turkey). Cretaceous Research, 4, 159-172. MERIt, E. & SENG(3LER,I. 1986. New observations on the stratigraphy of Upper Cretaceous-Paleocene around GGyniik (Bolu, northwest Anatolia). Jeoloji Mgthendisligi, 29, 61-64 [in Turkish]. MICHARD,A., BOUDIER, E & GOFFI~,B. 1991. Obduction versus subduction and collision in the Oman case and other Tethyan settings. In: PETEP,S, T., NICOLAS, A. & COLEMAN, R. G. (eds) Ophiolite genesis and evolution of the oceanic lithosphere. Ministry of Petroleum and Minerals in Oman, Masqat, Kluwer Acid. Publ., Dordrecht, 447-467. - - , GOFFE, B., CHOPlN, C. & HENRY,C. 1996. Did the Western Alps develop through an Oman-type stage ? The geotectonic setting of high-pressure metamorphism in two contrasting Tethyan transects. Eclogae Geologicae Helvetiae, 89, 43-80. MITCHEL, A. H. G. & GARSON, M. S. 1976. Mineralization at plate boundaries. Mineral Science and Engineering, 8, 129-169. MONOD, O.,ANDR1EUX,J., GAUTIER,Y. & KIENAST,J. R. 1991. Pontides-Taurides relationships in the region of Eskisebir (NW Turkey). Bulletin of the Technical University ]stanbul, 44, 257-278. - - , OKAY,A. I., MALUSKI,H., MONI[~,P. & AKKOK,R. 1996. Schistes bleus du Trias supGrieur en Turquie du NW: Comment s'est fermGe la Paleo-TGthys?. In: 16e ROunion des Sciences de la Terre, 10-12 April 1996, OrlOans, France, Abstracts, 43. MOORE, W. J., MCKEE, E. H. & AKINCI,O. 1980. Chemistry and chronology of plutonic rocks in the Pontid Mountains, northern Turkey. European Copper Deposits, 1, 209-216. NE~Em; K. 1961. Der Geologische Bau der Einzugsgebiete Kelkit Gay und Kmhrmak (NE-Anatolien). Bulletin of the Mineral Research and Exploration Institute of Turkey, 57, 1-51. NORMAN, Y. 1976. Paleo-current directions in the Lower Tertiary basin south of Bayburt. Tiirkiye Jeoloji Kurumu Biilteni, 19, 23-30 [in Turkish]. --& RAD, M. R. 1971. Changes in heavy mineral compositions and textural parameters in the Harhor Formation in the region of ~ayraz (Haymana). Tiirkiye Jeoloji Kurumu Biilteni, 14, 205-225 [in Turkish]. OKAY,A. I. 1984a. The geology of the A~vanis metamorphic rocks and neighbouring formations.
TETHYAN SUTURES OF N O R T H E R N T U R K E Y
Bulletin of the Mineral Research and Exploration Institute of Turkey, 991100, 16-36. 1984b. Distribution and characteristicg of the northwest Turkish blueschists. In: DIXON, J. E. & ROBERTSON, A. H. E (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 455-466. 1989a. Tectonic units and sutures in the Pontides, northern Turkey. In: SENGOR,A. M. C. (ed.) Tectonic evolution of the Tethyan region. Kluwer Academic Publications, Dordrecht, 109-115. 1989b. Alpine-Himalayan blueschists. Annual Reviews of the Earth and Planetary Sciences, 17, 55-87. 1996. Granulite facies gneisses from the Pulur region, Eastern Pontides. Turkish Journal of Earth Sciences, 5, 55-61. & KELLEY, S. 1994. Jadeite and chloritoid schists from northwest Turkey: tectonic setting, petrology and geochronology. Journal of Metamorphic Geology, 12, 455-466. & LEVEN,E. J. 1996. Stratigraphy and paleontology of the Upper Paleozoic sequence in the Pulur (Bayburt) region, Eastern Pontides. Turkish Journal of Earth Sciences, 5, 145-155. & MONIE, P. 1997. Early Mesozoic subduction in the Eastern Mediterranean. Evidence from Triassic eclogite in northwest Turkey. Geology, 25, 595-598. & MOSTLER, H. 1994. Carboniferous and Permian radiolarite blocks from the Karakaya Complex in northwest Turkey. Turkish Journal of Earth Sciences, 3, 23-28. -& ~AHINTURK, 1~). 1998. Geology of the Eastern Pontides. In: ROBINSON, A. (ed.) Regional and Petroleum Geology of the Black Sea and Surrounding Regions, A A P G Memoirs 68, 291-311. --& SIYAKO,M. 1993. The new position of the izmir-Ankara Neo-Tethyan suture between Izmir and Bahkesir (in Turkish). In: TURGUT, S. (ed.) Tectonics and Hydrocarbon Potential of Anatolia and Surrounding Regions. Proceedings of the Ozan Sungurlu Symposium, Ankara, 333-355. & TANSEL,I. 1994. New data on the upper age of the Intra-Pontide ocean from north of S,arkGy (Thrace). Bulletin of Mineral Research and Exploration, 114, 23-26. , HARRIS,N. B. W. & KELLEY,S. P. 1998. Blueschist exhumation along a Tethyan suture in northwest Turkey. Tectonophysics, 285, 275-299. , ,SAHINTCrRK,O. & YAKAR, H. 1997. Stratigraphy and tectonics of the Pulur (Bayburt) region in the Eastern Pontides. Bulletin of the Mineral Research and Exploration Institute of Turkey, 119, 1-24. , SATIR, M., MALUSKI, H., SIYAKO,M., MONIE, P., METZGER, R. & AKV0Z, S. 1996. Paleo- and NeoTethyan events in northwest Turkey: Geological and geochronological constraints. In: YIN, A. & HARRISON,M. (eds) Tectonics of Asia. Cambridge University Press: 420--441. , ~ENGOR,A. M. C. & GOROR, N. 1994. Kinematic history of the opening of the Black Sea and its -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
513
effect on the surrounding regions. Geology, 22, 267-270. - - - , SIYAKO,M. & BORKAN, K. A. 1991. Geology and tectonic evolution of the Biga Peninsula. In: DEWEY, J. F. (ed.) Special Issue on Tectonics. Bulletin of the Technical University of istanbul, 44, 191-255. t3NEN, A. E & HALL, R. 1993. Ophiolites and related metamorphic rocks from the KtRahya region, north-west Turkey. Geological Journal, 28, 399-412. OZGI3L, N. 1984. Stratigraphy and tectonic evolution of t h e Central Taurides. In: TEKELI, O. & GONC~O~LU, M. C. (eds), Geology of the Taurus Belt. Mineral Research and Exploration Institute of Turkey, Ankara, 77-90. & TURSUCU,A. 1984. Stratigraphy of the Mesozoic carbonate sequence of the Munzur mountains (eastern Turkey). In: TEKELI, O. & GONCt?O~LU, M. C. (eds) Geology of the Taurus Belt. Maden Tetkik ve Arama Enstittisti, Ankara, 173-181. OZKO~AK, O. 1969. Etude gOologique du massif ultrabasique d'Orhaneli et de sa proche bordure (Bursa, Turquie). PhD Thesis, Universit6 de Paris. PATRIAT, P., SEGOUFIN, J., SCHLICH, R., GOSLIN, J., AUZENDE, J.-M., BEUZART, P., BONNIN, J. & OLIVET, J.-L. 1982. Les mouvements relatifs de l'Inde, de l'Afrique et de l'Eurasie. Bulletin de la Socidtd GOologique de France, 34, 363-373. PECCERILLO,A. & TAYLOR,S. R. 1975. Geochemistry of Upper Cretaceous volcanic rocks from the Pontide chain, northern Turkey. Bulletin Volcanologique, 39, 1-13. PELIN, S. 1977. Geological study of the area southeast of Alucra (Giresun) with special reference, to its petroleum potential. Karadeniz Teknik Universitesi Yaym No. 87, Trabzon [in Turkish]. PICKETT,E. A. & ROBERTSON,A. H. E 1996. Formation of the Late Paleozoic-Early Mesozoic Karakaya Complex and related ophiolites in NW Turkey by Paleotethyan subduction-accretion. Journal of the Geological Society, London, 153, 995-1009. POISSON, A. & SAH1NCI,A. 1988. La s6rie mdsozoique de Kemalpasa et le flysch pal6ocene d'Izmir au Nord-Ouest du Menderes (Anatolie occidentale, Turquie). Un jalon du microcontinent taurique. Comptes Rendus de l'Academie des Sciences, Paris, 307, 1075-1080. , GUEZOU, J. C., OZTORK,A., YNAN, S., TEMIZ, H., GORSOY, H., KAVAK,K. S. & OZDEN, S. 1996. Tectonic setting and evolution of the Sivas basin, central Anatolia, Turkey. International Geology Review, 38, 838-853. R~cou, L.-E. 1994. Tethys reconstructed: plates, continental fragments and their boundaries since 260 Ma from central America to south-eastern Asia. Geodinamica Acta, 7, 169-218. RING, U., RATSCHBACHER,L. & FmSCH,W. 1988. Playeboundary kinematics in the Alps: Motion in the Arosa suture zone. Geology, 16, 696--698. ROBINSON, A. G., BANKS, C. J., RUTHERFORD,M. M. & HIRST, J. P. R 1995. Stratigraphic and structural -
-
514
A.I. OKAY & O. TOYSI]Z
development of the Eastern Pontides, Turkey. Journal of the Geological Society, London, 152, 861-872. ROWLEY, D. B. 1996. Age of initiation of collision between India and Asia: A review of stratigraphic data. Earth and Planetary Science Letters, 145, 1-13. SAHINCI, A. 1976, La s6rie du Boztepe au nord de Manisa (Anatolie occidentale-Turquie). Norien sup6rieur n6ritique et S6nonien inf6rieur p61agique. Comptes Rendus de l'Academie des Sciences, Paris, 283, 1019-1020. SANER, S. 1978. Geology and the environments of deposition of Geyve-Osmaneli-GOlpazarl-Tarakh area. istanbul {dniversitesi Fen Fakiiltesi Mecmuast, Series B, 43, 63-91. 1980. Paleogeographic interpretation of the Jurassic and younger sediments of the MudurnuG6yntik Basin based on the depositional features of Jurassic and younger ages. Tiirkiye Jeoloji Kurumu Biilteni, 23, 39-52 [in Turkish]. SARIBUDAK, M., SANVER, M. & PONAT, E. 1989. Location of western Pontides, NW Turkey, during Triassic time: preliminary palaeomagnetic results. Geophysical Journal, 96, 43-50. --, SENGOR, A. M. C. & GORUR, N. 1990. Paleomagnetic evidence for substantial rotation of the Almaclk flake within the North Anatolian fault Zone, northwest Turkey. Geophysical Journal International, 102, 563-568. SCHMID, S. M., AEBLI, H. R., HELLER, E • ZINGG, A. 1989. The role of the Periadric Line in the tectonic evolution of the Alps. In: COWARD, M. R, DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics, Geological Society, London, Special Publications, 45, 153-172. SCHULTZE-WESTRUM, H. H. 1962. Das geologische Profil des Aksudere bei Giresun (NordostAnatolien). Abhandlungen von Bayerische Akademie der Wissenschaften, MathematischeNaturwissenschafiliche Klasse, 109, 23-58. SCOTESE, C. R. & GOLONKA, J. 1979. Paleogeographic Atlas. P A L E O M A P project. Department of Geology, University of Texas at Arlington. SENSOR, A. M. C. 1979. The North Anatolian transform fault, its age, offset and tectonic significance: Journal of the Geological Society, London, 136, 26%282. -1984. The Cimmeride Orogenic System and the Tectonics of Euroasia. Geological Society of America Special Papers, 195. 1987. Tectonics of the Tethysides: Orogenic collage development in a collisional setting. Annual Reviews of Earth and Planetary Science, 15, 213-244. 1990. A new model for the late Paleozoic-Mesozoic tectonic evolution of Iran and implications for Oman. In: ROBERTSON,A. H. E, SEARLE,M. R & RIES, A. C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 797-831. -& YILMAZ,Y. 1981. Tethyan evolution of Turkey, a plate tectonic approach. Tectonophysics, 75, 181-241. ,
-
-
-
-
--.,
ALTINER,D., CIN, A., USTAOMER,T. & HS~, K. J. 1988. Origin and assembly of the Tethyside orogenic collage at the expense of Gondwana Land. In: AUDLEY-CHARLES,M. G. &. HALLAM,A. (eds) Gondwana and Tethys, Geological Society, London, Special Publications, 37, 119-181. , YILMAZ, Y. & KETIN, I. 1980. Remnants of a pre-Late Jurassic ocean in northern Turkey, Fragments of Permo-Triassic Paleo-Tethys? Geological Society of America, Bulletin, 91, 59%609. --, -- & 1982. Remnants of a pre-Late Jurassic ocean in northern Turkey: fragments of Permo-Triassic Paleo-Tethys?: Reply. Geological Society of America Bulletin, 93, 932-936. SENTORK, K. & OKAY, A. I. 1984. Blueschists discovered east of Saros bay in Thrace. Bulletin of the Mineral Research and Exploration Institute of Turkey, 97/98, 152-155. SERVAIS,M. 1982. Collision et suture tdthysienne en Anatolie Centrale, dtude structurale et m~tamorphique (HP-BT) de la zone nord Kiitahya. PhD Thesis, Universit6 de Paris-Sud, Centre d'Orsay. SEYMEN, i. 1981. Stratigraphy and metamorphism of the Klr~ehir massif around Kaman (Klrsehir Turkey). Tiirkiye Jeoloji Kurumu Biilteni, 24, 101-108 [in Turkish]. 1983. Tectonic features of the Kaman Group in comparison with those of its neighbouring formations around Tamada~ (Kaman-Klrsehir). Tiirkiye Jeoloji Kurumu Biilteni, 26, 89-98 [in Turkish]. 1984. Geological evolution of the metamorphic rocks of the Klr~ehir Massif. In: Proceedings of the Ketin Symposium. Geological Society of Turkey, Ankara, 133-148 [in Turkish]. SHERLOCK, S., KELLEY, S., INGER, S., HARRIS, N. B. W., CARSWELL,A. & OKAY,A. I. 1997. "Excess" argon in high-pressure metamorphic terrains: two contrasting case studies from NW Turkey and the Dabie Shan of China. In: Abstracts of the Geochemistry Group Meeting, London. SMITH, A. G. 1973. The so-called Tethyan ophiolites. In: TARLING, D. H. & RUNCORN,S. K. (eds) Implications of Continental Drift to the Earth Sciences. Academic Press, Cambridge, 2, 977-986. ,HURLEY, A. M. & BmDEN, J. C. 1981. Phanerozoic paleocontinental worm maps. Cambridge University Press, Cambridge. SPRAY, J. G., BEBIEN, J., REX, D. C. & RODDICK, J. C. 1984. Age constraints on the igneous and metamorphic evolution of the Hellenic-Dinaric ophiolites. In: DIXON, J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 619-627. STAMPFLI, G. M. 1996. The Intra-Alpine terrain: A Paleotethyan remnant in the Alpine Variscides. Eclogae Geologicae Helvetiae, 89, 13-42. STOCKLIN, J. 1968. Structural history and tectonics of Iran - a review. A A P G Bulletin, 52, 1229-1258. TANER, M. E 1977. Etude gOologique etpOtrographique de la rdgion de Giineyce-lkizdere, situde au sud de -
-
-
-
TETHYAN SUTURES OF NORTHERN T U R K E Y
Rize (Pontides Orientales, Turquie). PhD Thesis, Universit6 de Genbve, Switzerland. --& ZANINETTI, L. 1978. Etude pal6ontologique darts le Cr6tace volcano-s6dimentaire de Gtineyee (Pontides orientales, Turquie). Rivista Italiana di Paleontologia e Stratigrafia, 84, 187-198. TANSEL, [. 1980. Biostratigraphical investigation of the Nalhhan region. Yerbilimleri, 5/6, 31-47 [in Turkish]. 1989. The Paleocene biostratigraphy and the Late Cretaceous-Paleocene boundary in the region of A~va (Istanbul). Tiirkiye Petrol Jeologlan Dernegi Balt., 1, 211-228 [in Turkish]. TEKELI, O. 1981. Subduction complex of pre-Jurassic age, northern Anatolia, Turkey. Geology, 9, 68-72. THEVENIAUT,H. 1993. Evolution de la TOthys occidentale et de la Pang~e au Trias. PhD Thesis, Universit6 de Paris VII. ToPuz, G., SATIR,M., ALTHERR, R. & SADIKLAR,M. B. 1997. Thermobarometrische und geochronologische Untersuchungen zur Metamorphosegeschichte Granat - und Cordieritftihrender Gneise des Pulur Massivs, NE Ttirkei. In: Deutsche Mineralogische Vereinigung, Abstracts, 45. TRI21MPY,R. 1980. Geology of Switzerland. Webf & Co. Publishers, Basel. TOYsl3z, O. 1990. Tectonic evolution of a part of the Tethyside orogenic collage: the Kargl Massif, northern Turkey. Tectonics, 9, 141-160. 1993. A geo-traverse from the Black Sea to the central Anatolia: Tectonic evolution of northern Neo-Tethys. Ti~rkiye Petrol Jeologlart Derne~i Bi;tlteni, 5,1-33 [in Turkish]. 1996. Geology of Amasya and the surroundings. In: Proceedings of the llth Petroleum Cong. Turkey, Ankara, 45-61 [in Turkish]. -& YIGITBA$, E. 1994. The Karakaya basin: a Palaeo-Tethyan marginal basin and its age of opening. Acta Geologica Hungarica, 37, 327-250. , DELLALO~LU,A. A. & TERZIOGLU, N. 1995. A magmatic belt within the Neo-Tethyan suture zone and its role in the tectonic evolution of northern Turkey. Tectonophysics, 243, 173-191. ]~INALAN, G., YOKSEL, g., TEKELI, T., GONENff, 0., SEYIRT, Z. & H0SEWN, S. 1976. The stratigraphy and paleogeographical evolution of the Upper Cretaceous-Lower Tertiary sediments in the -
-
-
-
-
-
515
Haymana-Polath region (SW of Ankara). Tiirkiye Jeoloji Kurumu Biilteni, 19, 159-176 [in Turkish]. USTAOMER, T. • ROBERTSON, A. H. F. 1994. Late Paleozoic marginal basin and subduction-accretion: the Paleotethyan Kfire Complex, Central Pontides, northern Turkey. Journal of the Geological Society, London, 151, 291-305. WILHAMS, H. & SMV'CH,R. 1973. Metamorphic aureoles beneath ophiolite suites and Alpine peridotites: tectonic implications with west Newfoundland examples. American Journal of Science, 273, 594-621. YALINIZ, M. K., FLOYD, P. A. & GONCIJO~LU, M. C. 1996. Supra-subduction zone ophiolites of Central Anatolia: geochemical evidence from the Sankaraman ophiolite, Aksaray, Turkey. Mineralogical Magazine, 60, 697-710. YILMAZ, A., TERLEMEZ, I. & UYSAL, ~. 1988. Some stratigraphic and tectonic characteristics of the area around Hmls (southeast of Erzurum). Bulletin of the Mineral Research and Exploration, 108, 1-22. YILMAZ, O. & BOZTUG, D. 1986. Kastamonu granitoid belt of northern Turkey: First arc plutonism product related to the subduction of the paleoTethys. Geology, 14, 179-183. YmMAZ, Y. 1972. Petrology and structure of the Giimiishane granite and the surrounding rocks, N.E. Anatolia, PhD Thesis, University College London, England. - - , GENt, S. C., YI~ITBA$, E., Bozcu, M. & YILMAZ, K. 1995. Geological evolution of the Late Mesozoic continental margin of nortwestern Anatolia. Tectonophysics, 243, 155-172. - - , GOz(3BOL,A. M. & TOYSOZ, O. 1982. Geology of an area in and around the Northern Anatolian transform Fault Zone between Bolu and Akyam. In: ISIKARA,A. M. & VOGEL, A. (eds) Multidisciplinary Approach to Earthquake Prediction. Friedr. Vieweg & Sohn, Braunschweig, 45-65. YURTTA$-OZDEMIR, l~l. 1973, Biostratigraphy and macrofauna of the Tepek6y Triassic of the Kocaeli Peninsula. Maden Tetkik ve Arama Dergisi, 77, 57-98 [in Turkish]. ZANKL, H. 1962. Magmatismus und Bauplan des Ostpontischen Gebirges im Querprofil des HarsitTales, NE Anatolien. Geologische Rundschau, 51, 218-239.
Petroleum systems of Alpine-Mediterranean foldbelts and basins P. A . Z I E G L E R 1 & F. R O U R E 2
1Geological-Palaeontological Institute, University o f Basel, Bernoullistrasse 32, 4056 Basel, Switzerland 2Institut Franfais du POtrole, 1-4 Avenue de Bois-PrOault, B P 311, 92506 RueilMalmaison, France Abstract" The Alpine-Mediterranean area hosts a number of large and smaller hydrocarbon provinces in platforms, rifts, foreland basins and the frontal parts of thrust belts. These hydrocarbon provinces are related to a wide range of source rocks occurring in Palaeozoic, Mesozoic and Cenozoic series. Source rocks are alternatively localized in prerift, syn-rift or passive-margin sequences, the age of which is highly variable. Moreover, syn-orogenic foreland basin sequences can contain important source-rocks and are frequently the locus of biogenic gas generation. Most Mesozoic passive margins of the Western Tethys were destroyed during the Alpine orogenic cycle and were overthrusted by or incorporated into orogenic wedges, the external parts of which host a number of highly diversified petroleum systems. Late Variscan and Alpine compressional intraplate deformation had severe repercussions on the hydrocarbon habitat of some of the Peri-Tethyan platforms. Late Cretaceous and Palaeogene intracratonic and back-arc rifts host some important petroleum provinces; the Euphrates, Abu Gharadig and Sirte basins rely for hydrocarbon charge on syn-rift source rocks whereas post-rift series contain the source rocks of the Black Sea back-arc basin. Oligocene and younger rift systems host the petroleum provinces of the Gulf of Suez, Pelagian Shelf, Valencia Trough and Rhine Graben; these rely variably on pre- and syn-rift source rocks. The Neogene Pannonian system of transtensional back-arc basins relies for hydrocarbon charge mainly on syn-rift source rocks whereas the Vienna Basin is exclusively charged by Mesozoic post-rift source rocks. The complex evolution of the Western Tethys belt, contrasted thermal regimes and successive episodes of sedimentary and tectonic burial account for the great diversity of its petroleum systems. Each hydrocarbon province is characterized by a very distinct scenario for the accumulation of source rocks, timing of their maturation, hydrocarbon expulsion and migration from effective kitchens to potential trapping domains, and the preservation of hydrocarbon accumulations. Future exploration in the external parts of the Alpine orogenic belts, their forelands and beneath the Neogene fill of the Pannonian Basin is likely to yield further oil and gas discoveries. In contrast to the Black Sea Basin, the hydrocarbon potential of the AlgeroProvenqal, Tyrrhenian and Aegean back-arc basins is questionable. The thick sedimentary sequences of the deep East Mediterranean Basin may contain effective source rocks.
Evolution of the A l p i n e - M e d i t e r r a n e a n domain was controlled by the Late Permian and Mesozoic break-up of Pangaea, resulting in the opening of oceanic basins forming the Western Tethys, and by the convergence and suturing of A f r i c a - A r a b i a with Eurasia during the Late Cretaceous and Cenozoic Alpine orogenic cycle. The A l p i n e - M e d i t e r r a n e a n area hosts a number of large and smaller hydrocarbon provinces in platforms, rifts, foreland basins and in the frontal parts of thrust belts. These are related to a wide range of petroleum systems, variably tied to source rocks occurring in Palaeozoic, Mesozoic and Cenozoic series (Fig. 1). The concept of 'petroleum systems', as discussed in M a g o o n & D o w (1995), is very complex and covers a wide range of geological features and processes, including the age and
distribution of source-rocks and reservoirs, as well as the timing of hydrocarbon migration and trapping. However, for the purpose of the discussion presented in this paper, we feel that the n a m e and/or age of an effective source-rock that has p r o v i d e d the charge for h y d r o c a r b o n accumulations is sufficient to characterize a petroleum system, even if these accumulations are contained in reservoirs of different age. In this context, it should be kept in mind, that a given petroleum province may be related to a single or to multiple petroleum systems, each characterized by a specific, effective source rock. Effective source rocks are widely distributed within the Alpine orogen, in associated foreland and back-arc basins and in peri-Alpine platforms and rifts. Source rocks occur in pre-rift, syn-rift and passive margin sequences, the age of
ZIELGER,P. A. & ROURE,E 1999. Petroleum systems of Alpine-Mediterranean foldbelts and basins. In: DURAND,B., JOLIVET,L., HORVNTH,E & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 517-540.
518
P.A. ZIEGLER & F. ROURE
Fig. 1. Tectonic framework of the Alpine-Mediterranean area showing major hydrocarbon reserves per country or basin. Blue circles: main hydrocarbon provinces; arrows pointing to blue circles originate in hydrocarbon provinces that contribute to national reserves. Figures inside blue circles: ultimate recoverable oil and gas-condensate reserves in established accumulations in 109 bbls; red figures next to blue circles: ultimate recoverable gas reserves in established accumulations in TCF.
which is highly variable. Moreover, syn-orogenic foreland basin sequences can contain important source rocks and are frequently the locus of biogenic gas generation. P e t r o l e u m systems related to Palaeozoic source rocks occur in the foreland of the Variscan orogen and also in Permo-Carboniferous late and post-orogenic basins, either located within the Variscan orogen or in its foreland. Petroleum systems related to Mesozoic and Cenozoic source rocks are variably tied to rifts and post-rift basins that developed during the opening of the Western Tethys, as well as to flexural basins that evolved during the Alpine orogenic cycle. Late Cretaceous and Cenozoic rifts in the Alpine foreland, as well as intra-Alpine back-arc extensional basins, host i m p o r t a n t hydrocarbon provinces. This paper reviews the distribution of source rocks in the Alpine-Mediterranean realm and relates the development of associated petroleum systems to the evolution of the Western Tethys belt. For supporting palaeo-reconstructions of
the Western Tethys belt, the reader is referred to Ziegler (1988, 1989, 1990), Dercourt et al. (1993), Nairn et al. (1996) and Yilmaz et al. (1996).
Palaeozoic series and related hydrocarbon provinces Palaeozoic hydrocarbon provinces are delimited by the Palaeozoic Caledonian and VariscanScythian orogens and by major arches which developed in their forelands in conjunction with intraplate compressional and extensional deformations (Fig. 2). In northern Africa and on the stable Arabian platform, p e t r o l e u m systems are related to Ordovician, Silurian and D e v o n i a n source rocks. These were deposited on a vast shelf occupying the n o r t h e r n margin of G o n d w a n a (Klemme & Ulmishek 1991). Particularly the western parts of this shelf were disrupted during the Late Carboniferous-Early Permian terminal
MEDITERRANEAN PETROLEUM SYSTEMS
519
Fig. 2. Distribution of Palaeozoic source rocks and related hydrocarbon provinces.
phases of the Variscan-Appalachian orogeny by intraplate compressional deformation. A second phase of intraplate compression occurred during the Alpine orogeny, resulting in further disruption of the sedimentary cover of the Sahara platform (Eocene inversion of S a o u r a - O u g a r t a arch, Ziegler 1989; Ziegler et al. 1995). The distribution of hydrocarbon provinces, such as those of the Essaouira, Ghadames, Illizi and Murzuk basins, is controlled by the degree of preservation of Palaeozoic strata beneath the regional basal Mesozoic unconformity, and by the thickness of Mesozoic series. (Fig. 3, Clifford 1986; Jabour & Nakayama 1988; Husseini 1991; Sahagian 1993; van de Weerd & Ware 1994; O N A R E P 1995; Macgregor 1996; Makhous et al. 1997). Some deep pools in the westernmost parts of the Western Desert (Egypt) may be related to Devonian source rocks (Macgregor 1996). Whereas the Mesozoic cover of the northern parts of the Arabian Platform is generally too thin for Palaeozoic source rocks to attain maturity (Best et al. 1993), tectonic burial of Silurian source rocks accounts for the South Turkish thrust belt oil province (Gilmour & Mfikel 1996). Main Palaeozoic production on the Arabian platform comes from its eastern
Fig. 3. Subcrop map of Palaeozoic series beneath base-Mesozoic unconformity in Northern Africa (after SONATRACH 1995 and other sources). parts where a thick Mesozoic and Cenozoic cover provides for source-rock maturation (McGillivary & Husseini 1992; Mahmoud et al. 1992). In Europe, Cambro-Silurian series, located in the foreland of the Polish Caledonides, are productive in the Baltic oil province (Ulmishek 1990; Brangulis et al. 1993). In comparison, Palaeozoic series occurring along the southwestern margin of the East European craton in the foreland of t h e Carpathians, in the northern
520
P.A. ZIEGLER & F. ROURE
Dobrogea and on the Moesian Platform have a minor hydrocarbon potential (Popescu 1995; Pene 1996; Robinson et al. 1996; Moroz et al. 1997). The petroleum system of the prolific Pripyat-Dniepr-Donets Basin of Belorus and the Ukraine is controlled by source rocks contained in Devonian syn-rift and Carboniferous post-rift series (Ulmishek et al. 1994; Kabyshev et al. 1998). The very prolific Permian gas province of Northwestern and Central Europe is related to Westphalian coal measures, deposited in the foreland basin of the Variscan orogen, and to reservoirs provided by the Permian Rotliegend sands and Zechstein carbonate/ evaporite series that accumulated in a thermal sag basin post-dating the Stephanian-Autunian phase of wrench faulting and magmatism (Leeder & Hardman 1990; Ziegler 1990). Within the Variscan orogen, widespread wrench- and rift-induced Permo-Carboniferous successor basins contain coal measures and lacustrine kerogenous shales that provide potential source rocks in the European Alpine foreland (Kettel 1989). These source-rocks, depending on their burial beneath Tethyan passive margin sequences, or beneath synorogenic flexural sequences and the Alpine allochthon, have locally preserved part of their petroleum potential and thus can contribute to effective petroleum systems, as for instance in the Jura Mountains (Mascle et al. 1994; Philippe 1994), in the German Molasse Basin (Kettel & Herzog 1989) and in the subthrust play of the Polish Carpathians (Bessereau et al. 1996). Small oil and gas discoveries along the northern margin of the Pyrenean-Languedoc fold belt are related to coals and lacustrine shales contained in Permo-Carboniferous transtensional basins (Soler et al. 1981; Philippe 1994; Mascle et al. 1996; Roure & Colletta 1996; Vially & Tr6moli6res 1996). It is uncertain whether and to what extent Palaeozoic source rocks occur in the Mediterranean off-shore of Tunisia, Libya and Egypt (Fig. 2).
Mesozoic syn- and post-rift series and related hydrocarbon provinces Within the Western Tethys belt, rifting activity commenced during the Permian and accelerated during the Triassic. Step-wise opening of the different oceanic basins of the Western Tethys resulted in the development of passive margins during Mid-Triassic to Mid-Jurassic and Early Cretaceous times (Dercourt et al. 1993; O N A R E P 1995; Fourcade et al. 1996; Marcoux
& Baud 1996; Ricou 1996; Stampfli 1996; Yilmaz et al. 1996).
First marine connections between the Western Tethys and the Arctic Ocean via t h e Norwegian-Greenland Sea rift were established at the beginning of the Early Jurassic. These were interrupted during the Bajocian-Bathonian, but reopened on a permanent basis during the Callovian (Ziegler 1988). Moreover, as a result of the gradual break-up of Pangaea, a latitudinal current regime developed between the Western Tethys and the Pacific via the Central Atlantic and Gulf of Mexico, either as early as the Pliensbachian or as late as the Callovian (Baudin 1995). During much of Permian to Early Cretaceous times the Western Tethys belt was located in the northern trade-wind belt, between the Equator and 30~ In combination with cyclically rising sea levels, this permitted the establishment of extensive carbonate platforms in the Western Tethys belt (Philip et al. 1996). Syn-sedimentary extensional tectonics accounted for rapid lateral facies and thickness changes. Partly reefal and dolomitic shallow-water carbonates were deposited on rift flanks, the crests of tilted blocks and on little extended platforms, whereas shaly or cherty deeper water carbonates and shales accumulated in rapidly subsiding grabens and in sediment starved lagoons, partly under anoxic conditions. Under such a setting, stable water stratification in tectonically controlled submarine depressions, as well as on wide shelves during major transgressions, allowed for the repeated development of anoxic bottom water conditions and the deposition of kerogenous source rocks. Similarly, wind-driven upwelling conditions provided for the development of anoxic conditions, whereas strong bottom currents caused oxygenation of bottom waters. On the other hand, climatic conditions controlled the deposition of continental and paralic coal measures, providing essentially humic source rocks (Baudin & Herbin 1996). Triassic and Early Jurassic source rocks
During the Triassic and Early Jurassic, the Tethys rift system propagated westward and interfered in the North Atlantic domain with the southward propagating Arctic-North Atlantic rift system. During the Triassic, the Western Tethys had the configuration of a large embayment. By Early Jurassic times, a relatively narrow sea arm extended from the Mediterranean domain through the Central Atlantic into the Gulf of Mexico and communications were established
MEDITERRANEAN PETROLEUM SYSTEMS with the Arctic seas via the Arctic-North Atlantic rift systems (Ziegler 1988; Marcoux & Baud 1996; Fourcade et al. 1996). During Triassic and Early Jurassic times, the apparent absence of major intrabasinal currents favoured the development of stable anoxic bottom water conditions in rift-induced depressions of the Western Tethys belt, particularly during the Triassic. This is reflected by the deposition of important marine source-rocks on the carbonate dominated Apulian platform. Middle and Late Triassic, and in part also Early Jurassic source-rocks account for the petroleum systems of the Western Po Plain, the Southern Alps, the Adriatic, Sicily and the Albanides (Figs 4 and 5; Pieri & Mattavelli 1986; Brosse et al. 1990; Mattavelli & Novelli 1990; Casero et al. 1991; Mattavelli et al. 1993; Zappaterra 1994; Karakitsios 1995; Anelli et al. 1996; Frasheri et al. 1996; Krois 1998). Triassic source rocks occur also in the Austro-Alpine nappes of the Eastern Alps. Triassic and Early Jurassic effective source rocks have been identified in the substratum of the Neogene Pannonian Basin (K6kai & Pog~cstis 1991; Clayton & Koncz 1994; Mattick et al. 1996) and have been reported from the Balkanides (Georgiev & Atanasov 1993). An effective source rock of possibly Early Jurassic age occurs in Georgia (Robinson et al. 1996). Potential Triassic source rocks occur in the rifted Palmyra Trough (May 1991; Beydoun & Habib 1995.). In the northern Sinai, the Early Jurassic syn-rift, paralic to shallow-marine Shusha Formation contains potential source rocks (Mustafa & Khalil 1990; Abdel Aal & Lelek 1994). During the Early Jurassic, extensive marine source rocks were deposited on the tectonically unstable, clastic dominated shelves of Northwestern Europe. Main source-rock developments occur in areas where the Arctic and Tethys waters interfingered, providing for repeated stagnant bottom water conditions, particularly in the more rapidly subsiding parts of basins that were characterized by slow sedimentation rates (Farrimond et al. 1989; Baudin & Herbin 1996). Early Jurassic source rocks account for the oil provinces of the Paris and Channel basins, on- and off-shore Netherlands, North Germany and the Upper Rhine Graben (Ziegler 1990; Richard 1994; Baudin 1995; Bessereau et al. 1995). Early Jurassic source rocks contribute to the Aquitaine Basin petroleum province (Bourrouilh et al. 1995; Le Vot et al. 1996). They occur also in the rifted basin of southeastern France (Roure & Colletta 1996) and in the Lusitania Basin of Portugal where they are only marginally mature for oil generation. The rifted Tarfaya and Prerif basins of
521
Morocco host Early Jurassic petroleum systems (ONAREP 1995; Zizi 1996). Mid-Jurassic
source rocks
During the Mid-Jurassic, the Tethys-Pacific connection was well established and widened progressively in response to sea-floor spreading in the Central Atlantic (Fourcade et al. 1996). At the same time marine communications with the Arctic seas became constricted and even interrupted during the Bajocian and Bathonian (Ziegler 1988). Within the Western Tethys domain, rifting activity continued and culminated in the development of a transtensional plate boundary between Africa and Europe. Establishment of major intrabasinal current regimes within the gradually opening Tethys-Atlantic-Pacific seaway presumably inhibited the development of anoxic bottom conditions. This is reflected by the fact that sedimentary sequences deposited during the Mid-Jurassic rifting phases, preceding and accompanying the opening of the A l b o r a n Piemont-South Penninic Ocean, have a much lower source-rock potential than the Triassic and Early Jurassic series. Therefore, they play a smaller role in the hydrocarbon habitat of the Alpine-Mediterranean belt (Fig. 6). Mid-Jurassic effective source rocks occur along the margins of the Moesian Platform and the southeastern margin of the East European Platform (Georgiev & Atanasov 1993; Popescu 1995), as well as in the Western Desert of Egypt (Callovian-Oxfordian Khatatba Shale; Clifford 1986; May 1991). Whether the latter extends northeastwards into the Levante domain is uncertain. Possible Mid-Jurassic source rocks may also occur on the Pelagian Shelf (Macgregor 1996). On the carbonate dominated shelf of the Arabian Platform, the marine BajocianCallovian Dhruma and Sargelu shales represent important source rocks in northern Iraq and northern Saudi Arabia (Ayres et al. 1982). In contrast, no major Mid- Jurassic source-rock developments are evident on the shelves of Western Europe. However, in the Central and Northern North Sea, coals associated with the deltaic Rannoc/Etive and Brent Group have contributed to this prolific hydrocarbon province (Espitali6 et al. 1991; Cornford 1998).
Late Jurassic-Early
Cretaceous
source
rocks
Late Jurassic and Early Cretaceous progressive opening of the Central Atlantic was accompanied
522
P.A. Z I E G L E R & F. R O U R E
Fig. 4. Distribution of Triassic source rocks and related hydrocarbon provinces.
Fig. 5. Distribution of Lower Jurassic source rocks and related hydrocarbon provinces.
MEDITERRANEAN PETROLEUM SYSTEMS by northward propagation of the Atlantic rift system, Mid-Cretaceous isolation of the Iberian microcontinent and a sinistral translation of Laurasia relative to Africa-Arabia. This was paralleled by rifting activity in the North Penninic and East Mediterranean domains, and gradual closure of the Pelagonia and Vardar Ocean (Ziegler 1988; Ricou 1996). Rifting in the Black Sea back-arc basin apparently commenced during the Mid-Cretaceous (Robinson et al. 1995, 1996). Basins associated with the Arctic-North Atlantic rift systems contain major Late Jurassic source-rock developments. These are probably related to a generous nutrient supply by surface run-off, wind-driven upwelling systems and stagnant bottom water conditions in tectonically silled, rapidly subsiding and sediment starved depressions. Late Jurassic source rocks are the main contributors to the prolific Central and Northern North Sea, Mid-Norway, West Shetland and Jeanne d'Arc hydrocarbon provinces (Pegrum & Spencer 1990; Ziegler 1990; Espitali6 etal. 1991; Cornford 1998). In these basins, Early Cretaceous series have only locally a sourcerock potential, as, e.g., the Berriasian lacustrine shales of the Lower Saxony Basin, and the more regionally distributed transgressive marine Aptian shales (Ziegler 1990). On the other hand, the giant hydrocarbon province of the Middle East relies to a large extent on source rocks that were deposited in sediment-starved depressions on the vast carbonate/evaporite-dominated Arabian shelf during the late Callovian-early Kimmeridgian (Hanifa/Diyab/Tuwaiq/Gotnia Formation), the Berriasian-Valanginian (Sulaiy/lower Thamama Formation) and the Aptian (Shuaiba Formation) (Murris 1980; Stoneley 1990; Beydoun et al. 1992; Baudin 1995; Sadooni 1997). In the Western Tethys belt, Late Jurassic-Early Cretaceous series contain relatively few source rocks, presumably due to limited nutrient supply and current-induced good oxygenation of bottom waters (Moore et al. 1995), as suggested by the prevalence of 'Rosso Ammonitico' and red cherts in basinal areas and only local occurrences of organic shales associated with radiolarites (Baudin & Herbin 1996). However, there are a few exceptions. These are related to tectonically active rift basins and also to basins undergoing first phases of compressional deformation, providing a new mechanism for the development of tectonically silled submarine depressions. In comparison with the major hydrocarbon provinces of the Peri-Tethys shelves, the Late Jurassic-Early Cretaceous hydrocarbon
523
provinces of the Alpine-Mediterranean domain have a small, albeit still significant potential (Fig. 7). Amongst these, the Aquitaine Basin petroleum province is the most important one, it relies mainly on the Late Jurassic and Early Cretaceous source rocks that were deposited during the rifting stage preceding the opening of the oceanic Bay of Biscay Basin (Bourrouilh et al. 1995; Le Vot et al. 1996). In the Mediterranean off-shore of Spain, the Late Jurassic Salzedella Shale contributed locally to the petroleum province of the Valencia Trough. The Moroccan Essaouira Basin hosts an effective Oxfordian petroleum system (Broughton & Trepani6r 1993; O N A R E P 1995). Effective Late Jurassic source rocks occur also in the Vienna Basin (Ladwein 1988) and the Czech Carpathian subthrust (Ciprys et al. 1995; Brzobohaty et al. 1996; Picha 1996); the respective deep-water shales, that are offset by a Late Jurassic reef fringed carbonate platform, may extend beneath the eastern Carpathians (Izotova & Popadyuk 1996). On Peri-Adriatic platforms, bauxite deposits are often conformably overlain by euxinic phosphatic strata, as seen in the Ionian and Lago Negro basins. The petroleum accumulations occurring in coastal Israel are sourced by Late Jurassic pelagic carbonates of the Barnea Formation (Bein & Sofer 1987). In the northern Sinai, the Late Jurassic Gifgafa Shale has been identified as a potential source rock (Abdel Aal & Lelek 1994). Early Cretaceous marine source rocks are widely distributed within the Carpathian nappes and may have contributed to their hydrocarbon potential; however, clear geochemical evidence is lacking (Popescu 1995; Bessereau et al. 1996; Stefanescu & Baltes 1996). Deposition of these source rocks was contemporaneous with the influx of flysch, indicative of orogenic activity. Aptian-Albian source rocks have been reported from the Greater Caucasus area (Robinson et al. 1996). In contrast, the Aptian-Albian syn-rift series of the Black Sea, although containing black shales, do not have an obvious source-rock potential (Robinson et al. 1996). During the Early Cretaceous, the southeastern Mediterranean Basin had widened considerably and was characterized by an euxinic environment (Flexer et al. 1986; May 1991). However, the Neocomian Gevaram and Albian shales have apparently not contributed to the petroleum province of southwestern Israel (Flexer et al. 1986; Bein & Sofer 1987), whereas the time equivalent Sidi Barrani Shale is thought to have contributed to the petroleum province of the Western Desert (Clifford 1986; May 1991). Early Cretaceous lacustrine shales,
524
P.A. Z I E G L E R & F. R O U R E
Fig. 6. Distribution of Middle Jurassic source rocks and related hydrocarbon provinces.
Fig. 7. Distribution of Upper Jurassic-Lower Cretaceous source rocks and related hydrocarbon provinces.
MEDITERRANEAN PETROLEUM SYSTEMS deposited in an extensional basin underlying the southern parts of the Sirte Basin, are thought to be the source of waxy oils occurring in Libya (Macgregor 1996). Early Cretaceous rifts are also evident on the Lampedusa Shelf of Libya (Torelli et al. 1995). A p t i a n - A l b i a n shales may contribute to the petroleum province of Tunisia (Burrolet 1986; Bishop 1988). Oil accumulations in the Southern Apennines, such as the Costa Molina and Tempa Rossa fields, are related to Early Cretaceous source rocks. Late Cretaceous source rocks
During the Late Cretaceous opening of the Atlantic Ocean and northward propagation of sea-floor spreading into the area of the Labrador Sea, the sinistral translation between Laurasia and Africa-Arabia persisted. This was accompanied by rifting activity in the Black Sea, on the northern parts of the Arabian Platform, in the Western Desert and in Libya (Guiraud & Bosworth 1997). At the same time, AfricaArabia began to converge with Europe in a counter-clockwise, rotational mode, resulting in gradual closure of seaways, increasing orogenic activity in the Alpine-Carpathian domain and gradual westward propagation of subduction zones (Ziegler 1988; Ricou 1996). Moreover, cyclically rising sea levels, culminating in the Campanian high-stand, caused an expansion of carbonate platforms on the southern Tethyan shelves and the establishment of the Chalk platform in Europe (Masse et al. 1996). During the Late Cretaceous, multiple anoxic events occurred within the Alpine-Mediterranean domain, the most widespread of which corresponds to the nearly globally recognized 'Bonarelli Member' at the Cenomanian-Turonian boundary (Schlanger et al. 1987). The Late Cretaceous anoxic events reflect a climatically controlled increased marine productivity, decreased bottom in the world oceans, possibly due to the lack of polar ice caps, and shallowing of the oxygen minimum zone, providing for the preservation of organic matter (Hart & Leary 1989; Farrimond et al. 1990; Baudin 1995; Kuhnt & Wiedmann 1995; van Buchem et al. 1995; Baudin & Herbin 1996). In North Africa and on the northern parts of the Arabian platform, Late Cretaceous source rocks play a very important role (Fig. 8; May 1991; Macgregor 1996). Along the Atlantic coast of Morocco, immature Cenomanian-Turonian organic-rich chalks contain major shaleoil reserves in the Tarfaya-Aaiun Basin (Einsele & Wiedmann 1982; Leine 1986). In
525
Tunisia and on the Pelagian Shelf, the basal Turonian Bahloul Shale forms the principal source-rock of this petroleum province (Bishop 1988). Similarly, Cenomanian source-rocks contribute to the petroleum province of the rifted Sirte Basin (Clifford 1986; Macgregor 1996). The Turonian Abu Roash and the Senonian Khoman syn-rift series of the Abu Gharadig graben, located in the Western Desert, contain effective source-rocks (Mustafa a& Khalil 1990; May 1991). In the prolific Cenozoic Gulf of Suez rift, the Santonian 'Brown Limestone', forming part of the pre-rift platform series, is the principal source rock. Some bituminous layers occur in the Santonian-Coniacian Sudr Chalk of the Northern and Central Sinai (Jenkins 1990). In Israel, Turonian and Senonian source rocks are well developed but generally immature, whereas in the Jordanian Late Cretaceous to Eocene Azraq Graben minor amounts of oil have been produced from Late Cretaceous reservoirs charged by Turonian shales (Flexer et al. 1986; May 1991). The petroleum system of the Late Cretaceous Euphrates Graben, located in Southeast Syria, is related to the Coniacian R'Mah source rocks (de Ruiter et al. 1994). The Campanian-Maastrichtian Gurpi-Shiranish Shale contributes significantly to the hydrocarbon potential of northeastern Syria, northern Iraq and Iran (Bordenave & Burwood 1990; May 1991). During the Late Cretaceous, the Adriatic platform continued to be dominated by carbonates, although its northeastern parts were incorporated into the Dinarides-Albanides foreland basin. Widely distributed organic-rich shales were deposited during the Cenomanian-Turonian and Coniacian anoxic events; however, these have contributed only in Albania and the southern Apennines to effective petroleum systems (Jenkins 1991; Casero et al. 1991; Mattavelli et al. 1993; Jerinic et al. 1994; Frasheri et al. t996; Krois 1998). In the basin of southeastern France, Turonian lignites and Santonian kerogenous shales have been identified; these source rocks are only marginally mature under Tertiary grabens and have, if at all, not generated significant amounts of hydrocarbons (Mascle et al. 1996). In the Carpathian flysch nappes, Cenomanian-Turonian shales have a certain sourcerock potential but do not ,appear to have contributed major amounts of hydrocarbons (Bessereau et al. 1996; Dicea 1996; Stefanescu & Baltes 1996).
526
P.A. ZIEGLER & F. ROURE
Cenozoic series and related hydrocarbon provinces During Cenozoic times, Africa-Arabia continued to converge with Europe in a rotational counter-clockwise mode, controlling the main and late phases of the Alpine orogeny. However, following earliest Eocene crustal separation between Europe and Greenland, sinistral translations between Europe and Africa-Arabia ceased and gave way at the onset of the Oligocene to a dextral translation that appears to persists to the present (Ziegler 1988; Ricou 1996). During the Cenozoic phases of the Alpine orogeny, closure of remnant oceanic basins resulted in the collision of Alpine orogenic wedges with the Tethyan passive margins, their incorporation into foreland basins and the emplacement of nappes on forelands. Amongst all the passive margins that had developed during the Mesozoic opening of the Western Tethys, only the margins of northeastern Africa and the Levante are still preserved. Moreover, only the western margin of the oceanic Black Sea Basin was not affected by Alpine compressional deformation. From Oligocene time onwards, increasing collisional coupling between Europe and Africa-Arabia played a significant role in the development of indenter-induced 'escape tectonics', controlling the development of the extensional Aegean and Pannonian back-arc basins. On the other hand, Early Miocene opening of the oceanic Algero-Provenqal and Mio-Pliocene opening of the partly oceanic Tyrrhenian basins may be related to a combination of mechanisms, such as slab roll-back, dextral translation of Europe and Africa, and evolution of the European and African Cenozoic rift systems (Ziegler 1988, 1994). Palaeogene and Miocene effective sourcerocks occur in post-rift and passive margin sequences, in accretionary wedges and flexural foreland basin sequences as well as in Cenozoic syn-rift series. Source rocks in Cenozic post-rift and passive margin series During the Late Cretaceous, the western Black Sea Basin entered its post-rift phase. In contrast, the eastern Black Sea Basin opened by rifting during the Paleocene-early Eocene and thereafter entered the post-rift phase. During their subsequent evolution, both basins coalesced into the present basin configuration. The Western and Eastern Black Sea basins contain up to 13 km and 11 km, respectively, of post-rift
sediment. The southern margins of both basins were affected by compressional tectonics during the late Eocene and younger evolution of the Pontides, whereas the northeastern margin of the Black Sea was affected by the Caucasus orogeny, starting in the late Eocene (Robinson et al. 1995, 1996). In the Black Sea area, the late Eocene-early Oligocene Maykop Shale, an excellent kerogenous source rock, has a regional distribution, reflecting that the entire basin had turned anoxic (Fig. 9). The Maykop series controls the petroleum systems of the Romanian and Bulgarian off-shore, the northern Crimea, the Indolo-Kuban and the Terekcaspian basins and possibly the Gulf of Odessa (Robinson et al. 1996). Palaeocene and Eocene effective source rocks occur in northeastern Africa and in the northern parts of the Arabian Platform. The Palaeocene E1 Haria and the Eocene Bou Dabbous shales contribute to the hydrocarbon potential of the Pelagian shelf (Bishop 1988) and underlay the petroleum system of the Algerian Tellian Atlas (Macgregor 1996). Palaeocene and Eocene source rocks contribute to the hydrocarbon habitat of the Sirte Basin (Parsons et al. 1980; Clifford 1986). Early Eocene pre-rift and Miocene syn-rift series contribute to the petroleum province of the Cenozoic Gulf of Suez rift (Macgregor 1996; Alsharhan & Salah 1997). The source rocks from which the thermal gas and oil occurring in the Nile Delta was derived (Abu ElElla 1990; El Ayouty 1990) is unknown and could possibly be of Palaeogene age. Main oil source rocks of northeastern Syria and northern Iraq include the Palaeocene Geram Series (May, 1991).
Source rocks in accretionary prisms and syn-flexural series Accretionary prisms, as well as syn-flexural series, can contain important source-rocks that contribute towards the hydrocarbon potential of the external parts of Alpine fold- and thrustbelts (Ziegler & Horv~ith 1996). The principal source rock of the Carpathian petroleum system is the Oligocene Menilite Shale (Fig. 10; Koltun 1992; ten Haven et al. 1993; Lafargue et al. 1994; Popescu 1995; Bessereau et al. 1996). These shales were deposited in a tectonically silled flysch basin that was later incorporated into the Carpathian accretionary wedge and thrusted over the foreland in the form of major nappes (Ellouz & Rocca 1994; Dicea 1995, 1996; Kusmierek 1996; Sovchik & Vul 1996). Syn-flexural Molasse-type
M E D I T E R R A N E A N P E T R O L E U M SYSTEMS
Fig. 8. Distribution of Upper Cretaceous source rocks and related hydrocarbon provinces.
Fig. 9. Distribution of Palaeocene-Eocene source rocks and related hydrocarbon provinces.
527
528
P.A. ZIEGLER & F. ROURE
series of the Carpathian foreland basin contain significant amounts of biogenic gas (Fig. 11; Kotarba 1992). Within the Carpathian arc, the Neogene Transylvanian Basin corresponds to a piggy-back foreland basin that is bounded to the west by the Apuseni arc, together with which it was transported to the east during the Carpathian orogeny (Huismans et al. 1997). The hydrocarbon system of the Transylvanian Basin relies mainly on biogenic gas generated in late Miocene and Pliocene deeper water series. Oligocene kerogenous source rocks occur only beneath the northernmost parts of this basin and the adjacent Maramures area where they were deposited in a small foreland basin that developed along the Mid-Hungarian line (Ionescu 1994; Popescu 1995; Cranganu & Deming 1996). Similarly, the Central Carpathian Palaeogene basin of the Slovak Republic and the Ukraine must be regarded as a piggy-back basin; although source rocks contained in its Eocene sedimentary fill are mature, only small amounts of oil and gas have been discovered so far (Nemcok et al. 1996). Several Mesozoic source rocks contribute to the petroleum system of the external Albanides.
However, Mio-Pliocene series contain also biogenic gas (Frasheri et al. 1996; Krois 1998). The petroleum system of the Austrian and German Molasse Basin relies on basal Oligocene kerogenous shales. These were deposited during a phase of rapid thrust-loaded basin subsidence on top of trangressive Eocene clastics and carbonates, resting on partly deeply truncated Mesozoic series (Ziegler 1990; Roeder & Bachmann 1996; Zimmer & Wessely 1996a; Wagner 1996). In the eastern parts of the German Molasse Basin and in the Molasse Basin of Austria, Oligocene and early Miocene clastics are developed in a deep-water facies and contain biogenic gas. In the Northern Apennines, Miocene flysch has contributed some oil (Fig. 12; Mattavelli et al. 1996). The Plio-Pleistocene turbiditic fill of the Apennine foredeep contains major reserves of biogenic gas in the Po plain and the Adriatic Sea (Fig. 11; Anelli et al. 1996). Tertiary source rocks have contributed minor amounts of hydrocarbons to the petroleum system of the Aquitaine Basin (Le Vot et al. 1996). Minor biogenic gas occurs in Neogene sediments of the Gulf of C~idiz (Baraza & Ercilla 1996) and the Rif foredeep (Flinch 1996).
Fig. 10. Distribution of Oligocene source rocks and related hydrocarbon provinces.
M E D I T E R R A N E A N PETROLEUM SYSTEMS
Fig. 11. Occurrence of biogenic gas accumulations in Oligocene-Pliocene, mainly flysch-type series.
Fig. 12. Distribution of Miocene source rocks and related hydrocarbon provinces.
529
530
P.A. ZIEGLER & F. ROURE
C e n o z o i c rift s y s t e m s a n d their p e t r o l e u m systems
During the late Eocene, the Cenozoic rift system of Western and Central Europe began to develop. In its present configuration, it extends from the shores of the North Sea over a distance of some 1100 km into the western Mediterranean domain, from where an alkaline volcanic chain projects southwestwards across the Alboran Sea, the Rif fold belt and the Atlas ranges to the Atlantic coast and the Cape Verde islands. Including this volcanic chain, the entire rift system has a length of some 3000 km. Its southern elements cross-cut the Alpine chains of the western Mediterranean domain. Development of the Cenozoic rift system of Western and Central Europe was contemporaneous with the Eocene and later phases of the Alpine orogeny, during which the northwestern Alpine foreland was repeatedly subjected to horizontal intraplate compressional stresses, causing inversion of Mesozoic extensional basins at considerable distances from the Alpine thrust front (Ziegler 1990, 1994). Viewed on a broader scale, evolution of the Cenozoic Western and Central European rift system was essentially contemporaneous with the development of the East African-Red Sea, Libyan and Pelagian Shelf rift systems (Fig. 1). Moreover, Neogene development of these rift systems was paralleled by back-arc extension governing the subsidence of the oceanic AlgeroProvenqal Basin, the Alboran Basin, the partly oceanic Tyrrhenian Sea, the Pannonian Basin and the Aegean Sea. As such, the R h i n e Rh6ne-Valencia and the Red Sea-Libyan-Pelagian Shelf rift systems can be considered as forming part of the Neogene Alpine-Mediterranean collapse system. Most of the basins associated with the Alpine-Mediterranean rift and collapse system are characterized by thinned lithosphere and, correspondingly with relatively high geothermal gradients. These have contributed to the maturation of pre- and syn-rift source-rocks (Ziegler 1996a, b). Petroleum provinces associated with Cenozoic rifted basins often rely for hydrocarbon charge on one or more petroleum system. The European Cenozoic rift system hosts the Upper Rhine Graben, Gulf of Lions and Valencia Trough hydrocarbon provinces (Fig 10). The Rhine Graben petroleum province relies on a combination of Jurassic pre-rift and Oligocene syn-rift marine source rocks (Boigk 1981; Mauthe et al. 1993; Richard 1994). Small oil accumulations along the eastern margin of the
Bresse Graben are related to Permo-Carboniferous source rocks (Deville et al. 1989). Small oil discoveries in the rifted basins of the Gulf of Lion area are sourced by Oligocene lagoonal and lacustrine syn-rift deposits (Mascle et al. 1996; Vially & Tr6moli6res 1996). The oil province of the Valencia Trough is mainly related to the marine late Oligocene-early Miocene syn-rift Taraco Shale with the Late Jurassic Salzedella Shale contributing locally (Torres & Bois 1993; Torn6 et al. 1996; Salas et al. 1997). Limited exploration efforts in the Alboran Basin have yielded only gas shows that cannot be tied to a specific source rock (Chalouan et al. 1997). .The major Gulf of Suez hydrocarbon province relies principally on Late Cretaceous and Eocene pre-rift source rocks; the Miocene synrift Rudeis Shale has a rather low source potential and is probably only a secondary contributor to this important oil province (Fig. 12). However, hydrocarbon accumulations occurring in the northernmost parts of the Red Sea are exclusively charged by hydrocarbons generated by Miocene syn-rift source rocks (Macgregor 1996; Alsharhan & Salah 1997). On the other hand, the hydrocarbon potential of the prolific Sirte Basin is essentially related to Late Cretaceous and Palaeogene syn-rift source rocks (Clifford 1986, Macgregor 1996). The hydrocarbon potential of the Hungarian, Romanian and former Yugoslavian MioPliocene transtensional troughs, forming part of the Pannonian system of basins, is mainly related to Miocene marine syn-rift source rocks (Fig. 12); Oligocene and Mesozoic pre-rift series have contributed in some areas and may be locally the dominant source rocks (Kdkai & Pogficsfis 1991; Clayton & Koncz 1994; Milota et al. 1995; Popescu 1995; Mattick et al. 1996; Milicka et al. 1996). Among the Pannonian system of basins, the relatively small Vienna Basin is the most prolific; its hydrocarbon charge is exclusively derived from Late Jurassic shales preserved in the sub-thrust autochthon nappes (Krol11880; Ladwein 1988; Seifert 1996). Unlike the main parts of the Pannonian system of basins, the Vienna Basin is characterized by a relatively low geothermal gradient as extension affected only the Alpine nappe stack and not the underlying autochthonous basement (Tari 1996). A combination of pre-rift and Neogene synrift source-rocks controls the hydrocarbon occurrences in the transtensional Thrace Basin (Karahanoglu et al. 1995). In the northern Aegean, the small Prinos oil and Kavala gas accumulations are located in a rifted basin; they
MEDITERRANEAN PETROLEUM SYSTEMS are trapped beneath Messinian salt and were probably charged with hydrocarbons generated by pre-Messinian syn-rift sediments (Georgakopoulos 1995).
Effects of Alpine-Mediterranean dynamics on former passive-margin series and their hydrocarbon habitat Source-rocks contained in Palaeozoic platform series and in Mesozoic passive-margin prisms generally reached maximum burial, and thus peak maturity, either prior to or during their involvement into Alpine fold-and-thrust belts. However, some of the Tethys shelves were compressionally deformed prior to or during their collision with Alpine orogenic wedges (Ziegler et al. 1995, 1998). This resulted in deep erosion of their Mesozoic and Palaeozoic sedimentary cover and the destruction of their hydrocarbon potential. Moreover, development of the Cenozoic rift systems and the associated uplift of broad rift domes resulted in some areas in partial destruction of pre-existing Mesozoic shelves (e.g. Vosges-Black Forest arch, Ziegler 1994). Petroleum provinces of flexural foreland basins a n d external f o l d - a n d - t h r u s t belts
In the Alpine-Mediterranean area, flexural foreland basins developed on continental crust at different times during the Alpine orogenic cycle. For instance, the oldest and longest lived foreland basin system is associated with the Dinarides and Albanides. It evolved at the end of the Jurassic in response to. closure of the Vardar and Sub-Pelagonian oceanic basins and the obduction of ophiolites on the margin of the Ionian shelf. However, this Early Cretaceous to Palaeocene flysch basin (Krasta zone) was later incorporated into the allochthon. Syn-flexural series on the Ionian platform range in age from Oligocene to Pliocene, at least in its Albanian sector (Frasheri et al. 1996; Krois 1998). Foreland basins associated with the Balkans (Georgiev & Atanasov 1993; Doglioni et al. 1996), the South Turkish thrust belt (Gilmour & Mfike11996) and the northern Pyrenees (Le Vot et al. 1996) developed during the Senonian and Palaeogene. The Alpine and North Carpathian foreland basin evolved during early Eocene to Mid-Miocene times (Roure et al. 1992, 1994; Ziegler 1996; Bessereau et al. 1996; Roeder & Bachmann 1996; Wagner 1996) whereas the East Carpathian foreland basins developed only during the Mio-Pliocene (Ellouz & Rocca 1994)
531
and the South Carpathian Getic depression during the Eocene to Pliocene (Dicea 1996). On the other hand, the Apennine foreland basin evolved during the late Miocene and Pliocene (Anelli et al. 1996) and the flexural basins associated with the Betic Cordillera and Rif fold-andthrust belt during the Mid-Miocene and Pliocene (Flinch 1996). In contrast, earlierformed foreland basins associated with the Algerian Tellian nappes and the Western Alps were destroyed during late orogenic phases in conjunction with compressional foreland deformations. Also the Cretaceous syn-flexural flysch basins of the Albanides and Carpathians were heavily deformed during the Cenozoic orogenic phases. Petroleum provinces related to these flexural foreland basins and their associated fold-andthrust belts depend variably on source rocks contained in syn- and post-rift sequences of the respective passive margins, and/or in accretionary prism and syn-flexural series. Maturation of such source rocks can occur prior to and/or during thrust emplacement. Foreland basins, and their extensions beneath the external parts of fold-and-thrust belts, are generally characterized by low geothermal gradients. Therefore, source rocks contained in the passive-margin sequence and in syn-flexural series often retain their hydrocarbon generation potential, even in sub-thrust positions, to great depths (Roure & Sassi 1995; Ziegler & Route 1996). Foreland shelves that did not experience severe intra-plate deformations prior to their incorporation into flexural foreland basins, retained their source potential and developed, partly in conjunction with thrust-loaded maturation of source rocks and synorogenic structuration, into important hydrocarbon provinces. For example, the petroleum system of the South Turkish thrust belt relies exclusively on pre-rift Silurian source rocks (Gilmour & M~ikel 1996). The Aquitaine Basin petroleum province, located in the North Pyrenean foreland, relies largely on Late Jurassic-Early Cretaceous synrift source rocks (Le Vot et al. 1996). The petroleum system of the Balkan foreland basin is related to Triassic and Jurassic syn-rift source rocks (Georgiev & Atanasov 1993). Whereas Mesozoic post-rift series contribute to the hydrocarbon habitat of the German Molasse Basin (Bachmann et al. 1987) and are of major importance in the Czech foreland basin (Brzobohaty et al. 1996; Krejci et al. 1996; Picha 1996), the hydrocarbon potential of the Austrian Molasse Basin relies exclusively on syn-flexural series as its Mesozoic substratum was deeply
532
P.A. ZIEGLER & F. ROURE
disrupted during the Palaeocene phase of compressional foreland deformations (Zimmer & Wessely 1996a). In contrast, hydrocarbon accumulations in the Alpine nappes beneath the Vienna Basin and in its Neogene fill are exclusively charged by Late Jurassic source-rocks preserved in the subthrust autochthon (Zimmer & Wessely 1996b). The petroleum potential of the external Carpathian fold-and-thrust belt relies almost exclusively on the Oligocene Menilite Shale, contained in the accretionary wedge (Ellouz & Rocca 1994; Dicea 1995, 1996; Kusmierek 1996; Sovchik & Vul 1996). The hydrocarbon province of the external Albanides is mainly related to Triassic syn-rift and Jurassic and Late Cretaceous post-rift source rocks (Frasheri et al. 1996; Krois 1998). The hydrocarbon potential of the Po Plain and Adriatic foreland basin relies on Triassic-Early Jurassic syn-rift source rocks and biogenic gas generated in their late Miocene and Plio-Pleistocene synflexural wedges. Similarly, the petroleum provinces of the South Alpine and Apennine-Sicily thrust belts are largely related to Triassic and Early Jurassic syn-rift sourcerocks but lack a major contribution from synflexural biogenic gas (Brosse et al. 1990; Casero & Roure 1994; Anelli et al. 1996).
Foreland inversion and distribution of source rocks
During the Late Cretaceous and Cenozoic phases of the Alpine orogeny, collision-related horizontal compressional stresses reactivated in the European, African and Arabian forelands a broad spectrum of tensional and transtensional Mesozoic and Palaeozoic basins and fracture systems, causing basin inversion and upthrusting of basement blocks. Basin inversion can have severe repercussions on the hydrocarbon habitat of a basin, mainly by reversing its subsidence pattern, by re-configuration of pre-existing structural and stratigraphic traps, and by profound erosion, resulting in the loss of hydrocarbons to the surface. However, partly inverted basins can host important hydrocarbon provinces, such as those of the Lower Saxony and West Netherlands basins (Ziegler 1990, 1996b; Ziegler et al. 1995, 1998), particularly if post-inversion series account for further maturation of source rocks (e.g. Abu Gharadig Basin, Western Desert; Abdel Aal & Moustafa 1990). In the Alpine-Mediterranean domain, Late Cretaceous and Cenozoic basin inversion played an important role in the evolution and hydrocarbon potential of the rifted Mesozoic basins of
Morocco, Algeria and Tunisia (Vially et al. 1994; Zizi 1996), Egypt (Abu Gharadig Basin: Hantar 1990; north Sinai, Mustafa & Khalil 1990; Guiraud & Bosworth 1997), the Levante (Sinai fold belt: Quennell 1984; Palmyra Trough: Salel & S6guret 1994), Iberia (Ziegler 1988; Salas & Casas 1993) and the foreland of the Alps and Carpathians (Ziegler 1990; Roure & Colletta 1996), Of special interest is the latest Cretaceous and Palaeocene compressional deformation of the East Alpine and Northern Carpathian foreland, involving Rocky Mountain-style upthrusting of basement blocks in the Bohemian Massif and inversion of the Polish Trough, prior to the development of a flexural foreland basin (Ziegler 1990; Ziegler et al. 1995). These deformations, which extend deeply beneath the Carpathian-Alpine orogenic wedge, caused profound disruption of the sedimentary cover of the European passive margin, thus controlling the preservation of Mesozoic reservoirs and source rocks (Late Jurassic source rocks are preserved only beneath the Vienna Basin). Therefore, this phase of intra-plate deformations had a strong effect on the prospectivity of the Polish, Czech and Austrian foreland basin and subthrust plays (Bessereau et al. 1996; Brzobohaty et al. 1996; Wagner 1996; Zimmer & Wessely 1996a; Picha 1996). Similarly, Mid-Cretaceous inversion of the Dobrogean Mesozoic rift had repercussions on the subthrust plays of the Eastern Carpathians. Whether the inverted Polish and Dobrogean troughs link up beneath the Carpathian orogenic wedge is uncertain. On the Sahara Platform, the distribution of Palaeozoic petroleum provinces is largely controlled by Permo-Carboniferous and Palaeogene compressional foreland deformations that caused a profound disruption of its Palaeozic and Mesozoic sedimentary cover (Figs 2 and 3). The thickness of the Mesozoic strata controlled whether or not Palaeozoic series preserved beneath the regional 'Hercynian' unconformity entered during the Mesozoic a second, postinversion phase of source-rock maturation and hydrocarbon expulsion (Makhous et al. 1997). Cenozoic disruption of the Mesozoic sedimentary cover during the inversion of the Saoura-Ougarta chain (Ziegler 1988) changed the hydrodynamic system of Mesozoic reservoirs that were charged with hydrocarbons generated by Palaeozoic source rocks.
Conclusions The petroleum systems of the external parts of the Alpine orogen, its forelands and internal
MEDITERRANEAN PETROLEUM SYSTEMS basins are highly variable. They rely either on pre-rift, syn-rift or syn-flexural source rocks or on a combination thereof. Some of these source rocks attained maturity already during the passive-margin or flexural post-rift basin stage; others reached maturity only during the emplacement of allochthonous units (Ziegler & Roure, 1996). Late Cretaceous and Cenozoic rifted basins, which developed during the Alpine orogenic cycle, host a number of important hydrocarbon provinces. The petroleum habitat of each basin has its own case history. Contrasting thermal regimes and successive episodes of sedimentary and tectonic burial account for the great diversity of petroleum provinces identified in the Alpine orogen and associated basins. Each hydrocarbon province is characterized by a very distinct scenario for source-rock development, the timing of source-rock maturation, hydrocarbon expulsion, and migration from effective kitchens to potential trapping domains, and the preservation of hydrocarbon accumulations (Ziegler & Route 1996). The petroleum exploration and production history of the Alpine orogen and its associated basins commenced in the early nineteenth century. First hydrocarbon exploration efforts, based on surface seeps, were concentrated on the shallow, frontal anticlines of the Albanides and the Outer Carpathians of Poland, the Ukraine and Romania. In the course of time, improved geophysical methods provided new tools for imaging sub-surface structures. This resulted in new oil and gas discoveries in such foreland basins as the Po Plain (1960s), the Adriatic Sea, the German and Austrian Molasse Basin and the Aquitaine Basin (1960s), as well as in the Pannonian and Vienna neo-autochthonous basins (before World War II). With the improvement of static corrections and CDPmethods, and ultimately the advent of 3Dseismic technology, imaging of even structurally very complex areas became possible, thus providing access to complex thrusted structures and sub-thrust prospects of the Lopushnia (Ukrainian Carpathians, 1980s) and Tempa Rossa (Southern Apennines, 1990s) type (D'Andrea et al. 1993; Roure & Sassi 1995; Picha 1996). No doubt, future exploration efforts in the external parts of the Alpine orogen and beneath the Neogene fill of the Pannonian Basin will yield further oil and thermal gas discoveries, provided it can be established that efficient petroleum systems are available, have survived the most recent deformations, and that only minor hydrocarbon re-migration has occurred (Picha 1996; Wessely & Liebl 1996a).
533
On the other hand, the hydrocarbon potential of the deep-water Mediterranean basins must be regarded as speculative (Burrolet 1986). In the Gulf of Lion, untested Oligo-Miocene synrift series may have, in analogy with the Valencia Trough, a source potential (Vially & Trdmoli6res 1996). In the Algero-Provenqal Basin, sands associated with the Messinian drop in sea-level, possibly charged by source rocks occurring in the pre-salt sequences, may present viable objectives (Prinos and Kavala analogue, Georgakopoulos 1995); post-salt series may contain biogenic gas in turbiditic sands. The partly inverted extensional Alboran Basin, which is characterized by an up to 9 km thick Neogene sedimentary fill, mud diapirs and volcanic activity, is under explored; however, its hydrocarbon potential is uncertain, despite the occurrence of minor shows (Chalouan et al. 1997; Pdrez-Belzuz et al. 1997). The petroleum potential of the sediment starved, young Tyrrhenian and Aegean back-arc basins is questionable and probably limited as they evolved by extensional collapse of orogenic wedges, rendering the prediction of pre-rift petroleum systems extremely uncertain. Moreover, PlioPleistocene volcanic activity provides for a high CO2 risk. Exploration of the productive Pelagian-Lampedusa Shelf continues (Torelli et al. 1995). The Nile delta, which developed in response to Miocene thermal doming of the Red Sea-Gulf of Suez rift, contains several gas and oil accumulations in Mio-Pliocene sands and is the target of continued exploration activity (Abu El-Ella 1990; Harms & Wray 1990; E1 Ayouty 1990). The age of opening of the deep, partly oceanic Central and Southeast Mediterranean basins is uncertain (Permian, Stampfli 1996; Cretaceous, Dercourt et a11993); the hydrocarbon potential of their very thick sedimentary fill (Makris et al. 1983; Sage & Letouzey 1990; Mart 1994; Lallemant et al. 1994) is still an open question. However, gas shows in ODP wells drilled on mud volcanos of the Mediterranean Ridge thrust belt, as well as oil and gas shows obtained in gravity cores and grab samples, indicate the presence of mature source rocks of unknown age (Robertson et al. 1996; Cronin et al. 1997). In this deep-water frontier area, reservoir development is probably a major risk element. In contrast, the petroleum potential of the Black Sea Basin, which has undergone a complex evolution and is characterized by a very thick post-rift sedimentary prism, must be regarded as encouraging, particularly in the face of the regional distribution of the late Eocene Maykop source rocks (Robinson et al. 1996).
534
P.A. Z I E G L E R & F. R O U R E
The authors are indebted to the Institut Franqais du P6trole for preparing the text figures. Thanks are extended to G. Nesen, L. Mattavelli and A. Richard for constructive comments to an earlier version of this manuscript and also to numerous colleagues who have contributed to this paper through informal discussions.
References ABDELAAL, A. & LELEK,J. J. 1994. Structural development of the Northern Sinai, Egypt, and its implication on the hydrocarbon prospectivity of the Mesozoic. In: AL-HUSSEIM, M. (ed.) The Middle East Petroleum Geosciences. Geo'94. Gulf PetroLink, Manama, Bahrain, l, 15-30. & MOUSTEFA,A. R. 1990. Structural framework of the Abu Gharadig basin, Western Desert, Egypt. Proceedings of 9th Petroleum Exploration and Production Conference, Cairo, Nov. 1998. EGCP, 2, 23-50. ABU EL-ELLA, R. 1990. Maturation history of Neogene-Quaternary sediments, Nile Delta Basin, Egypt. Bulletin of the American Association of Petroleum Geologists, 74, 77-84. ALSHARHAN, A. S. & SALAH, M. G. 1997. A common source rock for Egyptian and Saudi hydrocarbons in the Read Sea. Bulletin of the American Association of Petroleum Geologists, 81, 1640-1659. ANELLI, L., MATTAVELLI,L. & PIERI, M. 1996. Structural-stratigraphic evolution and petroleum systems of Italy. In: ZIEGLER, P. A. & HORVATH,F. (eds) Structure and Prospects' of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2, M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 455--483. AYRES, M. G., BH~AL,M., JONES, R. W., SLEUTZ, L. W., TARTIR, M. & WOLSON, A. O. 1982. Hydrocarbon habitat in main producing areas, Saudi Arabia. Bulletin of the American Association of Petroleum Geologists, 66, 1-9. BACHMANN, G. H., MULLER, M. & WEGGEN, K. 1987. Evolution of the Molasse Basin (Germany, Switzerland). Tectonophysics, 137, 77-92. BARAZA, J. & ERCILLA, G. 1996. Gas-charged sediments and large pockmark-like features on the Gulf of Cadiz slope (SW Spain). Marine and Petroleum Geology, 13, 253-261. BAUD~N, E 1995. Depositional controls on Mesozoic source-rocks in the Tethys. In:. Huc, A.-Y. (ed.) Paleogeography, Paleoclimate and Source Rocks. American Association of Petroleum Geologists, Studies in Geology, 40, 191-211. --& HERBIN, J.-R 1996. Paleoenvironments and organic-rich facies deposition in the Tethyan realm. In: NAmN, A. E. M., RIcou, L.-E., VRIELYNK, B & DERCOURT, J. (eds) The Ocean Basins and Margins, Vol 8. The Tethys Ocean. Plenum Press, New York, 319-346. BEIN, A. & SOFER, Z. 1987. Origin of oils in Helez Region, Israel - Implications for exploration in the Eastern Mediterranean. Bulletin of the American Association of Petroleum Geologists, 71, 65-75. BESSEREAU, G., GUILLOCHEAU,E & Huc, A.-Y. 1995. -
-
Source rock occurrence in a sequence stratigraphic framework: the example of the Lias of the Paris Basin. In: Huc, A.-Y. (ed.) Paleogeography, Paleoclimate and Source Rocks. American Association of Petroleum Geologists, Studies in Geology, 40, 273-301. --, ROURE, F., KOTARBA, M., KUSMIEREK, J. d~; STRZETELSKI,W. 1996. Structure and hydrocarbon habitat of the Polish Carpathians. In: ZlEGLER, R A. & HORVATH, F. (eds) Structure and Prospects of Alpine Basins and Forelands'. Peri-Tethys Memoirs, 2, M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 343-373. BEST, J. A., BARAZANGI,M., AL-SAAD, D., SAWAF,T. & GEBRAN, A. 1993. Continental margin evolution of the North Arabian Platform in Syria. Bulletin of the American Association of Petroleum Geologists, 77, 173-193. BEYDOUN, Z. R. & HAmB, J. G. 1995. Lebanon revisited: new insight into Triassic hydrocarbon prospects. Journal of Petroleum Geology, 18, 75-90. --, HUGHES CLARK, M. W. & STONELY, R. 1992. Petroleum in the Zagros Basin: a Late Tertiary foreland basin overprinted onto the outer edge of a vast hydrocarbon-rich Paleozoic-Mesozoic passive margin shelf. In: MACOUEEN, R. W. & LECKIE, D. A. (eds) Foreland Basins and Fold Belts. American Association of Petroleum Geologists, Memoirs, 55, 309-339. B~SHOP,W. E 1988. Petroleum geology of East-Central Tunisia. Bulletin of the American Association of Petroleum Geologists, 72, 1033-1058. BOIGK, H. 1981. ErdOl und Erdgas in der Bundesrepublik Deutschland. ErdOlprovinzen, Felder, FOrderung, Vorriite, Lagerstiittentechnik. Ferdinand Elke Verlag, Stuttgart. BORDENAVE, M. L. & BURWOOD,R. 1990. Source rock distribution and maturation in the Zagros orogenic belt: provenance of the Asmari and Bangestan reservoir oil accumulations. Organic Geochemistry, 16, 369-387. BOURROUILH, R., RICHERT, J.-R & ZOLNAI, G. 1995. The North Pyrenean Aquitaine Basin, France: Evolution and Hydrocarbons. Bulletin of the American Association of Petroleum Geologists, 79, 831-853. BRANGULIS, A. R, KANEV, S. V., MARGULIS, L. S. & POMERANTSEVA,R. A. 1993. Geology and hydrocarbon prospects of the Paleozoic in the Baltic region. In: PARKER, J. R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. Geological Society, London, 621-656. BROSSE, E., RIVA, A., SARTI_JCCI, S., BERNON, M., LOREAU, J. R, FIXA, A. & LAGGOUN-DEFARGE,E 1990. Some sedimentogical and geochemical characters of the late Triassic Noto Formation, source rock in the Ragusa Basin, Sicily. Organic Geochemistry, 16, 715-734. BROUGHTON. P. & TREeANIER, A. 1993. Hydrocarbon generation in the Essaouira Basin of Western Morocco. Bulletin of the American Association of Petroleum Geologists', 77, 999-1015. BRZOBOHATY,J., BENADA, S., BERKA, J. & REHANEK,J.
M E D I T E R R A N E A N P E T R O L E U M SYSTEMS 1996. Hydrocarbon habitat of the Paleogene Nesvacilka Trough, Carpathian foreland basin, Czech Republic. In: ZIEGLER, P. A. & HORVATH,E (eds) Structure and Prospects of Alpine Basins and Forelands'. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 305-319. BURROLET,R F. 1986. Deep Mediterranean basins and their oil potential. In: HALBOUTY, M. T. (ed.) Future Petroleum Provinces of the WorM. American Association of Petroleum Geologists, Memoirs, 40, 545-557. CASERO,R & ROURE, F. 1994. Neogene deformation at the Sicily-North African plate boundary. In: ROURE, E (ed.) Peri-Tethyan Platforms. Editions Technip, Paris, 27-50. -& VIALLY,R. 1991. Tectonic framework and petroleum potential of the Southern Apennines. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons. Special Publications of the European Association of Petroleum Geoscientists, 1, 381-387. CHALOUAN,A., SAJI, R., MICHARD,A. & BALLY,A. W. 1997. Neogene tectonic evolution of the southwestern Alboran Basin as inferred from seismic data off Morocco. Bulletin of the American Association of Petroleum Geologists, 81, 1161-1184. CIPRYS,g., ADAMEK, J. & BENADA, S. 1995. Petroleum geology of the Carpathian foredeep and overthrust zone in the Czech Republic. Petroleum Geoscience, 1, 89-95. CLAYTON,J. L. & KONCZ, I. 1994. Petroleum geochemistry of the Zala Basin, Hungary. Bulletin of the American Association of Petroleum Geologists, 7 8 , 1-22. CLIFFORD, A. C. 1986. African oil - past, present and future. In: HALBOUTY,M. Y. (ed.) Future Petroleum Provinces of the WorM. American Association of Petroleum Geologists, Memoirs, 40, 339-372. CORNFORD, C. 1998. Source rocks and hydrocarbons of the North Sea. In: GLENNIE,K.W. (ed.) Petroleum Geology of the North Sea, Basic concepts and recent advances. 4th ed. Blackwell Science Ltd., Oxford, in press. CRANGANU, C. & DEMING, D. 1996. Heat flow and hydrocarbon generation in the Transylvanian Basin, Romania. Bulletin of the American Association of Petroleum Geologists, 80, 1641-1653. CRONIN, B. T., IVANOV,M. K., LIMONOV,A. E, EGOROV, A., AKHMANOV, G. G., AKHEMATJANOV,A. M., KOZLOVA, E. & SHIPBOARD SCIENTIFIC PARTY TTR-5 1997. New discoveries of mud volcanoes on the East Mediterranean Ridge. Journal of the Geological Society, London, 154, 173-182. D'ANDREA, S., PASU, R., BERTOZZI, G. & DATTILO, P. 1993. Geological model, advanced models help unlocking oil in Italy's Apennines. Oil and Gas Journal, Aug. 23, 3-56. DERCOURT, J., RICOU, L. E. & VRIELYNCK, B. (eds.) 1993. Atlas Tethys, palaeoenvironmental maps. Gauthier-Villars, Paris. DE RUITER, R. S. C., LOVELOCK,P. E. R. & NABLUSI,N.
535
1994. The Euphrates Graben of eastern Syria: a new petroleum province in the northeastern Middle East. In: AL-HUSSEINI, M. (ed.) The Middle East Petroleum Geosciences, Geo'94. Gulf PetroLink, Manama, Bahrain, 7, 357-368. DEVILLE, E., GUELLEC,S., TARDY,M., MUGNIER, J. L. & MASCLE,A. 1989. Structural styles of deformation in the foreland and external part of the Western Alps area: a view from ECORS seismic reflection to surface studies. Institut Fran~ais du P6trole, post-conference field trip, research conference "The Potential of Deep Seismic Profiling for Hydrocarbon Exploration", Aries. DICEA, O. 1995. Structure and hydrocarbon geology of the Romanian East Carpathian border from seismic data. Petroleum Geoscience, 1, 135-143. - 1996. Tectonic setting and hydrocarbon habitat of external Carpathian basins in Romania. In: ZIEGLER, V.A. & HORVATH,E (eds) Structure and Prospects of Alpine Basins and Forelands. PeriTethys Memoirs 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 403-425. DOGLIONI, G., BUSATrA, C., BOLIS, G., MARIANINI,L. & ZANELLA, M. 1996. Structural evolution of the eastern Balkans (Bulgaria). Marine and Petroleum Geology, 13, 225-251. EINSELE, G. & WIEDMANN, J. 1982. Turonian black shales in the Moroccan coastal basins. In: YON RAAD, U. ET AL. (eds) Geology of the Northwest African Continental Margin. Springer Verlag, 396-414. EL AYOUTY,M. 1990. Petroleum Geology. In: SAID,R. (ed.) The Geology of Egypt. A.A. Balkema, Rotterdam, Brookfield, 567-599. ELLOUZ, N. & ROCCA, E. 1994. Palinspastic reconstruction of the Carpathians and adjacent areas since the Cretaceous: a quantitative approach. In: ROURE, F. (ed.) Peri-Tethyan Platforms. Editions Technip, Paris, 51-78. ESPITALII~,J., LAFARGUE,E. & EGGEN, S. 1991. Petroleum potential of terrestrial and marine organic matter in Jurassic sequences of the northern North Sea and offshore mid-Norway. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production of Europe's" Hydrocarbons. Special Publications of the European Association of Petroleum Geoscientists, 1, 49-63. FARRIMOND,P., EGLINGTON, G., BRASSELL, S. C. & JENKINS, H. C. 1989. Toarcian anoxic event in Europe: an organic geochemical study. Marine and Petroleum Geology, 6, 136-147. & -1990. The CenomanianTuronian anoxic event in Europe: an organic geochemical study. Marine and Petroleum Geology, 7, 75-89. FLEXER, A., ROSENFELD, A., LIPSON-BEN1TAH, S. & HONIGSTEIN, A. 1986. Relative sea level changes during the Cretaceous in Israel. Bulletin of the American Association of Petroleum Geologists, 70, 1685-1699. FLINCH, J. F. 1996. Accretion and extensional collapse of the external Western Rif (Northern Morocco). In: ZIEGLER, RA. & HORVATH,E (eds) Structure
536
P.A. Z I E G L E R & F. R O U R E
and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs 2. Mdmoires du Mus6um National d'Histoire Naturelle, Paris, 170, 61-85. FOURCADE, E., AZEMA, J., BASSOULLET,J.-P., CECCA,E & DERCOURT,J. 1996. Paleogeography and paleoenvironment of the Tethyan realm during the Jurassic Breakup of Pangea. In: NAIRN, A. E. M., RICOU, L.-E., VRIELYNK,B. & DERCOURT,J. (eds) The Ocean Basins and Margins, Vol. 8, The Tethys Ocean. Plenum Press, New York, 190-214. FRASHERI, A., NISHANI, P., BUSHATI, S. & HYSENI, A. 1996. Relationship between tectonic zones of the Albanides, based on results of geophysical studies. In: ZIEGLER,P A. & HORVATH, F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Musdum National d'Histoire Naturelle, Paris, 170, 485-511. GEORGAKOPOULOS,a . N., PAPACONSTANTINOU,C. M. & BAPAIOANNOU,CH. A. 1995. Natural gas storage in the South Kavala field: geological and seismological characteristics. Petroleum Geoscience, 1, 129-133. GEORGIEV, G. & ATANASOV,A. 1993. The importance of the Triassic-Jurassic unconformity to the hydrocarbon potential of Bulgaria. First Break, 11, 489-497. GILMOUR, N. R. ~; MAKEL, G. 1996.3-D geometry and kinematics of the N.V. Turkse Shell thrust belt oil fields, Southeast Turkey. In: ZIEGLER, P. A. & HORVATH, F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Musdum National d'Histoire Naturelle, Paris, 170, 524-547. GUIRAUD, R . 8,: BOSWORTH,W. 1997. Senonian basin inversion and rejuvenation of rifting in Africa and Arabia: synthesis and implications to plate-scale tectonics. Tectonophysics, 282, 39-82. HANTAR, G. 1990. North Western Desert. In: SAID, R. (ed.) The Geology of Egypt. A.A. Balkema, Rotterdam, Brookfield, 293-319. HARMS, J. C. & WRAY, J. L. 1990. Nile Delta. In: SAID, R. (ed.) The Geology of Egypt. A.A. Bakema, Rotterdam, Brookfield, 329-343. HART, M. B. & LEARY, E N. 1989. Stratigraphic and palaeogeographic setting of the Late Cenomanian "anoxic" event. Journal of the Geological Society, London, 146, 305-310. HUISMANS,R. S., BERTOTTI,G., CIULAVU,D., SANDERS, C. A. E., Ct,OETINGH, S. & DINU, C. 1997. Structural evolution of the Transylvanian Basin (Romania): a sedimentary basin in the bend zone of the Carpathians. Tectonophysics, 272, 249-268. HUSSE1NI,M. I. 1991. Tectonic and depositional model of the Arabian and adjoining plates during the Silurian-Devonian. Bulletin of the American Association of Petroleum Geologists, 75, 108-120. IONESCU, N. 1994. Exploration history and hydrocarbon prospects in Romania. In: POPESCU,B. M. (ed.) Hydrocarbons of Eastern Central Europe. Springer Verlag, Berlin, 217-248. IZOTOVA, T. S. & POPADYUK, I. g. 1996. Oil and gas accumulations in the Late Jurassic reefal complex of the West Ukrainian foredeep. In: ZmGLER, P.A.
& HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 375-390. JABOUR, H. & NAKAYAMA,K. 1988. Basin modeling of Tadla Basin, Morocco, fdr hydrocarbon potential. Bulletin of the American Association of Petroleum Geologists, 72, 1059-1073. JENKINS, D. A. 1990. North and Central Sinai. In: SAID, R. (ed.) The Geology of Egypt. A.A. Balkema, Rotterdam, Brookfield, 361-380. JENKINS,H. C. 1991. hnpact of Cretaceous sea level rise and anoxic events on the Mesozoic carbonate platform of Yugoslavia. Bulletin of the American Association of Petroleum Geologists, 75, 1007-1017. JERINIC, G., JELASKA, g. & ALAJBEG, A. 1994. Upper Cretaceous organic-rich laminated limestones of the Adriatic carbonate platform, Island of Hvar, Croatia. Bulletin of the American Association of Petroleum Geologists, 78, 1313-1321. KABYSHEV, B., KRIVCHENKOV, B., STOVBA, S. & ZmGLER, R A. 1998. Hydrocarbon habitat of the Dniepr-Donets Depression. Marine and Petroleum Geology, in press. KARAHANOGLU, N., ERLER, A. & ILLEEZ, H. I. 1995. Mathematical approach to hydrocarbon generation history and source rock potential in the Thrace Basin, Turkey. Marine and Petroleum Geology, 12, 597-956. KARAKITSIOS, g. 1995. The influence of preexisting structure and halokinesis on organic matter preservation and thrust system evolution in the Ionian basin, NW Greece. Bulletin of the American Association of Petroleum Geologists, 79, 960-980. KETrEL, D. 1989. Upper Carboniferous source-rocks north and south of the Variscan front (NW and Central Europe). Marine and Petroleum Geology, 6, 170-181. - & HERZOG, M. 1989. The Permo-Carboniferous of the South German Molasse Basin. Oil-Gas European Magazine, 15, 29-32. KLEMME, H. D. ~: ULMISHEK, G. E 1991. Effective petroleum source rocks of the World: stratigraphic distribution and controlling depositional factors. Bulletin of the American Association of Petroleum Geologists, 75, 1809-1851. KOKAI, J. & POGACSAS,G. 1991. Hydrocarbon plays in Mesozoic nappes, Tertiary wrench basins and interior sags in the Pannonian Basin. First Break, 9, 315-334. Kot:rUN, Y. V. 1992. Organic matter in Oligocene Menilite Formation rocks of the Ukrainian Carpathians: paleoenvironment and geochemical evolution. Organic Geochemistry, 18, 423-430. KOTARBA,M. 1992. Bacterial gases in Polish part of the Carpathian foredeep and the Flysch Carpathians: isotopic and geologic approach. In: VIALLY, R. (ed.) Bacterial Gas. Editions Technip, Paris, 133-146. KREJCI, 0., FRANCU,J., PEOLCHAU,H. S., MULLER,P. STRANIK, Z. 1996. Tectonic evolution and oil and gas generation at the border of the North
M E D I T E R R A N E A N PETROLEUM SYSTEMS European Platform with the West Carpathians (Czech Republic). In: WESSELY, G. & LIEBL, W. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. Special Publication of the European Association of Petroleum Geoscientists Eng., 5, 177-186. KROIS,P. 1998. Geological evolution and hydrocarbon habitat of the External Albanides (Albania). Tectonophysics, in press. KROLL,A. 1980. Das Wiener Becken. In: BACHMAYER, E, BRIX, E & SCHULTZ,O. (eds) ErdOl und Erdgas in Oesterreich. Selbsverlag Nattirhistorisches Museum, Wien, 147-179. KUHNT,W. & WIEDMANN,J. 1995. Cenomanian-Turonian source-rocks: paleobiographic and palaeoenvironmental aspects. In: Huc, A.-Y. (ed.) Paleogeography, Paleoclimate and Source Rocks. American Association of Petroleum Geologists, Studies in Geology, 40, 213-231. KUSMIEREK, I. 1996. Evolution of the Central Carpathian oil basin: a quantitative interpretation. In: ROURE, E, ELLOUZ, N., SHEIK,V. S. & SKVORTSOV, I. (eds) Geodynamic Evolution of Sedimentary Basins. Institut Fran~ais du Pdtrole and Editions Technip, Paris, 281-304. LADWEIN,H.W. 1988. Organic geochemistry of Vienna Basin: model for hydrocarbon generation in overthrust belts. Bulletin of the American Association of Petroleum Geologists, 72, 586-599. LAFARGUE, E., ELLOUZ, N. & ROURE, E 1994. Thrust controlled exploration plays in the Outer Carpathians and their foreland (Poland, Ukraine and Romania). First Break, 12, 69-79. LALLEMANT, S., TRUFFERT, C., JOLIVET, L., HENRY, P., CHAMOT-ROOKE,N. & DE VOOGD,B. 1994. Spatial transition from compression to extension in the Mediterranean Ridge accretionary complex. Tectonophysics, 234, 33-52. LEEDER, M. R. d~cHARDMAN,M. 1990. Carboniferous geology of the Southern North Sea Basin and controls on hydrocarbon prospectivity. In: HARDMAN, R. E R & BROOKS, J. (eds) Tectonic Events Responsible for Britain's Oil and Gas Reserves. Geological Society, London, Special Publications, 55, 87-105. LEINE, L. 1986. Geology of the Tarfaya oil shale deposit, Morocco. Geologie en Mijnbow, 65, 57-74. LE VOT, M., BITEAU, J. J. & MASSET, J. 1996. The Aquitaine Basin: oil and gas production in the foreland of the Pyrenean fold-and-thrust belt: new exploration perspectives. In: ZIEGLER,P. A. ~; HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. Mdmoires du Mus6um National d'Histoire Naturelle, Paris, 170, 159-171. MACGREGOR, D. S. 1996. The hydrocarbon systems of North Africa. Marine and Petroleum Geology, 13, 329-340. MAGOON, L. B. &; DOW,W. G. (eds) 1995. The Petroleum System - From Source to Trap. American Association of Petroleum Geologists, Memoirs, 60, 644p. MAHMOUD, M. D., VASLET,D. t~ HUSSEINI,M. I. 1992.
537
The lower Silurian Qualibah Formation of Saudi Arabia: an important hydrocarbon source rock. Bulletin of the American Association of Petroleum Geologists, 76, 1491-1506. MAKHOUS, M., GALUSHKIN,Y. • LOPATIN, N. 1997. Burial history and kinetik modeling for hydrocarbon generation, Part II: Applying the GALO Model to Saharan Basins. Bulletin of the American Association of Petroleum Geologists, 81, 1679-1699. MAKRIS, J., BEN AVRAHAM,Z., BEHLE, A., GINZBURG, A., GIESE, P., STEINMETZ,L., WHITMARSH,R. B. d(z ELEFTHERIOU, S. 1983. Seismic refraction profiles between Cyprus and Israel and their interpretation. Geophysical Journal of the Royal Astronomical Society, 75, 575-591. MARCOUX, J. dgzBAUD,A. 1996. Late Permian to Late Triassic Tethyan Paleoenvironments. Three snapshots: Late Mugrabian, Late Anisian, Late Norian. In: NAIRN, A. E. M., RICOU, L.-E., VRIELYNK, B. & DERCOURT, J. (eds) The Ocean Basins and Margins, Vol. 8, The Tethys Ocean. Plenum Press, New York, 153-190. MART, Y. 1994. Ptolemafs basin: the tectonic origin of a Senonian marine basin underneath the southeastern Mediterranean Sea. Tectonophysics, 234, 5-17. MASCLE, A., BERTRAND, G. & LAMIRAUX, CH. 1994. Exploration for and production of oil and gas in France: a review of the habitat, present activity and expected development. In: MASCLE,A. (ed.) Hydrocarbon and Petroleum Geology of France. Special Publications of the European Association of Petroleum Geoscientists, 5, 3-27. - - , VIALLY,R., DEVILLE,B., BIJU-DUVAL,B. & ROY, J. P. 1996. The petroleum evaluation of a tectonically complex area: the western margin of the Southeast Basin (France). Marine and Petroleum Geology, 13, 941-961. MASSE, J.-E, PHILIP, J. & CAMOIN,G. 1996. The Cretaceous Tethys. In: NAIRN,A. E. M., RICOU, L.-E., VRIELYNK, B. & DERCOURT, J. (eds) The Ocean Basins and Margins, Vol. 8, The Tethys Ocean. Plenum Press, New York, 215-236. MATTAVELLI, L. & NOVELLI, L. 1990. Geochemistry and habitat of the oils in Italy. Bulletin of the American Association of Petroleum Geologists, 74, 1623-1639. - - , PIERI, M. & GROPPI, G. 1993. Petroleum exploration in Italy: a review. Marine and Petroleum Geology, 10, 410-425. MATTICK,R. E.,TELEKI, R G., PHILLrPS,R. L., CLAYTON, J. L., DAVID, G., POGACSAS, G., BARDOCZ, B. & SIMON,E. 1996. Structure, stratigraphy and petroleum geology of the Little Plain Basin, Northwestern Hungary. Bulletin of the American Association of Petroleum Geologists', 80, 1780-1800. MAUTHE, G., BRINK, H.-J. & BURRI, P. 1993. Kohlenwasserstoffvorkommen und Petroleum potential im deutschen Teil des Oberrheingrabens. Bulletin Vereinigung Schweiz Petroleum Geologie und Ingenieure, 60, 15-29. MAY, R R. 1991. The East Mediterranean Mesozoic Basin: evolution and hydrocarbon habitat.
538
P.A. Z I E G L E R & F. R O U R E
Bulletin of the American Association of Petroleum Geologists, 75, 1215-1232. MCGILIAVARY, J. G. & HUSSEINI, M. L. 1992. The Paleozoic petroleum geology of Central Arabia. Bulletin of the American Association of Petroleum Geologists, 76, 1473-1490. MILICKA, J., PERESZLt~NYI,M., FRANCU,J. & VITALOS, R. 1996. Organic geochemical appraisal of hydrocarbon potential in the Danube Basin, Slovakia. In: WESSELY,G. & LIEBL,W. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. Specical Publication of the European Association of Petroleum Geoscientists, Eng., 5, 431-439. MILOTA, K., KOVACS,A. & GALICZ, ZS. 1995. Petroleum potential of the North Hungarian Oligocene sediments. Petroleum Geoscience, 1, 81-87. MOORE, G.T., BARRON,E. J. & HAYASHIDA,D. N. 1995. Kimmeridgian (Late Jurassic) general lithostratigraphy and source rock quality from the Western Tethys Sea inferred from paleoclimate results using a general circulation model. In: Huc, A.-Y. (ed.) Paleogeography, Paleoclimate, and Source Rocks. American Association of Petroleum Geologists, Studies in Geology, 40,157-172. MOROZ, V., NEAGA,V., POLONIC,P. & ROSCA,V. 1997. The Predobrogean depression in Romania and Moldova and its hydrocarbon prospects. Petroleum Geoscience, 3, 141-144. MUSTAFA, A. R. & KHALIL, M. H. 1990. Structural characteristics and tectonic evolution of North Sinai fold belt. In: SAID, R. (ed.) The Geology of Egypt. A.A. Balkema, Rotterdam, Brookfieid, 381-389. MURRIS,R. J. 1980. Hydrocarbon habitat of the Middle East. In: M1ALL, A. D. (ed.) Facts and Principles of World Petroleum Occurrence. Canadian Society of Petroleum Geologists, Memoirs, 6, 765-800. NAIRN, A. E. M., RICOU, L. E., VRIELYNK,B. & DERCOURT, J. (eds) 1996. Ocean-Basins and Margins, 8, The Tethys Ocean. Plenum Press, New York. NEMCOK,M., KEITH,J. E & NEES,D. G. 1996. Development and hydrocarbon potential of the Central Carpathian Paleogene Basin, West-Carpathians, Slovak Republic. In: ZIEGLER, P. A. & HORVATH, F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 321-342. ONAREP 1995. Petroleum systems of Morocco (a preliminary report). ONAREP, Rabat, Morocco. PARSONS, M. G., ZAGAAR,A. M. & CURRY,J. J. 1980. Hydrocarbon occurrences in the Sirte Basin, Libya. In: A.D. MIALL (ed.) Facts and Principles of World Petroleum Occurrence. Canadian Society of Petroleum Geologists, Memoirs, 6, 723-732. PEGRUM, R. M. & SPENCER,A. M. 1990. Hydrocarbon plays in the northern North Sea. In: BROOKS, J. (ed.) Classical Petroleum Provinces. Geological Society, London, Special Publications, 50, 441-470. PENE, C. 1996. Hydrocarbon generation modelling in
the west Moesian Platform, Romania. Petroleum Geoscience, 2, 241-248. PEREZ-BELZUZ, E, ALONSO, B. & ERCILLA, G. 1997. History of mud diapirism and trigger mechanisms in the Western Alboran Sea. Tectonophysics, 282, 399-422. PHILIP, J., MASSE, J.-E & CAMOIS,G. 1996. Tethyan carbonate platforms. In: NAIRN,A. E. M., RICOU, L.-E., VRIELYNK,B. & DERCOURT, J. (eds) The Ocean Basins and Margins, Vol 8, The Tethys Ocean. Plenum Press, New York, 239-265. PHILIPPE, Y. 1994. Transfer zone in the southern Jura thrust belt (Eastern France): geometry, development and comparison with analogue modelling experiments. In: MASCLE, A. (ed.) Hydrocarbon and Petroleum Geology of France. Special Publications of the European Association of Petroleum Geoscientists, 5, 317-116. PICHA, E J. 1996. Exploring for hydrocarbons under thrust belts - a challenging new frontier in the Carpathians and elsewhere. Bulletin of the American Association of Petroleum Geologists, 80, 1547-1564. PIERI, M. & MATFAVELLI,L. 1986. Geologic framework of Italian petroleum resources. Bulletin of the American Association of Petroleum Geologists, 711, 103-130. POPESCU, B. M. 1995. Romania's petroleum systems and their remaining potential. Petrololeum Geoscience, 1, 337-350. QUENNELL,A. M. 1984. The west Arabian rift system. In: DIXON,J. E. & ROBERTSON,A. H. F. (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 775-788. RICHARD, A. 1994. Petroleum exploration in the French Rhine Graben. In: MASCLE, A. (ed.) Hydrocarbon and Petroleum Geology of France. Special Publications of the European Association of Petroleum Geoscientists, 4, 361-363. Rlcou, L.-E. 1996. The plate tectonic history of the past Tethys Ocean. In: NAIRN, A. E. M., RICOU, L-E., VRIELYNK, B. & DERCOURT, J. (eds.) The Ocean Basins and Margins, Vol. 8, The Tethys Ocean. Plenum Press, New York, 3-70. ROBERTSON, h. H. W.~; SHIPBOARD SCIENTIFICPARTY 1996. Mud volcanism on the Mediterranean Ridge. In: EMEIS, K. C., ROBERTSON, R. F. H. & RICHTER,C. ETAI~. (eds.) Proceedings of the Ocean Drilling Program, Initial Report, 160, 521-526. ROBINSON,A., SPADINI,G., CLOETINGH,S. & RUDAT,Ji 1995. Stratigraphic evolution of the Black Sea: inference from basin modelling. Marine and Petroleum Geology, 12, 821-835. - - - , RUDAT,J. H., BANKS,C. J. & WILES,R. L. F. 1996. Petroleum geology of the Black Sea. Marine and Petroleum Geology, 13, 195-223. ROEDER, O. & BACHMANN, G. H. 1996. Evolution, structure and petroleum geology of the German Molasse Basin. In: ZIEGLER, R A. & HORVATH,F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2, M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 263-284,
M E D I T E R R A N E A N P E T R O L E U M SYSTEMS ROURE, E & COLLETTA,B. 1996. Cenozoic inversion structures in the foreland of the Pyrenees and Alps. In: ZIEGLER, P. A. & HORVATH, F. (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoir, 2 M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 173-209. & SASSl,W. 1995. Kinematics of deformation and petroleum system appraisal in Neogene foreland fold-and-thrust belts. Petroleum Geoscience, 1, 253-269. , ROCCA, E. & SASSI,W. 1992. The Neogene evolution of the Outer Carpathian flysch units (Poland, Ukraine and Romania): kinematics of a foreland fold-and-thrust belt. Sedimentary Geology, 86, 177-203. --, BRUN, J. P., COLLETTA, B. & VIALLY, R. 1994. Multiphase extensional structures, fault reactivation and petroleum plays in the Alpine foreland basin of southeastern France. In: MASCLE, A. (ed.) Hydrocarbon and Petroleum Geology of France. Special Publications of the European Association of Petroleum Geoscientists, 5, 245-268. SADOONI, E N. 1997. Stratigraphy and petroleum prospects of Upper Jurassic carbonates in Iraq. Petroleum Geoscience, 3, 233-243. SAGE, L. & LETOUZEY, J. 1990. Convergence of the African and Eurasian plates in the eastern Mediterranean. In: LETOUZEY,J. (ed.) Petroleum and Tectonics in Mobile Belts. Editions Technip, Paris, 49-68. SAHAGIAN,D. L. 1993. Structural evolution of African basins: stratigraphic synthesis. Basin Research, 5, 41-54. SALAS,R. ~1;CASAS,A. 1993. Mesozoic extensional tectonics, stratigraphy and crustal evolution during the Alpine cycle of the eastern Iberian basin. Tectonophysics, 228, 33-55. , GUIMERA, J., GIMI~NEZ-MONTSANT,J., 1VIART[NCLOSAS, C. & ROCCA, C. 1997. Mesozoic rift structure, stratigraphy and Paleogene inversion of the Maestrat Basin, Iberian Range (NE Spain). IGCP Project 369, Field Trip Guide, 1-3 Sept, 1997. University of Barcelona. SALEL, J. E & SI~GURET,M. 1994. Late Cretaceous to Palaeogene thin-skinned tectonics of the Palmyrides belt (Syria). Tectonophysics, 234, 265-290. SCHLANGER,S. 0., ARTHUR,M. A., JENKYNS,H. C. & SCHOLLE, P. A. 1987. The Cenomanian-Turonian oceanic anoxic event. I. Stratigraphy and distribution of organic-rich beds and the marine g13C excursion. In: BROOKS, J. & FLEET, A. J. (eds) Marine Petroleum Source Rocks. Geological Society, London, Special Publications, 26, 371-399. SEIFERT, P. 1996. Sedimentary-tectonic development and Austrian hydrocarbon potential of the Vienna Basin. In: WESSELY,G. & LIEBL,W. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. Special Publications of the European Association of Petroleum Geoscientists, Eng., 5, 331-341. -
-
539
SOLER, R., LOPEZ-VICHEZ,J. d~;RIAZA, C. 1981. Petroleum Geology of the Bay of Biscay. In: ILLINGS,L. V. & HOBSON, G. D. (eds) Petroleum Geology of the Continental Shelf of North-West Europe. Institute of Petroleum, London, Hayden & Son, London, 474-482. SONATRACH 1995. Exploration in Algeria. SONATRACH Exploration Division, HusseinDey, Algier. SOVCHIK,YA.V. &; VUL, M. A. 1996. New data on structure and hydrocarbon prospects of the Ukrainian Carpathians. In: ZIEGLER, R A. &. HORVATH, F (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. Mdmoires du Musdum National d'Histoire Naturelle, Paris, 170, 391-401. STAMPFLI, G. M. 1996. The Intra-Alpine terrain: a Paleotethyan remnant in the Alpine Variscides. Eclogae geologicae Helvetiae, 89, 13-43. STEFANESCU, M. 0. & BAI~TES, N. 1996. Do hydrocarbon prospects still exist in the East-Carpathian Cretaceous flysch nappes? In: Z1EGLER, R A. & HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. Mdmoires du Mus6um National d'Histoire Naturelle, Paris, 170, 427-438. STONELEY,R. 1990. The Middle East Basin: a summary overview. In: BROOKS, J. (ed.) Classic Petroleum Provinces. Geological Society, London, Special Publications, 50, 293-298. TEN HAVEN,H. L., LAFARGUE,E. & KOTARBA,M. 1993. Oil/oil and oil/source-rock correlation in the Carpathian foredeep and overthrust, southeast Poland. Organic Geochemistry, 20, 935-959. TARI, B. 1996. Meso-Alpine tectonics of the Danube Basin (NW Panninian Basin, Hungary). In: Z1EGLER, P. A. & HORVATH,F. (eds) Structure and Prospects of Alpine Basins and Forelands. PeriTethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 439-454. TORELLI, L., GRASSO, M., MAZZOLDI, G., PEIS, D. & Gore, D, 1995. Cretaceous to Neogene structural evolution of the Lampedusa Shelf (Pelagian Sea, Central Mediterranean). Terra Nova, 7, 200-212. TORYE, M., BANDA E. & FERNANDEZ, M. 1996. The Valencia Trough: geological and geophysical constraints on basin formation models. In: ZIEGLER, R A. & HORVATH,E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 103-128. TORRES, J. & BOIS, CH. 1993. Deep structure and oil potential of the Valencia Trough (western Mediterranean) - contribution of deep seismic reflection profiling. First Break, 11, 87-97. ULMISHEK, G. E 1990. Geological evolution and petroleum resources of the Baltic Basin. In: LEIGHTON, M. W. ET AL. (eds) Interior Cratonic Basins. American Association of Petroleum Geologists, Memoirs, 51, 603-632. - - , BOGINO,W.A., KELLER,M. B. & POZNYAKEVICH, Z. L. 1994. Structure, stratigraphy, and petroleum geology of the Prypiat and Dnieper-Donets
540
P.A. Z I E G L E R & F. R O U R E
basins, Byelarus and Ukraine. In: LANDON, S. M. (ed.) Interior Rift Basins. American Association of Petroleum Geologists, Memoirs, 59, 125-156. VAN BUCHEM, E S. P., DE BOER, E L., MCCAVE, I. N. & HERreN, J.-P. 1995. The organic carbon distribution in Mesozoic marine sediments and the influence of orbital climatic cycles (England and the Western North Atlantic). In: Hue, A.-Y. (ed.) Paleogeography, Paleoclimate and Source Rocks. American Association of Petroleum Geologists, Studies in Geology, 40, 303-335. VAN DE WEERD, A. & WARE, P. L. G. 1994. A review of the East Algerian Sahara oil and gas province (Triassic, Ghadames and Illizi Basins). First Break, 12, 363-373. VIALLY, R. & TRIS~MOLII~RES,P. 1996. Geodynamics of the Gulf of Lions: implications for petroleum exploration. In: ZIEGLER, P. A. & HORVATH, F. (eds), Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 129-158. , LETOUZEY, J., BI~NARD, E, HADDADI, N., DESFORGES, G., ASKRI, H. & BOUDJEMA, A. 1994. Basin inversion along the North African margin: the Saharan Atlas (Algeria). In: ROURE, F. (ed.) Peri- Tethyan Platforms. Editions Technip, 9-117. WAGNER, L. R. 1996. Stratigraphy and hydrocarbons in the Upper Austrian Molasse Foredeep (active margin). In: WESSELY, G. & LIEBL, W. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. Special Publications of the European Association of Petroleum Geolscientists, Eng., 5, 217-235. WESSELY, G. & LIEBL, W. (eds) 1996. Oil and gas in Alpidic thrustbelts and basins of Central and Eastern Europe. European Association of Petroleum Geoscientists, Eng., Special Publications, 5. YILMAZ, P. O., NORTON, I. O., LEARY, O. & CHUCLA, R. J. 1996. Tectonic evolution and paleogeography of Europe. In: ZmGLER, R A. & HORVATH, F. (eds), Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 61-86. ZAPPATERRA,E. 1994. Source-rock distribution model of the Periadriatic region. Bulletin of the American Association of Petroleum Geologists, 78, 333-354. ZmGLER, P. A. 1988. Evolution of the Arctic-North Atlantic and Western Tethys. American Association of Petroleum Geologists, Memoirs, 43. -1989. Evolution of Laurussia. A study in Late Palaeozoic plate tectonics. Kluwer Acad. PUN., Dordrecht.
1990. Geological Atlas of Western and Central Europe, 2nd Edition. Shell Internationale Petroleum Mij. B.V. and Geological Society, London. 1994. Cenozoic rift system of Western and Central Europe: an overview. Geologic en Mijnbow, 73, 99-127. 1996a. Geodynamic processes governing development of rifted basins. In: ROURE, F., ELLOUZ, N., SHHN, V. S. & SKVORTSOV,I. (eds) Geodynamic Evolution of Sedimentary Basins. Institut Franqais du P6trole and Editions Technip, Paris, 19-67. 1996b. Hydrocarbon habitat in rifted basins. In: ROURE, E, ELLOUZ, N., SHEIN, V. S. • SKVORTSOV, I. (eds) Geodynamic Evolution of Sedimentary Basins. Institut Franqais du P6trole and Editions Technip, Paris, 85-94. --& HORVATH, F. (eds) 1996. Structure and Prospects of Alpine Basins and Forelands. PeriTethys Memoirs, 2. M6m., M6moires du Mus6um National d'Histoire Naturelle, Paris, 170. & ROURE, E 1996. Architecture and petroleum systems of the Alpine Orogen and associated basins. In: ZIEGLER, P. A. & HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoir, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 15-45. - - - , CLOETINGH,S. & VAN WEES, J.-D. 1995. Dynamics of intra-plate compressional deformation: the Alpine foreland and other examples. Tectonophysics, 252, 7-59. - - - , VAN WEES, J.-D. & CLOETINGH, S. 1998. Mechanical control on collision-related compressional intraplate deformation. Tectonophysics, in press. ZIMMER, W. & WESSELV, G. 1996a. Hydrocarbon exploration in the Austrian Alps. In: ZIEGLER, P. A. & HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 285-304. & -1996b. Exploration results in thrust and subthrust complexes in the Alps and below the Vienna Basin. In: WESSELu G. & LIEBL,W. (eds) Oil and Gas in Alpidic Thrustbelts and Basins of Central and Eastern Europe. Special Publications of the American Association of Petroleum Geologists, Eng., 5, 81-107. Zlzl, M. 1996. Triassic-Jurassic extension and Alpine inversion in Northern Morocco. In: ZIEGLER, P. A. & HORVATH, E (eds) Structure and Prospects of Alpine Basins and Forelands. Peri-Tethys Memoirs, 2. M6moires du Mus6um National d'Histoire Naturelle, Paris, 170, 87-101. --
-
-
-
-
-
-
On the origin of west-directed subduction zones and applications to the western Mediterranean C. D O G L I O N I ,
1 E. GUEGUEN,
2 P. H A R A B A G L I A
2 & F. M O N G E L L I
3
1Dipartimento di Scienze della Terra, Universitd La Sapienza, 00185 Roma, Italy 2Centro di Geodinamica, Universitd della Basilicata, 85100 Potenza, Italy 3Dipartimento di Geologia e Geofisica, Universitd di Bari, 70125 Bari, Baly Abstract: West-directed subduction zones show east-verging arcs of 1500-3000 km. They are usually younger than 50 Ma and are characterized by a frontal accretionary wedge and a back-arc basin propagating together toward the east. The accretionary wedge scrapes off superficial layers of the downgoing plate (thin-skinned tectonics) whereas the back-arc extension cross-cuts the entire subduction hanging wall (thick-skinned tectonics). The slab of this type of subduction is steep to vertical and the hanging wall of the subduction has a mean elevation of 1000 m below sea level. Trenches and foredeeps are the deepest basins of the Earth and the mean depth is of 5000 m below sea level. West-directed subduction occurs both in case of the highest E - W convergence rates among plates (e.g. W Pacific examples) and no or very low convergence (e.g. Carpathians). Following Atlantic Wdirected subduction examples, the W-directed subductions seem to develop along the backthrust belt of former E-directed subduction zones, where oceanic lithosphere occur in the foreland to the east with the narrowing of the American continents. This could be applied to the onset of the Apennines subduction along the back-thrust belt of the Alpine-Betic orogen where Tethys oceanic crust was present. The Alpine orogen was stretched and scattered in the Apennines back-arc basin. The back-arc extension is internally punctuated by necks (sub-basins) and boudins (horsts of continental lithosphere). Asymmetric extension in the back-arc basin appears controlled by differential drag between the eastward mantle flow and the overlying passively transported crustal remnants. Compression in the accretionary prism may be interpreted as the superficial expression of the shear occurring between the downgoing lithosphere and the horizontally moving mantle which compensates the slab roll-back. The area of the Apennines appears lower than the area of the sedimentary cover before subduction: this favours the idea that not significant crustal slices have been involved in the Apenninic accretionary prism, and the basement thrust sheets included in the western part of the belt are mainly relicts of the Alpine-Betic orogen.
This p a p e r aims to u n d e r l i n e the peculiar characters of W - d i r e c t e d s u b d u c t i o n z o n e s w h i c h present strong differences particularly when c o m p a r e d to the o t h e r settings such as the classic A l p i n e or A n d e a n s u b d u c t i o n zones. T h e m a i n k n o w n , p r e s e n t l y active or p r e s e r v e d Wd i r e c t e d s u b d u c t i o n zones of the w o r l d are the A p e n n i n e s , Carpathians, Barbados, Sandwich, Aleutians, Kurile, Japan; Nankai, Ryukyu, I z u - B o n i n , Marianas, Tonga, K e r m a d e c , B a n d a , P h i l i p p i n e s . We p r e f e r to use t h e t e r m Wd i r e c t e d r a t h e r than W - d i p p i n g b e c a u s e t h e west is the m e a n direction of an arcuate slab w h i c h m a y dip for e x a m p l e to the southwest ( N o r t h e r n A p e n n i n e s ) , or to the n o r t h w e s t (Calabria), or e v e n to the n o r t h (Sicily and M a g h r e b i d e s ) . T h e A l e u t i a n slab dips n o r t h w e s t w a r d , b u t the Pacific plate travels W N W oblique to the trench. W e s t - d i r e c t e d s u b d u c t i o n zones are h e r e differe n t i a t e d b e c a u s e t h e y p r e s e n t a n u m b e r of peculiarities such low elevation in the h a n g i n g wall, b a c k - a r c spreading to t h e west, h i g h - a m p l i t u d e
gravimetric signatures, short life, etc. M o r e o v e r their a c c r e t i o n a r y prisms have a d o m i n a n t vergence to the east a n d t h e y are followed by a w a v e of e x t e n s i o n a l faults. T h e a c c r e t i o n a r y prisms are m a i n l y c o m p o s e d of s e d i m e n t a r y rocks off s c r a p e d f r o m the top of the f o r e l a n d subducting plate and b a s e m e n t relicts of f o r m e r o r o g e n s r e l a t e d to E- or N E - d i r e c t e d s u b d u c t i o n z o n e s ( D o g l i o n i 1991, 1992). T h e w e s t e r n M e d i t e r r a n e a n system is used h e r e as the m a i n e x a m p l e for the description of the m a i n traits of the W - d i r e c t e d s u b d u c t i o n zones.
Main characters of west-directed subduction z o n e s West-directed s u b d u c t i o n zones have generally arcuate shape, with the convexity verging m a i n l y t o w a r d the east (Fig. 1). It was p r o p o s e d that any s u b d u c t i o n on a s p h e r e g e n e r a t e s an arc (e.g. F o w l e r 1990). H o w e v e r this g e o m e t r i c observation does not explain w h y E- or N E - d i r e c t e d
DOGLIONI,C., GUEGUEN,E., HARABAGLIA,P. • MONGELLI,E 1999. On the origin of west-directed subduction zones and applications to the western Mediterranean. In: DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE, M. (eds) The MediterraneanBasins: TertiaryExtension within the Alpine Orogen. Oeological Society, London, Special Publications, 156, 541-561.
542
C. DOGLIONI E T A L .
steep to vertical slab (both for oceanic and continental lithosphere)
fast roll-back of the subduction hinge (>3-10 cm/a)
arcuate shape (arc length 2000-3000 km)
arc migrating eastward (along-strike extension)
younger than 50 Ma (usually less than 30 Ma)
always back-arc basin (migrating in the same direction)
lithospheric boudinage in the back-arc basin (asymmetric extension, mainly E-dipping normal faults)
western margin of the back-arc basin with convexity opposed to the the subduction arc
fast subsidence rates (foredeep or trench 1 m m / a ; back-arc basin 0.6 m m / a )
low elevation of the subduction hanging wall (average 1000 m below sea-level)
deep trenches (average 5000 m below sea-level)
thin crust in the hangingwall of the subduction (20-30 km thick, lying on a new 'Moho' and a new mantle section)
magmatic pair (sourced by the slab and the back-arc asthenosphere)
thin-skinned compressive tectonics at the front and thick-skinned extensional tectonics in the back-arc
Fig. 1. Summary of the main characteristics of west-directed subduction zones. Plate boundaries in the upper figure are after Fowler (1990).
WEST-DIRECTED SUBDUCTION ZONES
543
Fig. 2. The western Mediterranean is composed by sub-basins that show an age progression from west to east. They developed in the hanging wall of the eastward retreating Apenninic subduction as back-arc basins. The sub-basins show triangular shapes. A shows the location of the section shown in Fig. 3. Ages with the asterisk indicate the age of the oceanic crust (after Gueguen et al. 1997).
subduction zones do not show such a shape (e.g. Cordillera, Himalayas). The arc of the Wdirected subduction zones has average length of 1500-3000 km both in the Pacific and in the Mediterranean (Fig. 2). West-directed subductions occur both in case of the highest convergence rates among plates (e.g. W Pacific examples, with rates even higher than 10 cm a -1) and no or very low rates of convergence (e.g. W Atlantic examples, A p e n n i n e s , Carpathians, with rates of 0-2 cm a-l). There are W-directed subductions where the slab retreats eastward without any convergence along margins of plates travelling independently toward the NE (e.g. the B a n d a arc at the n o r t h w e s t e r n Australian margin, and the Apenninic arc at the western margin of the Adriatic-African plate, Fig. 3). Thrust belts associated with W-directed subductions have a shallow new Moho beneath the thrust belt with respect to the deeper and pre-subduction Moho of the foreland (Fig. 1). The recent age of this 'new' Moho is constrained by the age of the back-arc spreading and by the kinematics of the subducting foreland presubduction Moho. In fact, the so-called 'new' Moho below the belt shows lower velocities (7.7-8.0 km s) than the foreland Moho (7.9-8.2
km s). Compare the Apennine data with those of the Alps where notoriously the crust is thicker (Locardi & Nicolich 1988; Kissling 1993; Scarascia e t al. 1994). M o r e o v e r the accretionary wedges have only one main vergence (generally speaking eastward) and only one foredeep, in contrast with Alpine or A n d e a n orogens which are characterized by double vergence and two foredeeps or trenches. Double vergence of such belts are only related to external geodynamic settings overprinting the W-directed subduction, like the N-S compression of the Southern A m e r i c a plate due to its clockwise rotation which is deforming the southern arm of the Barbados subduction zone and the Caribbean backarc. The accretionary wedges of W-directed subductions are mainly composed of sedimentary rocks scraped off the top of the subducting plate; basement rocks occur as pre-existing structural highs of the foreland truncated by the advancing thrusts or inherited in the hanging wall from earlier orogenic belts. In some sections of such belts the main crystalline basement and crust may be missing due to subduction. These peculiarities could be explained by the behaviour of the d6collement zones of the W-directed subduction
544
C. D O G L I O N I E T A L .
Fig. 3. Kinematic model of the Apennines subduction using an area balancing method. The back-arc extension in the hanging wall of the Apenninic subduction is characterized by lithospheric necking progressively moving to the east. It opened first in the Provenqal basin, then it j u m p e d to the east of Sardinia and developed the Tyrrhenian extension with the Vavilov and Marsili sub-basins. This appears to indicate a discontinuous process of extension in the hanging wall of the W-directed subduction where large slices of lithosphere are boudinated and dragged eastward, including the Alpine orogen. While the back-arc extension appears punctuated, the subduction roll-back seems to be a more continuous process. A b o u t 800 km of subduction roll-back are calculated in the area of maximum subduction and arcuature of the Apennines. The eastward retreat of the slab implies an eastward relative motion of the mantle to compensate the slab retreat at the west, and to allow the slab retreat to the east. Note the relative motions among the black reference points. Location of the section on Fig. 2. Viscosity values after B. Romanowicz and R. Sabadini (pets. comm.).
WEST-DIRECTED SUBDUCTION ZONES zones which are polarized by the westward drift of the lithosphere relative to the asthenosphere (Ricard et al. 1991). In fact the basal ddcollement of the downgoing lithosphere is warped and subducted along W-directed subduction zones (Doglioni 1992). Without the possibility of uplifting deep-seated crustal rocks of the foreland, only shallow upper layers of the lithosphere are accreted (Fig. 1). The accretionary wedge is accompanied by right-lateral transpression and clockwise rotations of the thrust sheets in the southern arm of the arc associated to a W-directed subduction zone, whereas the northern arm is characterized by diffuse left-lateral transpression and counterclockwise rotation of the thrust sheets (Fig. 1). Opposed versus of transtension is recorded in the extensional area to the west: right-lateral transtension in the northern arm of the back-arc basin and left lateral transtension in the southern arm of the arc. No significant rotations are described in the back-arc transtensional areas. This may be ascribed also to the interpretation of the depth of the d6collements: the accretionary wedge is characterized more by thinskinned thrust-tectonics where the lithostatic load is low. On the other hand the extensional belt to the west is cross-cut by thick-skinned normal faulting cross-cutting the entire crust; this could inhibit large rotations of the hanging wall due to the higher lithostatic load operating on the d6collement planes of the lower crust. Foredeeps and trenches are also very pronounced along W-directed subduction zones. They show the highest subsidence rates of any basin on Earth in the order of 1600 m Ma -1 (e.g. Apennines and Carpathians). This observation enables us to interpret the slow filling of foredeeps with huge flysch deposits (Apennines) or poor deposition (Marianas trough). In fact the cross-sectional area of thrust belts associated with W-directed subduction is smaller in comparison with the area of the foredeep or trench. All these observations appear to maintain their general validity both in the case of subduction of oceanic lithosphere and of thin continental lithosphere. Along the W-directed subduction zones the origin of the foredeep and trench appears to be mainly controlled by the eastward roll-back of the subduction hinge resulting from the eastward mantle push (Doglioni 1994). The subsidence is so fast in those foredeeps that the fold growth along the frontal part of the accretionary wedge can be negative when the regional subsidence exceeded the single fold uplift rate. In these cases, the envelope to the folds crest may dip toward the hinterland (Doglioni & Prosser 1997). Associated with a W-directed subduction there always occurs a back-arc basin, with fast
545
eastward propagation (30-50 mm a-1, e.g. the Tyrrhenian Sea). The subducted lithosphere is replaced by new asthenospheric material in the back-arc. The main geometric and kinematic characters of the W-directed subduction zones are summarized in Fig. 1. The thickness and shape of the accretionary wedge along the trench or foredeep is controlled by the depth of the ddcollement, which is also a function of the thickness and rheology of the sedimentary cover arriving at the subduction hinge. Thin or absent sedimentary cover on top of the oceanic crust will inhibit the development of appreciable accretionary wedge like in the Marianas trough where the basal d6collement is almost at the earth surface. The calcalkaline to shoshonitic and alkaline to tholeiitic magmatic suite pairs mark the W-directed subduction settings as indicators of the subduction plane and the back-arc asthenospheric spreading sources. Most of these peculiarities have been so far explained in terms of slab pull and age of the subducting oceanic lithosphere, or rates of convergence between plates (e.g. Royden & Burchfiel 1989). However, in the Mediterranean, the Adriatic continental lithosphere and the Ionian oceanic lithosphere (de Voogd et al. 1992) are subducting both under the Apennines (steep Wdirected subduction, Selvaggi & Chiarabba 1995) and under the Dinarides-Hellenides (shallow NE-directed subduction, Christova & Nikolova 1993). The two related thrust belts follow the east and west Pacific rules, without age and thickness variations of the subducting lithosphere. In the Pacific itself, the W-directed subductions are the fastest in the world and the slab is steep, while the Andean subduction is active since the Mesozoic and the slab is shallow. Those examples are clearly in contradiction with an age of the subduction, and the age and thickness of the subducting lithosphere as first order controls of the tectonic style. The westward drift of the lithosphere relative to the mantle could better explain these asymmetries. Focal mechanisms of intermediate and deep events are an important tool in understanding subduction processes and in delineating slab geometry. In particular W-directed slabs have dip angles ranging from 40 ~ of the fastest Western Pacific slabs to verticality of the blocked Carpathians slab. The Pacific slabs are characterized by active convergence (4-10 cm a-j) as well as slab retreat, conversely the Atlantic and European slabs have slab retreat as dominant mechanism. The difference in dip of the two groups is probably due to the fact that the induced flow (e.g. Turcotte & Schubert 1982) in the first one is much higher than in the second one causing an uplift of the slab. An analysis of deep earthquakes shows us that their
546
C. DOGLIONI E T A L .
activity i n the Atlantic and European slabs is much less prominent that in the Pacific slabs. In particular the Sandwich slab has recorded events up to a depth of about 140 km, the Barbados down to about 170 kin, the Carpathians down to 180 km (e.g. Oncescu 1984, 1987), and the Tyrrhenian slab down to 450 km (e.g. Frepoli et al. 1996). Conversely all the Pacific slabs show deep earthquake activity almost to the 670 km discontinuity. Moreover in the Atlantic and European slabs there is a prevalence of compressional events; within the Pacific slabs we observe an alternation of compressional and tensional mechanisms, combined with horizontal shear flow. On the base of this observation, Frepoli et al. (1996) question the role of the slab pull for the Tyrrhenian slab. In fact slowly subducting slabs reach a state of thermal equilibrium at shallower depth; the density difference between the slabs and the surrounding mantle becomes less relevant and might be the reason because the slabs apparently do not sink. In the Apennines the seismicity appears located along the northwestward palaeogeographic prolongation of the Ionian Mesozoic oceanic lithosphere; the shortening in the Apenninic accretionary wedge is maximum in northern Calabria and southern Apennines and decreases along the opposite arms of the Apenninic arc. However the shortening which is appreciable in the accretionary prism suggests that subduction has occurred all along the Apenninic arc. Moreover the Mesozoic shallow-water facies piled up in the belt indicate that they were lying on a continental crust at least 25 km thick. The missing crust in the Apennines and the shortening in the belt support the subduction of those volumes of continental lithosphere. The latter has lower temperature of brittle-ductile transition and therefore the paucity of deep seismicity along northern segments of the Apenninic arc could be attributed to the different rheology of the quartz-feldspar rich Adriatic continental lithosphere with respect to that of the olivine-pyroxene rich Mesozoic oceanic Ionian sea. These differences are supported also by the magmatism that shows clearly different sources (Peccerillo 1985; Serri et al. 1993). Recent tomographic investigations of the Apenninic slab show a much more continuous cold body underneath the Apennines than so far imaged (Amato et al. 1996). The map view and the section of an arc associated to a W-directed subduction zone shows how the stress trajectories follow the arc and compression in the accretionary wedge can occur without relative motion between the foreland
and the hinterland plates (Fig. 1). Therefore the seismicity associated to the offscraping of the sedimentary cover due to slab retreat along the arc of W-directed subduction zones is not a reliable indicator of relative plate motion among footwall and hanging wall plates (e.g. Barbados or Apennines arcs). In this frame, the Africa-Europe relative motion that has been interpreted on the basis of the northern Africa seismicity (e.g. McKenzie 1972), can also be interpreted as due to the migration of the subduction hinge in a context of no or low convergence. The observed compression-extension wave is generated by the 'eastward' roll-back of the subducting Adriatic-Ionian-African lithosphere and only marginally deformed by the relative Africa-Europe convergence. It is commonly believed that the extension determining the opening of the western Mediterranean developed in a context of relative convergence between Africa and Europe. However the direction of relative motion is still under debate. Most of the reconstructions (Albarello et al. 1995; Dewey et al. 1989; Mazzoli & Helman 1994; Campan 1995) show an amount of shortening of 150 km during the all Tertiary in the western Alboran area. The main difference between these models is the increase of convergence further east ranging from 300 km (Dewey et al. 1989) to 250 km (Campan 1995) in Tunisia. The amount of shortening in the two areas diminishes when computed for the last 20 Ma to 70-80 km in the Alboran and 100-165 km in Tunisia (Dewey et al. 1989; Campan 1995). It appears that the amount of relative N-S Africa Europe relative motion was in any case five to eight times slower with respect to the eastward migration of the Apenninic arc which migrated eastward about 800 km during the last 23 Ma (Fig. 3), i.e. 4-7 mm a-1 v. 30-50 mm a -1. (Patacca & Scandone 1989; Doglioni 1991; Gueguen et al. 1997). Recent geodetic data (Lageos, VLBI, GPS) confirm this main frame (Smith et al. 1994; Ward 1994). Therefore the Apenninic arc migrated eastward faster than the N-S convergence related to the counterclockwise rotation of Africa relative to Europe. A similar setting may be observed in the Caribbean region (Westbrook & McCann 1986; Mascle et al. 1986) where the arc due to the Barbados subduction zone migrated eastward faster than the N-S convergence of the South America with the Caribbean back-arc basin owing to the clockwise rotation of the South America plate. Geologically, W-directed subductions are less known because they are mainly below sea level. Average low topography and pronounced
WEST-DIRECTED SUBDUCTION ZONES gravimetric anomalies characterize them. A narrow arc-chain in the hanging wall of the subduction (about 200-300 km in section) typically rises to 2000-3000 m over the mean plate height. The subducting plate is generally only 1000-2000 m lower than the overriding one. If the subducting plate is oceanic, there is always a p r o n o u n c e d trench, whereas for continental plates, the trench may be filled (e.g., Carpathians) or partly filled (Banda) by sediments. The associated volcanic arc is usually well defined. West-directed subductions are characterized by
547
strong negative free-air gravimetric anomalies with an asymmetric shape (150-200 mgal) along the trench, by a prominent positive signature (over 100 mgal) corresponding to the arc-chain, and similar gravimetric values on both plates immediately off the t r e n c h - a r c system. The minimum gravimetric values are located along the trench or foredeep, displaced from the mountain range in the hanging wall (e.g. the Apennines, Mongelli et aL 1975). For further details and comparison with E-directed subductions see Harabaglia & Doglioni (1998).
Fig. 4. Note that Barbados and Sandwich arcs formed only where the American continents narrow. Their Wdirected subduction zones evolved only where there was Atlantic oceanic lithosphere in the foreland of the back-thrust belts of the E-directed subduction zones of the northern and southern Cordillera. Lines i and 2 refer to Fig. 5.
548
C. DOGLIONI E T A L .
Initiation of west-directed s u b d u c t i o n zones West-directed subduction zones appear to nucleate along back-thrust belts of former Edirected subductions, if oceanic or thinned continental lithosphere was present in the foreland of the E-verging back-thrust belt (Fig. 4). This seems the indication coming from the Atlantic W-directed subduction zones. Those two subductions, i.e. Barbados and New Scotia arcs, developed along the front of the E-verging cordillera back-thrust belt (e.g. R o c k y Mountains, Sub-Andean thrust belt) only where there was Atlantic oceanic lithosphere in the foreland to the east of the back-thrust belt. The W-directed subductions did not develop where to the east of the back-thrust belts of the Edirected subduction there was thick continental lithosphere like in the Western Interior of North America or in Brazil and Argentina in the South America continent. Therefore W-directed subductions developed only in central America and south of Patagonia, where the Northern and Southern American continents narrow (Fig. 4). The two subduction zones developed during
Tertiary and Quaternary times, and they have arcuate shapes and a length of about 2000 km. In order to explain the W-directed Atlantic subduction zones, Russo & Silver (1994) proposed an eastward mantle flow in the back-arc of the Barbados and Sandwich arcs, laterally deviated from the A n d e a n subduction zone. F r o m the a f o r e m e n t i o n e d indications, it appears that W-directed subductions, which are world-wide Tertiary and Quaternary features, possibly form only in the presence of particular geodynamic constraints, i.e. (1) along the backthrust belt of earlier E-directed subduction zones, and (2) in the presence of oceanic or thinned continental lithosphere in the foreland of the related back-thrust belt (Fig. 5). In this paper we propose that similar geodynamic constraints favoured the N e o g e n e Q u a t e r n a r y A p e n n i n e s development, i.e. the A p e n n i n e s W-directed subduction formed where Mesozoic oceanic Tethys lithosphere (Bernoulli & Lemoine 1980) was present in the foreland of the back-thrust belt of the pre-existing E-directed subduction-related Alpine orogen (Figs 6 & 7). Remnants of the former Alpine orogen were passively incorporated into
1 oceanic lithosphere W
~
2 thick continental lithosphere 1
................
E
W
g
2[11)km
FRONTAL THRUST BELT BACK-THRUST BELT 1
"E"-directed subduction rt~oNrAi THI,~LJS[HEI,I BACK-[ ItI,~US] BI:I1
old oceanic lithosphere
"Eastward" relative asthenospheric flow BACK-ARC BASIN stretching the orogen active frontal thrust-belt t
] 1
"W"-directcd subduction ,/ /
/~
Fig. 5. The E-directed subduction forms where there is convergence between two plates and oceanic or thin continental lithosphere is in the footwall of the subduction (1, e.g. Andes). The W-directed subduction develops along the back-thrust belt of the former E-directed subduction-related orogen, where the continent narrows and oceanic lithosphere is present in the foreland of the orogen to the east (2, e.g. Barbados). The back-arc extension of the W-directed subduction stretches the internal parts and the back-thrust belt of the pre-existing orogen (e.g. early stages of the Central Americas and Barbados opposite subduction zones which are scattered in Cube, Haiti and Caribbean islands). 1 and 2 refer schematically to Fig. 4.
WEST-DIRECTED SUBDUCTION ZONES
549
Geodynamic setting of the western Mediterranean and back-arc settings The western Mediterranean basin (Auzende et al. 1973; Boccaletti & Guazzone 1974; Scandone 1980; R6hault et aL 1984; Stanley & Wezel 1985; Malinverno & Ryan 1986; Kastens et al. 1988;
Fig. 6. Palaeogeographic reconstruction of the Western Mediterranean since the Late Oligocene, and application of the Atlantic example of Fig. 4 for the initiation of the Apenninic subduction zone. The W-directed subduction might have developed along the back-thrust belt of the Aipine-Betic orogen where a pre-existing Mesozoic oceanic relict of the Tethys should have been present. The back-arc basin of the Apenninic subduction stretched and scattered into the segmented basins the Alpine-Betic orogen. the internal parts of the accretionary wedge, boudinaged and cross-cut by the back-arc extension related to the younger W-directed subduction (Figs 8 & 9). A similar geodynamic evolution has been proposed for the W-directed Japanese subduction, which apparently formed during the Neogene to the east of an earlier Andean type orogen related to an E-directed subduction (Sillitoe 1977; Cadet & Charvet 1983).
Robertson & Grasso 1995) consists of a series of V-shaped sub-basins (Fig. 2) which developed from late Oligocene on in the context of backarc extension contemporaneous to the eastward roll-back of the westerly directed Apenninic subduction zone (Gueguen et al. 1997). Figure 3 shows jumps in the thinning process and rejuvenation of the basins toward the east or southeast. The jumps in the thinning process are responsible for a boudinage of the crust and the lithosphere which show large lateral thickness variations (Calcagnile & Panza 1981; Banda & Santanach 1992; Blundell et al. 1992; Torn6 et al. 1992; Scarascia et al. 1994; Ferntandez et al. 1995). Therefore the western Mediterranean has to be viewed as a whole basin dismembered into subbasins by the heterogeneous stretching of the back-arc setting. The isolation of boudins indicates a discontinuous process in the back-arc area. The large scale lithospheric boudinage shows a range of 100-400 km wavelength. The Sardinia-Corsica continental block represents the largest lithospheric boudin of the western Mediterranean. The boudinage arrived to complete thinning of the continental lithosphere with formation of new oceanic crust in the Provenqal, Algerian, Vavilov and Marsili basins. This is also supported by the back-arc magmatism that does not show a regular spatial and temporal evolution. It rather shows moments of larger manifestations located in the western side of the boudins (e.g. the Early Miocene western Sardinia and Valencia trough magmatism). The N e o g e n e - Q u a t e r n a r y magmatism of the western Mediterranean back-arc setting regularly shows paired calc-alkaline and alkaline suites, the first usually being older in every single magmatic province (e.g. Mart/et al. 1992). The back-arc extension and the related magmatism are younger moving eastward (e.g. in the Tyrrhenian, Savelli 1984). Gravity profiles across the central western Mediterranean (Cella et al. 1998) also support the evidence of an asymmetric boudinage of the lithosphere. The evidences for the Apenninic subduction have been extensively documented in the last decades by geophysical, geological and volcanological data that indicate W-directed subduction of the Adriatic plate underneath central and southern Italy (Mongelli et aL 1975; Peccerillo 1985; Royden et al. 1987; Channell &
550
C. DOGLIONI E T A L .
Fig. 7. W-directed Apenninic subduction started in the western Mediterranean in the Late Oligocene along the back-thrust belt of the Alpine-Betic orogen. Its arc migrated eastward up to the present position due to the eastward slab retreat. The western Mediterranean back-arc basins developed in the hangingwall of this retreating slab. Note the lengthening of the arc which should generate along arc extension. The back-arc initiation cross-cuts the Alpine-Betic front (Doglioni et al. 1997), suggesting that the extension is not simply the collapse of the Alpine-Betic orogen.
Mareschal 1989; Beccaluva et al. 1989; Spakman 1989; Doglioni 1991; Amato et al. 1993; Serri et al. 1993; Selvaggi & Chiarabba 1995; Faccenna et al. 1996). The subduction retreated 'eastward' and the associated back-arc basins have ages progressively younger (30 Ma to present) moving from west (Valencia, Provenqal, Alboran and Algerian basins), to east (Tyrrhenian sea) (Fig. 2). Roll-back of the slab likely caused loss of lithosphere which should have been replaced by asthenosphere (Fig. 3) responsible for the high heat flow values (>100 mW m -2) measured in the western Mediterranean. The W-directed Apenninic subduction started to the east of the pre-existing Cretaceous to Miocene Alpine-Betic orogen. The back-arc extension in the hanging wall of the Apenninic subduction stretched and deformed the inherited Alpine-Betic orogen related to an 'E'directed subduction. Remnants of the Alpine-Betic thrust belt have been dispersed throughout the western Mediterranean (Alvarez et al. 1974), both in the basins and in the swells. Relicts of metamorphic rocks emplaced by Alpine thrusts have been dragged in the Tyrrhenian (Kastens et al. 1988) and are
scattered around the back-arc basins (e.g. the Kabylie in northern Africa and the Aspromonte and Peloritani in southern Italy). The subduction underneath the Apennines consumed inherited Tethyan domains (Dercourt et al. 1986; Ziegler 1988). Similarly, boudinage of the preexisting Alpine-Dinaric orogens occurred in the Pannonian basin, which is the MiocenePliocene back-arc basin related to the coeval Wdirected Carpathian subduction zone which retreated during the Miocene and Pliocene (Oncescu 1984; Royden & Horvfith 1988; Horv~th 1993; Tomek 1993; Tomek & Hall 1993; Linzer 1996). In the Pannonian basin the extension isolated boudins of continental lithosphere thickened by the earlier Dinaric orogen, like the Apuseni mountains which separate the Pannonian basin s.s. from the Transylvanian basin to the east. The western Mediterranean back-arc setting is comparable with Atlantic and western Pacific back-arc basins that show similar large scale lithospheric boudinage, where parts of earlier orogens have been scattered in the backarc area, like the central America Cordillera relicts that are dispersed in the Caribbean domain (Donnelly 1989). Back-arc basins are typical features forming in
WEST-DIRECTED S U B D U C T I O N ZONES
551
.,..~
O
o
9 e:~t) O .,..~
< . r.~ ,...~
O
O . ,...~
r162
O
9
O
~"'~
9
e4~
.,,.~ ~ o'~
~<
552
C. DOGLIONI E T A L .
"11
thrust b e l t
2
k
E
Eocene- Oligocene Dinaric subduction ~-~
A
Alpine subduction
H ' F{v / f
~ ~\,--
~ ~ . . . .
_~
ol
__
~-Proven al basin
onset along the backthrust belt of the Alps
~
..
enmmc
LateOligocene-Earl>,Miocene
r
~-...
subduction
B
LateMiocene - Pliocene su uction h in ge rolib - ack
H.F.
T~
//f--~\
100"
60 F 0
- - ~ - Corsica
Pleistocene
Apennines
Dinarides
Adriatic plate j "
boudinage of the Alpine orogen
'J'
. . asthenosphere
/[
/ [
Fig. 9. Kinematic reconstruction of the gross evolution of the Northern Apennines. The back-thrust belt of the Alpine orogen is interpreted as the seat for the development of the west-directed Apenninic subduction. The Alpine belt was progressively boudinated and deformed by the back-arc extension of the Apenninic subduction. H. F. is the speculated (A-B-C) and observed (D) heat flow evolution across the area at the different times (after Doglioni et al. 1998). the hanging wall of subduction zones. They are notably well developed in the western Pacific (Karig & Sharman 1975; Zonenshain & Savostin 1981; Honza 1995), and they are particularly associated with west-directed subduction zones (Uyeda & Kanamori 1979; Doglioni 1991). The
back-arc basins have shape, subsidence rates and timing which strongly differ from linear, Atlantic type, rift zones. They are characterized by semicircular or triangular shape (Fig. 1), the highest subsidence rates for extensional environments (up to 600 m Ma-1), and they have
WEST-DIRECTED SUBDUCTION ZONES ages ranging mainly from early Tertiary to Recent (Doglioni 1995; Honza 1995). Westdirected subduction zones are always associated with a back-arc basin in the hanging wall of the subduction to the west. The extension in the back-arc propagates eastward like the roll-back of the subduction zone. The basins are often floored of oceanic crust which is also rejuvenating toward the east (e.g. Parece Vela basin, Caribbean sea and western Mediterranean, Westbrook & McCann 1986; Honza 1995). The western margin of the back-arc basin has a convexity opposed to that of the subduction arc, i.e. toward the west (Fig. 1). In other words the shape of the back-arc in some way mirrors the one of the main arc. The re-entrance of the western margin of the back-arc basin is particularly visible on the Asian margins of the Japan sea and the South China sea, and the Cordillera margin of the Caribbean sea. The westernmost initiation of the back-arc extension probably corresponds to the latitude of the location of the first onset of the W-directed subduction to the east and in the end to the largest amount of subducted slab and the widest back-arc extension. Back-arc basins have maximum width variable between 800 and 1500 km. The maximum and minimum widths of the back-arc basin correlate to the west at the maximum and minimum amount of subduction present to the east. For instance the largest opening of the Tyrrhenian sea corresponds to the deepest part of the Apennines slab. During back-arc spreading, blocks moved eastward and rotated both clockwise (southern arm) and counterclockwise (northern arm). See the well described counterclockwise rotation of the Sardinia-Corsica continental block (Montigny et al. 1981; Vigliotti & Kent 1990), and the clockwise rotation of the Balearic promontory (Par6s et al. 1992). Moreover the increasing length of the arc (Fig. 7) should be responsible for an extensional stress parallel to the arc direction (Doglioni 1991). This should have generated along strike boudinage of the lithosphere like for instance in the Balearic promontory which is stretched both in N W - S E direction (main back-arc boudinage) and along the N E - S W direction (boudinage along the strike of the arc). The southern back-arc settings of the western Mediterranean are characterized by diffuse left-lateral transtension, whereas the associated North African accretionary wedge developed in a context of right-lateral transpression. On the other side, the Provenqal and northern Tyrrhenian sea were controlled by a diffuse right-lateral transtension while the frontal Apenninic accretionary prism formed in a
553
regime of left-lateral transpression. These tectonic settings regularly occur in W-directed subduction systems (Fig. 1). There is more and more evidence that the eastward migration of the back-arc extension is accommodated by asymmetric rifting, with lowangle normal faults dipping to the east. These faults have been recognized in the western Mediterranean back-arc setting, e.g. in the western margin of the Provencal basin (Benedicto et al. 1996), along the eastern margin of Corsica-Sardinia (Jolivet et al. 1990), in the northern Apennines (Barchi et al. 1997) and in other western Pacific back-arc basins. Eastdipping normal faults are usually spaced (10-50 km) and sometimes they isolate large boudins of thicker continental crust generating a lithospheric boudinage both in the Pacific, Atlantic and Mediterranean back-arc basins (Daniel et al. 1996; Gueguen et al. 1997). Back-arc spreading associated with the Wdirected subductions may develop both within the former orogen or even far into the foreland of the frontal thrust-belt of the earlier Edirected subduction zone, probably as a function of the width of the orogen (e.g. Japan Sea, Valencia trough). Part or the entire orogen of the former 'E'-directed subduction zone is stretched and boudinated in the hanging wall of the W-directed subduction, in the back-arc region. H T - L P metamorphism associated with asthenospheric wedging in the back-arc underneath the former orogen commonly overprints HP-LT metamorphic assemblages (Jolivet 1993, and references therein). The western Mediterranean back-arc basins associated with the Wdirected Apenninic subduction provide a complete set of these variations even along strike in the opening of the back-arc basins oblique to the former Alpine orogen. Lifetime o f west-directed s u b d u c t i o n ZOIIeS Present W-directed subduction zones developed during the Neogene. Some of them started during the Palaeogene (e.g. Barbados). Generally speaking, they are younger than 50 Ma (usually less than 30 Ma). This does not imply that W-directed subduction zones did not exist before the Tertiary, but simply that they have a fast and short evolution in the geological record, and they may have been overprinted and incorporated by later tectonic settings. A MidPalaeozoic analogue of this type of subduction zone could be the Antler orogeny in the western US which has been compared to the Apennines arc (Burchfiel & Royden 1991) or to the Banda
554
C. DOGLIONI E T A L .
arc (Carpenter et aL 1994). A short life of the Mid-Palaeozoic Antler orogen has been proposed by Johnson & Pendergast (1981). They have shown that the main deformation occurred during Early Mississippian type. The short life of the W-directed subduction zones is striking when compared to the A n d e a n type of subduction which has been active for hundreds of millions of years. There may be two simple reasons why W-directed subduction zones terminate. (1) The encroachment of the arc with thick continental lithosphere in the foreland to the east is one case in which the subduction stops due to the high buoyancy values of a thick continental lithosphere which is no subductable. Modification of the subduction retreat might also occur with the arrival of an oceanic volcanic plateau at the subduction hinge (e.g. Ontong-Java). This is the reason for the termination of E-directed subduction zones as well, but in this last case we may observe a real collision between upper and lower plate, collision which does not occur along the arc of W-directed subduction zones because the upper plate always remains westward of the arc, with an eastward motion slower than the rollback of the subduction hinge. Slowing or termination of W-directed subduction zones due to the presence of thick continental lithosphere in the foreland are the Carpathians, the Antler orogen, and some segments of the Apennines where the heterogeneity of the foreland controlled the different rates of slab retreat (Doglioni et al. 1994). The forebulge uplift in the foreland is more developed where the lithosphere decreases its subduction rates along segments of thick continental lithosphere. (2) The second reason for the end of a Wdirected subduction zone or the switch to an Edirected subduction zone of the system is implicit in the kinematics of W-directed subduction zones (Doglioni 1991, 1993). In fact the generation of a back-arc basin introduces two new i m p o r t a n t lateral discontinuities in the lithosphere at the margins of the basin in the hangingwall of the subduction. Those weak zones are observed to be the main areas of initiation of subduction zones. The presence of those discontinuities can trigger the change in the subduction polarity, e.g. the eastern margin of a back-arc basin with thin or oceanic lithosphere to the west might become the seat for the development of an E-directed subduction zone, like it occurred in the eastern margin of the South China sea. The South China sea is the back-arc basin of the W-directed Philippine subduction, and since late Miocene it developed along its eastern margin an E-directed subduction zone from Taiwan in the north, to western Borneo in the south.
The self generation of those discontinuities in the back-arc basin seems to control the early self destruction of the W-directed subduction zones which alternate repetitively like a yo-yo from Wto E-directed subduction polarity, as it occurred in the western Pacific subduction zones during the Phanerozoic (Doglioni 1993). The young age of the western Pacific W-directed subduction zones is also suggested by the close location of the subduction zones to the Asiatic continental margin: if the subduction zones were older than the Tertiary, considering the high velocity of the eastward retreat of the subduction, they should be positioned far in the middle of the Pacific ocean.
West-directed subduction zones and the westward drift of the lithosphere On the basis of the hot-spot reference frame a 'westward' drift of the lithosphere relative to the asthenospheric mantle has been suggested (Le Pichon 1968; O'Connell et al. 1991; Ricard et al. 1991; C a d e k & Ricard 1992). Summing the vectors of plate motions in the hot spot reference frame a westward component of the lithospheric motion of a few centimetres per year remains. This 'westward' drift implies that plates have a general sense of motion and that they are not moving randomly. If we accept this kinematic frame, plates are moving along this trend at different velocities toward the 'west' relative to the asthenospheric mantle. R a t h e r than exactly west it would be better to say moving generally 'westward' (SW, WWN, etc.) along flow lines, which undulate and are not E - W parallel (Doglioni 1993). The roll-back of the slab along W-directed subduction zones implies a substitution of the lithosphere by the mantle. Since the retreat of the subducting lithosphere is eastward directed, an equivalent amount of mantle should move eastward to replace the lithospheric loss. We might argue that vertical motions of the mantle could compensate this loss without to invoke the lateral eastward migration of the mantle, but the steep attitude of the W-directed subduction zones and the strong increase of the viscosity value at the 670 km discontinuity between upper and lower mantle (Hager 1990) make the mantle adjacent to the seismically detectable slab as an isolated system, with inhibited communication among eastern and western sectors of the slab, and the upper and lower mantle underneath. This supports the notion that since the slab retreats vertically toward the east, we should kinematically expect a contemporaneous migration of the mantle located to the east of the slab toward the east as well. This does not exclude that the
WEST-DIRECTED SUBDUCTION ZONES
555
eastward moving mantle is the cause for the eastward slab retreat and not a consequence of it, but nevertheless a relative eastward mantle flow is kinematically unavoidable. An actively eastward pushing mantle agrees with the westward drift of the lithosphere relative to the asthenosphere detected in the hot-spot reference frame (Ricard et al. 1991), and it supports an eastward-oriented push at depth on the slab in order to generate the arcuate shape of the W-directed subduction zones like an obstacle does in a river (Fig. 4). Marotta & Mongelli (1997) demonstrated that the combination of slab-pull, induced flow and eastward relative mantle flow can account for the observed asymmetry among steep W-directed and shallow E-directed subduction zones. The Margheriti et al. (1996) seismic polarization data indicate an east-west anisotropy in the Tyrrhenian area. The anisotropy deviates to an Apenninic trend underneath the belt: these data might be an indication of an eastward mantle flow underneath the Tyrrhenian back-arc where the crystals should parallel the direction of mantle movement, and the encroachment with the subduction zone underneath the Apennines where the crystals should reorient due to the obstacle of the subduction.
sector of the basin, but it is laterally displaced eastward, toward the subduction hinge; (2) the heat flow distribution follows that of the subbasins and boudins with a series of highs and lows. This confirms the hypothesis that the extension of the lithosphere and the upwelling of the asthenosphere is a pulsating phenomenon which occurs following the roll-back of the subducting slab. Probably each sub-basin may open by pure-shear or simple-shear extension. An example is represented by the back-arc basin related to the Apenninic subduction, which is in fact the entire western Mediterranean, where clearly heat-flow data and thermal modelling show that the maximum heat-flow values are encountered in the sub-basins: 120 mW m -2 in the eastern Alboran (Polyak et al. 1996), 90-100 mW m -2 in the Valencia trough (Foucher et al. 1992), and more than 200 mW m -2 in the Tyrrhenian sea (R6hault et al. 1990; Mongelli et al. 1991). The eastward migration of the accretionary wedge and the extension following to the west implies that the heat flow continuously propagated with time during the migration of the tectonic fields, generating waves able to create alternances of relative 'cold' and 'hot' moments in the same areas.
Thermal state of west-directed subduction zones
Coexisting compression and extension along west-directed subduction zones
The W-directed subduction zones present among the lowest and the highest heat-flow values of the Earth respectively in the foredeep or trench and in the back-arc basin. The low values in the foredeep (down to 30 mW m -z) are due to the deflection of the isotherms with the subduction and to the sediments filling the foredeep, whenever they occur. The accretionary wedge is generally formed of superficial thrust sheets and piggy-back basins mainly made of sedimentary cover and therefore a low thermal gradient is expected. A rapid increase of heatflow is observed from the foredeep toward the back-arc basin (e.g. in the Southern Apennines, up to 80 mW m -2, Doglioni et al. 1996). Back-arc basins are sites of very high surface heat flow, up to 150 mW m -2, which is the consequence of the thinning of the lithosphere. This has generally been so far considered as due to pure-shear extension and therefore characterized by a dome-shaped symmetrical structure (e.g. McKenzie 1978, and later literature). Wernicke (1985) and Lister et al. (1991) proposed also an alternative asymmetrical thinning of the lithosphere. A more careful examination of the distribution of the heat-flow in a back-arc basin reveals two important features: (1) the maximum value is not located in the central
The active accretionary wedge is followed by an extensional wave cross-cutting the thrusts and folds previously formed in the accretionary wedge. The Apennines are a natural laboratory where to study the interplay between compression and extension (see for example Fig. 8). The extension has been very well described in the deep seismic profile Crop 03 of the northern Apennines (Barchi et aL 1998) and it shows an asymmetric character with mainly E-dipping undulated normal faults. It has always been a task to interpret the coexistence of this tectonic pair (Lavecchia et al. 1994). Here we propose a kinematic interpretation based on the shape and kinematics of the W-directed subduction zones. Back-arc extension develops as soon as the subduction starts, suggesting prevalent mechanical controlling factors. The subduction induces a mantle motion around the slab that enhances the shear stress and the mantle drag at the base of the lithosphere. This shear has a strong decrease moving westward away from the subduction hinge and it could explain why eastdipping normal faults in the hanging wall of the subduction nucleate on top of the mantle wedge at the subduction hinge (Fig. 10). This shear stress can be viewed either as a resistive force or as a driving force in plate dynamics depending
556
C. DOGLIONI E T A L .
Fig. 10. Kinematic model for the association compression-extension of W-directed subduction zones as the Apennines. The extension in the hangingwall of a W-directed subduction may be attributed to the differential drag between the eastward intruding asthenosphere and the overlying relict crust. The differential drag may be controlled by the slower horizontal component of the mantle flow induced by subduction near the mantle wedge, and by thermal constraints. The shortening in the accretionary wedge can be explained as related to the shear between the downgoing and retreating lithosphere and the eastward compensating upper mantle. The displacement is transferred upward and peels-out the cover from the foreland lithosphere. Note the different trajectory of the two black spots in the asthenosphere and at the top of the slab.
on its magnitude and direction with respect to plate motion. In case of a W-directed subduction, the relative eastward relative asthenospheric flow would add its effect to that of the induced flow in the back-arc region while in case of an E- or NE-directed subduction it would counteract the induced flow effect. On the other hand, the shortening in the accretionary wedge can be explained as related to the shear between the downgoing and retreating lithosphere and the eastward compensating mantle. This displacement is transferred upward to the east and peels off the cover from the foreland lithosphere (Fig. 10). This kinematic model for the eastward migrating compression-extension pair is triggered by the subduction roll-back and by the simultaneous asthenospheric replacement (Fig. 10).
Volume of the accretionary prism According to the proposed evolution, the Apennines developed to the east of the former Alpine-Betic orogen. We tried to compute the amount of crust involved in the accretionary
prism. The thickness of the Apenninic crust is paradoxically often smaller than the thickness of the foreland crust. In Calabria the crust can be 17 km thick (Scarascia et al. 1994). Assuming a roll-back of about 800 km and an average of 4.5 km of the sedimentary cover of the downgoing lithosphere, an area of about 3500 km 2 results for the cover alone. The present crustal area of the Apennines along the same transect is about 2900 km 2 (Fig. 11). In other words, the crustal area is smaller than the area predicted for the s e d i m e n t a r y cover, which suggests that the accretionary wedge cannot contain significant involvement of the crystalline basement since the cover itself is not reaching the entire expected area (Fig. 11). It can correctly be argued that part of the Apennines have been eroded, but the amount of area eliminated by the erosion has to be very little in this belt where the vitrinite reflectance data indicate that the Apennines were never significantly overburden (Corrado 1995). Part of the Apenninic volume might have been lost by along strike extension. Still the crustal thickness of the Apennines remains too small to invoke large slices of
WEST-DIRECTED SUBDUCTION ZONES
557
Fig. 11. The area of the Apennines accretionary prism is smaller than the area calculated for the sedimentary cover originally present on the subducted plate. Erosion and along strike extension contribute to this decrement. However the small area of the accretionary prism excludes significant involvement of the foreland basement. The thickness of the pre-subduction sedimentary cover is computed conservatively as a mean value of the different paleogeographic domains such oceanic, platform and basinal facies on the continental passive margin of the Apulia plate.
basement involved in the accretionary wedge. See Fig. 10 for an interpretation of the kinematics of the main basal d6collement zone of the accretionary wedge which shows the absence of a classic 'vertical' back-stop. The compression can be generated by the shear between the mantle moving horizontally and the downgoing lithosphere (Fig. 10). Unlike the delamination model, this model indicates that not only the lithospheric mantle is subducted in a westdirected subduction, but also the largest part of the crust, apart the main body of the sedim e n t a r y cover (Doglioni 1991), as also supported by the geochemical signatures in the
calkalcaline magmas of the volcanic arc (e.g., Peccerillo 1985; Serri et al. 1993).
Conclusions West-directed subduction zones are peculiar geodynamic settings that should be differentiated from the other subduction zones. They form very fast and are short lived. They have an internal kinematics shaped by the particular behaviour of the d6collement zones, probably due to the global polarity generated by the westward drift of the lithosphere with respect to the asthenosphere. The so-called upper plate may
558
C. DOGLIONI E T A L .
not actively thrust the footwall plate, particularly in those W-directed subduction zones that developed without E - W convergence. The Apenninic W-directed subduction zone nucleated along the east-verging back-thrust belt of the earlier Alpine-Betic orogen, where a flip in the subduction polarity from E- to Wdirected took gradually place probably during the Early Miocene. During the Neogene and Quaternary, the Apenninic arc migrated up to about 800 km on its most advanced northern Calabrian parts. With this scenario, the about 100 km N - S A f r i c a - E u r o p e relative convergence during the same time appears to be of about eight times lower importance with respect to the E- or W-directed subductions in shaping the western M e d i t e r r a n e a n . The western M e d i t e r r a n e a n is the Apenninic subductionrelated back-arc basin formed during the eastward retreat of the W-directed slab. Boudins of continental lithosphere were dispersed and stretched in the back-arc setting (Fig. 6). The boudinage occurred from the late Oligocene in the westernmost parts (Alboran, Valencia and Provencal basins) up to now to the east (eastern Tyrrhenian sea). These observations indicate the back-arc area is punctuated by jumps in the spreading position (Fig. 3) which have also been recorded by the magmatic suites. The w e s t e r n m o s t basins of the Mediterranean developed obliquely to the A l p i n e - B e t i c orogen because the extension nucleated both within the pre-existing Betic cordillera (e.g. Alboran sea) and in its foreland (Valencia and Provenqal troughs), being the direction of grabens oblique to the partly coeval orogen and showing its structural independence from the orogenic roots (Fig. 7). The kinematics imposed by the slab retreat can account for the tectonic pair thin-skinned compression in the accretionary wedge and thick-skinned extension in the belt to the west following and crosscutting the accretionary wedge (Fig. 10). The small amount of relic crust now observable in the hangingwall of the subduction suggests that the largest part of the foreland crust has been subducted (Fig. 11). The paper benefited from critical reviews by J. M. Daniel, L. Jolivet and A. Mascle. Many thanks to D. Bernoulli, F. Horvgth, M. Fernfindez, G. Pialli, G. Prosser, R. Sabadini and F. Shbat for helpful discussions. The European Community and the Italian CNR supported this study (HCM Research Networks, 'Geodynamic modeling of western Mediterranean'), grants ERBCHRXCT940607, and 96.00279.CT05, 97.00246.CT05.
References ALBARELLO,D., MANTOVANI,E., BABBUCCI,D. & TAMBURELLI, C. 1995. Africa-Eurasia kinematics: main constraints and uncertainties. Tectonophysics, 243, 25-36. ALVAREZ,W., COCOZZA,T. & WEZEL, E C. 1974. Fragmentation of the Alpine orogenic belt by microplate dispersal. Nature, 248, 309-314. AMATO, A. & 15 OTHERS 1996. Quadro geodinamico dell'Appennino settentrionale da dati di sismologia passiva. In: Crop 03 Proceedings Meeting. Societh Geologica Italiana, Rome, 1-2. , ALESSANDRINI,B., CIMINI, G., FREPOLI, A. & SELVAGGI,G. 1993. Active and remnant subducted slabs beneath Italy: evidence from seismic tomography and seismicity. Annali di Geofisica, 36, 201-214. AUZENDE, J. M., BONNIN,J. • OLIVET,J. L. 1973. The origin of the western Mediterranean basin. Journal of the Geological Society, London, 129, 607-620. BANDA,E. & SANTANACH,R 1992. The Valencia trough (western Mediterranean): an overview. Tectonophysics, 208, 183-202. BARCHI,M., MINELLI,G. & PIALL1,G. 1998. The Crop 03 profile: a synthesis of results on deep structures of the Northern Apennines. Memorie della Societd Geologica Italiana, 52, 383-405. BECCALUVA, L., BROTZU, P., MACCIOTrA, G., MORBIDELLI,L., SERRI,G. & TRAVERSA,G. 1989. Cainozoic Tectono-Magmatic Evolution and inferred mantle sources in the Sardo-Tyrrhenian Area. In: The Lithsphere in Italy. Accademia Nazionale dei Lincei, 229-248. BENEDICTO, A., LABAUME, P., SI~GURET, M. & SI~RANNE,M. 1996. Low-angle crustal ramp and basin geometry in the Gulf of Lion passive margin: Oligocene-Aquitanian Vistrenque graben, SE France. Tectonics, 15, 1192-1212. BERNOULLI,D. & LEMOINE,M. 1980. Birth and early evolution of the Tethys: the overall situation. Memoires du BRGM, 115, 168-179. BLUNDELL, D., FREEMAN, R. & MUELLER,S. 1992. A continent revealed, The European Geotraverse. Cambridge University Press, Cambridge. BO('CALETTI,M. & GUAZZONE,G. 1974. Remnant arcs and marginal basins in the Cenozoic development of the Mediterranean. Nature, 252, 18-21. BURCHFIEL, B. C. & ROYDEN, L. H. 1991. Antler orogeny: A Mediterranean-type orogeny. Geology, 19, 66-69. CADEK, O. & RICARD, Y. 1992. Toroidal/Poloidal Energy Partitioning and Global Lithospheric Rotation during Cenozoic Time. Earth and Planetary Science Letters, 109, 621-632. CADET, J.-P. & CHARVET,J. 1983. From subduction to paleosubduction in Northern Japan. In: HASmMOTO M. & UYEDA S. (eds) Accretion tectonics in the Circum-Pacific Regions', Terra Pub, Tokyo 135-148. CALCAGMLE, G. & PANZA, G. F. 1981. The main characteristics of the lithosphere-asthenosphere system in Italy and surrounding regions. Pure and Applied Geophysics, 119, 865-879.
WEST-DIRECTED SUBDUCTION ZONES CAMPAN,A. 1995. Analyse cin(matique de l'Atlantique Equatorial: implications sur l'~volution de l'Atlantique Sud et sur la frontiOre de plaques Am~rique du Nord/AmOrique du Sud. Thesis, Univ. P. & M. Curie Paris VI. CARPENTER, J. A., CARPENTER,D. G. & DOBBS, S. W. 1994. Antler orogeny: paleostructural analysis and constraints on plate tectonic models with a global analogue in southeast Asia. In: DOBBS, S. W. & TAYLOR, W. J. (eds) Structural and stratigraphic investigations and petroleum potential of Nevada, with special emphasis South of the Railroad Valley Producing Trend. Nevada Petroleum Society, Conference, 2, 187-240. CELLA, E, FEDI, M., FLORIO, G. & RAPOLLA,A. 1998. Gravity modelling of the litho-astenosphere system in the central Mediterranean. Tectonophysics, 287, 117-138. CHANNELL,J. E. Y. & MARESCHAL,J. C. 1989. Delamination and asymmetric lithospheric thickening in the development of the Tyrrhenian Rift. In: COWARD,M. P., DIETRICH,D. & PARK,R. G. (eds) Alpine Tectonics'. Geological Society, London, Special Publications, 45, 285-302. CHRISTOVA, C. & NIKOLOVA,S. B. 1993. The Aegean region: deep structures and seismological properties. Geophysical Journal International, 115, 635-653. CORRADO, S. 1995. Optical parameters of maturity of organic matter dispersed in sediments: first results from the Central Apennines (Italy). Terra Nova, 7, 338-347. DANIEL, J. M., JOLIVET, L., GOFFE, B. & POINSSOT, C. 1996. Crustal scale strain partitioning: footwall deformation below the Alpine oligo-miocene detachement of Corsica. Journal of Structural Geology, 18, 41-59. DERCOURT, J., ZONENSHA1N, L. R, RICOU, L. E., KAZMIN, g. G., LE PICnON, X., KNIPPER, A. L., GRANDJACQUET, C., SBORTSH1KOV, I. M., GEYSSANT, J., LEPVRIER, C., PECHERSKY, O. H., BOULIN, J., SIBUET, J. C., SAVOSTIN, L. A., SOROKHTIN,O., WESTPHAL,M., BAZHENOV,M. L., LAUER, J. R & BIJU-DUVAL, B. 1986. Geological evolution of the Tethys belt from the Atlantic to the Pamirs since the Lias. Tectonophysics, 123, 241-315. DE VOOGD, B., TRUFFERT, C., CHAMOT-ROOKE N., HUCHON,P., LALLEIvlANT,S. & LE PICHON,X. 1992. Two-ship deep seismic soundings in the basins of the Eastern Mediterranean Sea (Pasiphae cruise). Geophysical Journal International, 109, 536-552. DEWEY,J. E, HELMAN,M. L., TURCO, E., HUTTON,D. H. W. & KNOTt, S. D. 1989. Kinematics of the western Mediterranean. In: COWARD,M. E, DIETRICH, D. & PARK,R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. DOGLIONI, C. 1991. A proposal of kinematic modelling for W-dipping subductions - Possible applications to the Tyrrhenian-Apennines system, Terra Nova, 3, 423-434. 1992. Main differences between thrust belts. Terra Nova, 4, 152-164.
559
1993. Geological evidence for a global tectonic polarity. Journal of the Geological Society, London, 150, 991-1002. 1994. Foredeeps versus subduction zones. Geology, 22, 271-274. 1995. Geological remarks on the relationships between extension and convergent geodynamic settings. Tectonophysics, 252, 253-268. & PROSSER, G. 1997. Fold uplift versus regional subsidence and sedimentation rate. Marine and Petroleum Geology, 14, 179-190. - - . , GUEGUEN,E., SABAT,E & FERNANDEZ,M. 1997. The Western Mediterranean extensional basins and the Alpine orogen. Terra Nova, 9, 109-112. - - . , HARABAGLIA,P., MARTINELLI,G., MONGELLI,F. & ZITO, G. 1996. A geodynamic model of the Southern Apennines accretionary prism. Terra Nova, 8, 540-547. - - , MONGELLI,E & PIALLI,G. 1998. Apenninic backarc boudinage on the former Alpine belt. Memorie della Societgt Geologica Italiana, 52, 457-468. & P~ER~, P. 1994. The Puglia uplift (SEItaly): an anomaly in the foreland of the Apenninic subduction due to buckling of a thick continental lithosphere. Tectonics, 13, 1309-1321. DONNELLY, T. W. 1989. Geologic history of the Caribbean and Central America. In: The Geology of North America. Geological Society of America, DNAG, A, 299-321. FACCENNA, C., DAVY P., BRUN, J.-P., FUNICIELLO, R., GIARDINI, D., MATTEI, M. & NALPAS,T. 1996. The dynamics of back-arc extension: an experimental approach to the opening of the Tyrrhenian Sea. Geophysical Journal International, 126, 781-795. FERNANDEZ,M., FOUCHER,J. P. & JURADO, M. J. 1995. Evidence for the multi-stage formation of the south-western Valencia trough. Marine and Petroleum Geology, 12, 101-109. FOUCHER, J.-P., MAUFFRET, A., STECKLER, M., BRUNET, M.-E, MAILLARD, A., RI~HAULT, J.-E, ALONSO, B., DESEGAULX, P., MURILLAS, J. & OUILLON, G. 1992. Heat flow in the Valencia trough: geodynamic implications. Tectonophysics, 203, 77-97. FOWLER, C. M. R. 1990. The solid Earth. Cambridge Univ. Press, Cambridge. FREPOLI,A., SELVAGG1,G. CHIARABBA,C. & AMATOA. 1996. State of stress in the Southern Tyrrhenian subduction zone from fault-plane solutions. Geophysical Journal International, 125, 879-891. GUEGUEN, E., DOGLIONI, C. & FERNANDEZ,M. 1997. Lithospheric boudinage in the western Mediterranean back-arc basins. Terra Nova, 9, 184-187. HAGER, B. H. 1990. The Viscosity Profile of the Mantle: A Comparison of Models on Postglacial and Convection Time Scales. EOS, 71, 1567. HARABAGLIA, P. & DOGLIONI, C. 1998. Topography and gravity across subduction zones. Geophysical Research Letters, 25, 703-706. HONZA, E. 1995. Spreading mode of backarc basins in the western Pacific. Tectonophysics, 251, 139-152. HORVATH, E 1993. Towards a mechanical model for -
-
-
-
-
-
-
-
C. D O G L I O N I E T A L .
560
the formation of the Pannonian basin. Tectono-
physics, 226, 333-357. JOHNSON, J. G. & PENDERGAST,A. 1981. Timing and mode of emplacement of the Roberts Mountains allochthon, Antler orogeny. Geological Society of America Bulletin, 92, 648-658. JOLIVET, L. 1993. Extension of thickened continental crust, from brittle to ductile deformation: examples from Alpine Corsica and Aegean Sea. Annali di Geofisica, 36, 2, 139-153. - - , DUBOIS, R., FOURNIER, M., GOFFI~,B., MICHARD, A. & JOURDAN, C. 1990. Ductile extension in alpine Corsica. Geology, 18, 1007-1010. KARIG, O. E. & SHARMAN,III G. E 1975. Subduction and accretion in trenches. Geological Society of America Bulletin, 86, 377-389. KASTENS, K., MASCLE, J., AUROUX, C., ET AL. 1988. ODP Leg 107 in the Tyrrhenian Sea: Insigths into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KISSLING, E. 1993. Deep structure of the Alps - what do we really know? Physics of the Earth and Planetary Interior, 79, 87-112. LAVECCHIA, G., BROZZETT1, E, BARCHI, M., MENICHETTI, M. & KELLER, J. g. A. 1994. Seismotectonic zoning in east-central Italy deduced from an analysis of the Neogene to present deformations and related stress fields. Geological Society of America Bulletin, 106, 1107-1120. LE P~CHON,X. 1968. Sea-floor spreading and continental drift. Journal of Geophysical Research, 73, 3661-3697. LINZER, H. G. 1996. Kinematics of retreating subduction along the Carpathian arc, Romania. Geology, 24, 167-170. LISTER, G. S., ETHERIDGE, M. A. & SYMONDS, P. A. 1991. Detachment models for the formation of passive continental margins. Tectonics, 10, 1038-1064. LOCARDI, E. & NICOLICH, R. 1988. Geodinamica del Tirreno e dell'Appennino centro-meridionale: la nuova carta della Moho. Memorie della Societd Geologica Italiana, 41, 121-140. MALINVERNO,A. & RYAN,W. B. F. 1986. Extension in the Tyrrhenian Sea and shortening in the Apennines as a result of arc migration driven by sinking of the lithosphere. Tectonics, 5, 227-245. MARGHERITI, L., NOSTRO, C., Cocco, M. & AMATO,A. 1996. Seismic anisotropy beneath the Northern Apennines (Italy) and its tectonic implications. Geophysical Research Letters, 23, 2721-2724. MAROTrA, A. d~; MONGELLI, F. 1997. Flexure of subducted slabs. Geophysical Journal International, 132, 701-711. MARTI, J., MITJAVILA,J., ROCA, E. & APARICIO,A. 1992. Cenozoic magmatism of the Valencia trough (western Mediterranean): relationship between structural evolution and volcanism. Tectonophysics, 203, 145-165. MASCLE, A., BIJU-DUVAL, B., DE CLARENS, E & MUNSCH, H. 1986. The growth of accretionary prisms: tectonic processes from Caribbean examples. In: WEZEL, E-C. (ed.) Origin of the
Arcs.
Developments in Geotectonics, 21. Elsevier, 375-400. MAZZOLI, S., & HELMAN,M. 1994. Neogene patterns of relative plate motion for Africa-Europe: some implications for recent central Mediterranean tectonics. Geolgische Rundschau, 83, 464--468. MCKENZ~E, D. 1972. Active tectonics of the Mediterranean region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters, 40, 25-32. MONGELLI, E, LODDO, M. & CALCAGNILE, G. 1975. Some observations on the Apennines gravity field. Earth and Planetary Science Letters, 24, 385-393. --, ZITO, G., DELLA VEDOVA, B., PELLIS, G., SQUARCI, P. & TAFI, L. 1991. Geothermal Regime of Italy and surrounding Seas. In: Exploration of the deep continental crust. Springer-Verlag Berlin, 381-394. MONTIGNY, R., EDEL, J. B. & THUIZAT,R. 1981. OligoMiocene rotation of Sardinia: K-Ar ages and paleomagnetic data of Tertiary volcanics: Earth and Planetary Science Letters, 54, 261-271. O'CONNELL, R., GABLE, C. G. & HAGER, B. 1991. Toroidal-poloidal partitioning of lithospheric plate motions. In: SABADIN|, R. ET AL. (eds) -
-
Glacial Isostasy, Sea-Level and Mantle Rheology. Kluwer Academic Publisher, 334, 535-551. ONCESCU, M. C. 1984. Deep structure of the Vrancea region, Romania, inferred from simultaneous inversion for hypocentres and 3-D velocity structure. Annales Geophysicae, 2, 22-28. 1987. On the stress tensor in Vrancea region. Journal of Geophysics, 62, 62-65. PARt'S, J. M., FREEMAN, R. & ROCA, E. 1992. Neogene structural development in the Valencia trough margins from palaeomagnetic data. Tectonophysics, 203, 111-124. PATACCA, E. & SCANCDONE, P. 1989. Post-Tortonian mountain building in the Apennines. The role of the passive sinking of a relic lithospheric slab. In: BORIANI, A., BONAFEDE, M., PICCARDO, G. B. & VAI, G. B. (eds) The lithosphere in Italy. Accademia Nazionale dei Lincei, 80, 157-176. PECCERILLO, A. 1985. Roman comagmatic province (central Italy): Evidence for subduction-related magma genesis. Geology, 13, 103-106. POLYAK, B. G., FERNANDEZ, M. & FLUCALB GROUP 1996. Heat flow in the Alboran Sea (western Mediterranean). Tectonophysics, 263, 191-218. RI~HAULT,J. P., MASCLE, J. &; BOILLOT, G. 1984. Evolution geodinamique de la Mediterranee depuis l'Oligocene. Memorie delia Societd Geologica Italiana, 27, 85-96 [in French]. - - , TISSEAU,C., BRUNET, M. E & LOUDEN, K. 1990. Subsidence analysis on the Sardinian margin and the central Tyrrhenian basin: thermal modelling and heat flow control; deep structure implications. Journal of Geodynamics, 12, 269-310. RICARD, Y., DOGLIONI, C. & SABADINI, R. 1991. Differential rotation between lithosphere and mantle: a consequence of lateral viscosity -
-
WEST-DIRECTED S U B D U C T I O N ZONES variations. Journal of Geophysical Research, 96, 8407-8415. ROBERTSON,A. H. E & GRASSO,M. 1995. Overview of the Late Tertiary-Recent tectonic and palaeoenvironmental development of the Mediterranean region. Terra Nova, 7, 114-127. ROYDEN, L. H. & BURCHFIEL,B. C. 1989. Are systematic variations in thrust belt style related to plate boundary processes? (the western Alps versus the Carpathians). Tectonics, 8, 51-62. -& HORVATH, E 1988. The Pannonian Basin. American Association of Petroleum Geologists Memoirs, 45, 1-394. , PATACCA,E. (~; SCANDONE,P. 1987. Segmentation and configuration of subducted lithosphere in Italy: An important control on thrust-belt and foredeep-basin evolution. Geology, 15, 714-717. Russo, R. M. & SILVER,R G. 1994. Trench-parallel flow beneath the Nazca plate from seismic anisotropy. Science, 263, 1105-1111. SAVELLI, C. 1984. Evoluzione del vulcanismo cenozoico (da 30Ma al presente) nel Mar Tirreno e nelle aree circostanti: ipotesi geocronologica sulle fasi dell'espansione oceanica. Memorie della Societgt Geologica Italiana, 27, 111-119. SCANDONE, P. 1980. Origin of the Tyrrhenian Sea and Calabrian Arc. Bollettino della Societ~ Geologica Italiana, 98, 27-34. SCARASCIA,S., LOZEJ, A. & CASSINIS,R. 1994. Crustal structures of the Ligurian, Tyrrhenian and Ionian seas and adjacent onshore areas interpreted from wide-angle seismic profiles. Bollettino di Geofisica Teorica e Applicata, 36, 5-19. SELVAGGI,G. & CHIARABBA,C. 1995. Seismicity and Pwave velocity image of the Southern Tyrrhenian subduction zone. Geophysical Journal International, 121, 818-826. SERRI, G., INNOCENTI,E & MANETTIP. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism of central Italy. Tectonophysics, 223, 117-147. SILLITOE, R. H. 1977. Metallogeny of an Andean-type continental margin in South Korea: implications for opening of the Japan Sea. In: TALWAN1,M. & PITMAN W. M. IIl (eds) Island arcs, Deep sea trenches" and back-arc basins. AGU, Maurice Ewing Series, l, 303-310. SMITH, D. E., KOLENKIEWICZ,R., NEREM, R. S., DUNN, E J., TORRENCE, M. H., ROBBINS,J.W., KLOSKO, S. M.,WII,IJAMSON, R. G. & PAVL1S,E. C. 1994. Contemporary global horizontal crustal motion. Geophysical Journal International, 119, 511-520.
561
SPAKMAN,W. 1989. Tomographic images of the upper mantle below central Europe and the Mediterranean. Terra Nova, 2, 542-553. STANLEY, D. J. & WEZEL, E C. 1985. Geological Evolution of the Mediterranean Basin. SpringerVerlag, 1-589. TOMEK, C. 1993. Deep crustal structure beneath the central and inner West Carpathians, Tectonophysics, 226, 417-431. & HALL, J. 1993. Subducted continental margin imaged in the Carpathians of Czechoslovakia. Geology, 21, 535-538. TORNI~,M., PASCAL,G., BUHL R, WATTS,A. B. & MAUFFRET, M. 1992. Crustal and velocity structure of the Valencia Trough (Western Mediterranean). Part I. A combined refraction/wide angle reflection and near vertical reflection study. Tectonophysics, 203, 1-20. TURCOTTE, D. L. & SCHUBERT,G. 1982. Geodynamics: applications of continuum physics to geological problems. Wiley, New York. UYEDA, S. & KANAMORI, H. 1979. Back-arc Opening and the Mode of Subduction. Journal of Geophysical Research, 84, 1049-1061. VIGLIOTTI, L. & KENT, D. V. 1990. Paleomagnetic results of Tertiary sediments from Corsica: evidence of post-Eocene rotation. Physics of the Earth and Planetary Interior, 62, 97-108. WARD, S. N. 1994. Constraints on the seismotectonics of the central Mediterranean from Very Long Baseline Interferometry. Geophysical Journal International, 117, 441-452. WERNICKE, B. 1985. Uniform-sense normal simple shear of the continental lithosphere. Canadian Journal of Earth Sciences, 22, 108-125. WESTBROOK,G. K. & MCCANN,W. R. 1986. Subduction of Atlantic Lithosphere beneath the Caribbean. In: The Geology of North America. American Geological Society, M, 341-350. ZEYEN,H. & FERNANDEZ, M. 1994. Integrated lithospheric modeling combining thermal, gravity, and local isostasy analysis: Application to the NE Spanish Geotransect. Journal of Geophysical Research, 99, 18 089-18 102. ZmGLER, R A. 1988. Evolution of the Arctic-North Atlantic and Western Tethys. American Association of Petroleum Geologists Memoirs, 43. ZONENSHAIN, L. P. & SAVOSTIN,L . n . 1981. Movements of lithosphere plates relative to subduction zones: formation of marginal seas and active continental margins. Tectonophysics, 74, 5%87. -
-
Index Page numbers in italic, e.g. 430, signify references to figures. Page numbers in bold, e.g. 133, denote references to tables. Accident Paul Faillot see Paul Faillot transfer fault Adriatic plate 288-289, 291 Aegean Basin and Sea 4, 8, 432 bathymetric map 430 brittle deformation 440, 443,446, 448 crustal spreading 452 crustal thickness 432-433 ductile deformation 427-432, 431,434, 436, 440, 443, 446 ductile extension 469-471,470 extension, timing 432 faulting 423-424, 425 focal mechanisms 417, 418-421,420, 421,422,423 geodynamic background 415-416, 428 kinematic evolution 450-452, 451 magmatism 153,153 metamorphic core complexes 433 metamorphic rocks 44%450, 470 Olympos-Naxos transect Andros 438, 440 Evia (Evvia) 413,433,436, 437, 438, 444 geology 433,435 Mykonos 444, 446 Naxos 439, 443, 444, 446 Olympos, Mount 434-435,436, 444, 446 P - T-t evolution 433-434, 434 Paros 439, 443, 444 Planitis island 441,442 structure 447 Tinos 438-444, 440, 441,443, 445, 446 palaeomagnetic rotation 6 seismicity 416-418, 416, 419 strain field 450-452 strain pattern 421-422 surrounding high-pressure sites 458 tectonics 422-423,424 topographic map 429 Aeolian Archipelago, magmatism 151,156 Africa Algeria, magmatism 146 Anatolia, relative motion to 5 basins 9 Algerian Basin 71, 75, 76, 77 coast 72 convergence with Europe 4, 142, 169 Iberia, relative motion to 72, 73, 75 Libya, magmatism 146-147 magmatism 145-147 Morocco, magmatism 146 pre-Neogene margins 74 tectonic motion 31 Tunisia, magmatism 146 Alboran Basin 5, 7, 71 Alboran Sea 4, 9, 142, 146 Alcapa block rotation 319, 323-324 Alcapa unit transpression 300-309, 301-308
Al~s Basin 21,133, 133, 137 A16s depocentre 26 Algeria, magmatism 146 Algerian Basin 71, 75, 76, 77 Alg6ro-Proven~al Basin 20, 29 Algy6 high 227,228, 236, 238 Alicante transform fault 113-114 Aliera Basin 176, 177 Bouguer anomalies 180 Alpine-Mediterranean area alps, eastern 195, 197, 197 alps, formation 5 alps, Southern 197,197 collisional belt 113 geodynamics 110, 531-532 hydrocarbon provinces Palaeozoic series 518-520, 519 Mesozoic series 520-525,522, 524, 527 Cenozoic series 526-531,527, 528, 529 petroleum systems 532-534 tectonic framework 518 Alsdtold, NE Hungary, tectonic phases 301 Anatolia, relative motion to Africa 5 Anatolian block 2, 4, 5 Anatolide-Tauride Block 481,489, 499 Andros 438, 440 Ankara region 494, 495 Ankara transect 493 Ankara-Beypazarl section 491492 Apennines 4, 5,545-546, 552 subduction 544, 551,556-557 Apuseni mountains 198, 216, 236 40Ar/39Ar laserprobe dating 460-461,462,463, 464 Argentera Basin 131 Assas Basin 91, 92 Atlantic-type margins 26-28, 27 Balagne Basin 131 Balaton, Lake 204, 216 formation 370 Pleistocene-Holocenedeposits 382-383 seismic survey 368, 385 area 369 data acquisition and processing 369-372, 371, 371, 372, 372 profiles 373-379 stratigraphy 372-383,380 Balearic Basin 156 Balearic islands 38 Balearic promontary 64-68, 67, 76 Balearic thrust system 71 Bahkesir-Beypazan section 483-488 Bahkesir-[zmir section 482-483,483 Baltimore Canyon Trough, margin features 27 Barbados, W-directed subduction zone 546-547, 547 Barrot Basin 131
564 basins Aegean see Aegean Basin and Sea African 9, 71 Alboran 5, 7, 71 Albs 21,133, 133, 137 Algerian 71, 75, 76, 77 Alg6ro-Provenqa120, 29 Aliera 176, 177,180 Argentera 131 Assas 91,92 Balagne 131 Balearic 156 Barrot 131 Camargue see Camargue Basin Campidano 21 Capraia 175 Cornaglia 54 Corsica see Corsica Basin Danube 197,216 Decazeville 130 Durban 130 Empord~ 77 extensional, models 25-26 Formiche di Giglio 186 Gabian see Gabian Basin Gard 129-130, 130 Graissessac 130 Haymana 490, 491-492 H6rault 25 Largenti6res 131 Le Vigan 130 Les Matelles 26 Liguro-Provenqal see Liguro-Provenqal Basin Lod6ve 131 Manosque 21,133-134, 134 Maramures see Maramures Basin Marsili 54, 55, 156 Matelles see Matelles Basin Minervois 133, 133 Monte Glio 183, 184 Mulargia Lake 131 Mura 198 Narbonne-H6rault 133 Neffi6s 130 North Tyrrhenian see North Tyrrhenian Basin Pannonian see Pannonian Basin Prades 130 Punta Ala 184,185 Reyran 130 Rodez 131 Saint Affrique 131 Segure 130 Southeast see Southeast Basin Styrian 197 Transylvanian see Transylvanian Basin Tyrrhenian see Tyrrhenian Basin Var 131 Vavilov 54, 55, 156 Vienna see Vienna Basin Zala 197, 207,209 Bay of Biscay 18 Betic Cordillera 5 Betic Rif formation 117 nappe complexes 111
INDEX Orogeny evolution 146 sinking slab model 110,111 timing 111,112 Beypazan-Ankara section 491-492 Beypazan-Izmir section 488-491 Beypazarl-Tokat section 495 Biscay, Bay of 18 Black Sea, onshore continuation 504-505 Bursa-Nalhhan section 485, 486, 487 Calabria 114, 116-117, 116, 117 Camargue Basin 125, 133,134 formation 24 seismic section 127 rock analysis 133 Campidano Basin 21 ~anklrl loop 493-495 Capanne, Monte 178, 181-184, 189, 191 Capraia Basin 175 Capraia Island 169 Capraia-Monte Cristo ridge 170, 189 LISA cruise 175, 176,177, 178-184,179,182, 183
Carpathian mountain belt 195,197, 197, 216 see also Intra-Carpathian Area C6vennes Basin see Gard Basin C6vennes Fault 21, 23 Chendrea outcrop 258 Cialdi, Monte, LISA cruise 184, 187 coal deposits 129-130, 131 pyrolysis data 131,132 collision Alpine-Mediterranean collisional belt 113 between Africa and the Kabylies 75 Combaillaux fault 85, 86, 91, 93 Combaillaux syncline 89, 90, 91,101-103,100 compaction-driven fluid flow 403-410 compression Aegean Sea 449-450 Pannonian Basin 202 Tisza-Dacia unit 309-313 convergence between Africa and Europe 4, 16, 142, 169 between Africa and Iberia 72, 73, 75 Corconne fault 82, 87 Cornaglia Basin 54 Corsica Basin 169, 173, 188 Bouguer anomalies 180 formation 54, 184-187, 190 LISA cruise 175,175, 176-178, 177, 178, 179 Moho depth 38 onshore extension 170 Corsica Island 54, 54, 55, 184-185, 189 Corsica-Sardinia block 115,188, 189 magmatism 148-149 rotation 18, 53,113, 184-187 Crete 449-450, 450 P - T - t evolution 467,468 Cristo, Monte 169, 189,191 see also Capraia-Monte Cristo ridge crust brittle deformation 440, 443, 446, 448 collapse 8, 10
INDEX ductile deformation 427-432, 431,434, 436, 440, 443, 446 ductile extension 469-471,470 oceanic 47-53, 50 spreading 452 stretching 21-25, 56, 57 thickness Aegean Basin and Sea 432-433 Gulf of Lion 4 Ligurian Sea 47-48, 48-49, 51 Liguro-Provenqal Basin 4, 51, 58 Val6ncia Trough 64, 66, 68, 76 Western Mediterranean 142-143 Cyclades archipelago 6, 457-459, 469-471 Dalmatian coast 278, 280, 282 Danube Basin 197, 216 Danube rift 239 Decazeville Basin 130 ddcollement basin model 25-26 delta formation 10 Dinarides mountains 195,197, 216 Bouguer anomalies 229 elevation 197 faulting 282 heat flow 232 seismicity 277-278 stress 275,276, 280, 281,290-291 Dragos-Voda fault 253, 261,262, 265 drainage systems, model 95-98, 96 Drava Trough 197, 232 Durban Basin 130 Eastern Pontides 496-502, 497 ECORS-Gulf of Lion deep seismic profile 19, 21, 24, 25, 28 Elba 169, 174 ]~mile Baudot escarpment 71, 76 Emporda Basin 77 erosion, model 95-98, 96 Etna, Mount 150-151 Etruschi, Monte, LISA cruise 187 Europe convergence with Africa 4, 142, 169 magmatism of margin 143-145 Evia (Evvia) 413, 433,436, 437, 438, 444 Expanding Spread Profile (ESP) 38, 40, 47 extension, inception and dynamics 7, 8-10, 95 extensional basins, models 25-26 finite element modelling of tectonic stress 282-285, 285, 286, 287 fission-track age determination 242 Formiche di Giglio Basin 186 Fourier transform methods 38, 40, 44 France Massif Central 23,144 origin of Southern physiography 125-128 petroleum exploration history 128-129 Provence, magmatism 147-148 Gabian Basin 131 coal pyrolysis 132 oil field 128, 135 Gabon, South, margin features 27
565
Galicia, margin features 27 Gallacian oil field 134 Gard Basin 129-130, 130 geothermal flow Dinarides mountains 232 Gulf of Lion 46 Intra-Carpathian Area 233 Liguro-Proven~al Basin 45, 51, 55, 58 Pannonian Basin 232-237, fold-out between pages 230-231 Glio, Monte 183, 184 Graissessac Basin 130 gravity anomalies Aliera Basin 180 Corsica Basin 180 Dinarides mountains 229 Gulf of Lion 56 Intra-Carpathian Area 230 North Tyrrhenian Basin 180 Pannonian Basin 231, fold-out between pages 230-231 Vienna Basin 229 gravity constraints 228-232 gravity inversion procedure 38 Great Hungarian Plain 198,216, 265 chronostratigraphy 200 hydrocarbon fields 205,206 petroleum system 207, 208 Gulf of Genoa 55 Gulf of Lion 83 basin 25 Bouguer anomalies 56 continental shelf 23-24 crust 21-25, 56, 57 features 27 formation 19, 125 margins 21-25, 22, 23, 26-28, 29, 31, 57, 58 maturation 137 oil exploration 136-137 petroleum systems 31,129 P-wave velocity 57 structure 24-25, 121,122,123 surrounding basins 130, 131 tectonic evolution 25, 28, 30, 31 thermal modelling 46 G0mO~hane-Munzur section 500, 501 half-graben on basement ramps model 26 hanging-wall syncline 26, 81, 87-89, 93-95 Haymana Basin 490, 491-492 heat flow calculation 242 Hdrault Basin 25 Hodis outcrop 255-256, 255 Hungary Als6told, tectonic phases 301 Bouguer anomalies 230 hydrocarbon reserves and production 196, 205, 207 hydrocarbon provinces source rocks Mesozoic 520-525 Cenozoic 526-531 Iberia magmatism of margin 143-145
566 Iberia continued pre-Neogene margins 74 relative motion to Africa 72, 73, 75 Iberian plate Iblean area (Sicily) 150 Integrated Basins Project (IBS) 1, 16, 19, 220 Intra-Carpathian Area 199 Bouguer anomalies 230 Carpathian mountain belt 195, 197,197, 216 geological outline 252 heat flow 233 lithospheric thickness 221 major sub-basins 297 tectonic evolution 198-200, 359 Intra-Pontide suture 502, 504, 508-509 evolution of ocean 506 Istanbul-Sakarya section 502-504 onshore continuation of faults 504-506, 504 westward continuation 506 Istanbul continental block 477-479, 478 Italy Calabria 114, 116-117,116,117 magmatism Central-South 151-152 North 152-153 Tuscany magmatism 151,152 margin, LISA cruise 184,184,185, 186 Iza valley 258 Izmir-Ankara-Erzincan suture 482,507-508, 508 Ankara region 494, 495 Bahkesir-Beypazan section 483-488 Beypazarl-Ankara section 491-492 Beypazan-Tokat section 495 Bursa-Nalhhan section 485,486, 487 ~ankan loop 493-495 Eastern Pontides 496-502, 497 Gtimtishane-Munzur section 500, 501 /zmir-Bahkesir section 482-483,483 Izmir-Beypazan section 488-491 Nalhhan-Tokat section 492 Sivrihisar region 488 Japan Sea, margin features 27 Kabylies belt 71-77, 73, 113, 115 Kaman transect 493-495 Kefallonia fault 5 Kir~ehir Massif 481,496 Lake Balaton see Balaton, Lake Langedoc 23 Largenti6res Basin 131 coal pyrolysis 132 Le Vigan Basin 130 Les Matelles Basin 26 Levant 154 Lez fault 92, 94 Libya, magmatism 146-147 Ligurian Sea 5 ocean crust thickness 47-48, 48-49, 51 ocean floor dating 53 spreading model 51-53, 52, 55, 58 thermal model 46, 52 Liguro-Provenqal Basin 4, 5
INDEX formation 50-51, 54, 58, 69, 70, 142 heat flow 45, 51, 55, 58 magmatism 149-150 Moho depth 38, 41 ocean crust thickness 51, 58 plate separation rate 55 rifting 56 structure 58 Val6ncia Trough, relation to 70 Little Hungarian Plain 197,200, 209 Lod6ve Basin 131 coal pyrolysis 132 Macedonia, magmatism 153 Maghrebides belt 71 magmatism Aegean area 153,153 Aeolian Archipelago 151,156 Africa, Northern margin 145-147 Alboran Sea 146 Algeria 146 Etna, Mount 150-151 European margin 143-145 geochemical characteristics elements, major 157,158 elements, trace 157-159, 158,159,160,161 Sr-Nd-Pb isotopes 159-160,161 Iberian margin 143-145 Italy, Central-South 151-152 Italy, North 152-153 Libya 146-147 Macedonia 153 Massif Central 144 Mediterranean, central, geodynamics anorogenic 143, 156-157 Neogene-Quaternary cycle 155,155 Oligo-Miocene cycle 154-155 sea floor spreading 156 source regions 160-162 tectonism 156 Morocco 146 origins 141 Pannonian Basin 237-238 Provence 147-148 Sardinia-Corsica 148-149 Sicily 150-151 Tunisia 146 Turkey, western 154 Tuscany 151,152 Tyrrhenian Sea 150 Val6ncia Trough 149 Mallorca 67-68, 67, 71 Manosque Basin 21,133-134, 134 mantle thinning factor equation 236 Maramures Basin 253, 261,263, 265 Palaeogene basin (MPB) 253-254, 264 Marmara Sea 503 Marsili Basin, formation 54, 55, 156 Massif Central 23, 144 Matelles Basin clast composition 98-101 ramp-flat model 97, 98-103 setting 82, 83 stratigraphy 84 structure 84, 85, 86, 91-93, 92, 94
INDEX Matelles fault 84-87, 85, 86, 90-91, 92, 93, 94 Matelles syncline 86, 88-89, 90, 99 catchments 101-103 clast composition 100 faulting, drainage and sedimentation 93 palinspastic maps 102, 104 structure 90 Mecsek mountains 236 Mediterranean Sea basin categories 9-10 crustal structure 1-4 evolution, western Tertiary 17 magmatism, geodynamics anorogenic 156-157 Neogene-Quaternary cycle 155,155 Oligo-Miocene cycle 154-155 sea floor spreading 156 source regions 160-162 tectonism 156 origins 16 structure 16 Minervois Basin 133,133 modelling methods drainage systems and erosion 95-98, 96 finite element 282-285,285, 286, 287 fluid flow 403-410, 405, 406, 407, 408 Fourier transform 38, 40, 44 neural networks 408-410 non-uniform extension 244-245 sediment thickness variation 336-338, 339, 344-345,346, 347 tectonic stress 282-285,285, 286, 287 tectono-stratigraphy 397-403 Moho depth Balearic islands 38 Corsica 38 determination corrections for 43-47 from Bouguer anomalies 38-40, 39, 40, 43-47, 44, 45 from gravity inversion 38, 42, 43 from seismic data 37-38, 43 Liguro-Proven~al Basin 38, 41 Sardinia 38 Valbncia Trough 38 Monte Capanne 178, 181-184, 189,191 Monte Cialdi, LISA cruise 184,187 Monte Cristo 169, 189,191 see also Capraia-Monte Cristo ridge Monte Etruschi, LISA cruise 187 Monte Glio Basin, LISA cruise 183, 184 Montenegro 281 Montpellier thrust 82, 87, 93 Morocco, magmatism 146 Mount Etna 150-151 Mount Olympos 434-435,436, 444, 446 Mount Vulture 152 Mulargia Lake Basin 131 Munzur-Gt~mt~s, hane section 500, 501 Mura Basin 198 Mykonos 444, 446 Nalhhan-Bursa section 485, 486, 487 Nalhhan-Tokat section 492 Narbonne-H6rault Basin 133
567
Naxos 439, 443, 444, 446 P - T - t evolution 468, 469 Neffi6s Basin 130 neural networks 408-410 N~mes fault 24, 26 North Anatolian fault 5, 8 North Balearic Transfer Fault see Paul Faillot transfer fault North Pyrenean Fault Zone 18 North Tyrrhenian Basin Bouguer anomalies 180 geophysical setting 16%174 LISA cruise 171,174, 175-184,175 structure 172,173 tectonic framework 169,170 transfer faults 191 see also Tyrrhenian Basin numerical cycle analysis 336-338, 339, 344-345,346, 347 Numidian flysch 74, 75 ocean floor spreading 48-53, 55, 56 oceanic crust, product of mantle melt 48-49, 51 oceanic crust thickness 47-53, 50 oil fields Gabian Basin 128, 135 Gallacian 134 Hungarian 196, 205,207 Saint Jean de Maru6jo1135 oilwell, Sigoyer 1, section 129 Olympos, Mount 434-435,436, 444, 446 palaeomagnetic rotations 5-6 Pannonian Basin Alpine structure, early 201 Bouguer anomalies fold-out between pages 230-231 chronostratigraphy 204, 360, 365-368, 370, 384-385 compressional events 202 evolution 203,349-353, 351 extension 201-202 extrusion kinematics 201 faulting 393 finite element modelling of tectonic stress 282-285, 285, 286,287 fluid flow modelling 403-410, 405, 406, 407, 408 formation 198, 270 geodynamic model 319-326, 320, 321,322,323 geological setting 195-200, 358-359 geothermal data 232-237 heat flow fold-out between pages 230-231 isopach map 392 lithospheric evolution 238-240, 241-242 lithospheric thickness 4, 233 location 195 magmatism 237-238 magnetostratigraphy 338-340 Palaeogene basin formation 201 permeability of southern marls and sandstones 404, 405 petroleum systems 204-211 porosity of southern marls and sandstones 404, 405 rifting 313-317 SE lithospheric structure Algy6 high 227,228, 236, 238
568
INDEX
Pannonian Basin continued block diagram 229 Bouguer anomalies 231 fission-track data 220-224, 222 gravity constraints 228-232 pre-Tertiary 219 reflection seismic data 218-220 regional geology 217-218 regional transects 224-228,225, 226 sedimentary features 350, 351 sedimentary fill, southern 393-395,394 sedimentological analysis 338, 340-344 magnetostratigraphy 338-340 methodology 336, 337, 337 numerical cycle analysis 336-338,339, 344-345, 346, 347 spacial evolution 240 strain rate 241 stratigraphy lithostratigraphic framework 361 sequence 340-344, 361-365,362,363,364, 365, 384 subdivisions 359-361 stress analysis 275, 276, 277, 286-290, 310, 311,312, 313 active deformation 279-282, 279 Alcapa unit transpression 300-309, 301-308 basin inversion 317-319 borehole breakout 272-274, 272, 273 crustal deformation 277, 289 earthquake focal mechanisms 271-272, 272 extension and faulting 317-318 finite element modelling 282-285,286, 287 inferred stress 274-276 in-situ measurement 274 palaeomagnetic data 297-298 seismicity 277-278, 278 tectonic observations 298-300, 298 Tisza-Dacia unit compression 309-313 stress map 396 structural periods 216-217, 218 subsidence 395-397,397, 398, 403 analysis 345-349,348, 349 tectonic deformation 383-386 tectonic units 270 tectono-stratigraphic modelling 397-403,399, 400, 401,401,403,410 thermal model 233-237,234, 235,235, 237 unconformity-bounded units 365-368 water level 352, 394-395 Pannonian GeoTraverse (PGT-1) line 218, 218, 227 Pannonian GeoTraverse (PGT-4) line 221 Pannonian Sea 335 Paratethyan stages 336 Paros 439, 443, 444 Paul Faillot Transfer Fault (North Balearic Transfer Fault) 18, 70, 71 Pearl River Basin, margin features 27 Pelagonian Zone geological map 458 P - T - t evolution 467-469, 468 thermotectonic evolution 461-466,465, 466 Pelion Massif 4~ laserprobe dating 460-461,462, 463, 464
P - T - t evolution 466-467 structure 459-460 permeability, modelling equation 405 Petit RhSne graben 24 petroleum exploration 9 history in France 128-129 petroleum source rocks 128-134 Mesozoic 520-525 Cenozoic 526-531 petroleum systems examples Alpine-Mediterranean area 532-534 Great Hungarian Plain 207, 208 Gulf of Lion 31,129 Pannonian Basin 204-211 Southeast Basin 129 Zala Basin 209 stable areas 135,136 subsiding areas 134 thrust belts 134-135 uplifted areas 135, 136 Pianosa Island 169, 181 Pianosa Ridge 170, 181 LISA cruise 175,176,177, 179, 181-184,182, 18.3 Pic Saint-Loup anticline 82, 87 Planitis island 441,442 Poiana Blenchii gorge 256-257,256, 262 Poiana Botizii outcrop 257-258,257, 262, 264 Pontides, Eastern 496-502, 497 porosity, modelling equation 405 Prades Basin 130 Prades fault 91, 92 Provence, magmatism 147-148 Punta Ala Basin, LISA cruise 184, 185 Pyr6n6es formation 5, 18 thrust system 69
ramp-flat extensional fault systems 81, 93-95, 97, 98-103 Reyran Basin 130 Rhine graben 5, 19 Rodez Basin 131 coal pyrolysis 132 Saint Affrique Basin 131 Saint-Cl6ment fault 91, 92, 94 Saint Jean de Maru6jol oil field 135 Saint-Mathieu fault 86, 91 Saint-Mathieu syncline 87-88, 88, 89, 99 faulting, drainage and sedimentation 93 structure 90, 98 Sakarya Zone 479-480, 480, 499 Sandwich, W-directed subduction zone 547, 547 Sardinia East, margin features 27 Moho depth 38 palaeomagnetic data 54, 54 Sardinia-Corsica, magmatism 148-149 Sardinia-Menorca transform fault 113 Sava Trough 197,232 Segure Basin 130 seismic profiles
INDEX Balaton, Lake 373-379 Camargue Basin 127 ECORS-Gulf of Lion 19, 21, 24, 25, 28 Pannonian Basin 361-365 seismicity Aegean Basin and Sea 416-418, 416, 419 focal mechanisms 417,418-421,420, 421,422, 423 Dinarides mountains 277-278 Pannonian Basin 277-278, 278, 285-290 Sicily-Calabria 166, 167 Sicily, magmatism 150-151 Sicily-Calabria, seismic epicentres 166,167 Sifnos, P - T - t evolution 468, 468 Sigoyer 1 oilwell, section 129 Sivrihisar region 488 slab detachment 9, 117-118 rollback 9, 117-118 sinking slab model 110,111 South Gabon, margin features 27 Southeast Basin formation 125 maturation history 137 petroleum systems 129 sedimentation 125 structure 121,122,123,126 surrounding basins 130,131 Southern Alps 197,197 Strandja Zone 479, 479 strike-slip faulting Dinarides mountains 276 North Anatolian fault 10 Pannonian Basin 317-318 strike-slip tectonics Aegean Basin and Sea 422-423 Styrian Basin 197 subduction Apenines 544, 551,556-557 Barbados 546-547, 547 Betic-Ligurian 113, 115 Sandwich 547,547 west-directed subduction zones accretionary prism volume 555-556, 557 characteristics 541-546, 542, 543 compression/extension 555, 556 initiation 546-548,547, 548, 549, 550, 551,552 lithosphere drift 554 persistence 553-554 thermal state 554-555 sutures Intra-Pontide 502, 504, 508-509 evolution of ocean 506 istanbul-Sakarya section 502-504 onshore continuation of faults 504-506, 504 westward continuation 506 Izmir-Ankara-Erzincan 482, 507-508, 508 Ankara region 494, 495 Bahkesir-Beypazarl section 483-488 Beypazarl-Ankara section 491-492 Beypazan-Tokat section 495 Bursa-Nalhhan section 485, 486, 487 ~anklrl loop 493-495 Eastern Pontides 496-502, 497
569 Gtimti~ane-Munzur section 500, 501 Izmir-Bahkesir section 482-483,483 Izmir-Beypazarl section 488-491 Nalhhan-Tokat section 492 Sivrihisar region 488
tectonic chart 124 tectonic stress, finite element modelling 282-285,285, 286, 287 tectonics, continental extension modes 239 Tellian front 76 Teyran fault 92 thermal effect of sedimentation calculation 242-244 Tinos 438-444, 440, 441,443, 445, 446 Tisza rift 238, 239, 240 Tisza-Dacia block 262, 263-265 rotation 319, 323-324 Tisza-Dacia unit compression 309-313 Tokat-Beypazan section 495 Tokat-Nalhhan section 492 Transdanubia 197,204 Transdanubian Range 197,216, 232 Transylvanian Basin 198, 216 Chendrea outcrop 258 chronostratigraphy 200 faulting 259, 260, 261 geodynamics 262-265 geological setting 251-254 Hodis outcrop 255-256, 255 Iza valley 258 Palaeogene basin (TPB) 252-253,261-262, 263 Poiana Blenchii gorge 256-257, 256, 262 Poiana Botizii outcrop 257-258, 257, 262, 264 stratigraphy 254 structure 260-262 Tisza-Dacia block 262, 263-265 rotation 319, 323-324 Tisza-Dacia unit compression 309-313 Tunisia, magmatism 146 Turkey Ankara region Ankara transect 493 continental blocks Anatolide-Tauride Block 481,489, 499 lstanbul Zone 477-479,478 Kirsehir Massif 481,496 Sakarya Zone 479-480, 480, 499 Strandja Zone 479, 479 Haymana Basin 490, 491-492 Intra-Pontide suture 502, 504, 508-509 evolution of ocean 506 lstanbul-Sakarya section 502-504 onshore continuation of faults 504-506, 504 westward continuation 506 Izmir-Ankara-Erzincan suture 482, 507-508,508 Ankara region 494, 495 Ankara transect 493 Bahkesir-Beypazarl section 483-488 Beypazan-Ankara section 491-492 Beypazan-Tokat section 495 Bursa-Nalhhan section 485, 486, 487 ~anklrl loop 493-495 Eastern Pontides 496-502, 497 evolution 501-502
570 Turkey continued G~imt~shane-Munzur section 500, 501 Izmir-Ballkesir section 482-483,483 Izmir-Beypazan section 488-491 Kaman transect 493-495 Nalhhan-Tokat section 492 Sivrihisar region 488 magmatism 154 Palaeo-Tethyan suture 506-507 Tuscany magmatism 151,152 margin, LISA cruise 184,184,185,186 Tyrrhenian Basin formation 50-51 opening 7 rifting 55-56, 187 see also North Tyrrhenian Basin Tyrrhenian Sea 4, 5 crustal collapse 8 magmatism 150 sea floor spreading 156 slab rollback 9 tectonic evolution 184, 187, 189, 190 Valbncia Trough 5, 18 crustal thickness 64, 66, 68, 76 crustal area 68 deformation 68-72 extension 66~57, 77 formation 69-70, 75-76, 142 geometry 64 Liguro-Provenqal Basin, relation to 70 magmatism 149
INDEX Moho depth 38 pre-Neogene structure 64-68, 67 profile 63, 65 rifting 70 Var Basin 131 Vavilov Basin, formation 54, 55, 156 Vienna Basin 197,216 Bouguer anomalies 229 formation 198 Vistrenque graben 24 Vcring, margin features 27 Vulture, Mount 152 West European Rift 5, 7, 19, 21 west-directed subduction zones accretionary prism volume 555-556, 557 characteristics 541-546, 542, 543 compression/extension 555,556 initiation 546-548, 547, 548, 549, 550, 551,552 lithosphere drift 554 persistence 553-554 thermal state 554-555 Western Mediterranean crustal thickness 142-143 evolution 72-77 geodynamic setting 548-553 kinematic evolution 113-115,114 tectonics 64 volcanic activity 145 World Stress Map Project 274 Zala Basin 197, 207 petroleum system 209
The Mediterranean Basins: Tertiary Extension within the Alpine Orogen edited by Bernard Durand (IFP, France), Laurent Jolivet (Universite P & M Curie, France), Frank Horvath (EStv6s Lorand University, Hungary) and Michel Seranne (Universite Montpellier, France) The coexistence in space and time of growing mountain belts and actively extending basins poses a number of yet unsolved questions in terms of mechanics. This problem is particularly crucial in the Mediterranean regions, where all Cenozoic basins opened in the iflternal zones of mountain belts.TheTyrrhenian Sea opened in the back-arc region of the Apennines, the Aegean Sea in the back-arc domain of the Hellenides and Hellenic arc, the Pannonian Basin behind the Carpathians and the Alboran Sea between the Betics and the Rif. In some examples, such as the Tyrrhenian Sea and the Aegean Sea, extension is ongoing while peripheral compression and convergence are active.The Alboran Sea and Pannonian Basin are now in a compression stage. Several models have been proposed to explain this coexistence of compression and extension: slab retreat during subduction process, detachments of a deep lithospheric root under the internal zones leading to radial extension and peripheral compression and slab detachments. This volume brings together contributions from geologists and geophysicists in the quest to solve the complex dynamic problem posed by the Mediterranean region. It presents a wealth of new data on various topics centred on the Mediterranean region from the deep mantle structure to the detailed geometry of sedimentary basins. This book results from the Integrated Basins Studies Project, which was funded by the European Commission and which involved the collaboration of over 200 researchers across Europe. • 584 pages • 23 chapters • 350 illustrations • index A sister volume has been published - Cenozoic Foreland Basins of Western Europe - Geological Society Special Publication No. 134 - it can be ordered from our online bookshop: www.bookshop.geolsoc.org.uk
Cover Illustration: Tithonian Structure near to Marseille.
ISBN 1 - 8 6 2 3 9 - 0 3 3 - 9
!UI!!!! J!! Jl!