Palaeozoic Reefs and Bioaccumulations" Climatic and Evolutionary Controls
The Geological Society of L o n d o n
Books Editorial Committee Chief Editor BOB PANKHURST(UK)
Society Books Editors JOHN GREGORY(UK) JIM GRIFFITHS (UK) JOHN HOWE (UK) PHIL LEAT (UK) NICK ROBINS (UK) JONATHANTURNER (UK)
Society Books Advisors MIKE BROWN(USA) ERIC BUFFETAUT(FRANCE) RETO G1ER]~(GERMANY) JON GLUYAS(UK) DOUG STEAD(CANADA) RANDELL STEPHENSON(NETHERLANDS)
Geological Society books refereeing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society's Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society Book Editors ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Society. The referees' forms and comments must be available to the Society's Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. More information about submitting a proposal and producing a book for the Society can be found on its web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways:
/~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275. KERSHAW, S., LI, Y. & Guo, L. 2007. Micritic fabrics define sharp margins of Wenlock patch reefs (middle Silurian) in Gotland and England. In: ALVARO,J. J., ARETZ, M., BOULVAIN,F., MUNNECKE, A., VACHARD, D. & VENNIN, E. (eds) Palaeozoic" Reefs and Bioaccumulations." Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 87-94.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 275
Palaeozoic Reefs and Bioaccumulations" Climatic and Evolutionary Controls
EDITED BY J. JAVIER ,/~LVARO University of Zaragoza, Spain and University of Lille I, France MARKUS ARETZ University of Cologne, Germany FRI~DI~RIC BOULVAIN University of Li6ge, Belgium AXEL MUNNECKE University of Erlangen-Niirnberg, Germany DANIEL VACHARD University of Lille I, France and EMMANUELLE VENNIN University of Bourgogne, France
2007 Published by The GeologicalSociety London
THE G E O L O G I C A L SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the U K national learned and professional society for geology with a worldwide Fellowship (FGS) of over 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society's fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society's international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists' Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies' publications at a discount. The Society's online bookshop (accessible from www.geolsoc.org.uk ) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J 0BG: Tel. +44 (0)20 7434 9944; Fax +44 (0)20 7439 8975; E-mail:
[email protected]. For information about the Society's meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society's Corporate Affiliates Scheme, write to
[email protected]
Published by The Geological Society from: The Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Brassmill Lane, Bath BA1 3JN, U K Tel. +44 (0)1225 445046, Fax +44 (0)1225 442836) Online bookshop: www.geolsoc.org.uk/bookshop
(Orders:
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 of London 2007. 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 W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/07/$15.00.
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 978-1-86239-221-2 Typeset by Charlesworth & Co Ltd, U K Printed by Cromwell Press, Trowbridge, U K
Distributors North America For trade and institutional orders: The Geological Society, c/o AIDC, 82 Winter Sport Lane, Williston, VT 05495, USA Orders: Tel +1 800-972-9892 Fax +1 802-864-7626 Email
[email protected] For individual and corporate orders: AAPG Bookstore, PO Box 979, Tulsa, OK 74101-0979, USA Orders: Tel +1 918-584-2555 Fax +1 918-560-2652 Email
[email protected] Website http://bookstore.aapg.org India Affiliated East-West Press Private Ltd, Marketing Division, G-l/16 Ansari Road, Darya Ganj, New Delhi 110 002, India Orders: Tel. +91 11 2327-9113/2326-4180 Fax +91 11 2326-0538 E-mail
[email protected]
CONTENTS Foreword
~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. Fabric transitions from shell accumulations to reefs: an introduction with Palaeozoic examples CLAUSEN,S. &/~kLVARO,J. J. Lower Cambrian shelled phosphorites from the northern Montagne Noire, France GANDIN, A., DEBRENNE,F. & DEBRENNE,M. Anatomy of the Early Cambrian 'La Sentinella' reef complex, Serra Scoris, SW Sardinia, Italy /~kLVARO,J. J. & CLAUSEN,S. Botoman (Lower Cambrian) turbid- and clear-water reefs and associated environments from the High Atlas, Morocco HUNTER, A. W., LEFEBVRE,B., RI~GNAULT,S., ROUSSEL,P. & CLAVERIE,R. A mixed ophiuroid-stylophoran assemblage (Echinodermata) from the Middle Ordovician (Llandeilian) of western Brittany, France KERSHAW, S., LI, Y. & Guo, L. Micritic fabrics define sharp margins of Wenlock patch reefs (middle Silurian) in Gotland and England HUBMANN, B. & SUTTNER,T. Siluro-Devonian Alpine reefs and pavements MABILLE, C. & BOULVAIN,F. Sedimentology and magnetic susceptibility of the Upper Eifelian-Lower Givetian (Middle Devonian) in SW Belgium: insights into carbonate platform initiation BOULVAIN, F. Frasnian carbonate mounds from Belgium: sedimentology and palaeoceanography POTY, E. & CHEVALIER,E. Late Frasnian phillipsastreid biostromes in Belgium ARETZ, M. & CHEVALIER,E. After the collapse of stromatoporid-coral reefs - - the Famennian and Dinantian reefs of Belgium: much more than Waulsortian mounds ALMAZAN-VAZQUEZ,E., BUITRON-SANCHEZ,B.E., VACHARD,D., MENDOZA-MADERA,C. & GOMEZ-ESPINOSA, C. The late Atokan (Moscovian, Pennsylvanian) chaetetid accumulations of Sierra Agua Verde, Sonora (NW Mexico): composition, facies and palaeoenvironmental signals BUITRON-SANCHEZ, B.E., GOMEZ-ESPINOSA,C., ALMAZAN-VAZQUEZ,E. & VACHARD,D. A late Atokan regional encrinite (early late Moscovian, Middle Pennsylvanian) in the Sierra Agua Verde, Sonora state, NW Mexico VENNIN, E. Coelobiontic communities in neptunian fissures of synsedimentary tectonic origin in Permian reef, southern Urals, Russia WEmI.ICH, O. Permian reef and shelf carbonates of the Arabian platform and Neo-Tethys as recorders of climatic and oceanographic changes THI~RY, J. M., VACHARD,D. & DRANSART,E. Late Permian limestones and the Permian-Triassic boundary: new biostratigraphic, palaeobiogeographical and geochemical data in Caucasus and eastern Europe ZAPALSKI, M. K., HUBERT, B. & MISTIAEN, B. Estimation of palaeoenvironmental changes: can analysis of distribution of tabulae in tabulates be a tool? Index
vii 1
17 29 51 71
87 95 109
125 143 163 189
201
211 229 255
275 283
Foreword The difficulty of studying reefs and shell accumulations rests primarily on its multidisciplinary position crossing numerous disciplines, such as biostratigraphy, geochemistry, palaeobiology, palaeoecology, petrology, sedimentology and taphonomy. The facies characterization of these bioclastic-bearing strata is, like many other biosedimentary structures, a process that requires the acquisition and integration of a wide and multiscale diversity of observations, which include field (global geometries), sample (fabrics) and thin-section (textures) scales. When we organized an international meeting focused on 'Climatic and Evolutionary Controls on Palaeozoic Reefs and Bioaccumulations' (7-9 September 2005, Paris, France), it was our intention to provide a forum for discussing the evolution of reefs, shell accumulations and their transitional deposits. We invited specialists on a wide range of taxonomic groups, siliciclasticmixed carbonate platforms and Palaeozoic ages to introduce a number of topics that are the focus of current research in the world of Palaeozoic benthic communities. The result of their contributions is presented in this Special Publication, which shows the complexity of intrareef synecological relationships and the diversity of concepts used to characterize both reefs and bioaccumulations. The editors offer, in the introductory paper (,~lvaro et al.), a discussion about concepts and definitions related to reefs and bioaccumulations. The transition between the concepts of reef and shell accumulation is gradual, as illustrated by some Palaeozoic examples, where reworked coquinas were episodically stabilized by encrusting communities and/or early diagenetic cements, in some cases forming the sole for future frame-building fabrics. Three Cambrian works are focused on shell-rich phosphorites from the Montagne Noire, France (Clausen & ,iAvaro), and microbial and archaeocyathanmicrobial reef complexes from Sardinia, Italy (Gandin et al.) and the High Atlas, Morocco (Alvaro & Clausen), the latter directly controlled by volcanogenic turbidity. The increase in biodiversity recorded in Ordovician bioaccumulations is illustrated by the characterization of a distinct echinoderm assemblage rich in ophiuroids and stylophorans (Hunter et al.). Kershaw et al. show that some Silurian reefs from Gotland and the U K have sharp boundaries, with the surrounding sediments terminating abruptly against the reef edge, and the sharp margins made up of automicrites; the sharp reef edges indicate
coherence of the micritic fabric, interpreted as a lithified wall against which bedded limestones were deposited. Hubmann & Suttner provide a review of Alpine Late Silurian-Late Devonian reefs and pavements, spanning a wide range of different autochthonous carbonates (e.g. brachiopod pavements, algal reefs, stromatoporoidcoral patch reefs) as well as allochthonous accumulations (e.g. serpulid accumulations). Two related papers offer an updated synthesis of the establishment of a carbonate platform (Mabille & Boulvain) and the sea-level-controlled evolution of Devonian mounds and atolls in the Dinant Synclinorium from Belgium (Boulvain). Examples for the youngest reefs of the Middle Palaeozoic reef community of stromatoporoids and corals are described from the latest Frasnian of Belgium (Poty & Chevalier). The struggle to establish a successful reef assemblage in the aftermath of the Kellwasser events and the importance of microbial communities in Famennian and Carboniferous reefs is described from Belgium (Aretz & Chevalier). The influence of palaeobathymetry and local synsedimentary tectonics in the establishment of carbonate factories is discussed in two papers focused on the development of Carboniferous chaetetid 'reefs' (Almaz~n-Vazquez et al.) and neighbouring crinoidal thickets (Buitr6n-S~nchez et aL) in Sonora, Mexico, whereas the documentation of sea-floor instability related to the evolution of the Permian foreland basin recorded in the southern Urals is characterized by Vennin. The input of cool waters within tropical PermoCarboniferous seas is analysed in Oman by Weidlich, and opens a large field of future discussions. Th6ry et al. provide new insights into the latest Permian reefs and bioaccumulations from eastern Europe and the Caucasus. And, finally, Zapalski et al. offer an estimation of palaeoenvironmental changes based on the distribution of late Middle Devonian tabulae in tabulate corals from northern France. One of the messages of this collection of papers is the wide diversity of sedimentary geometries and facies displayed by reefs, shell accumulations and transitional composite deposits. Readers will find that the papers in this Special Publication cover specific nomenclatural problems, evidenced by the widespread terminology used to describe skeletal assemblages. Rather than attempt a complete revision of terms, we have touched on some of the major issues at this stage of development in the field: the major
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special
Publications, 275, vii-viii. 0305-8719/07/$15.00 9 The Geological Society of London.
viii
FOREWORD
climatic, environmental and evolutionary factors that controlled the Palaeozoic development of shell accumulations and reefs. We hope that this volume attracts the attention of everyone interested in the fascinating diversity of Palaeozoic reefs and shell accumulation. It will be useful to senior undergraduate and postgraduate students of Earth Sciences and engineering. We also hope that it may prove useful to professionals
who explore and economically exploit Palaeozoic skeletal-rich strata. The editors would like to thank all contributors and referees for their rapid and stimulating collaboration. Thanks also for their suggestions, discussions, editorial work and constructive reviews. J. J. _~lvaro, M. Aretz, F. Boulvain, A. Munnecke, D. Vachard & E. Vennin
CONTENTS Foreword
~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. Fabric transitions from shell accumulations to reefs: an introduction with Palaeozoic examples CLAUSEN,S. &/~kLVARO,J. J. Lower Cambrian shelled phosphorites from the northern Montagne Noire, France GANDIN, A., DEBRENNE,F. & DEBRENNE,M. Anatomy of the Early Cambrian 'La Sentinella' reef complex, Serra Scoris, SW Sardinia, Italy /~kLVARO,J. J. & CLAUSEN,S. Botoman (Lower Cambrian) turbid- and clear-water reefs and associated environments from the High Atlas, Morocco HUNTER, A. W., LEFEBVRE,B., RI~GNAULT,S., ROUSSEL,P. & CLAVERIE,R. A mixed ophiuroid-stylophoran assemblage (Echinodermata) from the Middle Ordovician (Llandeilian) of western Brittany, France KERSHAW, S., LI, Y. & Guo, L. Micritic fabrics define sharp margins of Wenlock patch reefs (middle Silurian) in Gotland and England HUBMANN, B. & SUTTNER,T. Siluro-Devonian Alpine reefs and pavements MABILLE, C. & BOULVAIN,F. Sedimentology and magnetic susceptibility of the Upper Eifelian-Lower Givetian (Middle Devonian) in SW Belgium: insights into carbonate platform initiation BOULVAIN, F. Frasnian carbonate mounds from Belgium: sedimentology and palaeoceanography POTY, E. & CHEVALIER,E. Late Frasnian phillipsastreid biostromes in Belgium ARETZ, M. & CHEVALIER,E. After the collapse of stromatoporid-coral reefs - - the Famennian and Dinantian reefs of Belgium: much more than Waulsortian mounds ALMAZAN-VAZQUEZ,E., BUITRON-SANCHEZ,B.E., VACHARD,D., MENDOZA-MADERA,C. & GOMEZ-ESPINOSA, C. The late Atokan (Moscovian, Pennsylvanian) chaetetid accumulations of Sierra Agua Verde, Sonora (NW Mexico): composition, facies and palaeoenvironmental signals BUITRON-SANCHEZ, B.E., GOMEZ-ESPINOSA,C., ALMAZAN-VAZQUEZ,E. & VACHARD,D. A late Atokan regional encrinite (early late Moscovian, Middle Pennsylvanian) in the Sierra Agua Verde, Sonora state, NW Mexico VENNIN, E. Coelobiontic communities in neptunian fissures of synsedimentary tectonic origin in Permian reef, southern Urals, Russia WEmI.ICH, O. Permian reef and shelf carbonates of the Arabian platform and Neo-Tethys as recorders of climatic and oceanographic changes THI~RY, J. M., VACHARD,D. & DRANSART,E. Late Permian limestones and the Permian-Triassic boundary: new biostratigraphic, palaeobiogeographical and geochemical data in Caucasus and eastern Europe ZAPALSKI, M. K., HUBERT, B. & MISTIAEN, B. Estimation of palaeoenvironmental changes: can analysis of distribution of tabulae in tabulates be a tool? Index
vii 1
17 29 51 71
87 95 109
125 143 163 189
201
211 229 255
275 283
Fabric transitions from shell accumulations to reefs: an introduction with Palaeozoic examples J. J A V I E R / i ~ L V A R O 1,2, M A R K U S A R E T Z 3, F R t ~ D I ~ R I C B O U L V A I N 4, A X E L M U N N E C K E 5, D A N I E L V A C H A R D 2 & E M M A N U E L L E V E N N I N 6
1Departamento Ciencias de la Tierra, Universidad de Zaragoza, 50009 Zaragoza, Spain 2Laboratoire LP3, UMR 8014 du CNRS, UniversitO de Lille I, 59655 Villeneuve d'Ascq, France (e-mail:
[email protected]) 3Institut fiir Geologie und Mineralogie, Universitiit zu K61n, Ziilpicher Strasse 49a, 50674 Kdln, Germany 4pktrologie skdimentaire, B20, Sart Tilman, Universitd de Likge, 4000 Likge, Belgium 5Institut fiir Paliiontologie, Universitiit Erlangen-Niirnberg, Loewenichstrasse 28, 91054 Erlangen, Germany 6UMR 5561 CNRS, Biogkosciences, Universitk de Bourgogne, 6 bd. Gabriel, 21000 Dijon, France Abstract: One unresolved conceptual problem in some Palaeozoic sedimentary strata is the boundary between the concepts of 'shell concentration' and 'reef'. In fact, numerous bioclastic strata are transitional coquina-reef deposits, because either distinct frame-building skeletons are not commonly preserved in growth position, or skeletal remains are episodically encrusted by 'stabilizer' (reef-like) organisms, such as calcareous and problematic algae, encrusting microbes, bryozoans, foraminifers and sponges. The term 'parabiostrome', coined by Kershaw, can be used to describe some stratiform bioclastic deposits formed through the growth and destruction, by fair-weather wave and storm wave action, of meadows and carpets bearing frame-building (archaeocyaths, bryozoans, corals, stromatoporoids, etc.) and/or epibenthic, non-frame-building (e.g. pelmatozoan echinoderms, spiculate sponges and many brachiopods) organisms. This paper documents six Palaeozoic examples of stabilized coquinas leading to (pseudo)reef frameworks. Some of them formed by storm processes (generating reef soles, aborted reefs or being part of mounds) on ramps and shelves and were consolidated by either encrusting organisms or early diagenesic processes, whereas others, bioclastic-dominated shoals in barrier shelves, were episodically stabilized by encrusting organisms, indicating distinct episodes in which shoals ceased their lateral migration.
A paramount quantity of information characterizes the formation and distribution of shell concentrations and reefs in modern and Cenozoic strata, but the knowledge of Phanerozoic shell concentrations and reefs is decreasing when increasing in age. The Palaeozoic is a key time span in the history of life owing to the wide occurrence of skeletonized metazoans (the so-called 'Cambrian explosion': Zhuravlev & Riding 2000) and the successive biodiversifications related to: (i) major extinctions, such as the end-Ordovician (Cooper 2004), the Frasnian-Famenian boundary (Buggisch 1991; Copper 2002; Racki & House 2002) and the endPermian extinction (Wignall & Hallam 1992); and (ii) major community replacements, such as the Lower-Middle Cambrian transition
(Debrenne 1991; Zhuravlev 1995), the Cambrian-Ordovician transition (Barnes et al. 1996) and the so-called 'Mid-Carboniferous extinction' (Tappan & Loeblich 1988; R a y m o n d et al. 1990; Vachard & Maslo 1996). Shell concentrations (also named bioaccumulations, coquinas or lumachelles) are defined as 'relative dense accumulations of biomineralized animal remains with various amounts of sedimentary matrix and cement, irrespective of taxa composition and degree of post-mortem modification' (Kidwell et al. 1986). By contrast, the concept of reef is more complex: reefs can form as a result of microbial growth, mixed microbial-skeletal growth, complex microbially devoid biomineralized metazoan assemblages, or the accumulation of reef-builder
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. PalaeozoicReefs and Bioaccumulations: Climaticand Evolutionary Controls'.Geological Society, London, Special Publications, 275, 1-16. 0305-8719107l$15.009 The Geological Society of London.
2
J.J. ALVARO E T AL.
remains (Webb 1996). Following Wood's (1998) definition, reefs 'develop due to the aggregation of sessile epibenthic marine organisms, with the resultant higher rate of in-situ carbonate production than in surrounding sites'. Although the relative capacity of some gregarious epibenthic organisms bearing mineralized skeletons to be preserved in living position is a common characteristic of reef organisms, this character is also shared with other non-reef metazoans adapted to soft substrates and forming bundles-like clusters. One group of these organisms, named 'secondary soft-bottom dwellers' by Seilacher (1984), differs from mud stickers (e.g. non-encrusting pelmatozoans and sponges) in the heavy weight of their shells. These metazoans, mainly brachiopods and molluscs, develop common cup-, boulder- and fan-shaped convergent morphologies favouring recliner strategies (Seilacher 1984; Savazzi 1999). Skeletal concentrations can be subdivided, according to their biostratinomic features, into biogenic, sedimentary, diagenetic and mixed concentrations (Kidwell et al. 1986). Biogenic concentrations can be intrinsic or extrinsic in character, the former generated by the organisms that produce the hard parts resulting from intrinsic gregarious behaviours of autochthonous and parautochthonous skeletal organisms (e.g. preferential colonization by larvae, and single colonization events of opportunistic species), and the latter produced by other organisms that interact with skeletonized organisms on their discarded hard parts leading to the formation of parautochthonous and allochthonous concentrations (e.g. hard-part-rich fecal masses and shell-filled pits). Sedimentary concentrations result from physical (usually hydraulic) processes of concentration, in which hard parts behave as sediment particles and non-bioclastic matrix is either reworked or fails to accumulate (e.g. shelly storm lags, aeolian beach pavements, channel lags in fluvial, intertidal and subtidal environments or shell-paved turbidities). Diagenetic concentrations form or are significantly enhanced by processes acting after burial, such as compaction or selective dissolution of matrix. And, finally, mixed concentrations form by the interplay of two or more kinds of the aforementioned processes, and can display both autochthonous, parautochthonous and allochthonous shell assemblages. One unresolved conceptual problem in some of these mixed shell concentrations is the boundary between the concepts of 'shell accumulation' and 'reef'. Some of them are transitional coquina-reef deposits, because either distinct
flame-building skeletons are not commonly preserved in growth position, or skeletal remains are episodically encrusted by 'stabilizer' (reef-like) organisms, such as calcareous and problematic algae, encrusting microbes, bryozoans, foraminifers and sponges. Some of these centimetre- to metre-thick, mixed shell assemblages are named biostromes, and are described as tabular or sheetlike, bioclastic beds. Biostromes are built up by alternating gregarious settlement and hydraulic reworking leading to multiple events of hard-part concentrations. Originally, the term biostrome was coined by Cumings (1932), who defined it as a stratiform bed composed 'mainly or exclusively of shell remains', so that it included the modern concepts of shell accumulation and reef. This was subsequently constrained by Kershaw's (1994) concept of 'parabiostrome', where the constructing organisms are less than 20% in place. Therefore, the (para)biostromes formed through the growth and destruction, by fair-weather wave and storm wave action, of meadows and carpets bearing flame-building (archaeocyaths, bryozoans, corals, stromatoporoids, etc.) and/or epibenthic, non-flame-building (e.g. pelmatozoan echinoderms, spiculate sponges, and many brachiopods) organisms. Another mixed shell accumulation results from the colonization of shell accumulations by bottom-dwelling, substrate-dependent organisms. This process, named 'taphonomic feedback' by Kidwell & Jablonski (1983), allows a better understanding of the influence of hard parts on the ecological success of living benthos. The reworking of skeletal material by waves and storms commonly provides scattered hard shells in otherwise soft-bottom habitats. This can facilitate colonization and reproductive success by species that require or prefer hard substrates (mainly epibenthic suspension feeders), and at the same time inhibit earlier species that can tolerate only the initial soft-bottom conditions (Kidwell 1991; Kidwell & Bosence 1991). In some cases, preferential colonization by shelled benthos of isolated shells, on otherwise disadvantageously soft sea floor, is the first stage of development in many reefs: these mixed accumulations are usually interpreted as reef soles and reef pioneer communities (Lecompte 1954). The aim of this paper is to outline the fabrics and geometries of Palaeozoic bioaccumulations and reefs, and to illustrate some transitions between both biosedimentary geometries. The role played by distinct encrusting organisms and early diagenetic cements, stabilizing the sea floor as a response to hydrodynamic fluctuations and palaeoecological relationships, is documented in detail.
PALAEOZOIC COQUINA-REEF TRANSITIONS
Palaeozoic shell concentrations Overlying the earliest occurrence of skeletonized microfossils in the Ediacaran and earliest Cambrian, shell concentrations are found in most Palaeozoic lithofacies, except in high-porosity sedimentary rocks (such as conglomerates and breccias) because of shell dissolution. Thickness, abundance, packing, internal fabric and fossil preservation are all affected by changes in sediment accumulation rate and vary locally (Kidwell 1991). The physical scale, frequency of occurrence and taphonomic attributes of shell beds in different sedimentary environments are controlled by intensity and frequency of storms, background current and wave agitation, and diagenetic processes (Aigner 1985; Brett & Baird 1986; Speyer & Brett 1988; Kidwell 1991; Speyer 1991) Patterns of shell accumulation varied over Palaeozoic time because of changes in the diversity and environmental distribution of skeletonized organisms and those that interact with skeletal parts. Kidwell (1990) recognized two different modes of shell concentrations: archaic and modern modes. The archaic mode of shell concentration is represented in Palaeozoic and Triassic strata, and is characterized by relatively thin concentrations dominated by brachiopods and other epifaunal and semi-infaunal organisms (except for crinoidal calcarenites). In contrast, the modern mode is primarily CretaceousQuaternary in age, and contains thin pavements and full three-dimensional bioclastic concentrations dominated by molluscs and other epifaunal and fully infaunal organisms. Li & Droser (1992, 1997) further recognized distinct differences across the Cambrian shell beds, and between Cambrian and Ordovician shell beds. Cambrian shell beds are dominated by trilobites and occur as thin pavements, whereas Kidwell's (1990) archaic mode first really occurred during the Early Ordovician with the onset of the Ordovician biodiversification. After the Ordovician, the shell beds became more common, thicker and taxonomically diverse (Kidwell & Brenchely 1994). As a result, Li & Droser (1997) proposed to distinguish three types of shell concentrations, named Cambrian, post-Cambrian (but still Palaeozoic) and modern modes. In addition, although fluctuations in biodiversity of shell concentrations are primarily related to the appearance of new biomineralized clades, they are also controlled by the ecospace occupation: Cambrian faunas utilized relatively little ecospace, occupying mostly epifaunal guilds, and subsequently diversified during the rest of the
3
Palaeozoic in association with a distinct diversification in infaunal and pelagic guilds (Bottjer et al. 1996).
Palaeozoic reef geometries and their nomenclatural problem The variety and complexity that make reefs and mounds so interesting is also responsible for long-lasting problems with concepts and definitions. It is not the aim of this introduction to review reef classifications, but to give some working definitions of the most frequently used terms. Recent and extensive reviews by James & Bourque (1992), Bosence & Bridges (1995), Wood (1998, 2001), Kiessling et al. (2002), Riding (2002), and Schlager (2003), and give full access to nature and characteristics of reefs and mounds. A major distinction between reefs (sensu stricto) and other kinds of buildups was highlighted by Lowenstam (1950): 'reefs and 'ecological reefs' of Dunham (1970) are organically produced wave-resistant topographic structures'. This concept was further developed by Heckel (1974), in which he proposed that reefs are buildups which display evidence of potential wave resistance or growth in turbulent water and of control over the surrounding environment. Although these characteristics may be relative and difficult to be established in ancient buildups, this definition satisfied a need to give a special status to reefs. Other buildups, usually lime-dominated, were grouped under the term 'mound' or 'mud mound'. This term was popularized by Wilson (1975), but attributed to so many different examples of buildups that it lost parts of its meaning. This probably led James (1978) to propose the term 'reef mound' for quiet-water buildups growing below the fairweather wave base, rich in poorly sorted lime mud. Later, James & Bourque (1992) simplified matter by referring to 'mounds'. They also suggested that there are three end members of mounds that would grade into reefs in a tetragonal diagram (Fig. 1). These three mound types are microbial or cryptalgal mounds, skeletal mounds and mud mounds. Microbial or cryptalgal mounds are formed by stromatolites and/or thrombolites; skeletal mounds (such also correspond to the 'biodetrital mounds' of Bosence & Bridges 1995) are those with organisms baffling, trapping or stabilizing mud; and mud mounds are those resulting from piling of lime mud with minor amount of benthic biota. Several case studies show that overlapping of this mound type is frequent as is vertical
4
J.J. J~LVARO E T AL. (1988) distinguished a kind of incomplete ecological succession, the so-called 'arrested reef successions', in which full climax stages are not formally attained. These pioneering stages are typical in environments in which strong external control limits the potential of reefs to reach the climax stage. Palaeozoic examples of coquina-to-reef transitions
Fig. 1. Schematic classification of reef and mounds, together with variation of some basic parameters. The dark arrow represents an evolution from mud mound to reef via skeletal mound and, possibly, cryptalgal mound due to bathymetric decrease.
evolution from one type to another, or from mound to reef (Textoris 1966; Bourque & Gignac 1983; Boulvain 2001). This mound-reef transition is usually related to shallowing (Hoffman & Narkiewicz 1977; Boulvain 2001) rather than to ecological maturation (Walker & Alberstadt 1975). It is accompanied by textural evolution, community replacements (Fagerstrom 1991) and change in the dominant early diagenetic process, from organomineralization in mounds to cementation in reefs (Neuweiler et al. 1999; Schlager 2003). The wide diversity of patterns preserved in reefs allow the application of Root's (1967) concept of community guilds, each consisting of several species competing for the same 'class of environmental resources'. These are called constructor, baffler, binder, destroyer (borers, raspers, etc.) and dweller (passive members). The three first guilds build the rigid framework of reefs at a rate of upwards accretion that exceeds the rate of deposition of the adjacent levelbottom sediments to give the reef positive topographic relief (Fagerstrom 1988). If the vertical succession of different kinds of shell concentrations is commonly related to community replacements, which are directly controlled by changing external factors (Copper 1988), reef systems can display radial facies changes associated with intrinsic ecological successions. An ecological succession is 'an orderly, directional, and predictable, pioneer-to-climax process of community and species development' (Copper 1988). Ecological successions take place in environments where external physicochemical constraints (responsible for the aforementioned community replacements) are not undergoing major changes. In addition, Copper
This section documents six examples of stabilized coquinas. Some of them formed by storm processes (generating reef soles, aborted reefs or being part of mounds) on ramps and shelves and were consolidated by either encrusting organisms or early diagenesic processes, whereas others bioclastic-dominated shoals in barrier shelves - were episodically stabilized by encrusting organisms, marking distinct episodes in which shoals ceased their lateral migration. The presence of microbial mats (commonly called stromatolites and thrombolites) is conspicuous, coating numerous erosive discontinuities where the crusts commonly fossilize substrate irregularities enhanced by skeletal fabrics. Where these erosive discontinuities are repeated vertically the episodic growth of microbial mats can be used as a time record of when they interrupted background-sedimentation patterns, which are characterized by the amalgamation of high-energy events that involved repeated shell accumulations. The biological response of microbial communities to coat stratigraphic discontinuities can be considered as an integral part of dynamic stratigraphy, as they enhance the preservation and identification potential of interruptions in the background sedimentation on substrates devoid of burrowing activity. A semantic problem is related to the use of apparent synonym words related to microbial fabrics, such as 'thrombolite', 'thromboid' or 'clotted fabric'. Kennard (1994) followed the general terminology proposed by Kennard & James (1986), except for the replacement of 'mesoclots' with 'thromboids' (Shapiro 2000, p. 168): 'I recommend that the term 'thromboid' should not be used because 'mesoclot' is a suitable word for meaning the mesostructural elements, and the similarity of 'thromboid' with 'stromatoid' (macrostructural form of a stromatolite in the revision of Grey, 1989) will be misleading' (Shapiro 2000, p. 169). As explained in this work and those included within this Special Publication, this nomenclatural problem is still open to discussion according to the experience of each research worker.
PALAEOZOIC COQUINA-REEF TRANSITIONS
Cambrian shell concentrations stabilized by microbial mats One example of successive shell accumulationmicrobial crust alternations is illustrated by the amalgamation of: (i) millimetre- to centimetrescale, high-energy sedimentary events, characterized by deposition of broken to disarticulated shells; (ii) interrupted by episodes with extremely low sedimentation rates that led to development of centimetre- to decimetre-thick, microbial reefs (Fig. 2a). The case study reported here is taken from the Micmacca Breccia, a member of the lowermost Middle Cambrian Jbel Wawrmast Formation from the Moroccan Atlas that deposited on temperate-water substrates. The member consists of reddish-brownish, volcanoclasticbearing, bioclastic limestones that alternate with shale or sandstone strata. Each limestone, up to 1.2 m thick, is composed of amalgamated units (up to 0.4 m thick), separated by erosive surfaces, which show a vertical modification of textures:
5
the lenses that directly cover the scour surfaces show chaotic lithoclast orientations, passing upwards into packstones and local grainstones and brecciated volcanoclast-dominated levels (Fig. 2b), in some cases forming low-angle laminae. The shell accumulations display a wide diversity of trilobite debris, echinoderm ossicles, and subordinate heteractinid and hexactinellid sponge spicules, chancelloriid sclerites, brachiopods and helcionellids (fidvaro & Clausen 2006). The aforementioned limestone units are separated by stromatolitic crusts, which are up to 8 cm thick, and vary from wrinkled to domal and columnar in shape (Fig. 2c). The stromatolites grew perpendicularly to substrate surfaces, and generally followed their irregularities. Although the stromatolitic microfabric was largely the consequence of in situ calcite precipitation, there is evidence of litho- and bioclastic incorporation into the structures, in a manner analogous to many modern microbial mats generating agglutinated stromatolites (Riding 1999).
Fig. 2. (a) Field aspect of the Micmacca Breccia limestones with marked intrabed erosive discontinuities covered by packstone-dominated shell concentrations capped by microbial mats, Tarhoucht, Anti-Atlas. (b) Photomicrograph of a packstone with polyphase bioclasts replaced by iron oxides: e, echinoderm ossicle; ch, chancelloriid sclerite; t, trilobite sclerite, Lemdad valley, High Atlas; scale = 250 ~tm. (e) Wrinkled (wm) to domal (dm) stromatolites encrusting a shell concentration and topped by a wacke-packstone; Tarhoucht, Anti-Atlas; scale = 1 mm. (d) Trapping of bioclasts and volcanoclasts within a microbial mat, subsequently covered by a packstone, Lemdad valley, High Atlas; scales = 1 mm.
6
J.J./~LVARO ET AL.
Sediment consisting of quartz, feldspar, corroded micas, volcanoclastic sand- and silt-sized grains, and skeletons was episodically trapped and bound into the surface, increasing upwards in some crusts from 10 to 60% in volume (Fig. 2d). No evidence of grazing and boring organisms was found. The limestones of the Micmacca Breccia represent low-relief shoals that recorded abrupt changes in hydrodynamic patterns. These are indicated by the presence of centimetre- to decimetre-thick, microbial (stromatolitic) reefs that reflect successive interruptions of the background sedimentation, represented by the aforementioned high-energy coquinas.
Ordovician shell concentrations stabilized by encrusting bryozoans Late Caradocian-early Ashgillian echinodermbryozoan thickets and bundles characterize temperate-water substrates located on the northern margin of Gondwana. The Ashgillian, highlatitude, mixed substrates of the Lower-Ktaoua
Formation, in the Alnif area (eastern Anti-Atlas, Morocco), have recorded three distinct carbonate factories, displaying some examples of coquina-reef transition, primarily controlled by accommodation space, distance from source areas and subsequent siliciclastic input, and benthic-community replacements (Alvaro et al. 2006). The first benthic community is preserved on the foresets of siliciclastic shoal complexes. It is volumetrically dominated by brachiopods and non-pelmatozoan echinoderms, whereas robust bryozoans and phosphate-shelled brachiopods are secondary (Fig. 3a, b). The brachiopods are dominated by heterorthids and drabovids, which are epibenthic, fixosessile and plenipedunculate, and rafinesquinids interpreted as ambitopic, liberosessile juvenile brachiopods capable to change into quasi-endobenthic adult stages. This brachiopod assemblage was apparently adapted to sandy substrates, where heterorthids and drabovids with highly developed diductor muscles were able to prevent the sharp closure of their valves in turbulent waters, and
Fig. 3. (a) Carbonaceous siltstone of the Lower-Ktaoua Formation rich in calcite-walledbrachiopods and fragile-ramified bryozoans paralleling the inclined foreset of a shoal; scale= 2 mm. (b) A robust-ramified bryozoan inclined to the foreset plane; scale = 2 mm. (c) Non-pelmatozoan echinoderm with flat base parallel to the foreset plane; scale = 2 cm. (d) Distal phosphatic limestone (top arrowed) with alternation of encrusting bryozoans (eb) and brachiopod-dominated tempestites (b); scale= 5 cm.
PALAEOZOIC COQUINA-REEF TRANSITIONS rafinesquinids would have been able to change their position on unstable substrates. Sphaeronitid and aristocystitid echinoderms are also abundant (Fig. 3c). Despite their lack of encrusting strategies, they display their flat bases parallel to the foreset laminae, indicating episodic quiescent episodes in the shoal migration. As a result, this brachiopod-echinoderm benthic community marks episodes of decreased energy in prograding sandy shoals. The second benthic community, preserved in distal phosphatic limestones, is dominated by disarticulated to fragmented bryozoans and brachiopods (Fig. 3d). Common growth forms of bryozoans are robust and delicate branching, and multilaminar and unilaminar encrusting (nomenclature after Nelson et al. 1988). Carbonate- and phosphate-shelled brachiopods, trilobites, bivalves, ostracodes, ortoconic nautiloids and conulariids are locally abundant. The widespread variety of bryozoans indicates the establishment of bryozoan thickets that helped to stabilize episodically the substrate but without trapping significant amounts of mud. Finally, the top of the Lower-Ktaoua Formation is recognizable as a phosphatized coquina, which contains the peak in biodiversity patterns of the whole studied succession. It includes trilobites, brachiopods, bivalves, conulariids, orthoconic nautiloids, disarticulated pelmatozoan stems and scattered encrusting bryozoans. Brachiopods consist of fixosessile, plenipedunculate heterorthids and rhynchonellids, as well as the liberosessile ambitopic, quasi-infaunal xenambonitid Aegiromena aquila aquila. The condensed level also contains abundant trilobite debris derived from calymenids, dalmanitids, cheirurids and illaenids. Other taxa are sinuitid, eotomariid and holopeid gastropods, and hyolithids. This phosphatized coquina indicates development of condensation associated with marine depositional hiatuses, which allowed repeated early diagenetic phosphate cementation and encrustation made up by bryozoans.
Silurian shell concentrations stabilized dur&g early diagenesis The strata of Gotland (Sweden) represents an extraordinary well-preserved example of tropical carbonate platform development during Silurian times (late Llandovery-late Ludlow), and reflects a series of stacked carbonate platforms (Calner et al. 2004). On Gotland, a huge variety of carbonate facies is developed, from extremely shallow-water deposits mostly on the eastern part of the island to open-marine shelf deposits dominating the western part (Samtleben et al.
7
1996, 2000). Accordingly, the variety of different shell accumulations is very remarkable. However, a common characteristic of most of these deposits is that cementation occurred early in diagenetic history. Except for some crinoidal limestones, most coarse-grained limestones on Gotland do not show any fitted fabrics (Fig. 4b, c). This indicates that the original pore space was filled by calcite spar prior to mechanical compaction (cf. Bathurst 1995). Stylolites, which according to Bathurst (1991) are defined as 'serrated interface between two masses of rock', are common, and represent structures developed during late diagenesis. However, stylolites form only on sediment that has been lithified before. An early diagenetic lithification is also typically observed in fine-grained limestones from Gotland, where undeformed trace fossils and organic-walled microfossils document an early stabilization (= cementation) of the original carbonate mud (Munnecke & Samtleben 1996; Westphal & Munnecke 1997). These observations point to one of the fundamental questions in carbonate petrology (Bathurst 1970): how to cement a carbonate sediment while it is still largely uncompacted. Where was the source of such an enormous quantity of CaCO3? On Gotland, most of the section is developed as alternation of limestones and marls, which exhibit a wide morphological variety from nodular marl-dominated alternations to well-bedded limestone-dominated alternations. According to Munnecke & Samtleben (1996), lithification of the limestones in these alternations took place just below the sea floor in the shallow-marine burial realm. The source of the carbonate cement lithifying the micritic limestones is a selective dissolution of aragonitic constituents in the intercalated marl layers. Consequently, the marls are not cemented and represent a residual sediment depleted in aragonite. Owing to ongoing sedimentation, the marl layers were more and more compacted. But, what about coarse-grained shell accumulations? Where is the source for their carbonate cement? Bathurst (1995) pointed out that 'the style of pressure dissolution in pure limestones, fitted-fabric or stylolite, depends strongly on the availability of aragonite in the original sediment'. This, however, was questioned by Railsback (1996). But how can we prove the former existence of aragonite as source for the carbonate cement when it is dissolved during diagenesis? And when - that is, in which diagenetic environment - does aragonite dissolution occur? Cherns & Wright (2000), who compared early silicified faunas from Gotland with non-silicified faunas, show that aragonite dissolution must
8
J.J./kLVARO E T AL.
Fig. 4. All thin sections come from the Silurian of Gotland, Sweden. (a) Irregular hard ground with thin pyrite-enriched layer, and reworked intraclast in the overlying sediment (Slite Group, Wenlock, proximal shelf facies, locality Hagan~s 1) ; scale = 1 cm. (b) Reworked intraclast (arrow) with sharply eroded bryozoan (left) and crinoid (right) debris. The fact that matrix as well as biogenic components are sharply truncated indicate that the reworked layer was already completely lithified prior to erosion (?H6gklint Formation, locality Gutev~igen); scale = 1 mm. (c) Shell accumulation dominated by bivalve shells, partly with geopetal fillings. All components show micritic envelopes, which is a common feature in shallow, warm water. The bivalve shells are completely recrystallized to clear, blocky calcite spar (Halla Formation, marginal marine to lagoonal facies, locality near Gothemhammar 1); scale = 2 mm. (d) Vertically stacked bivalve and brachiopod shells embedded in a fine peloidal-grainstone matrix. The sharp boundaries of the bivalve shells indicate that dissolution of the aragonitic shells occurred after cementation of the matrix. Interestingly, the bivalve shells are partly filled with fine-grained sediment. This indicates that after the aragonite dissolution fine-grained material was flushed into the pore network filling primary (left arrow) and secondary pores (the dissolved shells). Sometimes the shells are filled geopetally (right arrow) (Halla Formation, marginal marine-lagoonal facies, locality near Gothemshammar 1); scale = 2 mm.
have happened very early in diagenetic history. Their finding is confirmed by our results: Figure 4d illustrates that aragonite dissolution in the limestones of Gotland must have occurred very early, but after cementation. The sharp boundaries of the former aragonitic bivalve shells indicate that the first step in diagenesis was the lithification of the matrix. Later, the shells were dissolved and partly refilled with fine-grained sediment. This must have taken place where soft mud was still available, i.e. close to the former seafloor. The bivalve shells themselves, however, cannot be the source for the cementation of the matrix because: (a) the limestone was cemented
prior to dissolution of the aragonite shells; and (b) the amount of CaCO 3 provided by this process would be far too low. So, a source outside the limestone bed has to be considered (see above). An early cementation of the Silurian rocks on Gotland is also documented by the occurrence of abundant hard grounds and reworked intraclasts, mainly in proximal environments (Fig. 4a, b) (Samtleben et al. 2000). Hard grounds are sedimentary surfaces that have existed as hardened sea floor prior to the overlying sediment (Flfigel 2004). In general, hard-ground formation can be explained in two different ways:
PALAEOZOIC COQUINA-REEF TRANSITIONS (i) by an interruption in sedimentation resulting in cement precipitation directly from sea water; and (ii) by an exhumation of a limestone bed that was lithified close to the surface, i.e. a few decimetres or metres below the sea floor. Such sedimentary depth is assumed for the cementation of the limestones in limestone-marl alternations (Munnecke & Samtleben 1996). Changes in the local hydrodynamic regime can easily result in erosion of the topmost part of the sediment, which consists of soft, unlithified mud, and thus resulting in an excavation of the early cemented limestone layer. As long as the water energy is high enough to prevent accumulation of new carbonate sediment this layer represents a hardened sea floor, and might be either encrusted by hard-bottom communities, or corroded or bored by chemical processes or biological activities (bioerosion). As the internal character of hardground beds is indistinguishable from 'normal' limestone layers on Gotland, and no traces of early marine cements have been found (Fig. 4a), this second process is assumed to be the common process in hard-ground formation (Munnecke 1997). A similar process is proposed for the formation of so-called 'hiatus concretions' (Voigt 1968). It is more and more accepted that aragonite dissolution and early cementation is a very important process in early marine carbonate diagenesis, even in non-tropical carbonates (Munnecke & Samtleben 1996; Cherns & Wright 2000; Melim et al. 2002; James et al. 2005; Munnecke & Westphal 2005; Dix & Nelson 2006). However, there is an ongoing discussion as to whether the bulk of the carbonate cement is provided by dissolution of mud-sized aragonite derived from shallow carbonate platforms or by dissolution of benthic molluscs (Cherns & Wright 2000; Wright & Cherns 2004; see discussion in Munnecke & Westphal 2005).
Nebuloids: gel-stabilized shell concentrations & Devonian carbonate mounds Late Frasnian Petit-Mont Member mounds occur in the southern part of the Dinant Synclinorium and in the Philippeville Anticline (SW Belgium). These mounds are 30-80 m thick and 100-250 m in diameter. They are embedded in shale, nodular shale and argillaceous limestone. Mound growth typically initiated from below the photic and storm wave base zones and continued into shallow-water environments. Above an argillaceous limestone substrate, the
9
first mound facies consists of spiculitic wackestone with stromatactis, which becomes progressively enriched in crinoids and corals, then in peloids, stromatoporoids and cyanobacteria (named Pm3 in Boulvain 2001). The shallower facies consists of algal-coral-peloid wackestone and packstone with green algae and thick algal coatings. A core of algal and microbial bindstone sporadically occurs within large mounds (Boulvain 2001). Of particular interest is the pink limestone with corals, crinoids, stromatactis, fenestrae and lamellar stromatoporoids (Pm3): this facies marks the entrance of the mounds into the storm wave zone. Corals are generally tabular
(Alveolites,
Phillipsastrea,
Thecostegites),
branching (Thamnopora, Senceliaepora) or fasciculate (Thamnophyllum); solitary rugose corals are also present. Receptaculites is locally abundant. Cyanobacteria form partial coatings around particles, and peloids are common and irregular. This muddy facies includes enigmatic structures consisting of decimetre-scale pockets or decimetre-thick beds of dark grey fibrous cement containing sorted brachiopods and crinoids (Fig. 5b, c). These particular structures (called 'nebuloids' by Boulvain 2001) may pass laterally, by reduction in the proportion of spar, into a network of centimetre-scale stromatactis or fenestrae. In some mounds, the nebuloids are laterally uninterrupted over tens of metres and show a rhythmic pattern consisting of decimetrethick spar networks separated by layers of pink limestone (Fig. 5a). In these large-scale structures the fossil content changes laterally: brachiopods are prominent in the central part of the mounds, although crinoids are more abundant close to the flanks. This reflects the usual lateral zonation of the mounds at this growth stage. In thin section, the fibrous cement corresponds to isopachous crusts of radiaxial calcite, with micron-sized inclusions of dolomite, and are very rich in organic matter (Fig. 5d). Internal sediment is very rare, although it is common in other types of fenestrae. Sorting of bioclasts, rhythmic pattern, local hummocky cross-stratification and position of this facies in the shallowing-upwards mounddevelopment sequence indicate that these nebuloid structures are formed by storms. Periodic increasing of water agitation was responsible for lime mud erosion and the concentration of bioclasts in beds or pockets. After deposition, fossils were cemented by early fibrous cement. It is, however, difficult to date the precipitation of the fibrous spar. Was it truly synsedimentary, before settling of the muddy interlayers, or did it postdate mud accumulation? Several lines of
10
J.J. J~LVARO ETAL.
Fig. 5. All pictures come from the Late Frasnian Petit-Mont Member mounds, Philippeville Anticlinorium, Belgium. (a) Rhythmic succession of nebuloids in the middle part of the Tapoumont mound; So = bedding. (b) Nebuloid structures (grey) in reddish limestone; pencil (arrowed) for scale; Les Bulants mound. (e) Close-up of a nebuloid, cemented by dark organic matter-rich radiaxial spar; a Receptaculitesneptuniis included in the spar to the right of the picture; coin (50 eurocents) for scale. (fl) Thin section in a nebuloid, showing dark organic matter inclusions (arrowed) and crinoid ossicles; crossed nicols, Tapoumont quarry; scale = 2 mm.
evidence suggest that cement was precipitated into a gel-like microbial mat stabilizing the storm layer: first, the unusual richness of the radiaxial cement in organic matter; and, second, the lateral transition of the nebuloids into mud with stromatactoid fenestrae. This could be explained by progressive lateral reduction of the mat, allowing mud to settle between the bioclasts. Unusual absence of internal sediment could also be explained by the presence of an organic gel.
Carboniferous shell concentrations stabilized by microb&l communities Shell accumulations are common features of many Carboniferous carbonate platforms. Some of these concentrations display important lateral extension, and thus are of stratigraphic importance; for example, the Upper Vis6an Orionastrea bed of England (Cossey et al. 2003). The composition of shell beds varies significantly in time and space. 'Encrinites', which are limestones of
various textures dominated by crinoid debris, are ubiquitously distributed (e.g. in the so-called 'Petit-Granit' in Belgium). Many Mississippian shell beds are macroscopically composed of brachiopods and/or corals. The compositional complexity of Pennsylvanian shell beds is similar to that of the older counterparts, but chaetetid sponges and calcareous algae play a more prominent role. The case studies described herein are Mississippian in age, and it is our aim to provide examples of the involvement of microbial communities into the hydrodynamic context of these shell beds. The first example is taken from the Middle Vis~an Lives Formation of the Namur Syncline (Belgium). The formation consists of a set of shallowing-upwards parasequences (Chevalier & Aretz 2005). An ideal parasequence starts with subtidal marine bioclastic packstones-grainstones, passing upwards into stromatolitic boundstones or peloidal mudstones of intertidal-supratidal environments, finally
PALAEOZOIC COQUINA-REEF TRANSITIONS capped by palaeosols. In one of the parasequences, a decimetre-thick bed rich in brachiopods, conventionally named the Composita bed, marks the transition from bioclastic limestone to lime mudstone. Without record of major macroscopical changes, the bed can easily be laterally traced over several kilometres. However, a microfacies analysis reveals a very heterogeneous composition of the bed: texture and allochem composition differ over short distances, with most of the bed grading between bioclastic grainstone (Fig. 6a) and oncoidal packstone (Fig. 6b). Microbial coating, resulting in advanced stages in oncolite formation, is ubiquitous in the bed, although increasing vertically in abundance and thickness. However, although reworking seems to be an important process in the shell bed formation, this trend may be partly obscured. The microbial coatings are important for the stabilization of the mobile substratum in which the shell bed formed. The development of proper boundstone fabrics is not observed. The transitional character of the shell bed fabric is
11
further expressed by the formation of small patch reefs, which initiated upon this bed (Chevalier & Aretz 2005). However, the reef fabrics show fundamental compositional and structural differences compared to the shell bed facies. The second example is taken from the Upper Visdan of the Montagne Noire (southern France). In the quarry of Castelsec, a small coral patch reef is topped by three well-bedded units, which are composed of partly argillaceous bioclastic limestones (Aretz & Herbig 2003). The lowest of these units is 2.8 m thick and comprises abundant small clisiophyllid and axophyllid rugose corals. Associated with the small corals are productid brachiopods, and larger and/or colonial rugose corals. All small rugose corals are coated with microbial crusts, which are differentiated into two crust morphotypes'. (i) crusts forming through incrustation of coral skeletons, thus resulting in a more circular geometry; and (ii) crusts of more aligned habits, which do not necessarily encircle corals, but seem to bound a number of previously encrusted (?) corals
Fig. 6. (a) Bioclastic-dominated part of the Compositabed; larger bioclasts are mainly brachiopods and gastropods; note the presence of thin microbial coatings surrounding some clasts; Compositabed (Middle Visdan), Engihoul Quarry ; scale = 2 mm. (b) Oncoidal grainstone; note the high diversity of encrusted bioclasts; Compositabed (Middle Visdan), Engihoul Quarry; scale = 2 mm. (c) Inner circular crust (morphotype 1) around a small solitary rugose surrounded by more aligned crusts (morphotype 2); outcrop picture, Castelsec Quarry, Montagne Noire; scale = 5 mm. (d) Complex microbial-dominated crust on coral skeleton; note the incorporation of grains; thin section from below the shell bed, Castelsec Quarry, Montagne Noire; scale = 1 mm.
12
J.J./i.LVARO E T A L .
together. Both morphotypes are well preserved on weathered surfaces (Fig. 6c), but crusts of the first morphotype seem more abundant, as parts of the unit tend to develop a nodular appearance. The composition of these microbial crusts is complex, and involves dark micritic layers, calcimicrobes, trapped carbonate particles and sparry cement layers. The importance of the microbial crusts for shell bed formation is their bounding activity, which results in a wellpreserved coral rubble, and the homogenous appearance of the shell bed. However, the relative smallness of the corals indicates an originally somewhat unfavourable environment for corals. The abundance and composition of microbial crusts in this shell bed is very similar to other deposits in the Montagne Noire (Fig. 6d), but the failure to develop a substantial reef development, as seen in other outcrops (Aretz & Herbig 2003), remains enigmatic. It is important to remark that microbial communities played a key role in the formation and preservation of many Mississippian shell beds, although they seem to be absent in others. Therefore, the functional role of microbial communities, which is mainly stabilization through binding, either remains open or is fulfilled by other encrusting organisms, such as bryozoans and chaetetid sponges. However, shell beds with an important microbial community commonly tend to be more compact and massive than other shell beds.
Permian shell concentrations stabilized by richthofenid brachiopds, calcisponges and bryozoans In the Mixteco Terrane (SW Mexico), diverse Permian strata onlap the Acatlfi.n metamorphic substrate as a result of a Devonian (probably Ligerian) suture. The Middle-Upper Permian Olinal~i Formation (Guerrero state, Mexico) consists of three (lithostratigraphically unnamed) successions of, in ascending order, siliciclastic, carbonate and siliciclastic character (Flores de Dios & Buitr6n 1982; Vachard et al. 2004). The lower siliciclastic succession is Wordian in age, and is composed of a lower massive conglomerate, a middle black shale bearing middle-Wordian goniatites (Vachard et al. 2004), and an upper palaeodelta complex with terrestrial plants and brachiopods (Buitr6n et al. 2005). The overlying carbonate-dominated succession can also be subdivided into three levels: a lower stromatolitic level (Flores de Dios et al. 2000), probably early Capitanian in age, a middle part rich in channelized accumulations of giant
fusulinids (e.g. Polydiexodina capitanensis) and an upper part containing mud mounds, dated as late Guadalupian based on the presence of the fusulinid Codonofusiella extensa (Vachard et al. 1993); its genus is commonly Lopingian (=Late Permian) in the other parts of the world. The third siliciclastic succession consists of shales with scattered indeterminate, but apparently Permian, ammonoids (Flores de Dios & Buitr6n 1982). If the mud mounds of the Olinalfi. Formation are late Capitanian in age, and grew up into the Lopingian, they would represent the youngest Permian evidences of North American reefs. As in the Lopingian reefs of South China (e.g. Rigby et al. 1989; Weidlich 2002), the Olinal~ mud mounds are built in particular by calcisponges and richthofeniid brachiopods. These deposits are first composed of crinoid rudstones with rare Polydiexodina fusulinids, the whole fabric recording current structures and channel fillings. The richthofeniid and encrusting productid brachiopds occur directly attached to the surface of these encrinites (Fig. 7a, b). The space between the richthofenids is filled with micrite and/or crinoidal wackestone with thin carbonate cements (Fig. 7a, b). This benthic assemblage is subsequently replaced by sphinctozoan calcisponges (Fig. 7c) and encrusting bryozoans, in association with the aforementioned richthofeniids. At the top of the mud mounds, some stromatolites cap the buildups (Fig. 7d), which were finally drowned and buried by ammonoidbearing shales. In summary, the Olinal/t mud mounds display a vertical evolution that started by an accumulation of crinoidal ossicles, subsequently stabilized by richthofeniid brachiopods, secondarily replaced by sphinctozoan calcisponges and bryozoans, and finally capped by microbial mats.
Concluding remarks The occurrence of new biomineralized metazoans during the Palaeozoic, and their synecological relationships with microbial communities, have greatly increased the concepts and nomenclature necessary to understand the evolution of benthic communities through this key time span. Although the concepts of reef and shell accumulation are clear, numerous transitional coquina-reef deposits reflect fluctuations in hydrodynamic conditions and palaeoecological relations leading, in some cases, to the episodic stabilization of epibenthic non-flamebuilding meadows and carpets, and reef communities. Some of these skeletal-rich geometries
PALAEOZOIC COQUINA-REEF TRANSITIONS
13
Fig. 7. (a) Two ventral richthofeniid valves, probably Cyclacantharia sp., directly attached to an encrinitic or bioclastic, slightly sandy, wackestone; note the numerous vesicles; Late Capitanian. (b) Another contact, although slightly modified by a small stylolite, of an encrinitic substrate of well-sorted ossicles and richthofenids; Late Capitanian. (c) Sphinctozoan calcisponges and encrusting bryozoa, associated with richthofeniids. (d) Stromatolites (top right) participating in the constructions with calcisponges (bottom). All samples from the Olinalfi section, Guerrero, Mexico; Late Capitanian.
can be described using Kershaw's concept of 'parabiostrome', which takes into account the percentage of constructing organisms preserved in living position.
References AIGNER, T. 1985. Storm depositional systematics: Dynamic stratigraphy in modern and ancient shallow marine sequences. In: FRIEDMAN, G. M., NEUGEBAUER, H. J. & SEILACHER, A. (eds) Cyclic and Event Stratification. Lecture Notes in Earth Sciences, 3, 1-174. ALVARO, J. J. & CLAUSEN, S. 2006. Microbial crusts as indicators of stratigraphic diastems in the Cambrian Micmacca Breccia, Moroccan Atlas. Sedimentary Geology, 185, 255-265. /~LVARO,J. J., VENNIN,E., VILLAS,E., DESTOMBES,J. & VIZCMNO, D. 2006. Pre-Hirnantian (latest Ordovician) benthic community assemblages: controls and replacements in a siliciclastic-dominated platform of the eastern Anti-Atlas, Morocco.
Palaeogeography, Palaeoelimatology, Palaeoecology. (in press) doi: 10.10161j.palaeo.2005.09.035 ARETZ, M. & HERBIG, H. G. 2003. Contributions of rugose corals to Late Vis6an and Serphukovian bioreconstructions in the Montagne Noire (southern France). In: AHR, W. A., HARRIS, P. N., MORGAN, W. A. & SOMERVILLE,I. D. (eds) PermoCarboniferous Carbonate Platforms and Reefs. Society of Economic Paleontologists and Mineralogists, Special Publications, 78, American Association of Petroleum Geologists Memoirs 83, 119-132. BARNES, C. R., FORTEY, R. & WILLIAMS, S. H. 1996. The patterns of global bio-events during the Ordovician Period. In: WALLISER,O. H. (ed.) Global Events and Event Stratigraphy. Springer, Berlin, 139-172. BATHURST, R. G. C. 1970. Problems of lithification in carbonate muds. Geologist's Association Proceedings, 81,429-440. BATHURST, R. G. C. 1991. Pressure-dissolution and limestone bedding: the influence of stratified cementatio. In: EINSELE, G., RICKEN, W. & SEILACHER, A. (eds) Cycles and Events in Stratigraphy. Springer, Berlin, 450-463.
14
J.J./~LVARO ETAL.
BATHURST, R. G. C. 1995. Burial diagenesis of limestones under simple overburden. Stylolites, cementation and feedback. Bulletin de la SociOtO gkologique de France, 166, 181-192. BOSENCE, D. W. J. & BRIDGES, P. 1995. Origin and evolution of carbonate mud-mounds. In: MONTV, C. L. V., BOSENCE, D. W. J., BRIDGES, P. H. & PRATT, B. R. (eds) Carbonate Mud-mounds: Their Origin and Evolution. International Association of Sedimentologists, Special Publications, 23, 3-9. BOTTJER, D. J., SCHUBERT,J. K. & DROSER, M. L. 1996. Comparative evolutionary palaeoecology: assessing the changing ecology of the past. In: HART, M. B. (ed.) Biotic Recovery From Mass Extinction Events. Geological Society, London, Special Publications, 102, 15-39. BOULVAIN, F. 2001. Facies architecture and diagenesis of Belgian late Frasnian carbonate mounds (Petit-Mont Member). Sedimentary Geology, 145, 269-294. BOURQUE, P. A. & GIGNAC, H. 1983. Spongeconstructed stromatactis mud mounds, Silurian of Gasp6, Qu6bec. Journal of Sedimentary Petrology, 53, 521-532. BRETT, C. E. & BAIRD, G. C. 1986. Comparative taphonomy: a key to paleoenvironmental interpretation based on fossil preservation. Palaios, 1, 207-227. BUGGISCH, W. 1991. The global Frasne/Famenne 'Kellwasser-Event'. Geologische Rundschau, 80, 49-72. BUITRON, B. E., SILVA-PINEDA,A., FLORESDE DIOS, A. & VACHARD, D. 2005. New Permian macrofauna and macroflora from the Olinaki region, Guerrero State, Mexico. Annales de la SociOt~ Gkologique du Nord, 11, 169-176. CALNER, M., JEPPSSON, L. & MUNNECKE,A. 2004. The Silurian of Gotland - Part I: Review of the stratigraphic framework, event stratigraphy, and stable carbon and oxygen isotope development. Erlanger geologische Abhandlungen, Sonderband 5, 113-131. CHERNS, L. & WRIGHT, V. P. 2000. Missing molluscs as evidence of large-scale, early skeletal aragonite dissolution in a Silurian sea. Geology, 28, 791-794. CHEVALIER, E. & ARETZ, M. 2005. A microbebryozoan reef from the Middle Vis6an of the Namur Syncline (Engihoul Quarry). Geologiea Belgica, 8, 109-119. COPPER, P. 1988. Ecological succession in Phanerozoic reef ecosystems: is it real? Palaios, 3, 136-152. COPPER, P. 2002. Reef development at the Frasnian/ Famennian mass extinction boundary. Palaeogeography, Palaeoclimatology, Palaeoecology, 181, 27-65. COOPER, R. A. 2004. Measures of diversity. In: WEBBY, B. D., PARIS, F., DROSER, M. L. & PERCIVAL,I. G. (eds) The Great Biodiversification Event. Columbia University Press, New York, 52-60. COSSEY, P. J., ADAMS, A. E., PURNELL, M. A., WHITELEY, M. J., WHYTE, M. A. & WRIGHT, V. P. 2003. British Lower Carboniferous Stratigraphy. Joint Nature Conservation Committee. Geological Conservation Review Series, 29, 1-617. CUMINGS, E. R. 1932. Reefs or bioherms? Bulletin of the Geological Society of America, 43, 331-352.
DEBRENNE, F. 1991. Extinction of the Archaeocyatha. Historical Biology, 5, 95-106. DIx, C. R. & NELSON, C. S. 2006. Diagenetic potential for lithification of cool-water carbonate shelf mud. Sedimentary Geology, 185, 41-58. DUNHAM, R. J. 1970. Stratigraphic reefs versus ecologic reefs. AAPG Bulletin, 54, 1931-1932. FAGERSTROM, J. A. 1988. A structural model for reef communities. Palaios, 3, 217-220. FAGERSTROM, J. A. 1991. Reef-building guilds and a checklist for determining guild membership. Coral Reefs, 10, 47-52. FLORES DE DIOS, A. & BUITRON, B. 1982. Revisidn y aportes a la estratigrafia de la Montafia de Guerrero. Universidad Autdnoma de Guerrero, serie tOcnico-cientifiea, 12, 3-28. FLORES DE DIOS, A., VACHARD, D. & BUITRON, B. 2000. La cubierta sedimentaria p&mica superior (Formaciones Olinalfi, Ihualtepec, Yododefie) de los terrenos Mixteco y Zapoteco; evoluci6n sedimentol6gica e interpretaciones paleogeogrfificas. Geos, 2a. Reuni6n de Ciencias de la Tierra, Resfimenes y Programa, 20, 324-325. FLOGEL, E. 2004. Microfacies of Carbonate Rocks Analysis, Interpretation and Application. Springer, Berlin. GREY, K. 1989. Handbook for the study of stromatolites and associated structures. In: KENNARD,J. M. & BURNE, R. V. (eds) Stromatolite Newsletter, 14, Bureau of Mineral Resources, Geology and Geophysics, Canberra, Australia, 82-171. HECKEL,P. H. 1974. Carbonate buildups in the geologic record: a review. In: LAPORTE, L. F. (ed.) Reefs in Time and Space. Society of Economic Paleontologists and Mineralogists, Special Publications, 18, 90-155. HOFVMAN,A. & NARKIEWICZ,M. 1977. Developmental pattern of Lower to Middle Paleozoic banks and reefs. Neues Jahrbuchfiir Geologie und Paliiontologie Abhandlungen, 5, 272-283. JAMES, N. P. 1978. Facies models 10. Reefs Geoscience Canada, 5, 16-26. JAMES, N. P. & BOURQUE, P. A. 1992. Reefs and mounds. In: WALKER, R. G. & JAMES, N. P. (eds) Facies Models. Response to Sea Level Change. Geological Association of Canada Geotext, 1,323-347. JAMES, N. P., BONE, Y. & KYSER, T. K. 2005. Where has all the aragonite gone? Mineralogy of Holocene neritic cool-water carbonates, southern Australia. Journal of Sedimentary Research, 75, 454-463. KENNARD, J. M. 1994. Thrombolites and stromatolites within shale-carbonate cycles, Middle-Late Cambrian Shannon Formation, Amadeux Basin, central Australia. In: BERTRAND-SARFATI,J. & MONTY, C. L. V. (eds) Phanerozoic Stromatolites II. Kluwer, Dordrecht, 443-471. KENNARD, J. M. & JAMES, N. P. 1986. Thrombolites and stromatolites: two distinct types of microbial structure. Palaios, 1,492-503. KERSHAW,S. 1994. Classification and geological significance of biostromes. Facies, 31, 81-92. KIDWELL, S. M. 1990. Phanerozoic evolution of macroinvertebrate shell accumulations: preliminary data from the Jurassic of Britain. In: MILLER, W., III (ed.) Paleocommunity Temporal Dynamics." The
PALAEOZOIC COQUINA-REEF TRANSITIONS
Long-term Development of Multispecies Assemblages. Paleontological Society, Special Publications, 5, 305-327. KIDWELL, S. M. 1991. Taphonomic feedback (live/dead interactions) in the genesis of bioclastic beds: key for reconstructing sedimentary dynamics. In: EINSELE, G., RICKEN, W. & SEILACHER,A. (eds) Cycles and Events in Stratigraphy. Springer, Berlin, 268-282. KIDWELL, S. M. & BOSENCE, D. W. J. 1991. Taphonomy and time-averaging of marine shelly faunas. In: ALLISON,P. A. & BRIGGS, D. E. G. (eds) Taphonomy: Releasing the Data Locked in the Fossil Record. Topics in Geobiology. Plenum Press, New York, 115-209. KIDWELL, S. M. & BRENCHLEY,P. J. 1994. Patterns in bioclastic accumulation through the Phanerozoic: changes in input or in destruction? Geology, 22, 1139-1143. KXDWELL, S. M. & JABLONSKI, D. 1983. Taphonomic feedback: ecological consequences of shell accumulations. In: TEVESZ, M. J. S. & MCCALL, P. L. (eds) Biotic Interactions in Recent and Fossil Benthic Communities. Topics in Geobiology, 3, 195-248. KIDWELL, S. M., FURSICH, F. Y. & AIGNER, T. 1986. Conceptual framework for the analysis and classification of fossil concentrations. Palaios, 1,228-238. KIESSLING,W., FLI3GEL, E. & GOLONKA,J. (eds). 2002. Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72. LECOMPTE, M. 1954. Quelques donn6es relatives/: la gen6se et aux caract6res 6cologiques des 'r6cifs' du Frasnien de l'Ardenne. Institut royal de Sciences naturelles de Belgique, Volume jubilaire Victor van Straelen, 1, 153-181. LI, X. & DROSER, M. L. 1992. The development of Early Paleozoic shell concentrations: evidence from the Cambrian and Ordovician of the Great Basin. Fifth North American Paleontological Convention, Abstracts and Programs, 183. LI, X. & DROSER, M. L. 1997. Nature and distribution of Cambrian shell concentrations: evidence from the Basin and Range Province of the Western United States (California, Nevada, and Utah). Palaios, 12, 111-126. LOWENSTAM, H. A. 1950. Niagaran reefs of the Great Lakes area. Journal of Geology, 58, 430-487. MELIM, L. A., WESTPHAL, H., SWART, P. K., EBERLI, G. P. & MLrNNECI,:E,A. 2002. Questioning carbonate diagenetic paradigms: evidence from the Neogene of the Bahamas. Marine Geology, 185, 27-53. MUNNEEKE, A. 1997. Bildung mikritischer Kalke im Silur auf Gotland. Courier Forschungsinstitut Senckenberg, 198, 1-71. MUNNECKE, A. & SAMTLEBEN,C. 1996. The formation of micritic limestones and the development of limestone-marl alternations in the Silurian of Gotland, Sweden. Facies, 34, 159-176. MUNNECKE, A. & WESTPHAL, H. 2005. Variations in primary aragonite, calcite, and clay in fine-grained calcareous rhythmites of Cambrian to Jurassic age an environmental archive? Facies, 51, 611-626.
15
NELSON, C. S., HYDEN, F. M., KEANE, S. L., LEASK,W. I. & GORDON, D. P. 1988. Application of bryozoan zoarial growth-form studies in facies analysis of non-tropical carbonate deposits in New Zealand. Sedimentary Geology, 60, 301-322. NEUWEILER, F., GAUTRET, P., THIEL, V., LANGE, R., MICHAELIS, W. & REITNER, J. 1999. Petrology of Lower Cretaceous carbonate mud mounds (Albian, N-Spain): insights into organomineralic deposits of the geological record. Sedimentology, 46, 837-859. RACK/, G. & HOUSE, M. R. 2002. Late Devonian Biotic crisis: ecological, depositional and geochemical records. Palaeogeography, Palaeoclimatology, Palaeoecology, 181, 1-374. RAILSBACK,L. B. 1996. Observation sur la note: Burial diagenesis of limestones under simple overburden. Stylolites, cementation and feedback (Bathurst R. G. C. Bull. Soc. gdol. Fr., 1995, 166, 2, 181-192)stylolites in limestones that lacked significant primary aragonite. Bulletin de la Socidt~ g~ologique de France, 167, 181-183. RAYMOND, A., KELLEY, P. H. & LUTKEN, C. B. 1990, Dead by degrees: Articulate brachiopods, paleDclimate, and the mid-carboniferous extinction event. Palaios, 5, 111-123. RIDING, R. 1999. The term stromatolite: towards an essential definition. Lethaia, 32, 321-330. RIDING, R. 2002. Structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth-Science Reviews, 58, 163-231. RIGBY, J. K., FAN, J. & ZHANG, W. 1989. Inozoan calcareous porifera from the Permian reefs in South China. Journal of Paleontology, 63, 778-800. ROOT, R. B. 1967. The niche exploitation pattern of the blue-gray gnatcatcher. Ecological Monography, 37, 17-350. SAMTLEBEN, C., MUNNECKE, A. & BICKERT, T. 2000. Development of facies and C/O-isotopes in transects through the Ludlow of Gotland: Evidence for global and local influences on a shallow-marine environment. Facies, 43, 1-38. SAMTLEBEN, C., MUNNECKE, A., BICKERT, T. & PATZOLD, J. 1996. The Silurian of Gotland (Sweden): facies interpretation based on stable isotopes in brachiopod shells. Geologische Rundschau, 85, 278-292. SAVAZZI,E. 1999. Soft-bottom dwellers and the leaning tower of Pisa's explotation of unstable life positions. In: SAVAZZI,E. (ed.) Functional Morphology of the Invertebrate Skeleton. Wiley, Chichester, 123-128. SCHLAGER, W. 2003. Benthic carbonate factories in the Phanerozoic. International Journal of Earth Sciences, 92, 445-464. SEILACHER, A. 1984. Constructional morphology of bivalves: evolutionary pathways in primary versus secondary soft-bottom dwellers. Palaeontology, 27, 207-237. SHAPIRO, R. S. 2000. A comment on the systematic confusion ofthrombolites. Palaios, 15, 166-169 SPEYER, S. E. 1991. Trilobite taphonomy: a basis for comparative studies of arthropod preservation, functional anatomy and behaviour. In: DONOVAN,
16
J.J. ALVARO ETAL.
S. K. (ed.) The Process of Fossilization. Belhaven Press, New York, 194-219. SPEYER, S. E. & BRETT, C. E. 1988. Taphofacies models for epeiric sea environments: Middle Paleozoic examples. Palaeogeography, Palaeoclimatology, Palaeoecology, 63, 22-262. TAPPAN, H. & LOEBLICH, A. R. 1988. Foraminiferal evolution, diversification, and extinction. Journal of Paleontology, 62, 695-714. TEXTORXS, D. A. 1966. Algal cap for a Niagaran (Silurian) carbonate mud mound of Indiana. Journal of Sedimentary Petrology, 36,455-461. VACHARD, D. & MASLO, A. 1996. Pr6cisions biostratigraphiques et micropal6ontologiques sur le Bashkirien d'Ukraine (Carbonif6re moyen). Revue de Palkobiologie, 15, 357-383. VACHARD, D., FLORES DE DIOS, A. & BUITRON, B. E. 2004. Guadalupian and Lopingian (Middle and Late Permian) deposits from Mexico and Guatemala, a review with new data. Geobios, 37, 99-115. VACHARD, D., OVIEDO, A., FLORES DE DIOS, A., MALPICA, R., BRUNNER, P., GUERRERO, M. & BUITRON, B. E. 1993. Barranca d'Olinal~i (Guerrero): une coupe de r6f6rence pour le Permien du Mexique central; &ude pr61iminaire. Annales de la Sociktk Gkologique du Nord, 2, 155-162. VOIGT, E. 1968. Uber Hiatus-Konkretionen (dargestellt an Beispielen aus dem Lias). Geologische Rundschau, 58, 281-296. WALKER, K. R. & ALBERSTADT,L. P. 1975. Ecological succession as an aspect of structure in fossil communities. Paleobiology, 1,238-257. WEBB, G. E. 1996. Was Phanerozoic reef history controlled by the distribution of non-enzymaticallysecreted carbonates (microbial carbonate and
bilogically-induced cement)? Sedimentology, 43, 947-971. WEIDLICH, O. 2002. Middle and Late Permian reefs distributional patterns and reservoir potential. In: KIESSLING, W., FLUGEL, E. & GOLONKA, J. (eds) Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 339-390. WESTPHAL, H. & MUNNECKE, A. 1997. Mechanical compaction versus early cementation in fine-grained limestones: differentiation by the preservation of organic microfossils. Sedimentary Geology, 112, 33-42. WIGNALL,P. B. & HALLAM,A. 1992. Anoxia as a cause of the Permian-Triassic extinction: facies evidence from northern Italy and the western United States.
Palaeogeography, Palaeoclimatology,Palaeoecology, 93, 21-46. WILSON, J. L. 1975. Carbonate Facies in Geologic History. Springer, Berlin. WOOD, R. 1998. Reef Evolution. Oxford University Press, Oxford. WOOD, R. 2001. Are reefs and mud mounds really so different? Sedimentary Geology, 145, 161-171. WRIGI-IT, V. P. & CHERNS, L. 2004. Are there 'black holes' in carbonate deposystems? Geologica Acta, 2, 285-290. ZHURAVLEV, A. YU. 1995. Reef ecosystem recovery after the Early Cambrian extinction. In: HART, M. B. (ed.) Biotic Recovery from Mass Extinction Events. Geological Society, London, Special Publications, 102, 79-96. ZHURAVLEV, A. YU. & RIDING, R. (eds). 2000. The Ecology of the Cambrian Radiation. Columbia University Press, New York.
Lower Cambrian shelled phosphorites from the northern Montagne Noire, France SI~BASTIEN C L A U S E N 1 & J. J A V I E R .A,L V A R O 1'2
~Laboratoire de Palkontologie et Palkogkographie du Paldozofque, U M R 8014 CNRS, Universitk des Sciences et Technologies de Lille, 59655-Villeneuve d'Ascq, France (e-mail." Sebastien.
[email protected]) 2Departamento Ciencias de la Tierra, Universidad de Zaragoza, Ciudad Universitaria, 50009-Zaragoza, Spain
Abstract: Shelledphosphorites of Early Cambrian age are common in the Av6ne-Mendic autochthonous unit (Marcory Formation) and the M61agnes nappe ('Heraultia beds' of the Lastours Formation), northern Montagne Noire (France). Palaeogeographically, the concentration of phosphate took place along the shelf edge between a stable inner platform (southern Montagne Noire) and an unstable slope-to-basin sea floor preserved in the northern Montagne Noire. Petrography, back-scattered SEM (scanning election microscopy) and elemental mapping by EDS (energy dispersive system) show that the phosphorites were generated by repeated alternations of low sedimentation rates and condensation forming hardgrounds, in situ early diagenetic phosphogenesis, winnowing and polyphase reworking of previously phosphatized skeletons and hardground-derived clasts. The successionof repeated cycles of sedimentation, phosphate concentration and reworking led to multi event phosphate deposits rich in allochthonous particles. Associated accumulations of exhumed and reworked pyrite clasts reflect final deposition in a mainly dysaerobic substrate.
A critical aspect of the NeoproterozoicCambrian transition is the occurrence of phosphogenic events contemporaneous with the 'Cambrian explosion' of metazoans, some of which are characterized by phosphate-shelled skeletons. The origin of phosphate shells (original biomineralization v. secondary epigenesis) is key to a better understanding the palaeoceanographical and palaeogeographical factors that controlled Neoproterozoic and Lower Cambrian phosphorites (Brasier 1990). In the present day, phosphorites form in two distinct environments: (i) along west-coast margins, where dynamic upwelling of nutrientrich water enhances primary productivity and development of oxygen minimum zones, thereby favouring accumulation of organic-rich sediments and subsequent phosphogenesis; and (ii) along current-dominated margins, where low sediment accumulation rates allow intense biogenic mixing of surficial sediments, giving rise to supersaturation of pore waters with respect to apatite (F611mi et al. 1991). In addition, porewater concentrations of dissolved apatite may be enhanced by microbial activity, which can mediate apatite precipitation (Wilby & Briggs 1997; Xiao & Knoll 1999). The purpose of this paper is to analyse the occurrence of Lower Cambrian phosphogenesis
in the northern Montagne Noire, France. This study takes into account the geochemical, petrographical, sedimentary and palaeogeographical aspects of the phosphorites, which offer key data to understand the geodynamic patterns of this margin of Gondwana after the Cadomian orogeny.
Geological setting and stratigraphy The Montagne Noire (southern France) consists of a complex framework of tectonostratigraphic units (Fig. 1a), which are grouped into three main structural domains (G~ze 1949): (i) a metamorphic axial zone, made of complex domes of gneiss and migmatites surrounded by micaschists, which has yielded Precambrian acritarchs in the 'Schistes X' unit (Fournier-Vinas & D6bat 1970); (ii) the southern flank composed of large nappes involving Lower Cambrian-Carboniferous strata; and (iii) the northern flank made of imbricated tectonic nappes composed of Lower Cambrian-Silurian rocks. Our study is focused on two stratigraphic units located on the eastern part of the Lacaune Mountains, northern Montagne Noire: the lower part of the Marcory Formation from the Av~ne-Mendic autochthonous unit, and the 'Heraultia beds' (uppermost part of the Lastours Formation) from the
tkLVARO,J. J., AREXZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special From:
Publications, 275, 17-28.0305-8719/07/$15.00 9 The Geological Society of London.
18
S. CLAUSEN & J. J./i~LVARO
Fig. 1. (a) Pre-Hercynian rocks of France and neighbouring areas with location of the Montagne Noire. (b) Map of the tectonostratigraphic units that form the Montagne Noire. (e) Geological map of the M61agues nappe and the Av6ne-Mendic unit; modified from Gu6rang6-Lozes & Burg (1990).
Fig. 2. (Neoproterozoic?)-Lower Cambrian stratigraphic framework of the M61agues and Brusque nappes, and the Av6ne-Mendic unit; summarized from Cobbold (1935), Thoral (1935), Roques & Vachette (1970), Courtessole (1973), Hamet & All+gre (1973), Rolet (1973), Boyer-Guilhaumaud (1974), Alsac & Donnot (1978), Donnot & Gu6rang6 (1978), Gachet (1983), Courjault-Rad6 (1985), Debrenne & Courjault-Rad6 (1986), Kerber (1988) and Gu6rang6-Lozes & Burg (1990).
M61agues nappe (Fig. 1b). The Lower Cambrian of the M61agues nappe offers close lithostratigraphic similarities with the fossiliferous Minervois and Pardailhan nappes (southern Montagne Noire) where the Lower Cambrian litho- and,biostratigraphic units were formally defined (Alvaro et al. 1998a). By contrast, the Lower Cambrian(?) lithostratigraphy of the Av6ne-Mendic units is complicated by the presence of volcano-sedimentary complexes,
which are absent in the southern Montagne Noire. The lower part of the Marcory Formation in the M61agues nappe (Fig. 2), with a total thickness of approximately 1000 m, consists of alternations of sandstones and green and black shales that contain phosphatic limestone nodules and beds. These occur interbedded within slope-related deposits, and grade upwards into the shallower marine carbonate strata of the
CAMBRIAN PHOSPHORITES FROM FRANCE Pardailhan Formation (Rolet 1973). The presence of the earliest Cambrian ichnogenus Taphrelminthopsis at the uppermost part of the Marcory Formation, considered by Seilacher (1997) as a senior synonym of Psammichnites, allows a broad correlation of these strata with the Tommotian-Atdabanian transition (Dor6 1994; Alvaro & Vizcaino 1999). The archaeocyathanbearing limestones and dolostones of the overlying Pardailhan Formation in the Brusque nappe (northern Montagne Noire: Debrenne & Courjault-Rad6 1986), and the Minervois and Pardailhan nappes (southern Montagne Noire: last revision in Alvaro et al. 1998a) are Botoman in age. The Av6ne-Mendic unit contains an incomplete Lower Cambrian volcano-sedimentary succession, underlain by the Mendic granite (broadly dated as 600-507Ma: Roques & Vachette 1970; Hamet & All6gre 1973; Gu6rang6-Lozes & Burg 1990) and overlain by the tectonic contact with the M61agues nappe. In ascending order the stratigraphic units comprise the so-called 'blavi6rites' (rhyolitic tufts and breccias) and the Layrac volcano-sedimentary complex (both units with an undetermined Neoproterozoic-Early Cambrian age), and the Pardailhan and Lastours formations (Fig. 2). The upper part of the latter contains some bedded phosphatic limestones (named 'Heraultia beds'), which have yielded a phosphatized skeletal and shelly microfauna composed of annelid and chancelloriid sclerites, brachiopods, crustaceans, gastropods, halkieriids, trilobites and problematica (Cobbold 1935; Kerber 1988). The presence of the mollusk Heraultia, characterized by a pandemic distribution in the Early Cambrian (Miiller 1975; Runnegar 1981), was taken by Kerber (1988) as indicative of a Botoman age. This proposal was revised by Gubanov (2002), who dated them as early Tommotian. However, this is incompatible with the age of the underlying archaeocyathanbearing Pardailhan Formation. More Lower Cambrian phosphorites are reported from other nappes of the northern Montagne Noire and the neighbouring C6vennes (e.g. Orgeval & Capus 1978; Notholt et al. 1979; Prian 1979, 1980; Notholt & Brasier 1986; Southgate 1986), but they are not documented here.
Palaeogeographical context The tectonostratigraphic patterns displayed by the Cambrian sedimentary rocks of the southern and northern Montagne Noire are still a matter of debate. The Lower Cambrian of the southern Montagne Noire represents what was a
19
homogeneously subsiding, stable platform where volcanic activity was absent, except for giving rise to: (i) the increased feldspar content in arkoses of the Marcory Formation (CourjaultRad6 1985), interpreted as the volcanoclastic influence of a magmatic activity recorded in the northern Montagne Noire; and (ii) the presence of volcanic and hydrothermal deposits associated with the massive sulphide-ore stratabound of the Salsigne district (Minervois nappe: L6pine et al. 1984, 1988). The southern Montagne Noire lacks Cambrian phosphorites, black shales, reworked pyrite clasts and slope-related strata (Alvaro et al. 1998b; Debrenne et al. 2002). Conversely, the Lower Cambrian of the northern Montagne Noire is characterized by an episodic tectonic instability of the platform with development of slopes and localized dysoxic substrates, phosphogenic episodes and volcanogenic influence, the last related to oceanic extension and rifting recorded in the northernmost part of the northern Montagne Noire (Alsac et al. 1987). Therefore, the northern Montagne Noire represents a heterogeneously subsiding, unstable margin bordering the relatively stable inner platform preserved in the southern Montagne Noire. Relics of primary to early diagenetic evaporites, reported from Lower Cambrian sedimentary rocks of SW Europe, are also present in the Lastours Formation of the southern Montagne Noire (Alvaro et al. 2000, 2003). This suggests development of an arid subtropical belt in the Early Cambrian southern Hemisphere (Alvaro et al. 2000), which affected the margin of Gondwana in the Montagne Noire region coevally with the phosphogenesis that took place on this shelf margin.
Methods Both originally phosphatic and secondarily phosphatized shelly and skeletal microfossils were extracted from the limestone matrix by dissolving the rock in a dilute (10% by volume) acetic acid. Diagenetic processes were investigated by a combination of petrographic observations: replacements, overgrowths and cross-cutting relationships of cements were clarified via optical and cathodoluminescence microscopy of ultra-thin sections, and backscattered scanning election microscopy (SEM) on polished and etched surfaces. Geochemical analyses were made by an energy dispersive system (EDS) of elemental mapping attached to the SEM (see Martill et al. 1992). This approach allows differentiation of polyphase phosphatic encrustation and impregnation: owing to slight
20
S. CLAUSEN & J. J. ALVARO
Fig. 3. Phosphatic heterogeneous hardgrounds of the M61aguesnappe. (a) Crenulation schistosity. (b) Detail of boxed area with lineation of blavieritic and pyrite grains cemented by phosphate. (c) & (d) Cemented blavieritic clasts, pyrite and bioclasts; all scales= 2 mm. p, pyrite grain; b, blavieritic (rhyolitic) grain; sf, bioclast.
modifications in the geochemical composition of apatite cements.
The phosphorites of the M~lagues nappe The phosphatic limestones of the Marcory Formation are interbedded with grey-black, well-laminated-massive shales (Fig. 2). They occur in one of the thrust sheets that characterize the southernmost edge of the M61agues nappe. Their outcrop has a lenticular shape, up to 100 m long, and is laterally bounded by intra-Marcory Hercynian faults. Influenced by the tectonic deformation, the fabric of the limestones is dramatically affected by an asymmetric crenulation schistosity (Fig. 3a). Other deformation processes are illustrated by the wealth of solution seams and stylolites, concentrated along cleavage surfaces. The presence of millimetre- to centimetre-thick, phosphate-rich laminae, easily identifiable by their grey-black colour, is directly controlled by this schistosity: the phosphatic bands display microfolding lineation formed by stretching and necking (boudins) following
the crenulation (Fig. 3a). We consider these laminae 'hardgrounds' following Southgate's (1986) concept of 'phosphatic hardground', i.e. a surface of synsedimentary lithification including grainy pavements of submergent or semi-emergent origin. Two kinds of hardgrounds can be distinguished in the Marcory Formation: (i) thinner (less than 3mm), homogeneous crusts made up of amorphous collophane; and (ii) thicker (up to 2 cm thick), heterogeneous (polymictic) lag deposits where a complex mixture of allochems is cemented with collophane (Fig. 3b). Hardgrounds are laterally discontinuous, and show a sharp upper contact and a lower gradational phosphatic concentration. Hardgrounds are interbedded with limestones (up to 80cm thick) that commonly exhibit normal grading from packstones to mudstones rich in skeletons and extraclasts (up to 2 cm in size) embedded in a sparry calcite mosaic. Although their allochems commonly display a randomly oriented fabric, long axes of sparry calcite crystals and elongated clasts are in some cases aligned subparallel to the crenulation
CAMBRIAN PHOSPHORITES FROM FRANCE axial planes displayed by the hardgrounds, resembling flow structures. Cathodoluminescence shows that identification of the original fabric of the sparry cement was a clast-supported breccia, dominated by subangular (secondarily subrounded in volume) calcite clasts, skeletons, shells and extraclasts, with dull red luminescence. This breccia is cemented by blocky calcite cements up to 100 gm across, which display an orange-bright yellow luminescence. The heterogeneous hardgrounds contain silty to medium sand-sized particles, composed of quartz, feldspar, rhyolitic clasts ('bliov6rites'), mica flakes, pyrite and other opaques, rip-up mud clasts, skeletons, shells and reworked phosphatic intraclasts derived from both homogeneous and heterogeneous crusts. These components are also dispersed in the sparry mosaics of calcite (Fig. 3c). Intraclastic phosphatic clasts are subrounded-subangular, and contain variable amounts of inclusions such as silt-sized quartz, rhyolitic clasts and mica flakes. Skeletal and shell material is largely composed of abraded and fragmentary, calcitic conical microfossils, up to 8 mm long, of uncertain affinity and commonly telescoped. Skeletons and shells occur as both isolated debris and embedded in multiphase composite clasts. The latter contain first-generation bioclasts, accretionary mud-sized sediment and calcite and phosphate cements truncated at borders, indicating that at least parts of the sea floor were lithified. Lithic extraclasts consist of reworked pyrite and other opaques, quartz, feldspar, chert and rhyolitic ('blavieritic') fragments and mica flakes, all silt to medium sand sized (Fig. 3d). Reworked concentrations of pyrite consist of single and composite grains. Although they are concentrated in the phosphatic hardgrounds, they also occur dispersed throughout the mosaics of sparry calcite. Reworking of pyrite is shown by mechanical breakage of pyrite grains, and the polyphase nature of some compound pyrite agglomerates. In addition, the aforementioned skeletons, shells and extraclasts (blavierites) can be coated by single or multiple laminae of collophane. The polymictic character and diversity of the extra and brecciated intraclasts reflect a complex provenance from different sources. The clasts were eroded from an upslope composed of: (i) siliciclastic muddy substrates (reworked as consolidated rip-up mud clasts) and characterized by anaerobic or minimally aerobic conditions (due to reworking of pyrite grains); (ii) an inherited palaeotopography composed of 'blavieritic' rocks, which crops out in the neighbouring Av6ne-Mendic unit; and (iii) carbonate substrates.
21
The limestones display lenticular concentrations of intra- and extraclasts, overlying distinct erosive surfaces, alternating with phosphatic hardgrounds. The episodic development of hardgrounds is evidence for two kinds of sedimentation rates: (i) normal, or background, conditions represented by upslope carbonate productivity and reworking of carbonate substrates; punctuated by (ii) condensation episodes represented by fall in terrigenous input and cementation of phosphatic sediment. There are neither foreset nor laminated structures to document a current or wave mode of transport, whereas evidence of erosive surfaces and polyphase transport of hard parts is abundant. This is indicated by local size sorting of shells and allochems, high-energy textures (lags related to underlying erosive bases) and preservation of shells with exotic matrix. The telescoped nature of the conical calcite-shelled skeletons also shows the influence of current activity. The abundance of multiphase compound clasts suggests multiple depositional, cementation and erosive events. The mixture of rhyolitic ('blavieritic') extraclasts, breccia, phosphatic and skeletal intraclasts indicates reworking processes on a complex submarine slope, which favoured exhumation and reworking of different lithotypes. The first erosion and transport of the allochthonous clasts are believed to have been variably aerobic, whereas their final sedimentation took place under dysaerobic conditions. Exhumation and concentration of pyritic debris require a sediment-starvation in a normally dysaerobic environment. Completely anaerobic conditions are not necessary for preservation of reworked pyrite grains because, once formed, pyrite may remain stable in a minimally oxygenated environment for a considerable time, for example in distal turbidites and contourites (Baird & Brett 1986).
The phosphorites of the Av~ne-Mendic unit In the vicinity of Saint-Geni6s de Varensal and Marcou (Fig. 1), the 'Heraultia beds', up to 20 m thick, are located at the upper part of the Lastours Formation. They consist of mottled, irregularly laminated, dolomitic limestones (Fig. 4a). The mottling aspect of these phosphoritic limestones is the result of the abundance of millimetre- to centimetre-thick, nodules of dolomite embedded in grey-white limestone. The patches of orange-coloured ferroan dolomite are subelliptical sucrosic mosaics of euhedral dolomite rhombs, up to 300 gm across. The mottling fabric displays a disorganized arrangement of centimetre-thick, sheet-like, contorted beds locally interrupted by breccia channels.
22
S. CLAUSEN & J. J. ALVARO
Fig. 4. Phosphorites of the Av6ne-Mendic unit. (a) Mottled aspect of the 'Heraultia beds' with squeezed and folded structures (dolostone laminae in light grey colours). (b) Intraclasts filling a channel; scale = 5 cm. (c) Cross-laminated structure by arrangement of bioclasts and extraclasts; scale = 1 cm. (d) Polyphase clasts with attached cements (c), see longitudinal section of an hyolith (hy) and oblique sections of helcionellids (he); scale = 2 mm.
Truncation surfaces, intraformational slumps and synsedimentary folding are common. Slumping and sliding has resulted in deformation of entrained material, lithologically similar to the underlying sediments, which displays a range of brittle 'firm' and soft-sediment deformation, contorted and folded. On a smaller scale, sedimentary boudinage (Fig. 4a) resulted from extensional stresses within a poorly lithified to unlithified sediment mass. Interbedded clastsupported, intraformational breccia is largely oligomictic and channelized (up to 1.6 m thick). The elongated character of its angular clasts ('chips' of Fig. 4b), up to 5 cm long, suggests provenance from the sheet-like contorted beds cut by the channels. Both the contorted tabular beds and breccias are wackestones-packstones rich in polyphase clasts (Fig. 4d), trilobites (Micmacca? albesensis Cobbold 1935) and other arthropods, brachiopods, halkieriids, helcionellids, hyoliths, and other skeletonized microfossils (see systematics in Kerber 1988). Skeletons are disarticulated and commonly broken, and locally display low-angle laminae (Fig. 4c). Both the microsparitic matrix
and the intraparticle porosity of the skeletons contain a (calcarenitic) mixture of fragmented skeletons, opaque minerals and phosphate silt-sized clasts. The aforementioned structures document different types of downslope movements by basal sliding of plastic to semi-rigid sediment masses. The reworked material was developed on slope(s) close to the sedimentary environments from which they were derived. Slope deposition from gravity flows occurred when substrate was both cohesive (brecciation and development of minor channels) and unconsolidated (disturbed and slumped substrate rich in disarticulated and broken skeletons). The lack of rhythmic turbidites suggests low-angle slopes, in contrast with the conventional steepened submarine fan models, which imply base-of-slope to basin deposition dominated by turbidites.
Diagenetic cements The porosity-occluding phases recorded in the shelly and skeletal phosphorites of the Marcory and the 'Heraultia beds' are similar (Table 1).
C A M B R I A N PHOSPHORITES F R O M F R A N C E
23
"4D
o .<
0 0
0
e4o
r/
o
~S
r.,/3
o
,q.
~.
<5
e,i
m.
e-:.
e!. 0
0
s
0o 0
0
9
0
0
0 ,3o
o
*'q.
o
o
0
e!
o0,5
= 0 0
V
V
S
V
.2o -6
~o~ o
bs
zv#
0
,~
9~._, .~ ~ ~ b
.~ r4 r
~ ~.~+~ 9
0
&
&
o
~~
N
9~
r~
~
~
~
.o ~
24
S. CLAUSEN & J. J. ALVARO
Fig. 5. Diagenetic aspects of polyphase clasts under plane polarized light. (a) Telescoped conical shells (transverse section of hyoliths). (b) & (d). Telescoped conical shells (oblique sections). (e) & (h). Sketches of a~t showing relative chronology of cements. Scales: (a) and (c)= 100 gin; (b) and (d)= 200 gm.
Four major pre-compactional cement minerals are identified: fluorapatite, calcite, pyrite and quartz (see chronology of cements preserved in intraparticular pores of polyphase clasts in Fig. 5a-h). The first stage of phosphate precipitation, which is most widespread and volumetrically important, consists of a polyphase filling of intraparticle porosities and is generally preserved as featureless cements composed of collophane and microcrystalline fluorapatite. In some of the Marcory phosphorites, and dominantly in the Heraultia phosphorites, this collophane occluded an intraparticle porosity partly filled
previously by passive introduction of loose sediment composed of by quartz, mica, sponge spicules and sparry calcite grains. The second stage, differentiable by their luminescence and type of occluding porosity, consists of the earliest replacement (epigenesis) of skeletal walls mainly by amorphous (collophane) and microcrystalline fluorapatite, which lacks preserved remains of original wall textures. Tracing compositional changes by EDS elemental mapping (Table 1) reveals that density of phosphatization (CaO and P205 contents) is relatively homogeneous, indicating uniform geochemical compositions
CAMBRIAN PHOSPHORITES FROM FRANCE of shallow-burial pore fluids during phosphogenesis. Data from both infill of intraparticle pores and shell replacement reveal average P205, F and CaO concentrations of 35-37, 9-11 and 48-55 wt%, respectively. However, the two phosphogenic episodes exhibit a slight but significant difference in chemistry, as documented by minor components: distribution of MgO, SiO2 and A1203 shows a similar slight increase in percentage in amorphous (collophane) and microcrystalline fluorapatite, the latter representing epigenized walls and polyphase cements. The final diagenetic phase post-dating apatite cementation but predating solution was calcite precipitation in the remaining intra- and interparticle porosity. Calcite cements are commonly preserved at present as bladed and blocky calcite mosaics (Fig. 6a-e), and are differentiable by their luminescence (from red to orange and bright yellow). In some cases, their growth was interrupted by polyphase reworking. The increase in cathodoluminescence brightness exhibited by, at least, three generations of calcite cements suggests a progressive increase in luminescence activators in pore fluids, reaching peaks in Fe203, MgO and MnO of 0.7, 2.5 and 1.7 wt%, respectively. This geochemical trend may be related to changes from relatively oxidizing ground waters to deeper (more reducing) burial pore waters. The end of calcite precipitation is accompanied by solution phenomena (unsaturated waters in carbonate) as indicated by the presence of embayed contacts and millimetre-sized vugs. Solution predated occlusion of the vuggy porosity by pyrite and microquartz cements (Fig. 6f-h) as a result of pH modification associated with sulphate reduction. Finally, these cements predate pressure solution, grain breakage (compaction) and tectonic veining. Conclusions
Phosphorites are absent in the Lower Cambrian epicontinental platform preserved in the southern Montagne Noire, but occur in seawards shelf slope-related settings of the northern Montagne Noire. There, two Early Cambrian phosphogenic episodes are recognized in the Av6ne-Mendic unit and the M61agues nappe of the northern Montagne Noire. Phosphogenesis took place along the edges of a continental shelf, associated with slope-related facies and an unstable sea floor. Although detailed palaeogeographical reconstructions are not available for the whole Montagne Noire, owing to the complex Hercynian degree of deformation, this shelf edge may represent the lateral continuity of the stable
25
platform recorded on the southern Montagne Noire. Evidence of aridity in the inner platform is given by evaporitic relics, which suggest the setting of this margin of Gondwana in the arid subtropical belt on the Early Cambrian southern hemisphere. Both phosphogenic episodes are separated by a Botoman interval of carbonate productivity where bioclast-generating carbonate factories, represented by archaeocyathanmicrobial reef complexes (Pardailhan and lower part of Lastours formations), spread widely throughout the inner (southern Montagne Noire) and outer (northern Montagne Noire) shelf. Significant concentrations of phosphate were generated by repeated alternations of low sedimentation rates and condensation, leading to hardground formation, in situ early diagenetic phosphogenesis, winnowing, and reworking of previously phosphatized skeletons and hardground-derived clasts. The succession of repeated cycles of sedimentation, phosphate concentration and reworking led to multievent phosphate deposits rich in allochthonous particles. In the case of the Marcory phosphorites, a selective enrichment of diagenetic and reworked pyrite and silica (quartz and chert) took place after some phosphogenic episodes. Pyrite originally precipitated upslope at somewhat greater burial depths than phosphogenesis, when sulphate reduction started under stronger reducing pH values. Accumulations of exhumed and reworked, phosphatized skeletons, and reworked phosphate-hardground, chert and pyrite clasts concentrated during brief erosive episodes in a mainly dysaerobic environment. Four major porosity-occluding cement minerals are identified in the phosphorites: fluorapatite, calcite, pyrite and quartz. The density of fluorapatite was relatively homogeneous, indicating homogeneous geochemical compositions of shallow-burial pore fluids during phosphogenesis, although slight but significant changes in chemistry are documented by minor increases in percentage of MgO, SiO/, and A1203. Subsequent phases of calcite precipitation in the remaining intra- and interparticle porosity are differentiable by their luminescence (from red to orange and bright yellow). Both EDS analyses and cathodoluminescence brightness document a geochemical trend of calcite cementation related to a change from more oxidizing ground waters to deeper (more reducing) burial pore waters. The end of calcite precipitation is accompanied by solution phenomena (unsaturated waters in carbonate), which predated occlusion of the
26
S. CLAUSEN & J. J. ALVARO
Fig. 6. Diagenetic aspects under cathodoluminescence. (a) Same clast as Figure 5a. (b) Elemental map of previous clast. (c) Same clast than Figure 5b. (d) Same clast as Figure 5c. (e) Same clast as Figure 5d. (f) Elemental mapping showing distinct phosphate replacement phases. (g) & (h) Elemental mapping showing phosphate replacement phases.
CAMBRIAN PHOSPHORITES FROM FRANCE vuggy porosity by pyrite and quartz as a result of p H modification associated with sulphate reduction. Finally, these cements predate pressure solution, grain breakage and veining. The authors thank T. Flores de Dios, J. Rolet, N. Tormo and D. Vizca'ino for regional information and field-trip support, and B. Pratt and an anonymous referee for their constructive comments. This paper is a contribution to project CGL2006-13533. References ALSAC, C. & DONNO'r, M. 1978. Le volcanisme cambrien de l'unit6 de Brusque dans les monts de l'Est de Lacaune. Bureau de Recherches Gkologiques et MiniOres, Rdseau Scientifique et Technologique, 628, 1-17. ALSAC, C., CABANIS, B., GUI~RANGI~-LOZES, J. & BEZIAT, D. 1987. Caract6res magmatiques du volcanisme basique ordovicien de Lanagre de Saint-Salvi-de-Carcav6s dans l'Albigeois cristallin (Tarn-Aveyron, France). Comptes Rendus de l'Acadkmie des Sciences, Paris, 305, 1199-1205. ~i.LVARO, J. J. & VIZCAiNO, D. 1999. Biostratigraphic significance and environmental setting of the trace fossil Psammichnites in the Lower Cambrian of the Montagne Noire (France). Bulletin de la Sociktk gkologique de France, 170, 821-828. ALVARO, J. J., COURJAULT-RADI~, P. ET AL. 1998a. Nouveau d6coupage stratigraphique des s6ries cambriennes des nappes de Pardailhan et du Minervois (versant sud de la Montagne Noire, France). Gdologie de la France, 1998, 3-12. /~LVARO,J. J., ELICKI, O., GEYER, G., RUSHTON,A. W. A. & SHERGOLD, J. H. 2003. Palaeogeographical controls on the Cambrian trilobite immigration and evolutionary patterns reported in the western Gondwana margin. Palaeogeography, Palaeoclimatology, Palaeoeeology, 195, 5-35. ,~LVARO, J. J., ROUCHY, J. M. ET AL. 2000. Evaporitic constraints on the southward drifting of the western Gondwana margin during Early Cambrian times. Palaeogeography, Palaeoclimatology, Palaeoecology, 160, 105-122. /~LVARO, J. J., VENNIN, E. & VIZCAINO, D. 1998b. Depositional controls on Early Cambrian microbial carbonates from the Montagne Noire, southern France. Transactions of the Royal Society of Edinburgh: Earth Sciences, 89, 135-143. BAIRD, G. C. & BRETT, C. E. 1986. Erosion on an anaerobic seafloor: significance of reworked pyrite deposits from the Devonian of New York State. Palaeogeography, Palaeoclimatology, Palaeoecology, 57, 157-193. BOYER-GUILHAUMAUD, C. 1974. Volcanismes acides palkozo~'ques dans le Massif armoricain. ThGse doctorale, Orsay. BRASIER,M. D. 1990. Phosphogenic events and skeletal preservation across the Precambrian-Cambrian boundary interval. In: NOTHOLT, A. J. G. & JARWS, I. (eds) Phosphorite Research and Development.
27
Geological Society, London, Special Publications, 52, 289-303. COBBOLD, E. S. 1935. Lower Cambrian faunas from HGrault, France. Annals of the Magazine of Natural History, London, 10, 25-48. COURJAULT-RADE, P. 1985. Comparaison de l'Gvolution sGdimentaire des sGquences du Cambrien infGrieur et moyen (p.p.) dans les versant sud et nord (unit6 Brusque) de la Montagne Noire (Massif Central). Comptes Rendus de l'Acaddmie des Sciences, Paris, 301, 43-48. COURTESSOLE, R. 1973. Le Cambrien moyen de la Montagne Noire. Biostratigraphie. Imprim d'Oc, Toulouse. DEBRENNE, F. & COURJAULT-RADI~, P. 1986. DGcouverte de faunules d'ArchGocyathes dans l'Est des monts de Lacaune, flanc nord de la Montagne Noire. Implications biostratigraphiques. Bulletin de la Sociktk gkologique de France, 2, 285-292. DEBRENNE, F., GANDIN, A. & COURJAULT-RADI~,P. 2002. Facies and depositional setting of the Lower Cambrian archaeocyath-bearing limestones of southern Montagne Noire (Massif Central, France). Bulletin de la Sociktd g~ologique de France, 6, 533-546. DONNOT, M. & GUERANGE, J. 1978. Le synclinorium cambrien de Brusque. Implications stratigraphiques et structurales dans les monts de l'Est de Lacaune (Tarn, Aveyron, HGrault) - versant nord de la Montagne Noire. Bulletin du BRGM, France, 2, 333-363. DORE, F. 1994. Cambrian of the Armorican Massif. In: KEPPIE, J. D. (ed.) Pre-Mesozoic Geology in France and Related Areas. Springer, Berlin, 136-141. FOLLMI, K. B., GARRISON,R. E. & GRIMM, K. A. 1991. Stratification in phosphatic sediments: illustrations from the Neogene of California. In: EINSELE, G., RICKEN, W. & SEILACHER, A. (eds) Cycles and Events in Stratigraphy. Springer, Berlin, 492-507. FOURMER-VINAS, C. & DEBAT, P. 1970. Pr6sence de micro-organismes dans les terrains m&amorphiques pr&ambriens (~(schistes X,) de l'Ouest de la Montagne Noire. Bulletin de la Sociktk gkologique de France, 12, 351-355. GACHET, L. 1983. Contribution gt l'~tude du volcanisme cambrien des unitks de Brusque et du Merdellou ( Monts de l'Est de Lacaune ). Approches pktrographique et structurale. Th&e doctorale, Lyon. GEZE, B. 1949. Etude g6ologique de la Montagne Noire et des C6vennes m6ridionales. Mkmoires de la Sociktk gkologique de France (nouvelle skrie), 62, 1-215. GUI~RANGI~-LOZES,J. & BURG, J. P. 1990. Les nappes varisques du Sud-Ouest du Massif Central. GOologie de la France, 3--4, 71-106. GUBANOV, A. P. 2002. Early Cambrian palaeogeography and the probable Iberia-Siberia connection. Tectonophysics, 352, 153-168. HAMET, J. & ALLI~GRE, C. J. 1973. Datation 87Rb/s7Sr du massif granitique du Mendic et des porphyroides de l'Est de la Montagne Noire. Un exemple de relation entre pluton et volcans. Contributions Mindralogiques et Petrologiques, 38, 291-298.
28
S. CLAUSEN & J. J./kLVARO
KERBER, M. 1988. Mikrofossilien aus unterkambrischen Gesteinen der Montagne Noire, Frankreich. Palaeontographica, Abteilung A, 202, 127-203. LI~PINE, J., BEN AYAD, M. A. & BI~ZIAT,D. 1988. Mise en 6vidence de m6tabasites alcalines du Cambrien inf6rieur dans le district aurif6re de Salsigne (versant Sud-Ouest de la Montagne Noire). Comptes Rendus de l'Acadkmie des Sciences, Paris, 306, 1087-1092. LI~PINE, J., COURJAULT-RADI~, P., CROUZET, J. & TOLLON, F. 1984. Pr6sence d'une zone haute du Cambrien inf6rieur dans le secteur de Salsigne (versant Sud de la Montagne Noire, Aude). Manifestations volcaniques et hydrothermales associ6es: cons6quences m6tallog6niques. Comptes Rendus de l'Acadkmie des Sciences, Paris, 299, 347-350. MARTILL, D. M., WILBY, P. R. & WILLIAMS,N. 1992. Elemental mapping: a technique for investigating delicate phosphatized fossil soft tissues. Palaeontology, 35, 869-874. M(ILLER, K. J. 1975. 'Heuraltia' varensalensis Cobbold (Crustacea) aus dem unteren Kambrium der filteste Fall von Geschlechtsdimorphismus. Paliiontologische Zeitschrift, 49, 168-180. NOTHOLT, A. J. G. & BRASIER,M. D. 1986. Proterozoic and Cambrian phosphorites - regional review: Europe. In: COOK, P. J. & SHERGOLD, J. H. (eds) Phosphate Deposits of the World. Volume 1. Proterozoic and Cambrian Phosphorites. Cambridge University Press, Cambridge, 91-100. NOTHOLT, A. J. G., HIGHLEY, D. E. & SLANSKY, M. 1979. Raw Materials. Research and Development IV. Dossier on Phosphate. Commission of the European Communities, Brussels. ORGEVAL, J. J. & CAPUS, G. 1978. Existence d'un horizon phosphat6 uranif4re et thorif4re ~ la base des formations carbonathes cambriennes de la rhgion du Vigan (Gard, France): son remaniement karstique. Comptes Rendus de la Sociktk gkologique de France, 1978, 115-117. PRIAN, J. P. 1979. Caracthristiques des pal6oenvironnements des phosphorites cambriennes du versant septentrional de la Montagne Noire (Sud du
Massif Central), France. Documents du Bureau de Recherches GOologiques et Minikres, 24, 93-111. PRIAN, J. P. 1980. Les phosphorites cambriennes du versant septentrional de la Montagne Noire, au Sud du Bassin permien de Camarks (Aveyron). PhD thesis, Paris VI. ROLET, J. 1973. Contribution d l'~tude gkologique des monts de l'Est de Lacaune aautochtone du Mendic et Ocaille de Marcou~), Montagne Noire, France. PhD thesis, Orsay. ROQUES, M. & VACHETTE,M. 1970. Ages au strontium sur roches totales des migmatites de la zone axiale de la Montagne Noire et du massif de granite du Mendic (Massif Central fran~ais). Comptes Rendus de l'Acad~mie des Sciences, Paris, 270, 275-278. RUNNEGAR, B. 1981. Biostratigraphy of Cambrian mollusks. In: TAYLOR, M. E. (ed.) Short papers for the Second International Symposium on the Cambrian System. United States, Department, Geological Survey, Open-File Report, 81-743, 198-202. SEILACHER, A. 1997. Fossil Art: An Exhibition of the Geologisches Institut, Tuebingen University, Germany. Royal Tyrrell Museum of Palaeontology, Drumheller, Alberta. SOUTHGATE, P. N. 1986. Proterozoic and Cambrian studies: Middle Cambrian phosphate hardgrounds, phoscrete profiles and stromatolites and their implications for phosphogenesis. In: CooK, P. J. & SHERGOLD, J. H. (eds) Phosphate Deposits of the World. Volume 1. Proterozoic and Cambrian Phosphorites. Cambridge University Press, Cambridge, 327-351. THORAL, M. 1935. Contributions h l'6tude g6ologique des monts de Lacaune et des terrains cambriens et ordoviciens de la Montagne Noire. Bulletin du Service de la Carte Gkologique de France, 192, 318-637. WILBY, P. R. & BRIGGS, D. E. G. 1997. Taphonomic trends in the resolution of detail preserved in fossil phosphatized soft tissues. Geobios, M6moire Sp6cial, 20, 493-502. XIAO, S. & KNOLL, A. 1999. Fossil preservation in the Neoproterozoic Doushantuo phosphorite Lagerst~itte, South China. Lethaia, 32, 219-240.
Anatomy of the Early Cambrian 'La SentineHa' reef complex, Serra Scoris, SW Sardinia, Italy A N N A G A N D I N 1, F R A N ( ~ O I S E D E B R E N N E 2 & M A X D E B R E N N E t
1Dipartimento di Scienze della Terra, Via Laterina 8, 53100 Siena, Italy (e-mail." gandin@unisi, it) 2UMR 5143, CNRS, CP 38, 57 rue Cuvier, 75231 Paris COdex 05, France (e-mail."
[email protected])
Abstract: All bioherms from the Early Cambrian (Botoman) Matoppa Formation of the Nebida Group in SW Sardinia were previously thought to be dominated by Epiphyton. However, at 'La Sentinella' (Serra Scoris Hill), they are composed of Girvanella, Razumovskia, Botomaella and Renalcis, with Epiphyton and archaeocyaths as accessory components. This association forms two unusual types of crust boundstone, consisting of stacked flat or curved crusts and saucer-like archaeocyaths delimiting shelter cavities. Dendrolitic Renalcis archaeocyath-cement boundstone caps the bioherm. Analysis of the La Sentinella reef complex and comparison with similar constructions from Mongolia (Zuune Arts, Salaany Gol), Nevada (Stewart's Mill, Battle Mountain), Mexico (Sonora) and China (Tianheban Formation) suggest that episodic deposition of fine-grained siliciclastic or carbonate sediment followed by periods of non-deposition enabled the calcimicrobial rafts and crusts to colonize the substrate and then provide synoptic relief for the development of a dendrolitic Renalcis-cement framework. 'La Sentinella' is one of the rare examples of Cambrian reef complex displaying community replacement, from an initial stage of thrombolitic and/or flat-stacked microbial crusts on a muddy substrate to an arched microbial crust system, to a more resistant Renalcis-cement boundstone. Such bioherms reflect an open-shelf, shallow-marine environment of increasing energy.
The buildups already described in the Lower Cambrian (Early Botoman) Matoppa Formation, in SW Sardinia (Figs 1-3), consist of microbial reef complexes made of a massive core mainly built by Epiphyton with rare scattered archaeocyaths, that is commonly surrounded by pink-red, laminated flank facies, with laminae still built by Epiphyton, or by marly nodular beds, or by green and red shales containing carbonate lenses and floated archaeocyath cups. This complex in some localities rests on dark grey nodular, thrombolitic or oncolitic marly limestone mainly built by Girvanella (Debrenne et al. 1979, 1993; Gandin & Debrenne 1984b). So far, only one of the bioconstructions exposed in Matoppa Valley (labelled MT 1-2 in fig. 4 of Debrenne et al. 1993) results to be of different composition: branching archaeocyaths and Renalcis are the dominant framebuilders reinforced by abundant cement. However, the study of a new outcrop, which we call 'La Sentinella' (the Sentinel) because of its distinctive shape, revealed a completely different structure of the construction and a much more diverse calcimicrobial association, in which the main frame builders are Girvanella, Razumovskia, Botomaella
and Renalcis while Epiphyton appears to represent only an accessory component. The field investigations and the first draft of this paper were carried out in co-operation with the late David Gravestock, F. Debrenne's student on archaeocyaths and one of the most promising Australian Cambrian specialists. We miss him and his contribution to the completion of our common research.
Geological setting 'La Sentinella' is the most prominent of several limestone outcrops related to the upper carbonate horizon of the Matoppa Formation (Figs 2 & 3) situated on the southern side of Serra Scoris Hill near Gonnesa (Iglesiente) in SW Sardinia (Figs 1, 2 & 4a). The Matoppa Formation is composed mainly of green, fine-grained sandstone, siltstone and shale with two interbedded carbonate horizons (Fig. 3). The upper calcareous horizon consists of massive white or pale grey reef kalyptrae (loaf-like or pillow-shaped mounds, according to Rowland & Gangloff 1988), laterally and vertically associated with red/pink, more or less marly limestones. Locally, the kalyptrae rest directly on the green sandstones or on dark
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations." Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 29-50. 0305-8719/07/$15.00 9 The Geological Society of London.
30
A. GANDIN E T AL. inhomogeneous carbonate bed were re-folded. 'La Sentinella', corresponding to the intersection between a subvertical morphological plane and the normal limb of the anticline that re-folds older folds with E-W axes, exposes a flat vertical surface 2.0 m high and 1.3 m wide (Fig. 4b, c). The layers dip gently about 10~ ~ northward. The depositional structures and facies relationships are fairly well preserved, even though pressure-solution seams tend locally to conceal some of the boundaries.
'La Sentinella' reef complex
Fig. 1. Schematic geological map showing the outcrop pattern of Cambrian-Ordovician rocks in SW Sardinia. Legend of symbols: 1, Nebida Group (Lower Cambrian); 2, Gonnesa Group (Lower Cambrian); 3, Iglesias Group (Middle and Upper Cambrian); 4, Ordovician deposits; 5, Variscan granite; 6, post-Ordovician deposits (Palaeozoic, Mesozoic and Cenozoic). (After Debrenne et al. 1993, modified.)
grey, mostly thrombolitic or nodular, oncolitic, bioclastic limestone. The top of this carbonate interval is overlain by red and/or green shales, with archaeocyaths in carbonate nodules, or by fine-grained sandstone (Gandin 1987). 'La Sentinella' is one of the bowl-shaped outcrops described as an isolated buildup (Fig. 4a, b) by Gandin & Debrenne (1984b). However, investigations of the structural aspect of the Lower Cambrian rocks of Iglesiente (Carmignani et al. 1986; Galassi & Gandin 1992; Regoli et al. 1995) suggest that most of the Matoppa calcareous outcrops correspond to emergence of the hinges of folds derived from the interference of multiple deformational phases linked to the SardinianCaledonian and Variscan tectonic events. According to this reconstruction, the present morphology of the calcareous bodies can be related to tectonic boudinage when the more competent parts of the originally continuous but
Field analysis shows that 'La Sentinella' is a composite bioconstruction consisting of different lithofacies that can be grouped in two units (Fig. 4a), the basal grey, locally dolomitized, limestone and the overlying light, pale grey-pink and rusty-red limestone. The lithological, sedimentological and faunistic composition of the upper unit, chosen for its good exposition and relatively low imprint of the tectonic deformation, has been studied first in the field and sampled with a saw (Fig. 4b). The composition of the basal grey limestone could not be determined in the field because strong folding and local extensive dolomitization make components difficult to evaluate. Method
A 30 x 30 cm grid was traced on the flat, vertical surface of the Upper Bioherm (Fig. 4c), in which the different sedimentological and fossil components were located and quantified (Figs 5 & 6). Methodical field observations were carried out to evaluate the distribution of bioherm constructors and the relative proportions of sediments and cements within each square and through the vertical section of the outcrop (Fig. 5). In the laboratory, thin sections were cut in three perpendicular planes fixed by reference marks on sawn slabs collected in the field, in the unsuccessful tentative attempt at a three-dimensional reconstruction of the buildup. Sedimentological, palaeontological and palaeoecological studies aimed at determining the depositional conditions of one of the best preserved Matoppa bioconstructions were undertaken by the different authors according to their specialities. The results of this detailed analysis disclose that 'La Sentinella' calcareous body corresponds to a complex mosaic of facies that can be regarded as a reef complex, resulting from the growth activity of sessile benthic organisms and exhibiting topographic relief in 'La Sentinella'
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
31
',O
Cas~ ~. Antonio
(
i
tMedau Miniera
/J SERRA 9239 249
',
scoh~s
Case de ~=. Moddizzis
s
N
Fig. 2. Schematic geological map of the Serra Scoris area showing the upper carbonate layer of the Matoppa Formation deformed by interference folds derived from the superposition of Variscan and Caledonian tectonic phases. Legend of symbols: 1, Nebida Group sandstones and shales; 2, Matoppa Formation limestones; 3, Punta Manna Formation limestones; 4, Gonnesa Group; 5, Ordovician 'Puddinga'; 6, alluvial deposits; 7, faults. (Unpublished map by R. Regoli.)
Hill. It can be divided into two units, representing geometrically well-defined bodies, the 'Lower Bioherm' and the 'Upper Bioherm' (Fig. 4a), formed by the superposition of smaller buildups (facies A, C, D and G), resulting from the intergrowths of diverse assemblages of metazoans and calcified microfossils. These compound biohermal bodies are laterally associated with muddy interbiohermal facies (facies B and facies G ) with local coquinas of archaeocyaths.
Results The Lower Bioherm is made of grey massive and nodular or oncolitic marly limestones, in places extensively affected by fabric-destructive dolomitization. The massive facies (facies G) has a thrombolitic fabric and contains scattered broken archaeocyath cups (Fig. 7a-d). The rare archaeocyaths are often found in the cores of oncoids or in the bioclastic facies surrounded by microbial overgrowths. The nodular facies
(facies G') is locally bioclastic and characterized by irregularly alternating lenses and nodules of packstone with oncoids and yellowish dolomitic bands (Fig. 7e, f). Irregular yellowish, dolomitized seams and patches are common. The Upper Bioherm consists of pink laminated to light grey/white massive limestone surrounded by pink to red archaeocyath-rich marly lithofacies. The pink laminated limestone (facies A) contains interspersed archaeocyaths of various sizes (Figs 8 & 9a) and commonly grade to the massive facies C characterized by rather large growth and shelter cavities in some cases roofed by archaeocyath cups (Fig. 10a-c), lined by cement (Fig. 10c) or filled with internal sediment (Figs 10a & 13). In the upper part, the massive facies D lacks large cavities and appears more homogeneous and crystalline. Contact with the underlying and overlying muddy interbiohermal sediments is abrupt. Judging by the irregular but smooth
32
A. G A N D I N
E T AL.
texture, the skeletons are clearly in place or floated from the surrounding areas. u.I
BASIN
m rF iii 13_
OPEN S H E L F
,ii
f PERITIDAL
o
I.IJ
f/TIDAL-FLAT .~ SABKHA
,,_1 <
OOLITIC SHOAL
~ ~
Six microfacies were recognized from field observations and petrographic analysis. Two pertain to the dark grey lithofacies of the Lower Bioherm (facies G and G') and four to the pale grey Upper Bioherm and pink interbiohermal lithofacies (facies A-D).
I..1_ I.-._1 13_
LAGOON
Fabric and components
~co
< SHALLOW
o
MARINE
o
Fig. 3. Stratigraphic log of the Cambrian of Sardinia. The 'La Sentinella' reef complex occurs within the upper of the two carbonate horizons of the Matoppa Formation. Legend of symbols: 1, fine-grained sandstones and shales; 2, Matoppa limestones; 3, quartzarenites; 4, bedded and massive limestone; 5, laminated and massive dolostone; 6, calcareous breccia; 7, marly nodular limestone; 8, conglomerate ('Puddinga') and sandstone; 9, archaeocyath; 10, Epiphyton; 11, renalcids; 12, trilobites; 13, graptolites. (After Galassi & Gandin 1992, modified.)
shape of the upper surface of the buildup, synoptic relief at the upper contact was probably no more than 20-30 cm. Lateral to the Upper Bioherm is a wellbedded homogeneous rusty-red muddy-marly limestone (facies B) with numerous small- tomedium-sized cups and ribbon-like fragments of archaeocyaths (Fig. 9d, g). This red sediment passes transitionally into the pale pink-deeper rose laminated limestone (facies A) (Figs 8 & 9a). Although accumulations of archaeocyath cups in the mud-supported facies have a floatstone
Lower Bioherm Thrombolitic-oncolitic lithofacies. Dark grey marly limestone (Fig. 7a, b) resting on the Matoppa green sandstones. Facies G: spongy crust and raft boundstone (Fig. 7c, d). Facies G is a complex and irregular association of microbial communities and intervening irregular mud-filled cavities characteristic of the thrombolitic fabric. Superposed thick spongy masses of Razumovskia and/or Girvanella (Fig. 7d) include oncoidal agglomerates, dense sheets of Botomaella, scattered Epiphyton bushes and, less frequently, corroded and/or encrusted cups of archaeocyaths. The oncoids commonly consist of Girvanella filaments with an occasional outer irregular coating of Razumovskia (Fig. 7e). Rather large growth cavities are filled with dense micrite containing small siliceous sponge spicules or large Chancelloria sclerites. Less common are fragments/intraclasts of Epiphyton bushes, trilobite sclerites, echinoderm ossicles and bivalve shells. Cups of Protopharetra protea (Fig. 1 lb, d), probably preserved in life position, Anthomorpha margarita (Fig. 7b), Afiacyathus alloiteaui and encrusted fragmented remains of Erismacoscinus, Coscinocyathus, Gandinocyathus and Ichnusocyathus (Fig. 7d, f) are found entrapped in the microbial framework. Grey nodular-bioclastic lithofacies. Dark grey dolomitized marly limestone. Facies G': bioclastic wackestone-packstonefloatstone (Fig. 7f). This facies consists of calcareous nodules bounded by marly carbonate partings and interbeds, showing a high degree of dolomitization (Fig. 7b). The calcareous nodules mainly consist of bioclastic packstonefloatstone. Thick microbial crusts with interlaminar internal sediment are associated with oncoids, but given the nodular nature of the sediment it is not always clear if all of them represent depositional grains or interbedded pioneer bioconstructions rather than mechanically induced boudins. The packstone matrix is microsparitic, but relics of the micritic fabric are preserved in
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
33
Fig. 4. Views of the 'La Sentinella' outcrop. (a) On the southern side of Serra Scoris Hill 'La Sentinella' is one of the exposed white blocks of the upper carbonate horizon of the Matoppa Formation, isolated by tectonic boudinage. The arrows mark the relative position of the Lower Bioherm (LB) and Upper Bioherm (UB). (b) The Upper Bioherm of the 'La Sentinella' outcrop has been sampled with a saw (black arrows on left side of photograph mark the sampling scars) on the normal limb of a second-phase anticline. The underlying Lower Bioherm is not visible. (c) Sketch of the outcrop showing the method of sampling: samples were extracted from the flat surface of the Upper Bioherm after the different facies were identified and mapped along a 30 x 30 cm grid (see Fig. 5). The lithofacies recognized are: pink laminated limestone-facies A; red marly limestone-facies B; light grey massive limestone with shelter cavities-facies C; and light grey massive, crystalline limestone-facies D.
some places. The often dolomitized/recrystallized nodular sediment (Fig. 7f) usually contains angular quartz-silt, mica and scattered recrystallized skeletal particles. The skeletal particles are well-sorted angular skeletal fragments of echinoderms, trilobites, phosphatic brachiopods, molluscs (?) and Chancelloria, together with oncoids and laminar fragments of Girvanella crusts
and Razumovskia rafts. Clasts of archaeocyaths include some apex portions of Erismacoscinus (Fig. 7c) and large Coscinocyathus dianthus (Fig. 7f) cups, with abraded and corroded outer walls. Most of the skeletal particles are rimmed by a more or less developed microbial micritic coating, in some cases obviously produced by Girvanella filaments. The original structure of the
34
A. GANDIN ET AL.
Fig. 5. Detailed facies map of the exposed surface of the Upper Bioherm. (a) Location of the samples and the spatial distribution and relationships of the facies A - D resulting from: (b) the quantification in every grid square, of the sedimentological features and the fossil components recognized in the field. Legend of symbols: facies A, laminar facies (cement < matrix up to 50% of the construction); facies B, bioclastic floatstone/ packstone; facies C, crust boundstone with large shelter cavities (matrix >> cement up to 50% of the construction); facies D, dendrolitic facies (cement >> matrix up to 30% of the construction).
skeletal particles has been replaced by equant calcite except for the echinoderm, trilobite and phosphatic brachiopod fragments. Upper Bioherm Laminar lithofacies. Laminated, microbial or micritic limestone (Figs. 8 & 9a). Facies A: Botomaella-Girvanella microbial crust boundstone (Fig. 9b, c). It is formed by the superposition of nearly flat, thick sheets of
Botomaella (Fig. 9b, c, e), Girvanella (Fig. 9b, c) and, less frequently, Razumovskia (Fig. 9f); the sheets are commonly tightly stacked (Fig. 9b, c), in some cases are separated by narrow laminar cavities mainly filled by pink micritic sediment or by dark red marly residual mudstonewackestone (Fig. 9c) or skeletal packstonefloatstone (Fig. 9f) similar to facies B. Cement is poorly represented and usually less developed than the interlaminar sediment. Small pendant clusters of Renalcis (Fig. 9b, c) and Epiphyton
SARDINIAN EARLY CAMBRIAN REEF COMPLEX AJACICYATHIDS
ARCHAEOCYATHIDS
35
Some large cups of Afiacyathus alloiteaui (Fig. 12b) and stick-like cups of Tubicoscinus cupulosus (Fig. 9g) are present. Laminar fragments of microbial crusts of Botomaella, Girvanella and Razumovskia can be interpreted as intraclasts or as fractured slivers probably resulting from tectonically induced pressuresolution disruption of formerly contiguous crust boundstone.
fA
fB (~ = Number of taxa fC
fD !
I
I
I
I
I
|
150
100
50
0
50
100
150
'~
Number of specimens
)
Fig. 6. Repartition of the archaeocyath taxa (see Table 1) recognized in the different facies of the 'La Sentinella' Upper Bioherm.
line the roofs of cavities beneath the microbial crusts and are always embedded in a thin irregular band of fine-grained marine cement. Archaeocyaths are not involved in the framework construction, but are frequently found as small floated cups or as fragments of the same size in the red muddy interlaminar facies. Ajacicyathids are represented by seven genera. Archaeocyathids are less diverse in terms of recognizable genera, as they are mostly represented by juveniles (Table 1), but they equal the ajacicyathids in number of specimens. Ajacicyathina, Erismacoscinina and Coscinocyathina are represented by genera recognized in other localities of the same horizon. Tabulacyathus insperatus, the only Tabulacyathida known in western Europe, is present but rare (Fig. 12a, right).
Red bioclastic marly lithofacies. Skeletal dark red marly mudstone (Fig. 9d). Facies B: bioclastic wackestone, packstonefloatstone (Fig. 9g). The matrix consists of rustyred-stained microbioclastic microsparite rich in residual clay minerals and iron oxides. Unsorted, sometimes large, skeletal particles (Figs 9a, d, g & 14a) are represented by echinoderm ossicles, trilobite sclerites and fragments, and/or apex portions of archaeocyathan cups generally larger than in facies A. Although accumulations of archaeocyath cups in the mud-supported facies have a floatstone fabric, the skeletons are clearly in place or floated from the surrounding areas. The ajacicyathids, with the same generic composition as in facies A, are more diverse and more numerous than the archaeocyathids.
Massive lithofacies with large shelter~growth cavities. Microbial massive crust limestone with large shelter cavities in part also roofed by archaeocyaths and filled with pink limemudstone (Figs 10 & 13). Facies C: crust boundstone with accessory archaeocyaths and Renalcis or Epiphyton dendrolitic clusters. Curved, relatively thick microbial crusts are associated with Renalcis clusters and small archaeocyath cups. Large growth cavities with geopetal infill have developed mostly beneath Girvanella-Razumovskia crusts (Fig. 13b, e) and, less frequently, beneath saucerlike Anthomorpha (Fig. 10c) or bowl-shaped Coscinocyathus (Fig. 10b). Cavity roofs and walls are lined by festoon-like microcrystalline cement rinds that commonly enclose cryptobionts such as Renalcis in pendant clusters (Figs 10b, c & 13b, c, e) and, less frequently, Archaeopharetra (Fig. 10b). Fine-grained, locally recrystallized cement partly fills the cavities. It commonly encloses Renalcis clusters and in places Epiphyton dendrolitic clusters (Fig. 13b) and floating Razumovskia rafts (Fig. 10b, e). The internal sediment consists of pink micrite characterized by scattered calcite-filled microborings (Figs 10c & 13b-e). On the floor of the cavities, rare accumulations of particles in the matrix consist of skeletal remains of trilobites, bivalve shells and small Protopharetra cups, calcareous microspheres (Fig. 13b, d) and fragments of the roof pendant-cement. Botomaella fans and Epiphyton clusters sometimes hang from the walls or fill the cavities (Fig. 13b). The outer wall of some Protopharetra cups is thickened, apparently in response to microbial encrustation (Fig. 10f). Delicate crusts of Razumovskia and some Protophareta cups may support pendant Renalcis clusters (Fig. 10d, e). Archaeocyaths are relatively abundant as isolated cups in the boundstone, with dominant tabulate conical forms of Erismacoscinus elongatus (Fig. 12c), Erismacoscinus cancellatus (Fig. 1 l a) and stick-like cups of Tubicoscinus cupulosus, Gandinocyathus gravestocki (Fig. 12c, d), and Taylorcyathus vologdini (Fig. l lb) together with solitary Protopharetra protea (Fig. 1lb) and Dictyocyathus tenerrimus
36
A. GANDIN ET AL.
Fig. 7. Macro- and microfacies of the Lower Bioherm, consisting of dark grey thrombolitic-oncolitic limestone (facies G) and bioclastic-nodular wackestone-packstone-floatstone (facies G'). (a) Thrombolitic fabric of the dark grey limestone. (b) Thrombolitic-nodular facies with a mud-filled shelter cavity (arrow) roofed by a large saucer-like cup of Anthomorpha margarita Born. (c) microbial corrosion and encrustation (arrow) of the outer wall of a fragmented Ichnusocyathus cup entrapped in the thrombolitic framework (facies G). In the central cavity skeletal sediment with small cups of Erismacoscinina (facies GS. [SCS G3 II| - M84492.] (d) Thrombolitic fabric of the Girvanella-Razumovskia boundstone. Irregular growth cavities filled with microsparitic matrix and recrystallized cement. [SCS 95Senti.] (e) Micritic sediment with calcite-filled microborings embedding an oncoid composed of dense filamentous microbial layers (?Girvanellaand Botomaella?). [SCS G2B.] (f) Recrystallizeddolomitized marly floatstone (facies G'). Matrix contains silt-sized quartz and small skeletal fragments; outer walls of Coscinocyathus dianthus Born. corroded or locally encrusted. [SCS G3II.] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
37
Fig. 8. The unusual pink and white laminated limestone built by flat microbial crusts (facies A), passes upwards to the more massive crust boundstone buildup containing large saucer-like archaeocyath cups (facies C) that bound shelter cavities filled by pink mudstone.
(Fig. 11c). Rare pseudo-modular specimens of
Archaeopharetra ertashkaense are recognized here for the first time in the bioherm (Fig. 10b).
Biohermal lithofacies. White, massive crystalline limestone.
Facies D: cement-supported Renalcis dendrolitic boundstone with archaeocyaths (Fig. 14a~l). Renalcis is the main frame-builder (Fig. 14b-e), frequently associated with numerous budding cylindrical cups of branching Archaeopharetra (Fig. 14c, d). Epiphyton bushes (Fig. 14c) and Botomaella fans are occasional components. Stick-like cups of Erismacoscinus, Porocoscinus, Tubicoscinus and Coscinocyathus, solitary Dictyocyathus, apices of Anthomorpha with exothecal roots (Fig. 141) and large bowl-shaped cups of Porocoscinus flexibilis (Fig. 14b) are sporadically present. Pore spaces and some of the infrequent small growth cavities are commonly occluded by yellowish, fine-grained marine cement (Fig. 14b-1) while late diagenetic, clear prismatic/equant cements mainly infill the intraskeletal pores of archaeocyath cups (Fig. 14c-f). Less frequently the cavities are infilled by marine multilayered fibrous cements or by silt-sized internal sediment (Fig. 14d), coarser than that found in facies C. Microbial crusts are characteristically absent.
Bioherm components Fauna The faunal composition of the biohermal complex is dominated by archaeocyaths, while various other organisms are associated in the peri-biohermal deposits.
Archaeocyath assemblages are more diversified in the crust boundstone facies C and peribiohermal mudstone facies B. All the species previously recorded in the Matoppa Formation (Debrenne et al. 1993) were also found at 'La Sentinella' (Table 1); their distribution depends on the different depositional facies (Figs 6 & 15). In the interbiohermal wackestone-floatstone deposits of the Lower and Upper Bioherms the archaeocyaths are accumulated as transported skeletal particles. The faunal composition of the pink-rusty red facies B is diverse and similar to that of facies A, with a slight dominance of ajacicyathids over archaeocyathids, while only rare floated cups of small Erismacoscinina (Fig. 7c) and of skeletal fragments of Coscinocyathus dianthus (Fig. 7t) occur in the dark grey basal facies (facies G'). Archaeocyath remains are rare in the basal thrombolitic boundstone (facies G): only corroded and encrusted fragments of Ichnusocyathus (Fig. 8c) cups have been found entrapped in the Girvanella-Razumovskia microbial framework. In the pink laminated microbial boundstone (facies A) the archaeocyath assemblage consists of small cups or fragments of larger cups. It shows a well-balanced distribution (Table 1) between diversified solitary ajacicyathids (formerly 'Regular') and archaeocyathids (formerly 'Irregular'). In the Renalcis-crust boundstone (facies C), the ajacicyathids display greater generic diversification than the archaeocyathids, but the latter are much more diversified at the specific level and are occasional builders owing to the presence of modular budding Protopharetra and
Archaeopharetra.
38
A. GANDIN E T A L .
Fig. 9. Macro and microfacies of the microbial-crust boundstone (facies A) and rusty-red wackestone-packstonefloatstone (facies B). (a) Pink laminated limestone (facies A) with small scattered archaeocyath cups. (b) & (c) Microbial crust boundstone (facies A) composed of superposed plane laminae built by Botomaella, Girvanella and Razumovskia. Interlaminar sheet-like open spaces partially filled by muddy sediment (facies B), with roof lined by cement, pendant renalcids (e, arrow) and small archaeocyath cups (c). [SCS2B .1_+ SCS4C.DG.] (d) Disoriented archaeocyath cups enclosed in microbial crusts. (e) Detail of Botomaella tuft of filaments. [SCS 1B Ia.] (f) Detail of a Razumovskia crust embedded in skeletal wackestone. [SCS 1B Ia.] (g) Skeletal packstone-floatstone (facies B) produced by the accumulation of small cups of Tubicoscinus cupulosus Debr. and fragments of archaeocyaths, echinoderms, trilobites, sponge spicules and chancelloriid sclerites. [SCS R6 bas/M84485.] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
39
Fig. 10. Macro and micro-facies of large shelter cavities roofed by archaeocyath saucer-shaped cups (facies C). (a) Macroscopic view of the relationships among facies A, B and C: a large saucer-like cup of archaeocyath (arrow) forms a shelter cavity filled by pink micrite, resting on the pink laminated facies A, which support the light grey massive facies C; both are in lateral contact with the rusty-red interbiohermal facies fB. (b) Large growth cavity roofed by a bowl-cup of Coscinocyathus dianthus Born. supporting cryptobionts: Archaeophatretra ertashkaense Vol. and pendant Renalcis. Cement infills most of the cavity, enclosing arcuate rafts of Razumovskia, Renalcis clusters and small cups of A. ertashkaense Vol. [SCS 5 R• (e) Shelter cavity roofed by a saucer-like cup of Anthomorpha margarita Born. A festooned irregular band of finely crystalline marine cement encases pendant renalcids and drapes the inner wall of the archaeocyath cup. Microboring (arrow) in micritic internal sediment. [SCS 3G II| (d) & (e) Protopharetra (d) and a delicate Razumovskia crust (e) supporting pendant Renalcis clusters. [SCS 4D II; SCS 4A 12_] (f) Cups of Tubicoscinus (left) and Protopharetra (right) associated with Renalcis boundstone surrounding a mud-filled cavity. Protopharetra outer wall thickened in response to microbial encrusting. [SCS 4D II.] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
40
A. GANDIN ETAL.
Fig. 11. (a) Erismacoscinus cancellatus(Born.), fragment of a crenulate transverse section of the cup. [SCS R5 Q/MNHN 84479.] (b) Taylorcyathus vologdiniDebr. (centre) [SCS 3GIII/MNHN 84481], Protopharetra protea (Born.) (top left). [SCS 3GIII/MNHN 84482.] (e) Dictyocyathus tenerrimus(Born.), oblique transverse section. [SCS 3G'IV, 1/MNHN 84483.] (d) Protopharetraprotea (Born.), longitudinal section through the intervallum. [SCS G1/MNHN 84484.] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
Cement-supported Renalcis boundstone (facies D) shows lower diversity and a decreasing number of ajacicyathid genera, represented only by tabulate forms. Clustered Archaeopharetra budding-cups contribute to the bioconstruction
(Fig. 14c, d), while large bowls of Anthomorpha form shelter cavities (Fig. 14a, b). Among the other faunal components, rare sponge spicules, replaced by equant calcite mosaics, occur exclusively in the micritic infiUing of
SARDINIAN EARLY CAMBRIAN REEF COMPLEX Table 1. Facies distribution of the archaeocyath taxa
ARCHAEOCYATHA
'Regulares' Inessocyathus spatiosus (Born.) Afiacyathus alloiteaui Debr. Taylorcyathus vologdiniDebr. Gandinocyathusgravestocki Debr.* Erismacoscinus elongatus (Born.) Erismacoscinus calathus (Born.) Erismacoscinus cancellatus(Born) Porocoscinusflexibilis Debr. Tubicoscinus cupulosus Debr. Tabulacyathus insperatus Debr. Coscinocyathus dianthus Born. Coscinocyathusfornicatus Debr. 'Irregulares' Anthomorpha margarita Born. Dictyocyathus tenerrimus Born. Archaeopharetra ertashkaenseVol. Protopharetraprotea (Born.)
Interbiohermal facies B+G'
Lower Bioherm facies G
A
Upper Bioherm facies C
D
9 9 9 9 9 9 9 9 9 9
Gandinocyathus DEBRENNE F&M; [*G. gravestocki, OD; holotype (Debrenne et al., 1993, plate 3, Fig. 1), M84234 MNHN, Paris]. Outer wall with horizontal to upwardly projecting straight canals, bearing supplementary bracts externally (imparting overall inverted V-shaped appearance to outer wall); inner wall with one row of pores per intersept, bearing upwardly projecting cupped bracts; septa completely porous. Lower Cambrian (Bot.), Europe (Sardinia).
the shelter cavities of facies C and G, whereas in the interbiohermal facies G' they are mostly siliceous and smaller. Rare Chancelloria sclerites, also replaced by equant calcite, are found in the micritic infilling of facies C cavities. They appear more frequently in facies G', where they are sometimes corroded and filled with dense micrite. Rare sclerites are also entrapped in the Renalcis boundstone of facies C and D. Echinoderm ossicles, trilobite sclerites and disarticulated bivalve shells are the main components of the bioclastic packstones (facies B and G'). In the biohermal facies only fragments of trilobites occur, while echinoderm ossicles are notably absent.
Microbial assemblages Calcimicrobes are the primary constructors of the buildups. They form different framework morphologies according to their living requirements. The dominant forms in the 'La Sentinella' buildups are Girvanella-Razumovskia and Botomaella, which often form crusts or rafts up to 15 mm long, and Renalcis, which usually produces dense dendrolitic frameworks. Epiphyton bushes are frequently associated.
Girvanella-Razumovskia act as builders in facies A and C, commonly forming thick and dense sheets which delimit planar (Figs 9c, d & 13c, e) or curved (Fig. 13b) shelter cavities or irregular spongy frameworks characteristic of the thrombolitic facies G (Fig. 7). The fabric of the dense sheets results from the subparallel arrangement of the filaments, while the thrombolitic spongy fabric consists of a nest-like arrangement of fine (Girvanella-Razumovskia) or coarse (Cladogirvanella?) filaments. They outline rounded or irregularly shaped cavities filled by micritic mud and/or calcite spar. Girvanella also acts as an encrusting organism, coating most of the skeletal particles (facies B and G'), and as a builder of oncoids (Fig. 7e). Razumovskia also forms thin arcuate rafts (Figs 9f & 10e), commonly scattered in the cement (Fig. 10b) or in the micritic infilling of the cavities. Botomaella builds thick stacked sheets with narrow intralaminar cavities (facies A) (Fig. 9b, c, e). It can also occur as isolated fans or in dense irregular masses intergrowing with other calcimicrobes. Renalcis builds dense dendrolitic aggregates in facies D (Fig. 14b-d), while in facies A and C
42
A. GANDIN ETAL.
Fig. 12. (a) Tabulacyathus insperatusDebr., longitudinal section (right) and Coscinocyathusdianthus Born. (left). [SCS1BII (a)/MNHN 84475.] (b) Afiacyathus alloiteaui Debr., oblique longitudinal section. [SCS1BIa'/MNHN 84476.] (c) Gandinocyathusgravestocki Debr., oblique transverse section (left), [SCS2F/MNHN 84477]; Erismacoscinus elongatus (Born.), oblique longitudinal section (right). [SCS2F/MNHN 84478.] (d) Gandinocyathusgravestocki Debr., longitudinal section showing the upward bracts of the inner wall and inverted V-shaped canals of the outer wall. [SCS R5/MNHN84480.] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
43
Fig. 13. Macro and microfacies of the microbial-crust boundstone with large shelter cavities (facies C). (a) A square of the sampling grid on the pale-grey massive limestone (facies C). (b) Superposed shelter cavities roofed by Girvanella-Razumovskia crusts with different infills. The lower cavity houses thick clusters of Renalcis and Epiphyton cemented by finely crystalline marine calcite followed by late equant calcite; the roof has been partially destroyed by pressure-solution processes, which spared some remains of the microbial sheet (arrow). The upper cavity is infilled with bioturbated mud-matrix and concentration of spherulites with a micritic core (the enigmatic alga Acanthina?); its roof is delimited by a thick Girvanella-Razumovskia crust, downwards draped by a festoonshaped band of microcrystalline cement encasing pendant Renalcis clusters. [SCS 4AI.] (c) Large geopetal cavity roofed by a microbial crust partially destroyed by pressure solution, and draped by a festoon-shaped band of microcrystalline cement encasing pendant Renalcis clusters; bottom and lateral walls are lined by a rind of stubby blades and locally by multilayered cement. Micritic internal sediment with microboring, filled with late equant calcite. [SCS 4AII1.] (d) Growth cavity lined by a monolayered rind of stubby fibres and blades encasing neighbouring archaeocyath cups and Renalcis clusters. Spherulites with a micritic core (Acanthina alga?), apparently float in the homogeneous micritic matrix. [SCS R5.] (e) Open growth spaces outlined by thick sheets of Razumovskia and clusters of Archaeopharetra ertashkaense Vol. is filled by micritic matrix with scattered microborings. [SCS 1F1 .] [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
44
A. GANDIN ETAL.
Fig. 14. Macro and micro-facies of cement-supported Renalcis boundstone with modular archaeocyath clusters (facies D). (a) Anthomorpha cup supporting the Renalcis--cement boundstone (facies D) shelters floated skeletal material in rusty-red matrix (facies B). (b)-(e) Renalcis dendrolitic boundstone with small growth cavities filled with fine-grained cement. (b) Large saucer-like cup of Porocoscinusflexibilis Debr. [SCS 4E II | .] (c) Cluster of branching Archaeopharetra ertashkaense Vol. with small bushes of Epiphyton (arrow). [SCS 2F)/M 84489.] (d) Budding Archaeopharetra ertashkaense Vol. with irregular growth cavities filled by dense microsparitic internal sediment (arrow). [SCS 5R/M 84490.] (e) Erismacoscinus cancellatus (Born.) (right) and Protopharetra protea Born. (left). [SCS 3B base.] (f) Apex of Anthomorpha cup with exothecal roots enclosed in bladed cement and encrusted by Renalcis. [3G II| [SCS numbers correspond to field sampling, M84xxx numbers to the collection of figured archaeocyaths housed in the Mus6um National d'Histoire Naturelle, Paris.]
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
45
Fig. 15. Evolution of the depositional conditions of the 'La Sentinella' reef complex. As a response to a slow increase of energy and simultaneous decrease of siliciclastic input on the marginal zone of a shallow-marine setting, pioneer microbial communities and archaeocyaths gradually built up structures characterized by increasing synoptic relief: facies G, pioneer thrombolitic build-ups; facies A, planar microbial crusts, colonize the surrounding interbiohermal lime-muds and skeletal debris (facies G' and B); facies C, establishment of a relative relief built by superposed, arched microbial crusts and large saucer-shaped archaeocyaths, forming mud-filled sheltering cavities and supporting cryptobiont communities; facies D, growth to the maximum synoptic relief of 30 cm, of luxuriant Renalcis dendrolitic framework supported by thickets of modular archaeocyaths and early interskeletal marine cement.
it forms pendant clusters of small specimens attached to cavity roofs (Figs 9b, c, 10b, c & 13b-d); only occasionally is Renalcis found attached to hard substrates such as archaeocyath cups (Fig 10d) or microbial crusts (Fig. 10e). Epiphyton is represented by scarce but commonly large clusters, often interwoven with Renalcis (facies C and D) or Botomaella (facies G). Circular-subcircular spherulites, consisting of micritic cores surrounded by elongated fibres of calcite, may represent the enigmatic alga Acanthina (Mankiewicz 1992). They are found 'floating' within the matrix (Fig. 13d) or concentrated at the bottom of large mud-filled cavities (Fig. 13b). Matrix The matrix made of pink or grey homogeneous micrite, only partially fills the central cavities of
archaeocyaths or the elongate or irregular flamework cavities in the crust boundstone (facies C) and thrombolitic facies G to form geopetal structures. It corresponds to a mudstone with rare skeletal remains and spherulites, often found concentrated at the bottom of the cavity (Fig. 13b, d). It contains small elongate or irregular vugs filled by equant calcite (Figs 10c & 13b, c, e) that can be interpreted as being produced by softbodied burrowers and, more rarely, incipient stromatactis structures with geopetal infill. In some cases, this matrix, more than an original muddy internal sediment, can better be interpreted as normal shelf sediment that periodically interrupted the growth of the microbial crusts. On the other hand, the homogeneous matrix found in the small cavities of facies D, consisting of pink coarse microsparite, can be considered the product of diagenetic recrystallization. The matrix of the interbiohermal marly sediment, occurring only in the bioclastic facies B and G', is composed of microsparite formed by
46
A. GANDIN ETAL.
silt-sized skeletal particles associated with siltsized quartz, clay minerals and iron oxides. The iron oxides, locally concentrated by pressuresolution processes, impart to the sediment a characteristic rusty-red colour.
Cements Primary marine cements, which often are well preserved, are characterized by a distinctive yellowish colour. This feature can be related to the possible presence of microdolomite inclusions, which suggests an original high-Mg calcite composition for the cements (James & Klappa 1983). Limpid, equant calcite, interpreted as a late diagenetic cement of meteoric derivation owing to its blocky shape and lack of inclusions (Folk 1974), fills either residual syndiagenetic voids of geopetal cavities or syntectonic veins. The prismatic/blocky crystals are always deeply twinned as the result of intense tectonic stresses. The marine cements are variably developed in the different facies and in general display a small size of the component crystals. In facies C, stubby cone-shaped crystals or blades (type 2 of James & Klappa 1983) form thin monolayered rinds around calcimicrobial crusts and clusters (Renalcis-Epiphyton), or infill part of the geopetal/shelter cavity (Fig. 13d). Less commonly, multilayered crystalline coatings line the inner part of cavity walls (Fig. 13c) and locally encrust isolated archaeocyaths or envelope the micritic core of spherulites (Fig. 13b, d). Microcrystalline inequigranular mosaic cements form thick festoon-like rinds hanging from shelter cavity roofs (Figs 10b, c & 13b, c) and encase small Renalcis clusters (Figs 10b, c &
13b, c). Fine-grained inequigranular mosaic cements are dominant in facies D, infilling the skeletal pores of Renalcis dendrolites (Fig. 14c-f). The intraskeletal open spaces of archaeocyaths (Fig. 14c-f) and the small angular framework cavities, later occluded by coarse microsparite (Fig. 14d), are often lined by rinds of coneshaped blades.
Facies relationships The depositional setting and evolution of the 'La Sentinella' reef complex results from detailed analyses of the microfacies, their relationships and spatial distribution. The basal part of both the Lower and the Upper Bioherm, consisting of marly lime-muds composed of skeletal mudstone to wackestone to floatstone (facies B and G'), indicates a shallow-marine setting that in the
beginning appears to have been rather restricted (facies G'). This low-energy setting hosted a diverse association of archaeocyaths, dominated by small solitary forms. The marly sedimentation (facies G ) also documents the start of the carbonate deposition that can be connected with the decrease of the siliciclastic input consequent to the onset of more arid climatic conditions, on the shallow siliciclastic shelf of the Matoppa green sandstones. On the restricted mudstones (facies G), a pioneer construction was built by a GirvanellaRazumovskia-Botomaella association that formed a thrombolitic framework. Fragments and cups of archaeocyaths with corroded or encrusted outer walls appear to have been floated from the interbiohermal muddy bottom and entrapped within the microbial framework. These features, as well as the poor development of cements and the black colour of the rock, suggest a persistent restricted environment and general low-energy conditions (Fig. 15) that, however, may have been occasionally interrupted by storms of moderate energy. Pioneer bioconstruction, built by stacked thick, flat crusts of Botomaella and GirvanellaRazumovskia (facies A), grew in well-oxygenated conditions trying to stabilize the loose sediment of the bottom (facies B). Archaeocyaths are relatively rare and isolated. The microbial colonization, at least in the beginning, was repeatedly interrupted by the sedimentation of marly limemuds. The growth and increased thickness of the microbial crusts was enhanced by the disappearance of the clay input from the continent. The depositional conditions of this peculiar type of bioconstruction can be referred to a slight increase of the ambient energy due to steady bottom currents (Fig. 15). The presence of a hard substrate promoted the growth of the microbial crusts, which gradually evolved into arcuate thin crusts and rafts (facies C). The crust and the rare but large saucer-like archaeocyaths delimit large growth cavities, lined on the ceiling by thick festoon-like rinds of microcrystalline cement (Figs 10c & 13b) enclosing small clusters of pendant Renalcis and small cryptic archaeocyaths (Fig. 10b). The origin of this peculiar festoon-like microcrystalline cement is uncertain: it may have been produced by some other 'non-skeletal' microbial organism or may have resulted from precipitation of calcite on the EPS (extracellular polymeric substance) mucilages enveloping the renalcid clusters. The growth cavities can be filled with associations of Renalcis-Epiphyton dendrolites associated with fans of Botomaella,
SARDINIAN EARLY CAMBRIAN REEF COMPLEX but more frequently are filled with pink homogeneous mudstone containing rare skeletal remains and spherulites that are often concentrated at the bottom. Small scattered irregular voids filled with equant calcite probably originated from bioturbation. Archaeocyaths are relatively rare and isolated. The depositional features of this delicate type of buildup appear to reflect a welloxygenated setting (Fig. 15) with slow circulation of waters that deposited poorly developed cement rinds of short stubby blades on the walls of large cavities mainly supported by microbial crusts. Towards the top of the bioherm the microbial crusts (facies C) gradually decrease in number and size, and are replaced by cement-rich boundstone (facies D) containing a few small growth cavities that sometimes are lined by a rind of short stubby blade calcite and filled by rustyred coarse microsparite. Renalcis and modular archaeocyaths become the main framebuilders, associated with large clusters of Epiphyton. The microbial dendrolitic framework is sustained by the microcrystalline cement formed within the small interskeletal growth cavities (Fig. 14c-f). These features indicate a rather resistant framework that formed in response to the higher energy level (Fig. 15) reached by the 'La Sentinella' reefal environment.
Comparison with other Lower Cambrian bioconstructions In the basal part of the 'La Sentinella' reef complex, the Lower Bioherm, made up of dark grey limestone, rests directly on the green sandstone of the Matoppa siliciclastic shelf. The pioneer Girvanella and Razumovskia communities form a spongy fabric disrupted and disconnected by soft-bodied burrowers (facies G), which often contains corroded and encrusted cups of stranded archaeocyaths. The features of this construction (Fig. 7), which can be referred to a thrombolite (Kennard & James 1986; Kennard et al. 1989), and those of the associated skeletaloncoidal mudstones (facies G') can be compared to similar encrusting communities living in a muddy environment associated to oncoidal deposits, described in the Early Cambrian Tianheban Formation of China (Debrenne et al. 1991; Gandin & Luchinina 1993). Both units rest directly on marine sandstones and record the beginning of the carbonate deposition on a siliciclastic shelf. They represent the pioneer stage of the reef development in a restricted, low-energy shallow subtidal environment still polluted by land-derived input of fine-grained
47
detrital material. The lower part of the Upper Bioherm represents unusual facies in which crusts and rafts of Girvanella-RazumovskiaBotomaella are the major framework builders, with accessory dendrolitic Renalcis and Epiphyton (facies A and C). We apply the term 'crust boundstone' buildup to bioconstructions of this type. Similar crust fabrics, often built by different microbial communities and usually found as isolated occurrences enclosed in different frameworks, have been described in blocks of the Cow Head Breccia in Newfoundland (James 1981), and of the Shady Dolomite in Virginia (Read & Pfeil 1983), in Nevada from the Poleta Formation (Rowland & Gangloff 1988) and Battle Mountain (Debrenne et al. 1990), in Mexico from the Puerto Blanco Formation (type 2 buildups: Debrenne et al. 1989), in Mongolia at Zuune Arts (Wood et al. 1993) and Zavkhan Basin (Kruse et al. 1996). The crust boundstone buildup (facies A, Figs 8 & 9a-c) has a layered fabric that, according to Wood et al. (1993), represents a substrate surface that was alternately colonized by calcimicrobial crusts and then covered by sediments which temporarily stifled growth. This structure does not build synoptic relief but provides a hard substrate for the successive stage of development of the bioherm (facies C). It corresponds to the stabilization stage of the reef. The crust boundstone buildup (facies C, Figs 10 & 13) is characterized by superposition of shelter cavities, delimited by arched crusts of Girvanella-Razumovskia and saucer-like archaeocyaths supporting Epiphyton, Renalcis and Archaeopharetra as cryptic dendrolitic components. However, in each documented case cited above, there is an obligate microbial component with erect dendrolitic growth habit, represented by Tarthinia, Epiphyton and Gordonophyton in Mongolia, and by Epiphyton and Renalcis in Mexico. In the Sardinian 'La Sentinella', this component is represented by Renalcis or Epiphyton. The resulting construction grown in shallow, moderately turbulent waters, reached a very low relief of about 30 cm. The upper part of the Upper Bioherm (facies D, Fig. 14) is a typical Early Cambrian Renalcis-cement buildup with subordinated archaeocyaths, characteristic of low- to moderate-energy, open-shelf environments. Such buildups range from the Tommotian to the Toyonian in Siberia (James et al. 1990; Kruse et al. 1995) and are widely documented from Antarctica (Rees et al. 1989), Mexico (Debrenne et al. 1989) and Australia (James & Gravestock 1990). They may be kalyptrate, composite or
48
A.
GANDIN
E T AL.
,,,d
"~~
i ,' f 1 1 1
//////
I'/I'tll',t,/~it/ " i 9 :
< < <
/,/////,. ,
,
,
,
9
229
< <
<
~
>~ ~ "
9. :.
c0~
nZl ~.j
0
0
~n
0
0--
m m
\n @~
< < < < < < < <
< < < ~<
o
//~)/~/.~ //~..FP~:
i
,>
,I
< o ~ . ~
~.~..~.~..
> ) )
/
/
9
c~
:~ . ~
9~
.~. c,~~
u-~ ,-~ ~
I-~ ~
~.'~
:~.~-~ = 0
~ ~ • :~ ~t ,,.....: ~~ I:'
0
~
~
0
~
~ ~.:~ ~,~.~.~
~
~
o
///////////A / / // / / / / / / / // ////////]
i ,. ".
//////////////:: ".
/ // / / / / / / / / / / '
9~ , - ~
~
~
"
SARDINIAN EARLY CAMBRIAN REEF COMPLEX
49
involved with Franqoise's work. Cambrian research workers will miss him. S. Rowland and an anonymous reviewer made pertinent suggestions and language ameliorations that improved the final version of this paper. The substantial contribution of B. Terrosi and A. Mancini in the assemblage of figures and preparation of drawings is greatly appreciated.
massive, as determined by the relative proportions of Renalcis, archaeocyaths and cement, as well as by the growth habit of Renalcis (encrusting v. dendrolitic). The depositional conditions of the Upper Bioherm seem to have become progressively more energetic. In the last phase of evolution of the 'La Sentinella' reef complex (facies D), the increase in the water energy is one of the factors that can explain the development of the building activity of dendrolitic Renalcis and archaeocyaths, the disappearance of mud and microbial crusts, and the contemporaneous development of calcite cements. Thus, the complex intergrowth of calcimicrobes, archaeocyaths and cements, the builders of the superposed units of the 'La Sentinella' reef complex (Figs 15 & 16), appears to reflect a probable slight variation towards more arid climatic conditions, and a relative increase in energy from moderate in the Lower Bioherm to higher in the Upper Bioherm, within the marginal zone of an open shelf. We suggest that two conditions were necessary for the genesis of the flat- and arched-crust boundstones. The first was an obligate calcimicrobe with erect growth in sufficient quantity to provide synoptic relief. The second was an episodic supply of fine-grained calcareous sediment followed by periods of non-deposition to enable calcimicrobial crusts to re-cover the substrate. These conditions appear to be satisfied by the calcimicrobial boundstones of Zuune Arts bioherms, which grew up to 2 m in diameter (Wood et al. 1993). The Upper Bioherm at 'La Sentinella' also satisfies these requirements. This bioherm is up to 1 m thick with synoptic relief up to 30 cm, but, as a result of folding, the lateral dimensions are unknown. The Zuune Arts crust boundstone buildups are at the base of a succession interpreted to represent upwards-deepening conditions. Thus, the buildups grew in agitated, shallow subtidal conditions (Wood et al. 1993). The Mexican crust boundstone biostromes are interpreted to have developed in a lagoonal setting protected by an oolitc shoal complex (Debrenne et al. 1989). The peculiar types of crust buildups (facies A and facies C) found in the 'La Sentinella' reef complex have never before been recorded in the Lower Cambrian of Sardinia or, as far as we know, in other regions of the world. They represent the second stage of evolution in the ecological succession of the 'La Sentinella' reef complex, which developed on the Matoppa shelf during a temporal interruption of the siliciclastic deposition.
GALASSI, R. & GANDIN, A. 1992. New structural data and their bearing on the Cambrian stratigraphy of the Iglesiente region (Sardinia, Italy). Comptes Rendus de l'Acadkmie des Sciences, Paris, 314, 93-100. GANDIN, A. 1987. Depositional and paleogeographic evolution of the Cambrian in southwestern Sardinia. In: SASSI, F. P. & BOURROUILH,R. (eds). IGCP No. 5: Newsletter 7, 151-166. GANDIN, A. & DEBRENNE, F. 1984a. Paleoenvironmental features and paleoecology of the Lower Cambrian of Sardinia. In: Abstracts of the 5th Euro-
In memory of David Gravestock and of Max Debrenne. 54 years of everyday life and scientific collaboration ended with Max's sudden death. He was closely
pean Regional Meeting of Sedimentology, Marseille, 183-184. GANDIN, m. & DEBRENNE,F. 1984b. Lower Cambrian bioconstructions in Southwestem Sardinia (Italy). Gkobios, 8, Mdmoire Sp6cial, 231-240.
References CARMIGNANI, L., COCOZZA, T., GANDIN, A. & PERTUSATI, P. C. 1986. The geology of Iglesiente and description of stops. In: CARMIGNANI, L., COCOZZA, T., GHEZZO, C., PERTUSATI, P. C. & RIco, C. A. (eds) Guide-book to the Excursions on the Paleozoic Basement of Sardinia. IGCP No. 5: Newsletter, Special Issue, 31-49. DEBRENNE, F., GANDIN, A. & DEBRENNE, M. 1993. Composition faunique des calcaires du membre de Matoppa (Formation de Nebida), Cambrien inf6rieur du sud-ouest de la Sardaigne (Italie). Annales de Pal~ontologie, 79, 1-42. DEBRENNE, F., GANDIN, m. & GANGLOFF,R. A. 1990. Analyse s6dimentologique et pal6ontologique des calcaires organog6nes du Cambrien inf6rieur de Battle Mountain (Nevada, USA). Annales de Palkontologie, 76, 73-119 DEBRENNE, F., GANDIN, m. & ROWLAND,S. M. 1989. Lower Cambrian bioconstructions in northwestern Mexico (Sonora). Depositional settings, paleoecology and systematics of archaeocyaths. G~obios, 22, 137-195. DEBRENNE,F., GANDIN,A. & SIMONE,L. 1979. Studio sedimentologico comparato di tre 'lenti' calcaree ad archeociati dell 'Iglesiente e Sulcis (Sardegna Sud-occidentale). Memorie della Societh Geologica Italiana, 20, 379-393. DEBRENNE, F., GANDIN, A. & ZHURAVLEV, Yo. A. 1991. Paleoecological and sedimentological remarks on some Lower Cambrian deposits of the Yangtze Platform (China). Sociktk Gkologique de France, Bulletin, 162, 575-584. FOLK, R. L. 1974. The natural history of crystalline calcium carbonate: effect of magnesium content and salinity. Journal of Sedimentary Petrology, 44, 40-53.
A. G A N D I N ETAL.
50
GANDIN, A. & LUCHININA, V. 1993. Occurrence and environmental meaning of the Early Cambrian calcareous algae of the Tianheban Formation of China (Yangtze Area). In: BARATTOLO, F., DE CASTRO, P. 8~ PARENTE, M. (eds) Studies on Fossil
Benthic Algae. Societ3 Paleontologica Italiana, Special volume l, 211-217. JAMES, N. P. 1981. Megablocks of calcified algae in the Cow Head Breccia, western Newfoundland: vestiges of a Cambro-Ordovician platform margin. Geological Society of America Bulletin, 92, 799-81 I. JAMES, N. P. & GRAVESTOCK, D. I. 1990. Lower Cambrian shelf and shelf margin buildups, Flinders Ranges, South Australia. Sedimentology, 37, 455-480. JAMES, N. P. & KLAPPA, C. F. 1983. Petrogenesis of Early Cambrian reef limestones, Labrador, Canada. Journal of Sedimentary Petrology, 53, 1051-1096. JAMES, N. P., KRUSE, P. D. & ZHURAVLEV, A. Yu. 1990. Tommotian and Toyonian bioherms, the same basic plan. In: REPINA, L. N. & ZHURAVLEV, A.Yu. (eds) Abstracts" of the Third Intemational
Symposium on the Cambrian System, Novosibirsk, 118. KENNARD, J. M. & JAMES, N. P. 1986. Thrombolites and stromatolites: Two distinct types of microbial structures. Palaios, 1,492-503. KENNARD, J. M., CHOW, J. & JAMES, N. P. 1989. Thrombolite-Stromatolite Bioherm, Middle Cambrian, Port au Port Peninsula, western Newfoundland. In: GELDSETZER,H. H., JAMES, N. P. & TEBBUTT, G. E. (eds) Reefs, Canada and Adjacent
Area. Canadian Society of Petroleum Geologists, Memoir, 13, 151-155.
KRUSE, P. D., ZHURAVLEV,A.Yu. & JAMES,N. P. 1995. Primordial metazoan-calcimicrobial reefs: Tommotian (Early Cambrian) of the Siberian Platform. Palaios, 10, 291-321. KRUSE, P. D., GANDIN,A., DEBRENNE,F. t~; WOOD, R. 1996. Early Cambrian bioconstuctions in the Zavkhan Basin of western Mongolia. Geological Magazine, 33, 429-444. MANKIEWICZ, C., 1992. Proterozoic and Early Cambrian Calcareous Algae. In." SCHOVF, J. W. & KLEEN, C. (eds) The Proterozoic Biosphere. Cambridge University Press, Cambridge, 359-367. Read, J. F. & Pfeil, R. W. 1983. Fabrics of allochthonous reefal blocks, Shady Dolomite (Lower to Middle Cambrian), Virginia Appalachians. Journal of Sedimentary Petrology, 53, 761-778. REES, M. N., PRATT, B. R. & ROWELL,A. J. 1989. Early Cambrian Reefs, reef complexes, and associated lithofacies of the Shackleton Limestone, Transantarctic Mountains. Sedimentology, 36, 341-361. REGOLI, R., GANDIN, A. & ELTER, F. M. 1995. The Lower Cambrian Nebida Formations. In: CHERCHI, A. (ed.) Proceedings of the Sixth Paleobenthos International Symposium, Alghero, 167. ROWLAND, S. M. & GANGLOFF, R. A. 1988. Structure and Paleoecology of Lower Cambrian reefs. Palaios, 3, Reefs Issue, 111-135. WOOD, R., ZHURAVLEV,A.Yu. & CHIMED TSEREN, A. 1993. The ecology of Lower Cambrian buildups from Zuune Arts, Mongolia: implications for early metazoan reef evolution. Sedimentology, 40, 829-858.
Botoman (Lower Cambrian) turbid- and clear-water reefs and associated environments from the High Atlas, Morocco J. J A V I E R A L V A R O 1'2 & S l ~ B A S T I E N C L A U S E N 2
1Departamento Ciencias de la Tierra, Universidad de Zaragoza, Ciudad Universitaria, 50009-Zaragoza, Spain 2LP3, U M R 8014 CNRS, Universitd des Sciences et Technologies de Lille, 59655- Villeneuve d'Ascq, France (e-maik
[email protected]) Exposures of the Botoman (Lower Cambrian), Lemdad and Issafen formations on the Lemdad syncline, southern High Atlas, provide an excellent example of the interactions between tectonic events, magmatic activity and carbonate productivity. The major factors that controlled the nucleation of carbonate factories on the Botoman High Atlas platform were: (i) synsedimentary tectonism, as normal faulting resulted in tilting of fault blocks causing irregular topographies and subsequent sharp erosion; (ii) volcanism, because pyroclastic influx smothered carbonate factories except in distal areas of the platform or during quiescent episodes of volcanic activity; and (iii) the influence of successive shoaling parasequences. The Botoman reefs exhibit a wide range of external morphologies, including tabular (biostromes) and domal (bioherms and patches) boundstones, which do not exceed 3.5 m of thickness. Although archaeocyathan-microbial reefs only developed under clearwater conditions, microbial reefs grew also under turbid-water conditions. Domal and digitate stromatoids, Girvanella crusts, Epiphyton bushes and thromboid-stromatoid intergrowths document the ability of some microbial communities to develop heterotrophic strategies when submitted to a moderate terrigenous input. Turbidity was a major ecological factor that constrained development of filter/suspension-feeder and phototrophic organisms, but not necessarily of benthic non-phototrophic microbial communities. Abstract:
Although it is currently accepted that terrigenous input inhibits carbonate productivity (mainly frame-building factories), new research on modern and Phanerozoic reefs is increasing in siliciclastic shorelines and platforms recording volcanoclastic influence. There, although siliciclastic and volcanoclastic influx affects biotic development, some carbonate factories can exhibit a broad tolerance to continuous or episodic increases in water turbidity (Woolfe & Larcombe 1999; Larcombe et al. 2001; Wilson 2000, 2005; Wilson & Lockier 2002). Turbid waters can spread during rifting episodes, as they are associated with distinct volcanic activity, and are commonly accompanied by episodic instability of the sea floor and marked asymmetric changes in sedimentary architecture. Successive rifting episodes are reported from the Moroccan margin of West Gondwana across latest Neoproterozoic and Cambrian times (Boudda et al. 1975, 1979). They are associated with a distinct igneous activity recorded in the Atlas with alkaline and/or tholeiitic affinity (Ezzouhairi 2001). This magmatic activity reflects a tectonic inversion from a Neoproteorozic (Pan-African) compressive margin (Badra et al. 1992; Jouhari et al. 2001; Ouazzani
2001) to intraplate extension related to latest Neoproteorozic-Early Cambrian rifting processes (Piqu6 et al. 1995). Hup6 (1955) was the first to suggest a major episode of Cambrian reorganization of basinal geometries in Morocco. He tried to explain the apparent occurrence of discordant features associated with the setting of the Micmacca Breccia at the base of the Middle Cambrian 'Paradoxides Shales' in the Anti-Atlas. He associated the setting of this apparent discordance with a supposed 'Salairian' tectonic phase in Morocco. Subsequently, Boudda et al. (1975, 1979) proposed a widespread subaerial exposure and erosion of large parts of the basin at the end of this episode to explain abrupt modifications in thickness of facies and lithologies. Recently, Alvaro & Clausen (2005, 2006) have related the so-called 'Salairian' tectonic phase to a rifting interval with tilting and karstification. Other episodes of basin reorganization predating the Moroccan 'Sala'irian' phase took place in Botoman (Early Cambrian) times, as recorded in the southern rim of the High Atlas. This paper is focused on the analysis of the sedimentary factors that controlled the nucleation, growth and disappearance of the Botoman microbial
From:/~LVARO,J. J., ARETZ, M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 51-70. 0305-8719107/$15.00 9 The Geological Society of London.
52
J.J. JkLVARO & S. CLAUSEN
and archaeocyathan-microbial reefs that occur in the Lemdad and Issafen formations of the southern High Atlas. This tectonically and magmatically active area of the Moroccan platform favoured development of turbid waters, which directly affected the type of carbonate factories.
Geological setting and stratigraphy The Neoproterozoic(?)-Cambrian successions of the Moroccan Atlas are located in the Anti-Atlas and central High Atlas mountains, although some disconnected outcrops occur in the Jbilet and Rehamna regions, and the Meseta plateau (Fig. 1a). The axis Of the Late Proterozoic-Early Palaeozoic Souss Basin (Geyer 1990a) roughly coincides with the modern trend of the AntiAtlas (SW-NE), where common west-to-east facies changes throughout the Cambrian successions reflect the eastern setting of proximal areas (Destombes et al. 1985). To differentiate between the specific palaeogeographical patterns of the High Atlas and the Anti-Atlas, we will refer to them below as the High Atlas and Anti-Atlas platforms. This paper is focused on the analysis of the Lemdad and Issafen formations cropping out on the Lemdad syncline. The NW-striking Lemdad (from 'Oued el Mdad') syncline, called 'Ounein' by some authors, is situated at the southern margin of the west-central High Atlas (Fig. lb). We have selected for description herewith the westernmost and easternmost outcrops, named sections Lel and Lel 1, respectively, by Geyer & Landing (1995). Lel is the most complete Lower Cambrian section from the High Atlas (Fig. 2): it is the stratotype of the Lemdad Formation
(Geyer 1990a), and is considered as reference for Lower Cambrian stratigraphic correlations in the High Atlas (see Boudda et al. 1975, 1979; Siegert 1986; Geyer & Landing 1995 for lithological and biostratigraphical precisions). Numerous trilobite-bearing strata are known throughout the Lemdad and overlying Issafen formations, which allow identification of the Antatlasia guttapluviae and Sectigena zones (Geyer 1990b) (Fig. 2). In addition, the identification of two major archaeocyathan-bearing reefs in Lel (previously reported as Ounein A and B; see last revision in Debrenne & Debrenne 1995) allows improvement in biostratigraphic correlations. A third archaeocyathan-bearing reef is located in the Lemdad Formation of section Lel 1; it was previously reported as 'Ounein C' or 'bioherme de la cascade' (Fig. 2; see Debrenne & Debrenne 1995 for further information), and is included in the Sectigena Zone owing to the presence of the trilobite Berabichia cf. vertummnia (Heldmaier 1997). The Lemdad Formation (Geyer 1990a) is a heterolithic succession composed of marlstones, siltstones, limestones and dolostones (some of them microbial and archaeocyathan-microbial reefs), with volcanic ash and volcanoclastic intercalations. It is geographically restricted to a part of the central High Atlas, and its thickness is estimated to be 500 m in the Lemdad syncline. There, a 517__+1.5 Ma age is available (Landing et al. 1998) from volcanic ashes (sample Lel 1/ 10.8 of Geyer & Landing 1995) interbedded in the volcanoclastic sandstones of the Lemdad Formation, and lying in the upper Antatlasia guttapluviae Zone of the Moroccan Lower Cambrian Banian Stage (Botoman Stage according to the Siberian scale; Spizharski et al. 1986).
Fig. 1. (a) Geological sketch of the Neoproterozoic-Cambrian rocks in the High Atlas and Anti-Atlas, Morocco, with setting of the Lemdad syncline. (b) Geological map of the Lemdad syncline with situation of sections Le 1 and Lel 1 (modified from Destombes et al. 1985 and Geyer & Landing 1995).
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO
53
Fig. 2. (a) Cambrian stratigraphic framework of the Lower-Middle Cambrian transition in the Lemdad syncline. (b) Summarized sections Lel and Lel 1 (after Geyer & Landing 1995, Heldmaier 1997 and this work).
The Issafen Formation (26-180 m thick) consists of shales with secondary calcareous and/or sandstone beds in its lower part, and nodular limestones (named 'calcaires scoriac6s' by some authors) in its upper part on some areas of the western Anti-Atlas. The formation is widely exposed along the Anti-Atlas and the southern High Atlas. In the Lemdad syncline the formation is extremely thin by erosion. Following the
original definitions of 'S6rie schisto-calcaire' (Choubert 1958), 'Issafene substage' (Boudda et al. 1979), Issafen Shales (Siegert 1986) and Issafen Formation (Geyer 1990b), the base of the Issafen Formation consists of slightly micaceous, finely laminated shales (Geyer & Landing 1995, p. 20). As a result, we do not place its base at the erosive base of the reef complex 'Ounein B', coinciding with a biostratigraphic boundary (as
54
J.J. ALVARO & S. CLAUSEN
proposed by Geyer & Landing 1995), but at the occurrence of thick shales overlying the disappearance of volcanoclastic sandstones and conglomerates in Lel. The Issafen Formation shows a drastic decrease in thickness eastwards in the Lemdad syncline, disappearing in Lel 1 (Fig. 2). The following sections present description and environmental interpretation of the facies associations found in the Lemdad and Issafen formations of sections Lel and Lel 1 (see detailed logs and parasequences in Fig. 3). Facies associations are divided into: (i) volcanoclasticdominated associations; (ii) carbonate (non-reef) associations; (iii) microbial reefs; and (iv) archaeocyathan-microbial reef complexes.
Volcanoclastic-dominated facies associations Volcanoclastic sediments are subdivided in this paper into pyroclastic (volcanogenic) and epiclastic strata. The term 'epiclastic' is used to denote sediment where most clasts are derived from weathering of pre-existing sedimentary rocks (Fisher & Smith 1991), despite a minor presence of reworked pyroclasts. In some cases, differentiation of strata primarily composed of reworked pyroclasts and epiclasts can only be made based on matrix composition. For this reason, following Riggs et al. (1997), volcanic detritus is referred to as feldspar 'crystals', whereas sediment detritus is referred to as quartz or feldspar 'grains'. In addition, we will follow Smith's (1991) differentiation between: (i) syneruptive sediments, in which the dominant material is pyroclastic, mainly in the ash fraction; and (ii) intereruptive sediments, related to normal or background conditions lacking a significant pyroclastic content both in matrix and clasts.
Tidally influenced epiclastic strata This facies association occurs exclusively in the Lemdad Formation of Lel 1, and is made up of amalgamated, large-scale, planar and trough cross-stratified beds. They consist ofweUsorted, very-fine to medium-grained arkoseslithoarenites. The upper and lower boundaries of the individual lenticular sets (0.3-2 m thick) are sharp and some distinctly erosive. They are commonly bounded by millimetre-scale, clay laminae and shaly draps, and display both planar and trough cross-stratification, numerous reactivation surfaces, and parallel, low-angle and waveripple lamination. Trough cross-stratification surfaces have a sigmoidal configuration, with tangential bottomset cross-laminae. The lenticular sets are amalgamated into thickening- and coarsening-upwards sequences,
generally less than 4.4 m thick. Palaeocurrents, determined from dip azimuth of foresets (largescale cross-bedding) and plunge direction of trough axes, demonstrate reverse-flow polarities. Although the dominant flow trends towards the SE (120~176 thinner sets capping the thickening-upwards sequences exhibit N N W (290o-330 ~ palaeocurrents, characterized by less-steep foresets and ripple cross-lamination. The large-scale, planar and trough crossstratification, which commonly displays vertical palaeocurrent bimodality, represents part of a system where the dominant current regime changed rapidly from flood to ebb. Although distinct herringbone cross-bedding is absent, the trough cross-stratification shows bi-directional palaeocurrents, dominated by SE (flood) currents, and secondary NW (ebb) ones. Various diagnostic structures document the influence of tidal currents (Meyer et al. 1998; Myrow 1998), such as clay laminae (similar to present-day clay drapes occurring in tidal still-stand deposits), reactivation surfaces shown by cross-bedded sets (resulting from time-velocity asymmetry of tidal currents) and bi-directional foreset units. As a result, the reported thickening-upwards sequences may represent progradational trends of bars related to a tidal channel system, in which siliciclastic grains were primarily related to reworking of epiclastic material.
Shoaling epiclastic strata This facies association occurs in the Lemdad and Issafen formations of Lel. The fine- to mediumgrained arkoses and lithoarenites have erosive and abrupt lower and interbedded contacts, scouring into the underlying sediments. It comprises large-scale, trough and planar crossbeddings with sets ranging in thickness from 0.4 to 1.8 m, topped by symmetrical ripples and ripple lamination. Smaller scale sets of trough cross-bedding, of the order of 10-30 cm, are volumetrically less important and occur interbedded with the larger sets. Some intercalated centimetre-thick limestone nodules and layers have yielded trilobite, echinoderm and brachiopod debris, and other undetermined fossils. Palaeocurrent measures from the dip azimuth of foresets of large-scale cross-bedding (thicker than 0.5 m), and plunge directions of trough axes demonstrate unidirectional flow, dominantly north-south, with polymodal foreset migrations ranging from 240 ~ to 330 ~. The ubiquity of decimetre-scale cross-bedding indicates that the sandstones were originally deposited as large migrating bedforms. The predominance of trough shapes further shows
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO
55
Fig. 3. Representative parasequences of sections Lel and Lel 1 (see Fig. 2 for stratigraphic setting of A-A', B-B', C-C' and D-D').
56
J.J. ALVARO & S. CLAUSEN
that they were short-crested megaripples and sandwaves, where reactivation surfaces are rare. The association of sedimentary structures suggests a shallow platform dominated by wave activity, and minor storm reactivation related to interbedded erosive contacts. Episodes in which the bedforms ceased to migrate and were colonized by an open-sea benthic shelled community can be recognized by the presence of marine benthic fossils within carbonate and shale intercalations.
Pyroclastic channels and shoals Pyroclastic deposits display coarse planar bedding and laminae (Fig. 4a), wave-rippled, and small- to large-scale trough cross-bedding (0.21.8 m thick), with lenticular geometries, erosive bases and planar tops (Fig. 4b). Medium-scale cross-laminae, typically dipping subparallel to the imbrication of sigmoidal beds, form the common interval structure of the facies. Channels are filled with thinning- and fining-upwards sets, up to 2 m thick, with a lenticular geometry
and erosive bases. Palaeocurrents are similar to those reported in the preceding shoaling epiclastic strata. Pyroclastic strata consist of clast-supported, granule- to medium-grained lithoarenites, with a rounded to very angular, polymictic clast association (Fig. 4c) and an ash matrix. Unstable pyroclastic fragments are composed of microlithic (lathlike), microgranular felsitic or vitric textures, felsitic volcanic rock fragments of feldspar and mafic mineralogy, angular rhyolitic grains, reworked ooidal and bioclastic grainstones and packstones, and hematite clasts. The matrix, less than 10% in volume, consists of silty-sandy rock fragments locally cemented by hematite. Although pyroclastic dispersal was probably primarily related to sheetwash or hyperconcentrated volcanogenic flow, the pyroclastic material was subsequently reworked by marine wave and benthic currents. The aforementioned sedimentary structures, scarcity of matrix, geometry, grain size and shape suggest that these clast-supported pyroclasts were reworked as marine channels and shoals, induced by wave
Fig. 4. (a) Coarse planar to low-angle bedding and laminae in a pyroclastic sheet. (b) Amalgamation of pyroclastic shoals with erosive boundaries (arrowed), covered by lag imbrication (i). (e) & (d) Thin-section photomicrographs under plane-polarized light. (e) Mixed pyroclastic lithoarenite rich in mafic (mc), hematite (ht), polyphase calcite-cemented (c) clasts, and trilobite sclerites (t); scale= 6 mm. (d) Pyroclastic lithoarenite (bottom) encrusted by Epiphyton bushes (top); scale = 2 mm.
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO and unidirectional tractions. Locally, the channels and shoals were encrusted by microbial boundstones (Fig. 4d) during episodes of quiescence. The angular feldspars were not significantly transported, but probably yielded by volcanic explosion from neighbouring areas.
Shales and interlaminated claystones and siltstones These facies underlie, overlie (with onlapping geometries) and are interbedded with the aforementioned shoals, as well as with reef fabrics. They are dominated by fine quartz and angular feldspar silt (the latter locally up to 60% in volume), sericite, illite and muscovite flakes. Shales commonly pass gradually upwards into interlaminated clasystones and siltstones. The former are commonly massive, although their stratification is suggested by the layering of centimetre-thick carbonate nodules, which display bioclastic wackestone textures. Interlaminated claystones and siltstones are laminated at a millimetre scale, and include cross-laminated stringers of silty sand and symmetrical ripples of centimetre scale. Bioturbation, although commonly rare, can locally disrupt bedding. When present, partly articulated skeletons are trilobites, hyolithids and linguliformean brachiopods. This facies association was deposited on offshore substrates, under calm-water conditions as evidenced by the dominance of siliciclastic mud, and episodic preservation of a partially articulated shelly fauna. The interlaminated siltstones suggest fluctuations in energy, and accumulated above fair-weather wave base.
Carbonate (non-reef) facies associations
Grainstone shoals They occur as amalgamated beds and lenses, up to 2.4 m thick. They show vertical modifications in composition and texture, changing in ascending order from: (i) intraclastic lags and/or bioclastic packstone-siltstone alternations; to (ii) mixed peloidal-ooidal-pyroclastic grainstones; and (iii) oncoidal-dominated grainstone caps. Microbial reefs (described in next section) encrust some peloidal-ooidal-pyroclastic grainstone foresets and tops, and are also sometimes covered by oncoidal-dominated caps. (i) The lower, sand- to granule-sized, intraclastic lags (up to 0.1 m thick), which occur overlying scoured bases with flame structures, are clast-supported and chaotically arranged.
57
Clasts are randomly oriented, angular in shape, and commonly consist of packstones of peloids and peloidal aggregates. Lags pass vertically into millimetre-thick peloidal-ooidal packstonesiltstone rhythmites, rich in low-angle erosive surfaces and angular packstone intraclasts (less than 1 cm in size), where the content in siltstone decreases gradually upwards. (ii) The overlying strata (up to 1 m thick) show undulatory- and planar-laminar bedding, rare low-angle cross-laminae, irregular laminoid fabrics, and numerous low-angle erosive surfaces (up to 3 cm deep). These display gradual transitions between two different textures: peloidalpyroclastic and ooidal-pyroclastic grainstones. The peloidal-pyroclastic grainstones are composed of 50-70% peloids and peloidal aggregates (up to 1 mm in diameter), 10-30% subangular pyroclasts (up to 4 mm in size), and 5-10% quartz silt. The ooidal-pyroclastic grainstones contain well-preserved ooids (less than 2.5 mm in diameter) with complex radial-concentric cortices. Nuclei are commonly pyroclasts, feldspar and quartz grains, and skeletal fragments. Pyroclasts are very-coarse- to medium-grained sands (less than 2 mm across), subangular-subrounded in shape and some of them polyphasic in nature (Fig. 5a). Scattered bioclasts range from 10 to 40% in volume, and consist of echinoderm ossicles, brachiopod valves and trilobite sclerites. Pyroclasts and feldspars commonly exhibit single or double lamella, a coating process observed too around composite ooids and mixed aggregates of ooids and pyroclasts. The whole framework is cemented with a mosaic of fine, equant-bladed calcite cements (100-300 ~tm in size) occluding interparticle pores. (iii) The peloidal-ooidal-pyroclastic grainstones are arranged into thickening-upwards trends (2.0 m thick) capped by microbial reefs (described below) and/or thin (c. 0.3 m) oncoidaldominated grainstone lenses. The size average of the observed rounded-subrounded oncoids is about 1-3 mm (Fig. 5b), although they can reach in some levels 1 cm in diameter. The oncoids commonly show complex cortices evidencing changes in aggregation: for example (i) distinct, wrinkled laminae composed of alternating layers of dense, light-coloured carbonate-rich laminae and porous, darker, organic-rich laminae; and (ii) complex clotted and radial thread-like textures. The nucleus can be a calcite pebble, skeleton or, frequently, an ooid with complex radial-concentric cortices. The whole framework is also cemented with a mosaic of fibrous-equant calcite crystals (less than 300 ~m in size). Locally, V-shaped cracks, up to 2 cm deep, developed on the top of the amalgamated grainstones, which
58
J.J. J~LVARO & S. CLAUSEN
Fig. 5. Thin-section photomicrographs under plane-polarized light. (a) Mixed ooidal-pyroclastic grainstone; scale = 1 mm. (b) Mixed grainstone with occurrence of larger composite oncoids formed by stromatoid-thromboid intergrowths; scale = 1 cm. (e) Silty sparite with V-shaped cracks passively filled with a laminated (arrowed) calcarenite rich in oncoids (on) and ooids (oo); scale = 5 mm. (d) Hiatal shelled accumulation with light (parautochthonous) and darker (allochthonous) components, such as glauconite peloids (p), phosphatized oncoids (on), phosphatized chancelloriid sclerites (c), trilobites and echinoderm ossicles, and patches of microsparite (m); scale = 2 mm.
were subsequently passively filled with laminated mixed allochems and skeletons (Fig. 5c). The intraclastic lags of the amalgamated strata formed by disruption and erosion of coherent rhythmic packstone-siltstone alternations, which were accumulating cemented detritus derived from lateral peloidal and skeletal substrata. The tabular-lenticular shape of the overlying peloidal- and ooidal-pyroclastic grainstones and the preservation of cross-stratified and low-angle laminae suggest deposition as tabular blankets and low- to medium-relief shoals. The sorting, roundness, grading, alternating laminae and basal erosive surfaces of the peloidal-ooidal grainstones suggest intermittent reworking of micritic-peloidal sediments in a medium- to high-energy subtidal environment. The coarsening- and thickening-upwards trends of the amalgamated strata represents deposition in medium-energy, subtidal conditions, shallowing upwards into higher energy, shallowwater inner-platform (likely foreshore) shoal settings, probably as migrating bars and sheets
that incorporated extraclasts related to active volcanogenic input. Microbial activity during quiescence episodes of bar migrating is recorded as encrusting microbial reefs (described below), and/or microbial coating of clasts (oncoids). The grainstone sets of Lel 1 exhibit distinct trough cross-stratification with centimetre-thick sets bounded by shaly laminae and clasts, in which opposite palaeocurrents, probably tidally induced, have been deduced: the thicker and lower parts are ESE-directed, whereas the uppermost thinner parts are WNW-orientated. By contrast, the grainstone sets of Lel display unidirectional palaeocurrents towards the ESE.
Shelled accumulations A single bed (30-50 cm thick) is recognized in section Le 1 (see top of 11 th parasequence in section A - A ' of Fig. 3). It is lined with a lag concentration, less than 5 cm thick, of epiclastic, pyroclastic and limestone debris. The granule clasts are dominantly angular, whereas the sand-sized
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO clasts are subangular-subrounded in shape. The lag changes upwards into packstone-breccia textures, the latter rich in pyroclastic granules, locally displaying low-angle laminae. They contain sand-sized glauconitic peloids, and disarticulated to broken echinoderm ossicles and trilobite debris, and secondary sponge spicules, chancelloriid sclerites, brachiopods and helcionellids. Both litho- and bioclasts can be subdivided into primary (without epigenetic replacements) and reworked (polyphase) clasts (Fig. 5d). The allochthonous clasts contain cores of firstgeneration granules and skeletons with accretionary sediment. Hematite and phosphate typically occur as thin coatings or impregnations around and within selective debris, and also stain mineral (feldspar) and skeletal heterogeneities. Matrix varies from a mixture of calcite micrite and microsparite, detrital dolomite and terrigenous silt and sand in variable proportions. Micrite occurs in scattered pockets and lenses (up to 1 cm long), sometimes filling intraparticular primary geopetal structures discordant with shelter fillings and stratification. Spar-filled voids have irregular walls and a primary skeletal support. Primary shelter pores commonly formed beneath elongated trilobite sclerites and brachiopod valves. Bioturbation is absent. The polyphase character of some clasts suggests multiple depositional, cementation and erosive events. The shelled accumulation is interpreted as derived from washing and reworking of skeletons and other allochems from the surrounding sea floor generating low-angle shoals under high-energy conditions. The presence of a selective diagenetic precipitation of iron oxides and phosphate allows differentiation in parautochthonous (unaffected) and allochthonous (secondarily reworked after hematite and/or phosphate epigenesis) skeletons, because polyphase litho- and bioclasts are selectively coated or replaced by hematite and phosphate, a cement that generally occluded the primary (intraparticular) porosity. This facies association has been previously described in the overlying Micmacca Breccia, described and interpreted by Alvaro & Clausen (2005, 2006) as composite event-concentration, low-relief shoal complexes composed of parautochthonous and allochthonous skeletal assemblages representing hiatal shelled accumulations (sensu Kidwell & Bosence 1991).
Microbial reefs The term 'reef' is used here because the framebuilding fabrics are dominated by peloidal, thromboid and stromatoid textures, rather than
59
homogeneous mud fabrics. The fabrics correspond to 'Frame Reefs' (sensu Riding 2002), in which in-place skeletons (including calcified microbes) are in contact and microbial clottedpeloidal fabrics are considered as microframe structures.
Stromatoid crusts They show crinkled, bulbous, domal (up to 1 cm in diameter and 4 cm high), and digitate columnar morphologies embedded in a silty microsparite. Stromatoids consist of alternating layers of dense, light-coloured carbonate-rich laminae, extremely rich (up to 60% in volume) in silty, quartz, mica and feldspar crystals and grains, and darker, organic-rich laminae with pyroclastic material less than 10% in volume (Fig. 6a). Interdomal areas are either filled with finegrained bioclastic wackestones (with trilobite sclerites and skeletonized microfossiles), rich in silty quartz and feldspar grains and crystals, or host smaller domal and crinkled stromatolites that are gradually enveloped by the larger accreting domes. These stromatoids represent accretion of microbial mats and biofilms developed under turbid waters, and fit with Riding's (1991) concept of 'agglutinated stromatolites'. Girvanella crusts This calcimicrobe shows a variable preservation, from well-defined filaments to threads of micrite. It occurs as sheets and crusts of intertwined filaments (Fig. 6b), generally horizontal but also inclined at low to high angles with the stratification plane. Dispersed Girvanella bundles form biodictyon structures (sensu Krumbein et al. 2003) embedded in a micritic matrix where distribution of silty quartz and/or feldspar grains/ crystals is highly variable, ranging from 5 to 40% in volume. These microbial sheets and biofilms encrusted and stabilized soft carbonate muddysilty substrates. Pyroclastic influence, when present, did not affect their development. E p i p h y t o n bushes
Epiphyton is the most prolific calcimicrobe and form dense masses of grey clumps (Fig. 4d). Its shrubs commonly branch and bifurcate upwards and sideways, reaching up to 10 cm in thickness. The boundstones are massive, and composed of irregular layers of Epiphyton, although Renalcis clots are locally present, but secondary in volume. This boundstone is embedded in a siltstone that changes laterally into a silty sparry cement, composed of fibrous and bladed calcite
60
J.J. ,/~LVARO & S. CLAUSEN
Fig. 6. Thin-section photomicrographs under plane-polarized light. (a) Agglutinated stromatoid digitate columns rich in silt trapped and bound by the former microbial communities; scale ---5 mm. (b) Girvanellafilaments and threads reflecting episodic encrusting of a variable substrate, ranging from carbonate muddy (m) to silty granular (s); top on the right; scale = 1 mm. (e) Thromboid-stromatoid intergrowth where successive agglutinated stromatoid generations (st) are differentiated by their onlapping geometries, encrusted by several Epiphyton (E) and Renalcis (R) bushes, the whole framework embedded in a silty (s) matrix; scale =4 ram. (d) Replacement of Epiphyton (E) to Renalcis (R) bushes related to a sharp contact from silty (s) to muddy (m) substrate; top on the right; scale = 3 mm.
cements (100-300~tm across). The siltstone component, locally up to 40% in volume, consists of slightly abraded, euhedral feldspar crystals, euhedral and reworked quartz crystals and grains, and minor disarticulated fossil debris (mainly trilobite sclerites). The siltstone not only occurs dispersed in the grainstone embedding the microbial boundstone, but is also covered by boundstone textures forming both silty patches within the microbial boundstone and boundstone patches embedded in a silty matrix. This evidences growth of frame-building textures related to turbid waters.
Thromboid-stromatoid &tergrowths These microbial reefs are a complex intergrowth of thromboids, with Epiphyton as dominant calcimicrobe (up to 40% in volume), and local development of crinkled and domal stromatoids (Fig. 6c). The whole framework forms irregular crusts, changing laterally from 0 to 20 cm in
thickness, which covers the top of some of the aforementioned grainstone shoals, as well as some of their foresets. Archaeocyathan debris is extremely rare or absent. The microbial framework is composed of subunits bounded by erosive discontinuities. Each subunit changes vertically from an intraformational lag, rich in reworked microbial clasts embedded in a peloidal-microbial packstone, to a complex crust, composed of an intergrowth of Epiphyton, Girvanella, Renalcis and rarer domal stromatoids, preserved in growth position (Fig. 6d). Growth cavities between the thromboidstromatoid consortium have calcarenitic infillings and are encrusted by fibrous to bladed, calcite cement (100-300~tm in size). Internal sediment consists of a silty sparitic calcarenite rich in feldspar crystals (reaching locally 50% in volume), quartz and mica grains, minor disarticulated skeletal debris (rich in micritized, trilobite sclerites and echinoderm ossicles), and reworked (intraformational) microbial debris.
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO These fabrics developed as centimetre-scale bushes or clumps above the surrounding sea floor forming a depositional relief with ongoing physical deposition of sediment between the locally digitate boundstones. The cap of the frame-building subunits commonly consists of microbial packstones, dominated by Proaulopora tubes (commonly fragmented), peloids, intraformational lumps and silty feldspar crystals. Proaulopora filaments, 40-130gm in diameter and up to 500 gm in length, are straight to slightly curved tubes, sometimes bifurcated in longitudinal section. The possible binding role of Proaulopora was constrained by high-energy conditions, and the lack of self-supported structures and microbial biofilms able to stabilize the sediment. In summary, these thromboidstromatoid intergrowths episodically recorded the influence of high-energy benthic currents, which reworked part of the microbial framework and formed the described subunits bounded by erosive discontinuities. This microbial consortium commonly grew under turbid waters as evidenced by the wealth of embedded pyroclastic and epiclastic material.
Archaeocyathan-microbial reef complexes Three reef complexes have been identified: Ounein A, B and C (the latter also named 'bioherme de la cascade'; Fig. 3). The systematics of the archaeocyaths was updated in Debrenne & Debrenne (1995), whereas their trilobites and skeletonized microfauna were reported in Geyer & Landing (1995). As core textures are similar in the three reef complexes described below, they are described in this section in order to avoid repetitions. The boundstone consists of a complex intergrowth of Girvanella, Epiphyton and Renalcis, with variable numbers of archaeocyaths, rarely exceeding 20% in volume. Some of the archaeocyaths are attached as outgrowths fixed on other archaeocyathan walls (Fig. 7a), the whole framework encrusted by luxuriant Epiphyton and Renalcis overgrowths. Dweller organisms were trilobites, echinoderms, calcite- and phosphate-shelled brachiopods, hyoliths, chancelloriids and other skeletonized microfossils. Burrowing is nearly absent. Growth cavities between this framework are lined by fibrous-bladed calcite (100-300 gm across), and lack significant percentage of silty component (different from the flanks, described below, where it is abundant; Fig. 7b, c). Internal sediment is generally a microbioclastic wackestone with scattered microbial fragments. Three distinct matrix textures are preserved: peloidal,
61
clotted and homogeneous, which can interchange both vertically and laterally. Peloidal textures consist of irregular silt- to sand-sized peloids of microcrystalline calcite embedded in a groundmass of calcite microspar and pseudospar. Peloidal matrix grades into clotted micrite. Finally, homogeneous micrite occurs as millimetre-thick lenses and discontinuous laminae. Reef soles (underlying the reef cores), up to 10 cm thick, consist of irregular skeletal-peloidal wackestones and packstones, which directly overlie the pyroclastic strata and breccia beds that acted as nucleation surfaces for reef growth. By contrast, reef covers, up to 20 cm thick, commonly contain silty packstones rich in microbial and shelled accumulations composed of Proaulopora filaments, trilobites, hyoliths and other microfossils, similar to the aforementioned cover of the thromboid-stromatoid intergrowths. ' Oune~'n A ' The archaeocyathan-microbial reef complex occurs in a lateral ravine of the Lemdad oued (Fig. 7e). Three geometries are identified vertically. The lower part consists of archaeocyathanmicrobial patches, recognized as isolated pillow-shaped masses, ranging in size from 0.2 to 2 m and from 0.2 to 1 m high, underlain, surrounded and onlapped by the aforementioned pyroclastic shoals and laminated shales. This volcanosedimentary unit, up to 1.6 m thick, is directly covered (in the lateral ravine) by a tabular archaeocyathan-microbial reef, up to 2.6 m thick, dominated by archaeocyathan-rich floatstone textures, although some decimetre-thick boundstone lenses can be recognized towards the top. The biostrome exhibits numerous fractures dipping nearly vertically, and striking E-W. Finally, two lobated megabreccia accumulations occur along the ravine, up to 5.2 m thick, which are in contact with a synsedimentary fault (Fig. 7e). They are characterized by metre-sized olistoliths, and a mixture of angular fragments, granule-boulder in size (up to 2 m in diameter). These have chaotic orientations and boundstone to floatstone textures. The clasts are embedded in a matrix composed of thinly bedded and contorted shales, laminated marlstones, calcareous or silty claystone and deformed claystone clasts, and pyroclastic debris lying the boulders. Both megabreccia bodies represent redeposited fragments of the eastern neighbouring reef biostrome. These accumulations along the margin of a synsedimentary fault are not related to reef collapse by overbuilding owing to the thin relief of the source reef, and the presence of
62
J.J. A L V A R O & S. C L A U S E N
Fig. 7. (a) Archaeocyathan-microbial boundstone from Ounein B with longitudinal sections of encrusting archaeocyathan cups; scale = 2 mm. (b) Archaeocyathan floatstone rich in siltite matrix from Ounein B; scale = 5 mm. (c) Contact between archaeocyathan-microbial boundstone (left) and flank composed of siltite floatstone (right); scale = 5 mm. (d) Reef cover composed of a packstone rich in Proaulopora, and trilobite and hyolithid debris; scale = 1 mm. (e) Patch reef capping the reef complex Ounein B; scale = 10 cm. (f) Field aspect of Ounein A: b, breccia; s, shales with interbedded ashes and pyroclastic lithoarenites; is, in situ reef patches and biostromes; I-II-II, three breccia lobes; arrows, sliding palaeocurrents; 'top' marks the burial of the synsedimentary palaeotopography by pyroclastic strata. (g) Geometry of the bioherms of Ounein B finally capped by patch reefs (arrowed): ps, pyroclastic shoal; p, packstone sole; br, breccia; c, core; f, flank. (h) Breccia underlying Ounein C rich in carbonate and skeletal clasts; scale = 5 cm.
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO pyroclastic tufts and ashes interbedded with the lobes and reworked within them, which reflect perturbations related to volcanic activity. ' Ounei'n B '
This reef complex can be subdivided into four stratigraphic levels (Fig. 7g), in ascending order: (i) a lower bioherm displaying distinct core and flanks; (ii) an irregular breccia; and (iii) a second bioherm where flanks are not well preserved, finally capped by (iv) numerous patch reefs (Fig. 7f), subsequently onlapped by laminated shales. Both bioherms (i and iii), which display a marked tendency for aggregation, are less than 1.5 m thick and can be followed laterally up to 4 m. The flanks of the lower bioherm are generally thinly bedded (0.1-0.3 m), limestones with floatstone textures that show primary depositional slopes dipping up to 40 ~ They thin away from the core, passing laterally, over a distance of 2 m, into debris beds. Debris textures and grain size are highly variable with larger, centimetresized clasts of boundstone textures common in proximal positions. The clasts are chaotically arranged and commonly set in a matrix of bioclastic wackestone or poorly sorted packstone. Some core-facies material is reworked into the proximal flank sediments. Scours within the cores attest to episodic erosion of the surface, responsible for yielding the clasts present in the debris beds. In some cases in situ calcimicrobes grew as small bushes or clumps above the surrounding archaeocyathan-microbial floatstone. The lower bioherm is topped with a distinct erosive channel (Fig. 7g). This is filled with a thin (less than 1.4 m), clast-supported breccia (rich in pyroclasts and boundstone-floatstone clasts up to 2 cm across), with intrabreccia erosive surfaces lined with centimetre-thick siltstone layers. The bottom of the channel cuts down into the bioclastic sole of the lower bioherm exposing its volcanoclastic substrate. The upper bioherm, up to 1.6 m thick and 3 wide, is only recognized by its core. Its margin is not exposed in outcrop, so the nature of flanking strata is unknown. It is directly covered with numerous patch reefs (less than 0.8 m in diameter and height; Fig. 7e). The onlapping geometries of the overlying shales indicate that the patches were indeed topographic irregularities on the sea floor. 'Ounefn C'
The two-dimensional exposure of the reef complex prevents determination of its broad shape.
63
Anyway, as a whole, the reef complex shows three stratigraphic levels, from bottom to top: (i) a lower breccia body (up to 40 cm thick; Fig. 7h) related to a distinct underlying erosive surface; (ii) a stratiform frame-building body, up to 3.4 m thick; and (iii) a series of laterally related patch reefs, with flat bottom and convex tops, up to 0.5 m thick and 1.4 m in diameter. The reef complex in underlain by a sharp erosive surface, planar-deeply scoured, which exhibits sharp changes of its steeped slopes that make angles up to 80~ with the horizontal stratification. The surface is clearly demarcated truncating the underlying tidally influenced epiclastic strata, passing northwards to oblique and paralleling the overlying carbonate strata. The overlying breccia displays a coarse trough cross-bedded stratification with foreset migration towards the NNW. Both the erosive surface and the overlying cross-bedded breccia show N N W orientations, so that, according to the palaeogeographical models of the Souss Basin (Destombes et al. 1985; Geyer & Landing 1995), the brittle fracturation that induced the discontinuity was sloping landwards. There is a variety ofclast sizes and types, some of them polyphase in character, such as angular-subrounded pyroclasts (e.g. mafic and rhyolitic clasts less than 6 cm across, although most clasts do not reach 2 cm in diameter), cemented by a sparry calcite (c. 0.2 mm across), locally dolomitized. Angular feldspars are common, whereas fossils skeletons (such as micritized archaeocyaths, echinoderm ossicles and other microfossils of uncertain affinity) and breccia composed of boundstone and floatstone clasts (up to 4 cm across) are locally abundant (Fig. 7h). Carbonate and skeletal clasts are typically angular-subangular in shape, poorly to very poorly sorted, and were secondarily dolomitized, silicified or simply recrystallized. The middle part of the reef complex, overlying the aforementioned breccia, consists of a distinct glauconitic marker bed, up to 0.5 m thick. The marker bed is texturally a siltstone that contains glauconitic peloids (up to 2 mm in diameter and 70% in volume), and secondary pyroclastic fragments (up to 6 mm across), fossil debris, mostly trilobites, and siliciclastic grains (less than 0.6 mm across) of subangular feldspars, quartz and mica. The whole framework is locally cemented by sparry calcite. Glauconites occur as subrounded and medium-sorted peloids, in some cases somewhat fragmented and reworked. Although the original stratification is not well preserved, the fact that the glauconitic peloids commonly have their long axes at low angles to bedding seems to rule out compaction as the
64
J.J. ALVARO & S. CLAUSEN
main mechanism to explain their grain orientation. This marker bed displays an upwards increase of micrite, associated with a fall in pyroclastic material, leading to the nucleation of an archaeocyathan-microbial biostrome. Archaeocyaths increase in abundance upwards and become the major constituent at the top. Frame-builder skeletons are commonly micritized, preserved upright, although secondarily parallel to bedding, and their primary intraparticular porosity locally occluded by bladed sparite (100-200 gm in size). The upwards decrease in pyro- and epiclasts and glauconitic peloids is replaced by an increase in microbial and archaeocyathan debris, whose erect preservation (boundstone) becomes dominant towards the top. The top of this apparent biostrome is covered with metre-scale patch reefs. The whole reef complex is onlapped by millimetre- to centimetre-thick alternations of siltstones and bioclastic packstones. The latter are rich in trilobites covering sparry shelter structures. The terrigenous fraction comprises angular feldspars, pyroclasts and micas, petrographically assigned to an arkosic and lithoarenitic greywacke. The unoriented, unsorted and poorly rounded to angular clasts, and the lack of both bioturbation and lime-mud in the lowermost breccia body, suggests the possibility of mixing by mass flow along a palaeoslope. Intraclasts were directly derived from erosion of the lateral continuity of either this reef complex or from other ones. The clast-supported breccia is interpreted as debrites deposited by viscous mass flows, often with a lenticular geometry indicating that the breccia unit had an original relief above the sea floor. The breccia deposited likely as a result of reworking by traction processes of local lag deposition from bypassing high-concentration gravity flows. Owing to the lack of grading arrangements, it should be better interpreted as scars or gullies in a slope, rather than channels filled by tectonically induced deposits generated by different types of flow. There are no morphological criteria to interpret it as a submarine fan, but as an episodically reworked palaeoslope that acted as its own source of sediments. Glauconites formed as sea floor peloids. Their deep emerald green colour corresponds to mature, highly evolved, K20-rich glauconite mica (Odin & Fullagar 1988; Amorosi 1995). These authigenic minerals are interpreted as deposited in open-marine environments with very slow deposition rates, once the brecciadeposition events ended. The glauconitic peloids probably formed as a result of dissolutionprecipitation processes (described by Chafetz &
Reid 2000), and were reworked short distances although not derived by erosion of older rocks. The gradual decrease in sand content upwards and the increase in micrite favoured nucleation of an archaeocyathan-microbial biostrome, although its morphology is based on reduced outcrop exposure and its whole dimensions are unknown. The influence of volcanic activity is irregularly recorded across the whole reef complex as silty feldspar, quartz and pyroclasts. This environment ended sharply, and the frame builders developed metre-thick patch reefs. Their flanks recorded high-energy alternations of packstones and siltstones reflecting a progressive deepening-upwards trend accompanied by shale onlapping, which was sharply interrupted by the input of coarse-grained pyroclasts.
Parasequence framework The southern High Atlas platform recorded across Botoman times a mixed (carbonateepiclastic) sedimentation, characterized by stacked, small-scale, shallowing-upwards trends (parasequences s e n s u Van Wagoner e t al. 1988). These were episodically interrupted by tectonic perturbations and the input of large amounts of pyroclastic material (see Fig. 3), which directly affected the patchy occurrence of microbial and shelled carbonate factories. Correlation of smallscale parasequences across the Lemdad syncline is difficult owing to the development both of stratigraphic gaps related to erosion (e.g. the whole Issafen Formation in section Le 11; Fig. 2) and hiatal shelled accumulations, drastic lateral changes in thickness of pyroclastic material according to the distance to magmatic sources, the seawards dissapearance of the tidal influence (dominant in Lel but absent in Lel 1), episodes of local tectonic tilting and subsequent lateral changes in erosive depths, and the discontinuous distribution of carbonate productivity. Synsedimentary structural control during deposition of the Lemdad Formation is evidenced by abrupt, lateral and vertical changes in carbonate facies (reflecting differential subsidence of faultbounded blocks) and thickness, and widespread development of slope-related facies associations rich in conglomerates, breccia lobes, erosive cicatrices, and slumping and sliding structures (see Fig. 3). Tectonically induced subsidence with concomitant base-level change (related to emplacement and reworking of volcanic flows: Smith 1991) probably created accommodation space farther than it could be filled by pyroclastic input inducing syneruptive aggradation of pyroclastic material.
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO Although the input of allochthonous pyroclastic material cannot be related to distinct parasequences, the Botoman mixed (carbonateepiclastic) facies associations of the Lemdad syncline are arranged into two kinds of shallowing-upwards parasequences: mixed and epiclastic shoaling parasequences (Fig. 3). The bases of the parasequences are flooding surfaces, whereas their tops are bounded by erosive discontinuities that have undergone substantial channelling and/or slumping, and are directly overlain by reworked intraclasts. The Botoman platform of the Lemdad syncline recorded an evolution from carbonate- to epiclastic-dominated sediments, punctuated by a major episode of pyroclastic blanketing related to magmatic activity in neighbouring areas. This evolution is more complete in the distal part of the studied outcrops (Le 11) because the proximal ones (Lel) contain an erosive hiatus that erased the sedimentary record of a part of the Lemdad Formation and the whole Issafen Formation. In Le 1, the parasequences of the Lemdad Formation display a distinct tidal influence, which is absent in Lell. The older (stratigraphically lower) part of the distal Lemdad Formation (section A-A' of Lel; Fig. 3) is characterized by the vertical superposition of mixed shoaling parasequences. These, 1.2-10 m thick, are composed, from bottom to top, of: (i) offshore shales and bioclastic wackestones; (ii) millimetre-thick grainstone-siltstone alternations; and (iii) prograding ooidal-peloidal-pyroclastic grainstone shoals that record high-energy water conditions above fair-weather wave base. Some parasequences also pass vertically into (iv) microbial boundstones, and/or (v) oncoidal grainstones, partly dolomitized and covered by ferruginous crusts and reworked intraclasts. The tops of the parasequences are bounded by minor discontinuities, easily recognized by the development of microbial crusts, erosive surfaces, V-shaped cracks, dolomitized beds and reworked intraclasts. As a result, the parasequences are interpreted to have formed as prograding carbonate shoals on an open-marine platform (suggesting deposition by storms and waves from upper shoreface to foreshore) that reworked ooids, peloids, oncoids, aggregates, intraclasts and fossils. The setting of microbialites encrusting the foresets of grainstone shoals indicates episodes in which the shoal bedforms ceased to migrate and were locally stabilized by microbial biofilms and crusts. However, the aforementioned stromatoid crusts, Epiphyton bushes and thromboidstromatoid intergrowths mainly nucleated: (i) on protected (back-shoal) settings; and (ii) upon the topographic highs yielded by the foreset shoals
65
during episodes of quiescence. In both cases, they developed under turbid waters. Proaulopora debris capped some microbial reefs forming microbial bioaccumulations in high-energy environments. Subsequently, a major episode of pyroclastic input dramatically affected this platform (section A'-C of Lel; Fig. 2). The aforementioned mixed shoaling parasequences developed incompletely among the pyroclastic pulses, which were associated with tectonic instability of the platform. These perturbations are linked with nucleation and growth of the metre-sized archaeocyathanmicrobial reef complexes of the area. The latter (i) occur sandwiched between pyroclastic blanketing pulses, (ii) are related to tectonic events and erosive surfaces, and (iii) disappeared by flooding and shale onlapping. The second point is documented by the breccia lobes associated with the tectonically induced footwall highs that laterally bound the partially preserved biostrome of Ounein A, the breccia that underlies the upper bioherm of Ounein B (the lower one grew directly on a pyroclastic substrate), and the breccia that underlies the biostrome of Ounein C. A flooding of the platform is recorded across the Lemdad-Issafen transition, characterized by a fall in pyroclastic blanketing, and the establishment of epiclastic shoaling parasequences in an offshore-dominated platform. The succeeding epiclastic shoaling parasequences consist of coarsening- and thickening-upwards sequences, 1-12 m thick. From bottom to top, almost all the sequences display a systematic succession in which finer-grained storm deposits are covered gradationally or sharply by coarser-grained, cross-stratified sandstones. The top of some parasequences display erosive and truncating surfaces, partly stained by ferruginized hardgrounds, and locally highly bioturbated by Skolithos and Arenicolites. Shoaling (prograding trends related to shallowing-upwards fluctuations of the relative sea level) during times of rapid sediment influx offers a simple and repeatable mechanism for creating these coarseningupwards sequences, topped by hardgrounds widely bioturbated.
Palaeogeographical setting and turbidity Synsedimentary tectonism across the Botoman is manifest in the Lemdad syncline, and deeply influenced regional bathymetric changes and local submarine topography. The Lemdad reefs developed on the southern High Atlas platform did not accumulate on an intact, uniformly subsiding segment of the High Atlas platform, but instead on a mosaic of differentially subsiding,
66
J.J./i~LVARO & S. CLAUSEN
Fig. 8. Main types of microbial reefs in the Lemdad Formation of the Lemdad syncline.
fault-bounded crustal blocks. Synsedimentary deformation at the onset of reef sedimentation suggests that local topography/bathymetry was controlled by displacements on nearby faults. The Botoman microbial reefs of the High Atlas platform grew under turbid waters associated with low-medium pyroclastic input. Reefs and crusts nucleated on protected (back-shoal) substrates and on low-relief bathymetric highs associated with migrating carbonate grainstone shoals and tectonic highs. The pioneer stabilization phase, which represents the sole of numerous reefs, was made by encrusting microbial biofilms and mats (e.g. Girvanella crusts). Tectonic instability and pyroclastic blanketing changed this environmental mosaic, favouring development of archaeocyathan-microbial reef complexes. These appear to have formed episodically on isolated topographic highs protected from pyroclastic input, grew under clear waters, and their demise coincided with flooding and shale onlapping. Bathymetry alone did not control where the reef complexes formed, as most of the potentially contemporaneous platform lithofacies were deposited within the same depth range. Probably more important were local hydrodynamic conditions determined in part by surrounding structurally controlled submarine topography. Formation of such reef complexes may have been favoured when peculiarities of local submarine topography damped or blocked effects of prevailing wind-generated waves and tidal currents. Loose pyroclastic material frequently reworked and redeposited by marine processes. It mixed with shoal (grainstone barrier) and back-shoal (protected) microbial reef cores,
and archaeocyathan-microbial flanks, whereas archeocyathan-microbial cores developed under low pyroclastic input (Fig. 8). Some microbial communities tolerated a continuous or episodic pyroclastic influx, approximately equal to their production rates. Reduced water clarity resulting from turbidity, associated with pyroclastic input, is interpreted as a key control on local nucleation of reefs. Nutrient input, often associated with increased land-derived epiclastic input, was not a primary factor as the terrigenous input in the southern High Atlas was dominantly pyroclastic. The thromboid and stromatoid microbial crusts and reefs developed under turbid waters are not very different from better-studied clear-water counterparts from the Moroccan Anti-Atlas or the Montagne Noire (Monninger 1979; Schmitt 1979; Alvaro et al. 1998, 2006). Patterns that enabled these microbial organisms to inhabit areas affected by pyroclastic input may have included changes into heterotrophic feeding. Morphological changes among photoautotrophic mats and biofilms to maximize the surface area available to incident light (Wilson 2000; Wilson & Lockier 2002) are not recognized in the microbial reefs of the High Atlas. On a larger scale, the progressive demise of carbonate factories across Botoman times (Lemdad and Issafen formations) was sharply affected by the onset of the global Hawke Bay Regression. This is represented in the High Atlas and Anti-Atlas platforms by the prograding Asrir and laterally equivalent formations (/klvaro et al. 2003), which mark the disappearance of both microbial and archaeocyathan-microbial reefs.
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO
Comparison of reef geometries: High Atlas v. Anti-Atlas platforms The outcrops of the Lemdad syncline provide a key opportunity to compare the Botoman reef geometries and components between stable (Anti-Atlas) and unstable (High Atlas) platforms, where the effects of tectonic and volcanic activity directly controlled the development and subsequent demise of carbonate factories. The Botoman interval marks the peak for Early Cambrian archaeocyathan-microbial reefs in West Gondwana (Alvaro et al. 2003). The dimensions achieved by these reefs in the Montagne Noire, southern Francc (Pardailhan and Lastours formations), Ossa-Morena, southern Iberian Peninsula (Sierra Gorda, La Hoya and Pedroche formations), central Iberian Peninsula (Navalucillos Formation), and Sardinia (Matoppa and Punta Manna formations) were largely exceeded by the Great Atlasian Reef Complex (GARC). The GARC was the largest reef complex of archaeocyathan-microbial reefs recorded on the western Gondwana margin. It extended, at least, for over 400 km along the Moroccan margin of West Gondwana, with a thickness peak in the Anti-Atlas platform. The GARC, in some cases more than 100 m thick, did not develop as an Australian-type Barrier Reef because the latter is separated from the shore by a wide lagoon, which is incompletely developed in the Anti-Atlas. Four geometries were distinguished by Alvaro et al. (2006) in the archaeocyathan-microbial reefs of the Botoman Amouslek Formation in the Anti-Atlas: (i) patches ranging in size from 0.2 to 2 m and from 0.2 to 1 m high, surrounded by well-beddedmassive shales; (ii) bioherms of greater dimensions (but less than 3.0 m thick), in which flank beds are well developed, and their margins commonly intercalate with lithologically variable interreef sediments; (iii) biostrome or low-relief structures (up to 1.2 m thick) comprising tabular or sheet-like beds, more than 10 m wide; and (iv) patch-reef complex, kalyptrae or kalyptrate complexes (sensu Rowland & Shapiro 2002), in which patches and bioherms occur stacked together (some of them more than 30 m thick) bounded by clay-marl-silt, millimetre- to centimetre-thick seams and discontinuities. Vertical kalyptrate complexes resulted by vertical accretion of numerous closely spaced bioherm and patch (grading laterally into biostromes) reefs. During the same time span, contemporary rifting perturbations related to faulting and
67
volcanic activity produced a distinct palaeotopography in the neighbouring High Atlas platform (Fig. 9). Faulting in the High Atlas platform caused localized tilting, subsidence perturbations and sedimentation of slope-related facies. The episodic volcanic activity became the source of vast quantities of pyroclastic material, which induced the spread of turbid waters and limited the northwards development of the GARC carbonate factories, dominantly composed of clear-water archaeocyathan-microbial reefs. Despite this, scattered reef complexes nucleated on selective substrates throughout the southern High Atlas platform, but did not reach more than 3.5 m in thickness. The pyroclastic influx close to volcanic centres rapidly inhibited, and even buried, most of the few remaining areas of shallow-water carbonate productivity. Shelled and ooidal carbonate factories contemporaneous with episodic volcanism locally flourished incorporating pyroclastic material. Exclusively microbial carbonate factories relatively supported a moderate pyroclastic influx, as documented by their pyroclastic-rich textures.
Conclusions This paper documents how an interplay of geodynamic factors controlled the carbonate productivity through time and space in the Botoman, carbonate-mixed High Atlas platform. The main factors that influenced the Botoman nucleation of carbonate productivity there were as follows. 9 S y n s e d i m e n t a r y t e c t o n i s m . Normal faulting resulted in the formation of tilting, drowning of downfaulted blocks, and breccia sedimentation on adjacent footwall areas in the High Atlas platform. Subsidence in hanging-wall areas invariably caused irregular topographies and subsequent sharp erosion leading to incision of erosive discontinuities and deposition of lobe breccias in hanging-wall depocentres. Faulting drastically altered the morphology of the High Atlas platform and significantly reduced the available area for shallow-water nucleation of microbial and shelled carbonate factories. 9 Volcanism. The mass influx of pyroclastic debris on the High Atlas platform rapidly smothered carbonate factories in many of the remaining habitable shallow-water areas of the High Atlas platform. Only in more distal areas of the basin (Anti-Atlas platform) and during episodes of low volcanic activity (High Atlas) was carbonate productivity renewed. When the rate of influx of pyroclastic debris
68
J.J. ALVARO & S. CLAUSEN
Fig. 9. Comparative stratigraphic framework of Lower Cambrian rocks from the southern High Atlas and Anti-Atlas; based on Leblanc & Lancelot (1980), Clauer et al. (1982), Latham & Riding (1990), Comptson et al. (1992), Debrenne & Debrenne (1995), Geyer & Landing (1995), Piqu6 et al. (1995), Landing et al. (1998), ,~lvaro et al. (2000, 2003, 2006), Gasquet et al. (2005), Alvaro & Clausen (2006) and this work. was less than the accommodation rate of shallow-water carbonates, carbonate factories occurred coevally with volcanism. Then reefs nucleated on faulted highs, back-barrier (shoal) protected settings or directly over skeletal accumulations and pyroclastic substrates. 9 Relative sea-level fluctuations. Numerous reported parasequences contain at their upper (shallower) part microbial reef patches and complexes. The end of their shallowingupwards trends ended with either erosive contacts (filling by breccia and/or pyroclasts), probably related to syneruptive events, or sharp flooding and shale onlapping geometries, in both cases interrupting reef growth. The scattered archaeocyathan-microbial reefs of the Lemdad Formation exhibit a wide range of external morphologies in the High Atlas platform, including tabular (biostromes) and domal (bioherms and patches) strata that vary considerably in size. Reef nucleation and growth
were probably primarily controlled by local tectonism, as initial nucleation and accretion would have been facilitated by peculiar, hydrodynamic conditions imposed by irregular submarine topography perpetuated by faulting. Archaeocyathan-microbial reefs grew exclusively under clear waters, whereas microbial stromatoids, Girvanella crusts, E p i p h y t o n bushes and thromboid-stromatoid intergrowths probably developed heterotrophic strategies when submitted to a moderate terrigenous input. Turbidity was a major ecological factor that constrained development of filter/suspensionfeeder and phototrophic organisms, but not necessarily of benthic non-phototrophic microbial communities. The authors thank F. Debrenne and O. Weidlich for their constructive reviews. This research was predominantly undertaken at the University of Lille I as part of an 'Innovative Thematic Action' project entitled 'The
CAMBRIAN TURBID-WATER REEFS FROM MOROCCO First Metazoan Reefal Communities Across the Neoproterozoic-Cambrian Transition in Morocco and South-western Europe'. This project also tried to improve the stratigraphic control on condensation and hiatus levels of chronostratigraphic stratotypes potentially candidates for future GSSP of the International Subcommission on Cambrian Stratigraphy and is a contribution to project CGL2006-13533.
References ALVARO, J. J. & CLAUSEN, S. 2005. Major geodynamic and sedimentary constraints on the chronostratigraphic correlation of the Lower-Middle Cambrian transition in the western Mediterranean region. Geosciences Journal, 9, 145-160. JkLVARO, J. J. & CLAUSEN, S. 2006. Microbial crusts as indicators of stratigraphic diastems in the Cambrian Micmacca Breccia, Moroccan Atlas. Sedimentary Geology, 185, 255-265. /~LVARO, J. J., VENNIN, E. & VIZCA]NO, D. 1998. Depositional controls on Early Cambrian microbial carbonates from the Montagne Noire, southern France. Transactions of the Royal Society of Edinburgh." Earth Sciences, 89, 135-143. /~LVARO, J. J., ROUCHY, J. M. ET AL. 2000. Evaporitic constraints on the southward drifting of the western Gondwana margin during Early Cambrian times. Palaeogeography, Palaeoclimatology, Palaeoecology, 160, 105-122. ALVARO, J. J., ELICKI, O., GEYER, G., RUSHTON, A. W. A. & SHERGOLD, J. n . 2003. Palaeogeographical controls on the Cambrian trilobite immigration and evolutionary patterns reported in the western Gondwana margin. Palaeogeography, Palaeoclimatology, Palaeoecology, 195, 5-35. /~kLVARO,J. J., CLAUSEN,S., EL ALBANI,A. & CHELLAI, E. H. 2006. Facies distribution of the LowerCambrian cryptic microbial and epibenthic archaeocyathan-microbial communities in the western Anti-Atlas, Morocco. Sedimentology, 53, 35-53. AMOROSI, A. 1995. Glaucony and sequence stratigraphy: a conceptual framework of distribution in siliciclastic sequences. Journal of Sedimentary Research, B65, 419425. BADRA, L., POUCLET,A., PROST, A. E. & TOURAY, J. C. 1992. Mise en 6vidence d'une extension intra-plaque tardi-panafricaine d'int6r& m&allog6nique dans le Haut-Atlas occidental (Maroc). Comptes Rendus de l'Acad6mie des Sciences, Paris, 314, 703-709. BOUDDA, A., CHOUBERT, G. & FAURE-MURET, A. 1975. Coupe gdologique de l'Ounein. Description lithologique. R6union sur la limite Pr6cambrienCambrien, Agadir-Rabat, Documents, A+B. BOUDDA, A., CHOUBERT, G. 8z FAURE-MURET, A. 1979. Essai de stratigraphie de la couverture s6dimentaire de l'Anti-Atlas: Adoudounien, Cambrien inf6rieur. Notes et M6moires du Service G6ologique du Maroc, 271, 1-96. CHAFETZ, H. S. & REID, A. 2000. Syndepositional shallow-water precipitation of glauconitic minerals. Sedimentary Geology, 136, 2942.
69
CHOUBERT, G. 1958. L'Adoudounien et le Pr6cambrien III dans l'Anti-Atlas. Colloque International du CNRS, Paris (1957), 76, 143-157. CLAUER, N., CABY,R., JEANETTE,D. & TROMPETTE,R. 1982. Geochronology of sedimentary and metasedimentary rocks of the West African Craton. Precambrian Research, 18, 53-71. COMPSTON, W., WILLIAMS, J. L., KIRSCHVINK,J. L., ZHANG, Z. • MA, G. 1992. Zircon U-Pb ages for the Early Cambrian time scale. Journal of the Geological Society, London, 127, 319-332. DEBRENNE, F. & DEBRENNE, M. 1995. Archaeocyaths of the Lower Cambrian of Morocco. Beringeria, 2, Special Issue, 121-145. DESTOMBES, J., HOLLARD, H. & WILLEFERT, S. 1985. Lower Palaeozoic rocks of Morocco. In: HOLLARD, C. H. (ed.) Lower Palaeozoic Rocks of the World, Volume 4. Lower Palaeozoic of North-western and West Central Africa. Wiley, Chichester, 157-184. EZZOUHAIRI, H. 2001. Le magmatismepost-collisionnel panafricain (tardi ~ post-orog6nique) des rkgions d'Aghbalou, Sidi Flah Bouskour et Oued Imini (Ouarzazate, Anti Atlas central, Maroc). Lithostratigraphie, g6ochimie, p6trogenkse et contexte gkodynamique. PhD thesis, University of Choua'l~O Doukkali. FISHER, R. V. & SMITH, G. A. 1991. Volcanism, tectonics and sedimentation. In: FISHER, R. V. & SMITH, G. A. (eds) Sedimentation in Volcanic Settings. Society of Economic Paleontologists and Mineralogists, Special Publications, 45, 1-5. GASQUET, D., LEVRESSE, G., CHEILLETZ, A., AzIzISAMIR, H. R. & MOUTTAQUI,A. 2005. Contribution to a geodynamic reconstruction of the Anti-Atlas (Morocco) during Pan-African times with the emphasis on inversion tectonics and metallogenic activity at the Precambrian/Cambrian boundary. Precambrian Research, 140, 157-182. GEYER, G. 1990a. Proposal of formal lithostratigraphical units for the terminal Proterozoic to early Middle Cambrian of southern Morocco. Newsletters on Stratigraphy, 22, 87-109. GEYER, G. 1990b. Revised Lower to lower Middle Cambrian biostratigraphy of Morocco. Newsletters on Stratigraphy, 22, 53-70. GEYER, G. & LANDING, E. (eds). 1995. Morocco '95. The Lower-Middle Cambrian standard of Gondwana. Beringeria, 2, Special Issue, 1-171. HELDMAIER, W. 1997. The Lower-Middle Cambrian transition along the northwestern margin of Gondwana (Anti-Atlas and High Atlas, Morocco). PhD thesis, Wfirzburg University. Hupt~, P. 1955. Iindices d'une phase tectonique sala'irienne dans l'Anti-Atlas marocain. Comptes Rendus de l'Academie des Sciences de Paris, 241, 971-973. JOUHARI, A., EL ARCHI, A., AARAB, M., EL ATTARI, A., EMNIH, N. & LADURON, D. 2001. G6ochimie et cadre g6odynamique du volcanisme n6oprot6rozoique terminal (Vendien) du Haut Atlas occidental, Maroc. Journal of African Earth Sciences, 32, 695-705.
70
J.J. ,~LVARO & S. CLAUSEN
KIDWELL, S. M. & BOSENCE, D. W. J. 1991. Taphonomy and time-averaging of marine shelly faunas. In: ALLISON,P. A. & BRIGGS, D. E. G. (eds) Taphonomy: Releasing the Data Locked in the Fossil Record. Topics in Geobiology. Plenum Press, New York, 115-209. KRUMBEIN, W. E., BREHM, U., GERDES, G., GORBUSHINA, A. A., LEV1T, G. & PAL1NSKA,K. A. 2003. Biofilm, biodictyon, biomat microbialites, oolites, stromatolites geophysiology, global mechanism, parahistology. In: KRUMBEIN, W. E., PATERSON,D. M. & ZAVARIN,G. A. (eds) Fossil and Recent Biofilms. A Natural History of Life on Earth. Kluwer, Dordrecht, 1-28. LANDING, E., BOWRING, S. A., DAVIDEK, K. L., WESTROP, S. R., GEYER, G. & HELDMAIER, W. 1998. Duration of the Early Cambrian: U-Pb ages of volcanic ashes from Avalon and Gondwana. Canadian Journal of Earth Sciences, 35, 329-338. LARCOMBE, P., COSTEN, A. & WOOLFE, K. J. 2001. The hydrodynamic and sedimentary setting of nearshore coral reefs, central Great Barrier Reef shelf, Australia: Paluma Shoals, a case study. Sedimentology, 48, 811-835. LATHAM, A. & RIDING, R. 1990. Fossil evidence of the location of the Precambrian/Cambrian boundary in Morocco. Nature, 344, 752-754. LEBLANC, M. & LANCELOT, J.R. 1980. Interpr6tation g6odynamique du domain panafricain (Pr6cambrien terminal) de l'Anti-Atlas (Maroc) ~ partir de donn6es g6ologiques et g6ochronologiques. Canadian Journal of Earth Sciences, 17, 142-155. MEYER, R., KRAUSE, F. & BRAMAN, D. 1998. Unconformities within a progradational estuarine system: the upper Santonian Virgelle Member, Milk River Formation, Writing-on-stone Provincial Park, Alberta, Canada. In: ALEXANDER, C. R., DAVIS, R. A. & HENRY, V. J. (eds) Tidalites: Processes and Products. Society of Economic Paleontologists and Mineralogists, Special Publications, 61,129-142. MYROW, P. 1998. Transgressive stratigraphy and depositional framework of Cambrian tidal dune deposits, Peerless Formation, central Colorado, USA. In: ALEXANDER, C. R., DAVIS, R. A. & HENRY, V. J. (eds) Tidalites: Processes and Products. Society of Economic Paleontologists and Mineralogists, Special Publications, 61, 143-154. MONNINGER, W. 1979. The section of Tiout (Precambrian/Cambrian boundary beds, Anti-Atlas, Morocco): an environmental model. Arbeiten aus dem Paliiontologie der Universitiit Wiirzburg, 1, 1-289. ODIN, G. S. & FULLAGAR, P. D. 1988. Geological significance of the glaucony facies. In: ODIN, G. S. (ed.) Green Marine Clay. Elsevier, Amsterdam, 295-332. OUAZZANI, H. 2001. Le volcanisme d'arc massif ancien du Haut Atlas occidental (Maroc), un t6moin de la convergence de la branche occidentale de l'oc6an panafricain. Bulletin de la SociOtd gkologique de France, 172, 587-829.
PIQUI~,A., BOUABDELLI,M. & DARBOUX,J. R. 1995. Le rift cambrien du Maroc occidental. Comptes Rendus de l'AcadOmie des Sciences, Paris, 320, 1017-1024. RIDING, R. 1991. Classification of microbial carbonates. In: RIDING, R. (ed.) Calcareous Algae and Stromatolites. Springer, Berlin, 21-51. RIDING, R. 2002. Structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth-Sciences Review, 58, 163-231. RIGGS, N. R., HURLBERT, J. C., SCHROEDER, T. J. & WARD, S. A. 1997. The interaction of volcanism and sedimentation in the proximal areas of a MidTertiary volcanic dome field, central Arizona, USA. Journal of Sedimentary Research, 67, 142-153. ROWLAND, S. M. & SHAPIRO,R. S. 2002. Reef patterns and environmental influences in the Cambrian and earliest Ordovician. In: KIESSLING,W. & FL~GEL, E. (eds) Phanerozoic Reef Trends. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 95-128. SCHMITT, M. 1979. The section of Tiout (Precambrian/ Cambrian boundary beds, Anti-Atlas, Morocco): stromatolites and their biostratigraphy. Arbeiten aus dem Paliiontologie der Universitiit Wiirzburg, 2, 1-188. SIEGERT, R. 1986. Die klastischen Serien an der Wende yon Unterzum Mittelkambrium im Anti-Atlas (Marokko). PhD thesis, Erlangen University. SMITH, G. A. 1991. Facies sequences and geometries in continental volcanoclastic sequences. In: FISHER, R. V. & SMITH,G. A. (eds) Sedimentation in Volcanic Settings. Society of Economic Paleontologists and Mineralogists, Special Publications, 45, 109-122. SPIZHARSKI,T. N., ZHURAVLEVA,I. Y., REPINA, L. N., ROZANOV, A. YU., TCHERNYSHEVA, N. YE. & ERGALIEV, G. H. 1986. The stage scale of the Cambrian System. Geological Magazine, 123, 387-392. VAN WAGONER,J. C., POSAMENTIER,H. W., MITCHUM, R. M., VAIL, P. R., SARG, J. F., LOUTIT, Y. S. & HARDENBOL,J. 1988. An overview of the fundamentals of sequence stratigraphy and key definitions. In: WILGUS, C., HASTINGS,B., ROSS, C., POSAMENTIER, H., VAN WAGONER, J. & KENDALL, L. G. St. C. (eds) Sea-level Changes - An Integrated Approach. Society of Economic Paleontologists and Mineralogists, Special Publications, 42, 39-45. WILSON, M. E. J. 2000. Tectonic and volcanic influences on the development and diachronous termination of a Tertiary tropical carbonate platform. Journal of Sedimentary Research, 70, 310-324. WILSON, M. E. J. 2005. Development of equatorial delta-front patch reefs during the Neogene, Borneo. Journal of Sedimentary Research, 75, 114-133. WILSON, M. E. J. & LOCKIER, S. W. 2002. Siliciclastic and volcaniclastic influences on equatorial carbonates: insights from the Neogene of Indonesia. Sedimentology, 49, 583-601. WOOLFE, K. J. & LARCOMBE, P. 1999. Terrigenous sedimentation and coral reef growth: a conceptual framework. Marine Geology, 155, 331-345.
A mixed ophiuroid-stylophoran assemblage (Echinodermata) from the Middle Ordovician (Llandeilian) of western Brittany, France A A R O N W. H U N T E R 1'2, B E R T R A N D
L E F E B V R E 1, S E R G E R l ~ G N A U L T 3,
PHILIPPE ROUSSEL 4& ROLAND
CLAVERIE 5
~Centre National de la Recherche Scientifique, U M R 5561, Biogkosciences, Universitk de Bourgogne, 6 boulevard Gabriel, 21000 Dijon, France (e-maiL"
[email protected]) 2Research School o f Earth Sciences, Birkbeck & University College London, Gower Street, Bloomsbury, London WC1E 6BT, UK 3Musdum d'Histoire naturelle de Nantes, 12 rue Voltaire, 44000 Nantes, France 44 rue Arthur Lemoine de la Borderie, 56000 Lorient, France 5Rksidence le clos des ch6nes, 145 route de Grasse, 06600 Antibes, France Abstract: In the abandoned slate quarry ofGuernanic, Gourin (Morbihan, France), a single horizon (Upper Member of the Schistes de Postolonnec Formation) has yielded an exquisitely preserved Llandeilian (Middle Ordovician) echinoderm assemblage composed of the ophiuroid Taeniaster armoricanus sp. nov. and the mitrate Mitrocystella incipiens. These two groups of echinoderms represent the first fossils formally described from the Middle Ordovician of the Gourin area. The brittlestar T. armoricanus sp. nov. is the third and oldest ophiuroid reported so far in the Palaeozoic of the Armorican Massif. The mitrate Mitrocystella is also described for the first time from western Brittany. Taphonomic features of this ophiuroid-stylophoran aggregation suggest that it probably corresponds to the rapid burial of a life assemblage in an otherwise quiet and moderately deep setting (shelf) below, but close to, storm wave base. This echinoderm association represents the oldest evidence for a gregarious mode of life for ophiuroids, as well as the oldest indisputable example of a mixed ophiuroid-stylophoranmeadow.
Large concentrations of non-colonial organisms such as, for example, ophiuroids and crinoids are good sources of taphonomic, palaeoecological and taxonomic information (see Donovan 1991). Observations of modern communities from submersibles demonstrate that such accumulations could reach a great extent, not only in tropical waters but also on shallow temperate shelves, abyssal plains and even around the fringes of Antarctica (Aronson & Blake 1997; Juterzenka & Soltwedel 2004). Understanding such assemblages allows us to reconstruct ancient palaeocommunities and isolate the factors that control biofacies and taphofacies (Hunter & Donovan 2005). Recent work on modern environments has allowed us to understand mass accumulations and their relevance to fossil bioaccumulations as far back as the Palaeozoic (e.g. Brett et al. 1997). In modern seas, ophiuroid meadows probably represent one of the most significant and widely distributed examples of such accumulations of echinoderms (Aronson & Blake 1997). They correspond to extensive, dense populations (several tens to several hundreds of individuals per m 2)
of infaunal or epibenthic ophiuroids forming an actual living 'carpet' in or on the sea floor (Fujita & Ohta 1989, 1990; Aronson & Blake 1997). For example, the sea floor off the northern coast of Japan is nearly totally blanketed by dense beds of the modern ophiuroid Ophiura sarsii, between depths of 200 and 600 m (Fujita & Ohta 1990). Such assemblages are usually monospecific or largely dominated by a single species (Brun 1969; Fujita & Ohta 1990; Hily 1991). They are also typically characterized by a rare and poorly diversified associated fauna (Fujita & Ohta 1989, 1990; Ball et al. 1995). Several examples of comparable dense beds of ophiuroids have been also documented in the fossil record (Fujita 1992; Aronson & Blake 1997). The oldest indisputable occurrences of such ophiuroid meadows were reported from the Zaho~any and Bohdalec formations (Caradoc, Upper Ordovician) of Bohemia (Petr 1989; Mikulfis et al. 1995). In the Middle Ordovician of Spain (Montes de Toledo), H a m m a n n & Schmincke (1986) also reported the occurrence of several specimens of the ophiuroid
From: ~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECK~,A.,
VACHARD,D. & VENNIN,E. (eds) 2007.
Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 71-86. 0305-8719107l$15.009 The Geological Society of London.
72
A.W. HUNTER ET AL.
Taeniaster ibericus in a single horizon. However, this aggregation more probably results from the fortuitous local accumulation of several individuals transported by currents over a short distance, rather than an actual dense assemblage (Hammann & Schmincke 1986). In this paper, we report on the discovery of an autochthonous assemblage consisting of a dozen, well-preserved, articulated individuals of the ophiuroid Taeniaster in the Upper Middle Ordovician (Llandeilian) of Gourin, Brittany (western France). This assemblage thus represents the oldest indisputable evidence of a gregarious mode of life for ophiuroids. Interestingly, in the Middle Ordovician of Gourin, abundant remains of the mitrate stylophoran Mitrocystella were also sampled in the same horizon as well as specimens of ophiuroids. Stylophorans are an extinct class of free-living echinoderms (Middle Cambrian-Upper Carboniferous), characterized by a flattened theca and the possession of a single feeding arm (aulacophore: Ubaghs 1968; Chauvel 1981; Parsley 1988; Lefebvre 2003a). Their mode of life is supposed to be comparable to that of ophiuroids (Lefebvre 2003a, 2006). The occurrence of a dozen, fully articulated, autochthonous individuals of the mitrate Mitrocystella in a single horizon is indicative of a gregarious mode of life for these echinoderms. This interpretation is in good accordance with several previous records suggesting a gregarious mode of life for stylophorans: such stylophoran-dominated meadows are documented from the Middle Cambrian (Ubaghs & Robison 1988) to at least the Devonian (Ruta & Bartels 1998). However, the co-occurrence of abundant ophiuroids and mitrates in the same horizon in the Middle Ordovician of Gourin is an extremely significant observation from a palaeoecological point of view. Several examples of such mixed ophiuroid-stylophoran meadows have been reported in younger deposits of the Upper Ordovician (e.g. Lefebvre 2006) to Lower Devonian (e.g. Caster 1954). Consequently, this paper reports on the oldest definitive example of a mixed ophiuroid-stylophoran assemblage documented so far, and it demonstrates that the co-occurrence of these two echinoderm groups is a not uncommon occurrence in the fossil record. This study incorporates data from recent research on brittlestar preservation, palaeoecology and observations of modern communities from submersibles, along with numerous recent works on the probable palaeoecology of stylophorans (see Lefebvre 2003a) to track the early evolution of such bioaccumulations.
Finally, this paper describes the oldest remains of ophiuroids reported so far in the Armorican Massif and the third species of this Class ever described in the Palaeozoic of this region. The two previously described brittlestars are both early Devonian in age, and have been reported from Daoulas, Finist+re (Protaster daoulasensis: Davy 1886), and from Saulges, Mayenne (Ophiurina armoricana: Morzadec & Ubaghs 1969). The comparable Middle Ordovician successions and faunas observed in Spain, Portugal and the Armorican Massif are considered by most authors as evidence for a direct geographic continuity between these three regions (e.g. Paris & Robardet 1977; Romano & Henry 1982; Young 1990; Guti6rrez-Marco et al. 1999). Consequently, the rarity of ophiuroids in the Middle Ordovician of Brittany is intriguing, as at least three different forms have been reported in coeval levels from elsewhere (Llandeilian) in Spain (Chauvel & Mel6ndez 1978; Guti6rrezMarco et al. 1984; Hammann & Schmincke 1986). This paper also presents the first report of the mitrate Mitrocystella in the Chateaulin Basin (western Brittany). In the Armorican Massif, this stylophoran was previously only documented from several localities in eastern Brittany (e.g. Guichen), where most specimens were collected in siliceous concretions of the Traveusot Formation (Llandeilian: Chauvel 1941, 1981; Lefebvre 2000a). In eastern Brittany, this type of fossil occurrence severely limits palaeoecological interpretations (especially as far as a putative gregarious mode of life is concerned). Apart from the Armorican Massif, the mitrate Mitrocystella was also reported from the Middle Ordovician (Llandeilian) of Bohemia (Barrande 1887; Jefferies 1967, 1984; Parsley 1994), Spain and Portugal (Guti6rrez-Marco & Mel~ndez 1987; Couto & Guti6rrez-Marco 2000).
Repositories The material referred to in this paper is deposited in the following public institutions: M H N N (Mus6um d'Histoire naturelle de Nantes, France) and NMP (Narodni Muzeum Prague, Czech Republic).
Geological setting and associated fauna Both the ophiuroids (consisting entirely of the genus Taeniaster) and the mitrate stylophoran (Mitrocystella) were sampled from a single unit within the Upper Member of the Schistes de Postolonnec Formation (Llandeilian) in the abandoned slate quarry of Guernanic, near Gourin, Morbihan (grid ref. (GR) 157210688),
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS
73
Fig 1. (a) Location of the Armorican Massif. (b) Geological map of the Armorican Massif focusing on the areas of Ordovician outcrop. (c) Synthetic stratigraphic column of Middle Ordovician sections in Gourin area, showing identified fossiliferous horizons (star signifies the echinoderm bed).
in the central 'synclinorium' system of Brittany (western France). The occurrence of Middle Ordovician fossils in the Gourin area was first documented by Tromelin & Lebesconte (1876), who mentioned the presence of bivalves (Redonia deshayesiana, R. duvaliana) and trilobites ( Caly-
mene aragoi, C. salteri, C. tristani, Dalmanites macrophtalma, Illaenus beaumonti, I. giganteus) in the 'schiste ardoisier'. Later on, Barrois (1884, 1885) reported a comparable, although more diverse, fossil assemblage in the 'schistes ardoisiers d'Angers' (Middle Ordovician) of RochArvran (north of Gourin), including bivalves (Redonia deshayesiana, R. duvaliana), brachiopods (Orthis berthoisi), graptolites and trilobites
( Calymene aragoi, C. tristani, Dalmanites macrophtalma, Illaenus giganteus). The presence of Middle Ordovician echinoderm remains in the area of Gourin was quoted by Chauvel (1941, p. 65), who mentioned the probable occurrence of diploporite cystoids (Aristocystites cf. bohemicus). However, contrary to the situation in other regions of the Armorican Massif (e.g. Crozon peninsula, synclinals south of Rennes), where Middle Ordovician successions and faunas have been intensively studied (e.g. Kerforne 1901; Chauvel 1941, 1980, 1981; Babin 1966; Henry 1980, 1989; Henry et al. 1993), those
occurring in the Gourin area remained very poorly investigated (Pruvost & Le Maitre 1949; Le Gall et al. 1992). Consequently, this study represents the first systematic description of Middle Ordovician fossils collected in the eastern part of the Chateaulin Basin. In the Gourin area, the Schistes de Postolonnec Formation (AbereiddianLlandeilian) corresponds to a thick succession (at least 450 m thick) of shales and slates lying conformably above the sandstones of the Gr6s Armoricain Formation (Arenig). The uncertain presence of the Middle Member of the Schistes de Postolonnec Formation (Gr~s de Kerarvail) has not yet been definitively established in the Gourin area (Le Gall et al. 1992). In more western outcrops (Crozon peninsula), the Kerarvail Member corresponds to a relatively thick sandstone intercalation (up to 30 m thick) between the shales of the Lower Member (Abereiddian) and those of the Upper Member (Llandeilian) of the Schistes de Postolonnec Formation (see Barrois 1889; Kerforne 1901; Babin et al. 1976; Henry 1989). In spite of a difficult geological context (numerous faults, absence of continuous sections), intensive field studies lead since the mid 1990s by two of the authors (P. Roussel and R. Claverie) has made it possible to recognize at
74
A.W. HUNTER ET AL.
least four distinct fossiliferous intervals (typically 10-50 cm thick) within the Schistes de Postolonnec Formation (Fig. 1). These four intervals are characterized by their different faunal content and can be correlated between outcrops. They are designated here, from bottom to top, as the 'Pont Saint-Herv6', 'Monaco', 'Guernanic' and 'Pontvern Nord' units. Fossils collected in these four intervals of the Schistes de Postolonnec Formation are flattened and some of them are preserved in pyrite. The Pont Saint-Herv6 unit occurs about 40 m above the contact between the Gr6s Armoricain and Schistes de Postolonnec formations. It has yielded a rich Abereiddian assemblage, with abundant remains of cystoid echinoderms (Calix), graptolites and trilobites (e.g. Neseuretus, Plaesiacomia). The second fossiliferous interval (Monaco) is located about 350 m above the pont Saint-Herv6 unit, and it corresponds to a Llandeilian assemblage, dominated by trilobites (e.g. Placoparia, Prionocheilus). The third fossiliferous unit (Guernanic) occurs about 4 0 m above the Monaco unit. It is characterized by the presence of numerous micas (absent in the underlying Monaco unit). The Guernanic unit has yielded a rich and diverse fossil assemblage, dominated by articulated trilobites (e.g. Salterocoryphe, illaenid indet., Zeliskella), associated with echinoderms (e.g. crinoids, diploporite cystoids, mitrate stylophorans), brachiopods, bivalve and cephalopod molluscs. The presence of tarphicerid cephalopods is indicative of a possible late Llandeilian age. In one locality (abandoned slate quarry of Guernanic), a well-defined echinoderm-rich horizon occurring within this fossiliferous interval has yielded a dozen specimens of ophiuroids (Taeniaster) associated with a comparable proportion of mitrate stylophorans (Mitrocystella). It is this assemblage that will be described herein. It is apparent that this echinoderm horizon contains very little associated fauna. Finally, the fourth fossiliferous unit (Pontvern Nord) is located about 10 m above the Guernanic unit. It consists in a well-defined, mica-rich interval. The Pontvern Nord unit has yielded a late Llandeilian assemblage dominated by bivalves and trilobites (e.g. Phacopidina micheli). The composition of the four fossiliferous assemblages recognized so far in the Middle Ordovician of Gourin shows clear affinities with coeval faunas reported from elsewhere in Brittany (e.g. western part of the Chateaulin Basin, synclinals south of Rennes: Chauvel 1941, 1980, 1981; Henry 1980, 1989) but also in Portugal and Spain (see Chauvel & Mel6ndez 1978; Henry
1980; Guti6rrez-Marco et al. 1984; Guti6rrezMarco & Mel6ndez 1987). However, in the absence of a complete survey of the Gourin fauna (and especially of the trilobites), it is beyond the scope of this paper to elaborate further on its biostratigraphic and/or palaeobiogeographic significance.
Taphonomy Both ophiuroids and stylophorans are regarded universally as having echinoderm skeletons (e.g. Jefferies 1967; Ubaghs 1968; Chauvel 1981). In the taphonomic classification proposed by Brett et al. (1997), these two groups can be classified as 'type 1 echinoderms' in that they possess a skeleton composed of loosely articulated plates, and require exceptional preservation to prevent fragmentation. In spite of comparable taphonomic attributes, brittlestars are much better understood owing to being present in modern environments and will thus be discussed first. However, the preservation and palaeoecology of stylophorans have been compared with those of ophiuroids (Lefebvre 2006). The fossil record of ophiuroids is considered so patchy that their former diversity is often misunderstood. This is primarily owing to their calcareous endoskeleton being composed of ossicles, which completely disarticulate and disassociate in a very short period after death. Such ossicles have long been thought to be unclassifiable (see Mortensen 1938). However, subsequent studies using fragmentary remains have tried to readdress this balance in the Mesozoic (see Jagt 1999, 2000; Hunter 2004). As yet only Hotchkiss et al. (1999) have applied this treatment to Palaeozoic echinoderms. This study looks at the exceptional instances when ophiuroids are preserved in obrution deposits. Owing to the apparent similarities between extant and fossil ophiuroids it would be expected that the history of preservation would be well understood, especially in such obrutions. However, this is not the case, although Sch~ifer (1972) noted that the arms of extant ophiuroids begin to disintegrate within 15 h of death and noted that catastrophic burial would be reasonable for articulated brittlestar assemblages, such as those that are observed in the Lower Jurassic in Dorset, UK. More rigorous studies were not conducted until recently by Kerr & Twitchett (2004). It is this study that is critical to understanding the Middle Ordovician assemblage from Gourin. Using experiments analogous to those carried out by Briggs & Kear (1994), specimens of the extant ophiuroid Ophiura texturata were decayed for 14 days at temperatures ranging from 15 to 25 ~ with
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS frequent tumbling of specimens to represent the turbidity of the environment. These experiments produced several stages of decay that could be quantified ranging from fully articulated carcases to complete disarticulation. The ophiuroids in the present study could be classified in two of the decay stages identified in Kerr & Twitchett's (2004) study. Some specimens (including the paratype) could be classified as stage 1 (i.e. loss of ventral plates, oral shields, tooth papillae and jaws, and full articulation of the arms) (Figs 2a, d & 3a, MHNN.P.025955 and MHNN.P.025956), while most of them (including the holotype) would belong to stage 2 (i.e. arms begin to break off at/near the disk, and at least one arm remains attached) (see Figs 2b, c & 3c, MHNN.P.025954b and MHNN.P.025954a). However, interpretation of stage 2 preservation is difficult, as ophiuroids can sometimes autotomize their arms under environmental stress, so that forms lacking arms could have been buried before decay or shed arms during the catastrophic burial event itself. Kerr & Twitchett (2004) observed this type of preservation in brittlestars of the British Lower Jurassic and considered that burial occurred around the time of death, which implies the existence of a single community. However, such an observation could be a result of collector bias (see Hunter & Donovan 2005). In this study we have to assume an error caused by dealing with more distant Palaeozoic forms where the Recent brittlestar morphology is not strictly analogous with those in the Lower Palaeozoic. The most fossiliferous starfish bed known in the Lower Palaeozoic is the Lady Burn starfish beds of Girvan, Scotland, UK. This assemblage is so diverse that in the most recent survey of Palaeozoic asterozoans (Shackleton 2003) approximately 60% of the taxonomic data came from this locality. The Lady Burn starfish beds have lenticular thinning from 15 to 1 cm within 1.5m. This occurrence was interpreted by Goldring & Stephenson (1972) as resulting from the rapid entombment by storm deposits of a living, shallow-water assemblage, with limited transport and redeposition in a region where the fossilization potential is much higher than the environment of origin. However, more recent analyses based on the regional geological context, the composition of the faunal associations, as well as available sedimentological and taphonomic features of the Lady Burn starfish beds, all suggest that these assemblages more probably correspond to the rapid downslope transportation (by turbiditic currents), accumulation and burial of shallower, but relatively deep (i.e. outer shelf), living assemblages (e.g. Ingham 1978; Harper 1982, 2001; Daley 1992). In the
75
Ordovician, the geological context of Gourin (i.e. continental shelf, with relatively shallow deposits, at or below storm wave base) was clearly distinct from that documented in Girvan (i.e. outer edge of the Laurentian plate continental shelf). Moreover, contrary to the situation in the Lady Burn starfish beds, the Guernanic unit does not contain any evidence of lenticular thinning or several episodes of transport. Instead the whole horizon (including the echinoderm level) seems to be laterally continuous suggesting that the fauna was preserved in situ. Consequently, it is proposed here that the Guernanic fauna must have been entombed by a rapid influx of sediment as seen in ophiuroids from the Lower Lias of SW England (Kerr & Twichett 2004). It is most probable that the ophiuroid-stylophoran association represented the termination of a period of echinoderm colonization extinguished by sediment influx (as possibly suggested by the local abundance of floated micas). This would best explain the odd faunal content of the isolated echinoderm level atypical of the sampled horizon, primarily composed of trilobites, associated with rare brachiopods and molluscs (see Table 1). This trilobite-dominated assemblage would be more akin to the assemblages of the Middle Ordovician of Spain, where T. ibericus occurs at the base of a channel (5 x 2 cm). These channels are considered to be traps for the ophiuroids caught in the channel and covered quickly by channel fill (Hammann & Schmincke 1986). Although there is an absence of such channels in Guernanic, there are considerable similarities both in the monogeneric nature of the assemblage and the sedimentology. In Guernanic, the stylophorans are well preserved, and occur in 'clusters' on the rock surface (see Fig. 3b). Like the ophiuroids, stylophorans can also occur in deep low-energy distal shelf or slope environments (Lefebvre 2006). In most Gourin specimens, the preservation of the whole theca is likely to be the result of a quick burial to prevent disarticulation seen in type 1 skeletons. It is also likely that the plates of the theca are preserved in the same manner as regular echinoid tests (see Aslin 1968), where the theca is quickly filled with sediment through the openings in the cavity, thus preventing crushing of the theca. Another indicator of swift preservation is the absence in several specimens of the typical post-mortem kinking of their appendage (e.g. MHNN.P.025958 and MHNN.P.025960, see Fig 3b). In mitrates, such a flexed position of the aulacophore is generally interpreted as a distressed posture and/or as resulting from the postmortem contraction of this structure (Dehm 1932; Jefferies 1984; Parsley 1988; Ruta & Bartels
76
A.W. HUNTER E T AL.
Fig. 2. (a)-(d) Morphology of the ophiuroid Taeniaster armoricanus and (e)-(g) of the mitrate stylophoran Mitrocystella incipiens, Upper Member, Postolonnec Formation (Llandeilian), Guernanic quarry, Gourin (Morbihan), Brittany, France. (a) MHNN.P.025955. (b) MHNN.P.025954b. (c) MHNN.P.025954a. (d) MHNN.P.025956. (e) MHNN.P.025957, partly disarticulated lower thecal surface (upper aspect) showing the zygal crest. (f) MHNN.P.025961, articulated lower thecal surface (upper aspect), with adorals and some supracentrals. (g) MHNN.P.025960, articulated upper thecal surface, with adorals, supracentrals, and aulacophore (in extended position). Scale bar represents 1 cm.
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS
77
Fig. 3. (a) & (e) Taphonomy of the ophiuroid Taeniaster armoricanus and (b) of the mitrate stylophoran Mitrocystella incipiens, Upper Member, Postolonnec Formation (Llandeilian), Guernanic quarry, Gourin (Morbihan), Brittany, France. (a) MHNN.P.025955, articulated specimen. (c) MHNN.P.025955a-b, semi-articulated specimens. (e) MHNN.P.025958, cluster of articulated specimens. Scale bar represents 1 cm.
1998; Lefebvre 2003a). Consequently, the preservation of several specimens showing both their arm in the extended position and their cover plates ajar strongly suggests that these individuals were probably buried alive. In Gourin, the presence of some slightly disarticulated specimens of Mitrocystella (e.g.
MHNN.P.025957, see Fig. 3b) suggests that all mitrates were not alive when buried. However, the low degree of disarticulation of the Gourin Mitrocystella, combined with the absence of epibionts on their thecal plates, are both indicative of a short time of exposure on the sea floor. This preservation is clearly different from that
78
A.W. HUNTER ET AL.
Table 1. Faunal list of macrofossils found in Guernanic quarry, Gourin (Morbihan, France), in the Guernanic unit, Upper Member of the Schistes de Postolonnec Formation (Llandeilian), with an indication of their abundance (C, common; R, rare)
Brachiopods lingulids indet. (R) A egiromena (R) Eorhipidomella (Crozonorthis)?(R) articulate forms indet. (R) Echinoderms Calix (C) crinoids (C) Diamphidiocystis (R) Mitrocystella (C) Taeniaster (C)
Molluscs Trocholites (R) nautiloids indet. (R) Glyptarca (R) Praenucula (R) Redonia (R) bivalves indet. (R) Ribeiria (R)
observed in specimens of Mitrocystella from the Dobrotiva Formation (Llandeilian) of Bohemia. Although they are usually exquisitely preserved, and not much disarticulated, Bohemian mitrates are frequently relatively intensively bioeroded (e.g. numerous Arachnostega-like tubes on both the internal and external side of their lower thecal surface: see Bruthansovfi & Kraft 2003; Lefebvre 2006). These taphonomic features suggest that carcasses of Bohemian Mitrocystella were exposed for a long time on the seafloor, in quiet palaeoenvironmental conditions. Consequently, contrary to the situation in Gourin, the Bohemian assemblage clearly represents a taphocoenosis (accumulation of dead organisms). To sum up, several lines of evidence suggest instantaneous preservation of the Gourin echinoderm specimens (fully articulated, with even the most fragile portions still intact) and are suggestive of rapid burial in an otherwise low-energy environment. With such quick preservation, it seems extremely likely that both ophiuroids and stylophorans occurred together in the same environment.
Autecology of ophiuroids and stylophorans The above taphonomic section has established that fully articulated specimens of Taeniaster and Mitrocystella from Gourin are very probably
Trilobites Colpocoryphe (C) Coplacoparia (R)
Illaenid indet (C) Isabelinia (C) Neseuretus (C) Nob iliasaphus (C) Panderia (C) Parabarrandia (R) Phacopidina (R) Prionocheilus (R) Salterocorvphe (C) Selenopeltis (R) Zeliszkella (R)
Other fossils conularilds (R) hyolithids (R)
preserved in life position and correspond to an in situ assemblage, suddenly and deeply buried by obrution deposits, in a moderately deep-shelf environment, below, but close to, storm wave base. In Palaeozoic sediments, the frequent simultaneous occurrences of abundant stylophorans and ophiuroids in the same layers could indicate that these two groups of echinoderms shared comparable palaeoecology or functional morphology, or at least inhabited the same habitat. Mixed assemblages of ophiuroids and stylophorans are relatively common in the late Ordovician-early Devonian time interval. Notable examples that will be discussed here are the Izegguirene Formation of Anti-Atlas, Morocco (early Caradoc: Lefebvre 2006), the South Threave Formation of Girvan, Scotland, UK (Ashgill: Spencer 1950; Donovan et al. 1996; Jefferies & Daley 1996), the Ponta Grossa Formation of Parana, Brazil (Emsian: Caster 1954) and the Hunsrfick Slate, Bundenbach, Germany (Emsian: Dehm 1932; Ruta & Bartels 1998; Glass 2005). Such an association is not entirely unexpected, as both stylophorans and ophiuroids were small, flattened, free-living, unattached, gregarious, epibenthic or infaunal echinoderms, which were using their flexible arms for feeding and anchoring to the substrate for limited mobility (Lefebvre 2006). Comprising approximately 1800 species in 250 genera, ophiuroids are the largest group of
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS extant echinoderms and today range across all modem environments from the Arctic regions to the Equator (Simms et al. 1993). In terms of ecology, modern ophiuroids can be epifaunal suspension feeders or infaunal detritus feeders. However, the morphology of even the modern fauna is not mutually exclusive to one feeding habit, although many interpretations have been proved to be ambiguous. For instance, Spencer (1951) suggested that aboral paxillae in Palaeozoic asteroids could indicate an infaunal behaviour. However, in modern brittlestars, such a character occurs in both epifaunal and infaunal forms (Gale 1987). Although it is possible that Palaeozoic ophiuroids were probably low-level suspension feeders, with some of them adapted to occasional scavengers (see Petr 1989), they however lacked the mobility of the arms seen in more recent forms, so that a more advanced filter-feeding habit may have not been possible (Shackleton 2005). With many anatomical characters common to both ophiuroids and stylophorans (Nichols 1972; Lefebvre 2006), it is probable that comparable feeding techniques were used in these two groups of echinoderms. Thus, observations of modern ophiuroid meadows can probably help to understand not only the palaeoecology of fossil brittlestars, but also that of stylophorans. Indisputable ophiuroid remains have not been documented before the Lower Ordovician (Tremadocian: Thoral 1935; Spencer 1950, 1951; Fell 1963), and a possible gregarious or social mode of life for this echinoderm group (so typical of their modern relatives) was not evident before the Late Ordovician (Caradoc: Petr 1989; Fujita 1992; Mikul~s et al. 1995). The oldest dense assemblages of ophiuroids known so far have been documented in the Caradoc of Morocco and Bohemia (Petr 1989; Lefebvre 2006). On the other hand, stylophorans have been described as far back as the Middle Cambrian, and a gregarious mode of life seems to be the rule for most of them (Ubaghs 1968; Ubaghs & Robison 1988). The distribution of modern dense beds of epibenthic ophiuroids is relatively substrateindependent, but strongly controlled by depth (and water temperature). Extensive, dense populations of modern epibenthic ophiuroids are nearly exclusively restricted to the bathyal zone (depths of between 200 and 3000 m: Blaber et al. 1987; Fujita & Ohta 1989; Fujita 1992; Gage & Brey 1994). In shallower environments, very rare, small, patchy aggregations of epibenthic ophiuroids have been documented in the infralittoral zone (depths of between 10 and 50 m), as for example on the SW coast of England (Warner
79
1971; Broom 1975), or in a salt lake of the Bahamas (Aronson & Harms 1995; Aronson & Blake 1997). As argued above, ophiuroids can occur in dense aggregations (also known as 'brittlestar beds'), in which representatives of other echinoderm groups (e.g. echinoids, asteroids) are frequently present (Hunter 2004), whilst members of other major groups of marine invertebrates (e.g. crustaceans, molluscs, sea anemones, sponges) are generally rare and poorly diversified (Fujita & Ohta 1989). A gregarious mode of life, with very high densities of individuals and lowdiversity-associated faunas are features, which are not only observed in modern and fossil ophiuroid assemblages (see Petr 1989; Mikul~s et al. 1995) but seem to be a conservative characteristics of most echinoderm dense populations as far back as the early Palaeozoic (Lefebvre 2006). Such a gregarious mode of life can be highly advantageous not only for reproductive success (Lawrence 1987; Emson 1998), but also for a greater efficiency of filter feeding: in large aggregations of modern ophiuroids the presence of numerous arms facing the current locally slows down the current flow, and thus increases the deposition of food particles into the feeding structures (Warner 1971; Fujita 1992). A gregarious mode of life may also provide a better resistance against both high current flows (Warner 1971; Broom 1975) and predation ('principle of strength in numbers': Warner 1971; Fujita & Ohta 1989; Fujita 1992; Aronson & Blake 1997). Several studies based on modern and fossil ophiuroids have indicated that the presence and spatial extension of brittlestar beds were both strongly inversely correlated with the abundance of putative predators (e.g. sea-stars, teleost fishes: Aronson 1992; Fujita 1992; Aronson & Blake 1997). More generally, the distribution of dense populations of both fossil and modern sessile, suspension-feeding, epibenthic echinoderms (e.g. crinoids, ophiuroids) appears to be strongly influenced by the intensity of predation pressure (Bottjer & Jablonski 1988; Fujita 1992; Aronson & Blake 1997; Meyer 1997). This important interpretation suggests that, in the Palaeozoic, dense assemblages of stylophorans and/or other free-living epibenthic suspensionfeeding echinoderms (e.g. cinctans, ophiuroids, solutes) were probably also restricted to palaeoenvironments where predators were rare and even absent (Lefebvre 2006). However, observation of modern and fossil brittlestar beds indicate that in such echinoderm meadows not only predators are rare, but the whole associated fauna is also scarce and poorly diversified. Consequently, the comparison with modern
80
A.W. HUNTER E T A L .
brittlestar beds demonstrates that the low diversity observed in most dense assemblages of ophiuroids and stylophorans (e.g. in the Middle Ordovician of Gourin or Traveusot, or in the Upper Ordovician of Bohemia: see Petr 1989; Mikulas et al. 1995) probably does not result from a preservational bias, but corresponds to a general feature of most gregarious echinoderm faunas (Lefebvre 2006). Echinoderm meadows represent major bioaccumulations and are in urgent need of further consideration. Such bioaccumulations are not only important in tropical waters, but in m o d e m Arctic waters as well. This study shows that such gregarious occurrences of echinoderms are not uncommon in the fossil record.
Conclusions 9 In the Llandeilian Guernanic unit (Upper Member of the Schistes de Postolonnec Formation) of Gourin (Brittany), sedimentological, taphonomic (e.g. bivalves preserved with their valves in connection) and palaeontological evidence (e.g. composition of trilobite assemblages) all suggest that deposition was at or below storm wave base. 9 In Gourin, the fossils are largely complete and some of them in life position (e.g. mitrates with extended aulacophore), suggesting rapid sedimentation. ~ An application of experimental taphonomy on ophiuroids has allowed the greater understanding of the origin of bioaccumulations and whether they are allochthonous (e.g. Girvan) or autochthonous (e.g. Gourin). Thus, establishing that the ophiuroids like the mitrates were preserved quickly after death 9 Large echinoderm bioaccumulations are recorded early in the fossil record and could have been at least as extensive as those observed in modern temperate and Arctic waters. 9 It is becoming increasingly apparent that stylophorans occurred in association with ophiuroids, and that the Gourin assemblage represents the oldest indisputable evidence of such an association. This paper is a contribution to IGCP project No. 503 'Early Palaeozoic Palaeography and Palaeoclimate', and of the team 'Macrodvolution et dynamique de la biodiversit6' of the UMR CNRS 5561 Biog6osciences (Universit6 de Bourgogne, Dijon). The authors are particularly grateful to P. Jean (Mus6um d'Histoire naturelle de Nantes) and J. Thomas (Biog6osciences, Universit~ de Bourgogne) for photographs, D. Lewis
(Natural History Museum, London), J. Plaine (G6osciences Rennes), R. Prokop (Narodni Muzeum, Prague) and D. Vizcaino (Carcassonne) for access to specimens of Mitrocystella, and P.-A. Jaouen (Riecsur-Belon), L. Le Berre (Morlaix) and P. R6audin (Le Faouet) for access to the material they have collected in Gourin. The authors are also very grateful to D. le Solliec, maire de Gourin, for providing authorization for sampling, and to F. Barazer (Gourin) for providing detailed maps and various important documents on the former quarries of Gourin. The Conseil R6gional de Bourgogne is greatly acknowledged for funding A.W. Hunter's research through a postdoctoral fellowship. The authors thank J.J. Alvaro (Zaragoza), D.A.T. Harper (Copenhagen) and an anonymous reviewer for their informed and useful comments. Finally, A.W. Hunter's wishes to thank E. Clarkson (Edinburgh) for introducing him to the importance of Palaeozoic asterozoans and in particular the Ordovician of the Girvan district, Scotland.
Appendix: Systematic palaeontology Class OPHIUROIDEA Gray, 1840 (see Simms et al. 1993) Order OEGOPHIUROIDEA Matsumoto, 1915 Suborder LYSOPHIURINA Gregory, 1897 Family PROTASTERIDAE S. A. Miller, 1889 (Diagnosis of family modified by Petr 1989) Genus TAENIASTER Billings, 1858 Type species', Palaecoma spinosa Billings, 1857 Diagnosis. Modified from Hammann & Schmincke (1986) and Petr (1989): interradial outline of the disc straight or concave (but may be convex). Oral and aboral faces of the disc with small scales of uncertain outline covered with granulated integument. Oral face covered with spines. Probably no true marginalia present. Disc margins often thickened on aboral faces (possible post-mortem folding of the skin after death). Mouth angle plates short, with oral side covered with large papillae. Ambulacrals with typical boot-shaped ridge on the oral side of arms. Median suture between alternate ambulacrals straight, with sinuous appearance between neighbouring boot-shaped ridges. Excavations for the dorsal longitudinal muscles not very broad. Aboral outline of ambulacrals trapezoidal to quadrate. Laterals ear-shaped, with vertical and groove spines. Remarks. Taeniaster is considered a valid genus in this study despite being placed by Shackleton (2005) within the genus Protaster along with Bohemura. Despite good arguments by Shackleton (2005) we consider that Taeniaster has too many distinct characters to be included within a single Protaster genus. The oral sides of arms possess a straight median suture between the alternate ambulacrals. Straight median suture is unique among the protasterid ophiuroids. A post-mortem condition allows the tips of the arms to appear whip-like. However, following Petr (1989), this character is considered here as an unreliable character, and not
ORDOVICIAN O P H I U R O I D - S T Y L O P H O R A N BEDS included in the diagnosis above. Although Hammann & Schmincke (1986) noted that the area where the open ambulacral groove grades into the closed part does not vary within one species, Kesling & Le Vasseur (1971; see Petr 1989) showed considerable variation. The area where the open ambulacral groove grades into the closed part was considered to be very stable at species level by Hammann & Schmincke (1986), but highly variable in position by Kesling & Le Vasseur (1971). The degree of closure of the ambulacram possibly represents a post-mortem feature; this is also true in Taeniaster lacking a whip-like tail, as this is simply not preserved in most specimens (Petr 1989).
Taeniaster armoricanus sp. nov. Figs 2a-d & 3a, c.
Derivation of name. Both, the holotype and paratype are from the Department of Morbihan, in the ancient Gallo-Roman territory of Armorica (modern Brittany). Holotype. Specimen MHNN.P.025954b. Paratype. Specimen MHNN.P.025955. Material. Six specimens preserved on four slabs, housed in the Musfum d'Histoire Naturelle de Nantes. Diagnosis. Medium-sized pentagonal-round disc. Shortened arms with small short straight ambulacra with square boots. Very large spaces between the ambulacra and laterals, weakly lobed bowed reduced laterals that tend to be larger than the ambulacra. Description. The disc small and lightly pentagonal curved edges (Fig. 2a, MHNN.P.025955), these edges range from straight to very round in some specimens making the disk appear almost circular (Fig. 2b, MHNN.P.025954b). Only aboral surface is observed in the specimens, the thickening of the disc is observed in one specimen (Fig. 2d, MHNN.P.025956). Like many specimens of ophiuroids it is difficult to see the detail of the integument covering the aboral sides of the disc, although the remains of the papillae can be seen. The arms are well preserved. In most specimens the arms taper evenly (Fig. 2a, MHNN.P.025955), but there is no evidence of whip-like terminal section on any of the arms. From five to seven pairs of the ambulacral ossicles are incorporated into the disc (Fig. 2b, MHNN.P.025954b). About 16-17 pairs are present in the free arm; no ambulacra in the arm tips are recorded (Fig. 2a, MHNN.P.025955). Ambulacral plates of the left and right of the arm alternate and the median suture (also known as the median groove) is straight and narrow. The ambulacra are boot-shaped in morphology (Fig. 2c, MHNN.P.025954a). In the boot, the toe or foot is thinner and shorter and not bluntly developed compared to the leg section making the foot portion to appear larger in most specimens. Laterals are curved and are thinner and slightly reduced in size compared with the ambulacrals. Ambulacrals are not ear-shaped (i.e. depressed down) with clear space in between the lateral plates and those of the ambulacra (Fig. 2b, MHNN.P.025954b)
81
Remarks. Like in many early ophiuroid species, the differences seen are very slight but significant to diagnose a new species (without having to infer significant geographical separation as a proxy for new species diagnosis). Having a smaller and more rounded disk with shorter arms, this species has more in common with Taeniaster ibericus with reduced and more rounded disk and reduced laterals or adambulacrals. However, T. armoricanus is separated by the further reduction of laterals and a more pentagonal disk than that observed in Taeniaster ibericus. The laterals are not ear-shaped as seen in Taeniaster bohemicus specimens, where laterals are larger than ambulacrals, while the short arms seems to be a unique feature of T. armoricanus. Occurrence. Upper Member, Postolonnec Formation (Llandeilian), Guernanic quarry, Gourin, Morbihan, Brittany, France. Class STYLOPHORA Gill & Caster, 1960 Order M I T R A T A Jaekel, 1918 Suborder M I T R O C Y S T I T I D A Caster, 1952 Genus MITROCYSTELLA Jaekel, 1901 Type species, Anomalocystites incipiens Barrande, 1887
Remarks. Two species were assigned to the genus Mitrocystella by Jaekel (1901, 1918): M. incipiens (type species) and M. barrandei. In his cladistic analysis of mitrocystitid mitrates, Lefebvre (2000b) placed M.
barrandei (with 13 marginals, Z, and two infracentrals) in a distinct genus (Promitrocystites), so as to avoid the paraphyly of the genus Mitrocystella (defined by the possession of 12 marginals, Z, and two infracentrals). In his original description of 'Anomalocystites' incipiens, Barrande (1887) illustrated eight specimens from the S~trka and Dobrotivfi formations (Middle Ordovician) of Bohemia, but no type was implicitely designated. In his revision of mitrocystitid mitrates, Chauvel (1941) re-examined Barrande's original material of Mitrocystella incipiens and chose one specimen as lectotype (NMP.L9292; illustrated in Barrande 1887, pl. 5, fig. 1.3-4). Chauvel (1941, p. 174) also pointed out that two of the specimens figured by Barrande (1887) do not correspond to Mitrocystella incipiens. Re-examination of the original material of Barrande by one of the authors (B. Lefebvre) confirms that the two specimens NMP.L9297 (Barrande 1887, pl. 5 fig. 1.5-6) and NMP.L9300 (Barrande 1887, pl. 5 fig. 1.15-16) should be assigned to Mitrocystites mitra, and not to Mitrocystella incipiens. Moreover, the re-examination of Barrande's original material also shows that the morphology of the lectotype of Mitrocystella incipiens differs from that of all other specimens assigned to this species, and reported so far from Bohemia, Brittany, Portugal and Spain: four plates (instead of three) are present in central position on the lower thecal surface of NMP.L9292 (two posterior infracentral platelets, instead of one, are located between Z, PP1, PP2 and M'5). Although relatively rare, irregularities in plating sometimes occur in mitrocystitid mitrates (Lefebvre
A.W. HUNTER ET AL.
82
2003b). Consequently, the specimen chosen as lectotype of M. incipiens by Chauvel (1941) more probably corresponds to an abnormal (teratological) individual, rather than to a different species.
Mitrocystella incipiens (Barrande, 1887) Figs 2e-g & 3b
Material. Nine specimens preserved on five slabs are deposited in the collections of the Mus6um d'Histoire Naturelle de Nantes (MHNN.P.025957-61). Additional material, present in private collections, was also examined. Description. Measurements made on the theca of complete specimens follow the definitions proposed by Chauvel (1941, p. 156). Thecal length (L) is comprised between about 21 and 27 mm (L is estimated between the M1-M'I suture, at the aulacophore insertion, and the posterior edge of PP2). Thecal width (/) is comprised between about 15.7 and 19 mm (l corresponds to the largest width measured perpendicularly to L). The ratio llL is between 0.70 and 0.77. In some individuals, the theca is relatively elongate and narrow (llL< 0.75), with gently curved lateral margins, a straight to slightly concave posterior edge, and asymmetric outlines (e.g. MHNN.P.025960 and MHNN.P.025961). Thecal asymmetry is due to: (1) the forward extension of the right anterior lobe (right of the aulacophore insertion), clearly longer and more prominent than the left one; and (2) the course of the left thecal side, straighter than that of the right one. In other specimens, the theca is apparently less elongate and larger (l/L>0.75), with more parallel-sided lateral margins, less asymmetric outlines (left and right anterior lobes of comparable extension), and a gently curved, rounded posterior extremity (e.g. MHNN.P.025957 and MHNN.P.025959). Specimens from Gourin can be grouped into three categories based on their preservation: (1) individuals with partly disarticulated lower thecal surface, no upper surface elements (adorals, supracentrals) and no aulacophore (e.g. MHNN.P.025957); (2) specimens with fully articulated lower thecal surface, partly preserved upper surface (adorals and a few supracentrals) and no aulacophore (e.g. MHNN.P.025959 and MHNN.P.025961); and (3) individuals with wellpreserved lower and upper thecal surfaces, and complete, articulated aulacophore in extended position (e.g. MHNN.P.025958 and MHNN.P.025960). Although they are flattened and pyritized, most specimens are relatively well preserved, and most external and internal features generally documented in three-dimensional individuals preserved in silicic-aluminous concretions can be observed. Two large adorals of comparable shape and dimensions form the anterior edge of the upper surface. The external surface of each adoral bears, anteriorly, a strong anterior transverse crest, and posteriorly to this crest, numerous (about 12-15), sinuous, transverse cuesta-shaped ribs (e.g. MHNN.P.025960 and
MHNN.P.025961). The hydropore (cross-shaped slit anterior to the adoral crest) and the internal surface of adorals have not been observed. Posteriorly to the pair of adorals, the upper thecal surface is paved by numerous (more than 70), slightly imbricated, unorganized supracentral platelets (e.g. MHNN.P.025960). The size of supracentrals regularly decreases in a posterior direction. Several (at least 12), elongate and narrow anal plates are preserved in some specimens (e.g. MHNN.P.025958 and MHNN.P.025960). The lower thecal surface is composed by 12 large plates in marginal position (M1, M2, M3, M4, G and PP1 on the right; M'I, M'2, M'3, M'4 and M'5 on the left; PP2 in terminal position), and three elements in central position (the zygal plate Z, and two left infracentrals). No orifices are present on the lower thecal surface (e.g. lateripores, paripores). Several features of the internal (upper) side of the lower thecal surface can be observed in the Gourin specimens: (1) on M1 and M'I, a deep, well-marked furrow located on the posterior wall of the two aulacophore apophyses : the transverse anterior groove (e.g. MHNN.P.025957 and MHNN.P.025958); (2) on M'I and Z, a strong oblique ridge, asymmetrical in cross-section (right flank steeper than the left one): the zygal crest or septum (e.g. MHNN.P.025957 and MHNN.P.025959); (3) on M'I, a small circular depressed area located above the contact between the left apophysis and the zygal crest: the left scutula (e.g. MHNN.P.025957 and MHNN.P.025958); (4) on M'I, a short vertical ridge located left of the zygal crest and connected to the left scutula: the accessory septum (e.g. MHNN.P.025958); and (5) on M'5 and PP1, traces of insertion for rectal muscles (e.g. MHNN.025958 and MHNN.P.025959). Morphology of the aulacophore insertion is apparently variable within Gourin specimens: relatively deep and narrow (e.g. MHNN.P.025958), or shallower and more widely open (e.g. MHNN.P.025957). In bestpreserved specimens, at least seven-eight imbricated rings have been observed in the proximal portion of the aulacophore (e.g. MHNN.P.P.025958 and MHNN.P.025960). Each proximal ring is made of four plates (two inferolaterals and two tectals). The lower surface of the stylocone bears two strong, laterally compressed spines, connected by a sharp, longitudinal ridge (e.g. MHNN.P.025960). Ornamentation rapidly decreases in height and strength on the lower surface of more distal brachials (ossicles). In some specimens, ajar cover plates are preserved articulated to underlying ossicles (e.g. MHNN.P.025958). Remarks. The morphology of all mitrocystitid mitrates collected in Gourin is identical to that of specimens of Mitrocystella incipiens reported from elsewhere in Bohemia, Brittany, Portugal and Spain, and thus totally supports their assignment to this species. Chauvel (1937, 1941, 1981) placed all specimens of M. incipiens from Brittany in a different subspecies (M. incipiens miloni) from those from Bohemia (M. incipiens
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS
mcipiens). This distinction is mostly based on: (1) the aspect of thecal outlines (in 34. incipiens incipiens, lateral margins are parallel-sided, rather than curved, and the posterior thecal edge is not straight, but rounded); (2) the morphology of the aulacophore insertion (large and widely open in M. incipiens incipiens, narrower and deeper in M. incipiens miloni); and (3) the asymmetry of the two anterior lobes (right anterior lobe more developed in M. incipiens miloni). For the same reasons, all specimens collected in Portugal and Spain were also assigned to M. incipiens miloni (e.g. Guti6rrez-Marco & Mel6ndez 1987). Re-examination of the six Bohemian specimens ofM. incipiens illustrated by Barrande (1887) confirms that their morphology appears slightly different from that of most specimens collected in Brittany, Portugal and Spain. However, it should be pointed out that: (1) the mean size of Barrande's specimens is smaller than that of most individuals sampled in Britanny, Spain; and Portugal; and (2) the morphology of small Breton specimens is closely comparable to that of Bohemian forms. It is thus possible that the differences pointed out by Chauvel (1941) more probably reflect growth allometries, rather than actual regional differences between Bohemian forms on one side, and Ibero-Armorican forms on the other side. The slight morphological differences observed within the specimens collected in Gourin seems to support this interpretation. Occurrence. Upper Member, Postolonnec Formation (Llandeilian), Guernanic quarry, Gourin (Morbihan), Brittany (France); Traveusot Formation (Llandeilian), Bain-de-Bretagne, Guichen, Guignen and Langon (Ille-et-Vilaine), Brittany (France); Valongo Formation (Llandeilian), San Pedro de Cova (Portugal); Guindo Shales (Llandeilian), Almaden-Campos de Calatrava, Almodovar del Campo and Calzada de Calatrava, Ciudad Real (Spain); Sarka Formation (Abereiddian), Praha-Sarka and Rokycany (Osek), Bohemia (Czech Republic); Dobrotiva Formation (Llandeilian), Mal6 Prilepy and Svata Dobrotiva (='Sancta Benigna' in Barrande 1887), Bohemia (Czech Republic).
References ARONSON, R. B. 1992. Biology of a scale-independent predator-prey interaction. Marine Ecology Progress Series, 89, 1-13. ARONSON, R. B. & BLAKE, D. B. 1997. Evolutionary paleoecology of dense ophiuroid populations. Paleontological Society Papers, 3, 107-119. ARONSON, R. B. & HARMS, A. 1995. Ophiuroids in a Bahamian saltwater lake: the ecology of a Paleozoic-like community. Ecology, 66, 1472-1483. ASLIN, C. J. 1968. Echinoid preservation in the Upper Estuarine Limestone of Blisworth, Northamptonshire. Geological Magazine, 105, 506-518. BABIN, C. 1966. Mollusques bivalves et cOphalopodes du Pal~ozo~'que armoricain. Th6se Rennes (1965). Imprimerie Commerciale et Administrative, Brest.
83
BABIN, C., CHAUVEL,J. J., LARDEUX, H., PARIS, F. & ROBARDET, M. 1976. Lexique des Formations de l'Ordovicien armoricain. Bulletin de la SociOtk gdologique et mindralogique de Bretagne, sdrie C, fascicule sp6cial, 1-31. BALL, B. J., COSTELLOE, J., KONNECKER, G. & KEEGAN, B. F. 1995. The rocky subtidal assemblages of Kinsale Harbour (south coast of Ireland). In: ELEFTHERIOU,A., ANSELLA. D. & SMITH C. J. (eds) Biology and Ecology of Shallow Coastal Waters. Proceedings 28th European Marine Biology Symposium, 1993, Iraklio, Crete, 239-301. BARRANDE, J. 1887. Systkme Silurien du centre de la Bohdme. Volume 7. Classe des Echinodermes, Ordre des Cystidkes. Rivnac, Prague, and Gerhard, Leipzig. BARROIS, C. 1884. Le granite de Rostrenen, ses apophyses et ses contacts. Annales de la Sociktk gbologique du Nord, 12, 1-119. BARROIS, C. 1885. L6gende de la feuille de Chfiteaulin. Annales de la Socidtd gdologique du Nord, 13, 49-65. BARROIS, C. 1889. Observations sur la constitution g6ologique de l'Ouest de la Bretagne (3~me article). Annales de la Socidtd gdologique du Nord, 16, 1-11. BILLINGS, E. 1857. Descriptions of new fossils from Silurian rocks of Canada. In: Report of Progress for Years 1853-1856. Geological Survey of Canada, 256-345. BILLINGS, E. 1858. On the Asteridae of the Lower Silurian rocks of Canada, Figures and descriptions of Canadian organic remains. Geological Survey of" Canada, 3, 75-85. BLABER,S. J. M., MAY, J. L., YOUNG, J. W. & BULMAN, C. M. 1987. Population density and predators of Ophiacantha fidelis (KOEHLER, 1930) (Echinodermata: Ophiuroidea) on the continental slope of Tasmania. Australian Journal of Marine and Freshwater Research, 38, 243-247. BOTTJER, D. J. 8r JABLONSKI, D. 1988. Paleoenvironmental patterns in the evolution of postPaleozoic benthic marine invertebrates. Palaios, 3, 540-560. BRETT, C. E., MOFFAT, H. A. & TAYLOR, W. L. 1997. Echinoderm taphonomy, taphofacies, and Lagerst~itten. Paleontological Society Papers, 3, 147-190. BRIGGS, D. E. G. & KEAR, A. J. 1994. Decay and mineralization of shrimps. Palaios, 9, 431-456. BROOM, D. M. 1975. Aggregation behaviour of the brittle-star Ophiotrix fragilis. Journal of the Marine Biological Association of the United Kingdom, 55, 191-197. BRUN, E. 1969. Aggregation of Ophiotrix Jmgilis (ABILDGAARD) (Echinodermata: Ophiuroidea). Nytt Magasin for ZoologL 17, 153-160. BRUTHANSOVA, J. & KRAFT, P. 2003. Pellets independent of or associated with Bohemian body fossils. A cta Palaeontologica Potonica, 48, 437-445. CASTER, K. E. 1952. Concerning Enoploura of the Upper Ordovician and its relation to other carpoid Echinodermata. Bulletins of American Paleontology, 34, 1-47. CASTER, K. E. 1954. A new carpoid from the Parana Devonian. Anais da Academia Brasileira de Ci~ncias, 26, 123-147.
84
A.W. HUNTER ET AL.
CHAUVEL, J. 1937. Les H6t6rost616es armoricaines. Compte Rendu Sommaire des Skances de la Sociktk gkologique et min~ralogique de Bretagne, 2, 1-3. CHAUVEL, J. 1941. Recherches sur les cystoides et les carpoides armoricains. Mkmoires de la Sociktk gdologique et minkralogique de Bretagne, 5, 1-286. CHAUVEL, J. 1980. Donn6es nouvelles sur quelques cystoides diploporites (Echinodermes) du Pal6ozoique armoricain. Bulletin de la Sociktk gdologique et min~ralogique de Bretagne, ( C), 12, 1-28. CHAUVEL, J. 1981. Etude critique de quelques 6chinodermes stylophores du Massif armoricain. Bulletin de la Sociktk gOologique et minOralogique de Bretagne, (C), 13, 67-101. CHAUVEL, J. & MELENDEZ, B. 1978. Les 6chinodermes (cystoides, ast6rozoaires, homalozoaires) de l'Ordovicien moyen des Monts de Tol6de (Espagne). Estudios geologicos, 34, 75-87. COUTO, H. & GUTII~RREZ-MARCO, J. C. 2000. Alguns f6sseis raros ou pouco conhecidos da F o r m a ~ o de Valongo (Ordovicico M6dio, Norte de Portugal). In: I Congresso Ibkrico de Paleontologia & XVI Jornadas de la Sociedad espa~ola de Paleontologia. Publica9oes da Universidade de Evora, 191-192. DALEY, P. E. J. 1992. The anatomy of the solute Girvanicystis batheri (?Chordata) from the Upper Ordovician of Scotland and a new species of Girvanicystis from the Upper Ordovician of south Wales. Zoological Journal of the Linnean Society, 105, 353-375. DAVY, M. 1886. Note sur un ophiure (Protaster daoulasensis) du D6vonien inf6rieur de la rade de Brest. Bulletin de la SociOtk gkologique de France, 14, 182-187. DEHM, R. 1932. Cystoideen aus dem rheinischen Unterdevon. Neues Jahrbuch fiir Mineralogie, Geologie und Paliiontologie, 69, 63-93. DONOVAN, S. K. 1991. The taphonomy of echinoderms: calcareous multi element skeletons in the marine environment. In: DONOVAN, S. K. (ed.) The Processes of Fossilization. Belhaven Press, London, 241-269. DONOVAN, S. K., PAUL, C. R. C. & LEWIS, D. N. 1996. Echinoderms. In: HARPER,D. A. T. & OWEN, A. W. (eds) Fossils of the Upper Ordovician. Palaeontological Association, London, 202-267. EMSON, R.H. 1998. Making the best of it: the ecology of deep sea echinoderms. In: Mool, R. & TELFORD, M. (eds) Echinoderms: San Francisco. Balkema, Rotterdam, 452. FELL, H. B. 1963. The phylogeny of sea-stars. Philosophical Transactions of the Royal Society of London, Series B, 246, 381-435. FUJITA, T. 1992. Dense beds of ophiuroids from the Paleozoic to the Recent: the significance of bathyal populations. Otsuchi Marine Research Centre Reports, 18, 25-41. FUJITA, T. & OHTA, S. 1989. Spatial structure within a dense bed of the brittle star Ophiura sarsii (Ophiuroidea: Echinodermata) in the bathyal zone off Otsuchi, northeastern Japan. Journal of the Oceanographical Society of Japan, 45, 289-300. FUJITA, T. & OHTA, S. 1990. Size structure of dense populations of the brittle-star Ophiura sarsii
(Ophiuroidea: Echinodermata) in the bathyal zone around Japan. Marine Ecology Progress Series, 64, 113-122. GAGE, J. D. & BREY, T. 1994. P/B ratios in deep-sea brittle stars. In: DAVID,B., GUILLE,A., FLORAL,J.-P. & Roux, M. (eds) Echinoderms through Time. Balkema, Rotterdam, 421-426. GALE, J. E., 1987. Phylogeny and classification of the Asteroidea (Echinodermata). Zoological Journal of the Linnean Society of London, 89, 107-132. GILL, E. D. & CASTER, K. E. 1960. Carpoid echinoderms from the Silurian and Devonian of Australia. Bulletins of American Paleontology, 41, 5-71. GLASS, A. 2005. Two isorophid edrioasteroids (Echinodermata) encrusting conulariids from the Hunsrtick Slate (Lower Devonian, Emsian; Rheinisches Schiefergebirge) of Germany. Senckenbergiana lethaea, 85, 31-37. GOLDRING, R. & STEPHENSON, D. G. 1972. The depositional environment of three starfish beds. Neues Jahrbuch fiir Geologic und Paliiontologie, Monatschefte, 10, 611-624. GRAY, J. E. 1840. A synopsis of the genera and species of the class Hypostoma (Asterias Linn.). Annals' and Magazine of Natural History, 6, 275-290. GREGORY, J. W. 1897. On the classification of the Palaeozoic echinoderms of the group Ophiuroidea. Proceedings of the Zoological Society of London, 1896, 1028-1044. GUTII~RREZ-MARCO, J. C. & MELI~NDEZ, B. 1987. Nuevos hallazgos de Estil6foros (Homalozoos) en los materiales ordovicicos de la zona Centroib6rica. Coloquios de Paleontologia, 41, 41-50. GUTII~RREZ-MARCO,J. C., CHAUVEL,J., MELt~NDEZ,B. & SMITH, A. B. 1984. LOS equinodermos (Cystoidea, Homalozoa, Stelleroidea, Crinoidea) del Paleozoico inferior de los Montes de Toledo y Sierra Morena (Espafia). Estudios geologicos, 40, 421-453. GUTII~RREZ-MARCO, J. C., RABANO, I. ET AL. 1999. Faunal dynamics between Iberia and Bohemia during the Oretanian and Dobrotivian (late Middle-earliest Upper Ordovician), and biogeographic relations with Avalonia and Baltica. Acta Universitatis Carolinae, Geologica, 43, 487-490. HAMMANN, W. & SCHMINCKE, S. 1986. Depositional environment and systematics of a new ophiuroid, Taeniaster ibericus n. sp., from the Middle Ordovician of Spain. Neues Jahrbuch fiir Geologie und Paliiontologie, Abhandlungen 173, 47-74. HARPER, D. A. T. 1982. The stratigraphy of the Drummuck Group (Ashgill), Girvan. Geological Journal, 17, 251-277. HARPER, D. A. T. 2001. Late Ordovician brachiopod biofacies of the Girvan district, SW Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 91, 471-477. HENRY, J. L. 1980. Trilobites ordoviciens du Massif armoricain. Mkmoires de la Socidtk-gdologique et min&alogique de Bretagne, 22, 1-250. HENRY, J. L. 1989. Pal6oenvironnements et dynamique de faunes de trilobites dans l'Ordovicien (Llanvirn sup6rieur-Caradoc basal) du Massif armoricain (France). Palaeogeography, Palaeoclimatology, Palaeoecology, 73, 139-153.
ORDOVICIAN OPHIUROID-STYLOPHORAN BEDS HENRY, J. L., VIDAL, M. & LACOMBE,L. 1993. Apparition soudaine de Protolloydolithus (Trilobita, Trinucleidae) dans l'Ordovicien (Llandeilo) du Massif armoricain: un essai d'interpr6tation. Geobios, 26, 201-206. HILY, C. 1991. Is the activity of benthic suspension feeders a factor controlling water quality in the Bay of Brest? Marine Ecology Progress Series, 69, 179-188. HOTCHKISS, F. H. C., PROKOP, R. J. & PETR, V. 1999. Isolated skeletal ossicles of a new brittlestar of the Family Cheiropterasteridae SPENCER, 1934 (Echinodermata: Ophiuroidea) in the Lower Devonian of Bohemia (Czech Republic). Journal of the Czech Geological Society, 44, 189-193. HUNTER, A. W. 2004. Facies control on distribution of crinoid habitats in the Middle Jurassic (Bathonian) of England: implications for other echinoderm groups. In: HEINZELLER, T, & NEBELSICK, J. H., (eds) Echinoderms: Miinchen, Proceedings of the 11 th International Echinoderm Conference, Munich, Germany, 6-10 October 2003. Balkema, Leiden, 585. HUNTER, A. W. & DONOVAN, S. K. 2005. Field sampling bias, museum collections and completeness of the fossil record in echinoderms from the Campanian-Maastrichtian of Northern Europe. Lethaia, 38, 10. INGHAM, J. K. 1978. Geology of a continental margin. 2. Middle and late Ordovician transgression, Girvan. In: BOWLS, D. R. & LEAKE, B. E. (eds) Crustal Evolution in Northwestern Britain and Adjacent Regions. Seel House Press, Liverpool, 63-176. JAEKEL, O. 1901. Uber Carpoideen; eine neue Klasse von Pelmatozoen. Zeitschrift der deutschen geologischen Gesellschaft, 52, 661-677. JAEKEL, O. 1918. Phylogenie und System der Pelmatozoen. Pa liiontologische Zeitschrift, 3, l - 128. JAGT, J. W. M., 1999. Late Cretaceous-Early Palaeogene echinoderms and the K/T boundary in the southeast Netherlands and northeast Belgium-Part 1: Introduction and stratigraphy. Scripta Geologica, 116, 1-57. JAGT, J. W. M. 2000. Late Cretaceous-Early Palaeogene echinoderms and the K/T boundary in the southeast Netherlands and northeast Belgium-Part 3: Ophiuroids; (with a chapter on: Early Maastrichtian ophiuroids from Rfigen (Germany) and Mon (Denmark) by M. Kutscher & J.W.M. Jagt). Scripta Geologica, 121, 1-179. JEFFERIES, R. P. S. 1967. Some fossil chordates with echinoderm affinities. In: MILLOT, N. (ed.) Echinoderm Biology. Academic Press, London, 163-208. JEFFERIES, R. P. S. 1984. Locomotion, shape, ornament, and external ontogeny in some mitrate calcichordates. Journal of Vertebrate Paleontology, 4, 292-319. JEFFERIES, R. P. S. & DALLY, P. E. J. 1996. Calcichordates. In: HARPER, D. A. T. & OWEN, A. W. (eds) Fossils of the Upper Ordovician. The Palaeontological Association, London, 268-276. JUTERZENKA, K. V. & SOLTWEDEL, T. 2004. Echinoderm assemblages in the deep Greenland Sea: patterns in space. In: HEINZELLER,T. & NEBELSICK, J. H. (eds) Echinoderms." Miinchen. Balkema, Leiden, 221-223.
85
KERFORNE, F. 1901. Etude de la rkgion silurique occidentale de la presqu'fle de Crozon (Finist&e). Imprimerie Fr. Simon, Rennes. KERR, T. J. V. & TwrrCHETr, R. J. 2004. Experimental decay and disarticulation of Ophiura texturata: implications for the fossil record of ophiuroids. In: HEINZELLER, T. & NEBELSICK,J. H. (eds) Echinoderms: Miinchen. Balkema, Leiden, 439-446. KESLING, R. V. & LE VASSEUR, D. 1971. Strataster ohioensis, a new Early Mississipian brittle-star, and the paleoecology of its community. Contributions from the Museum of Paleontology, University of Michigan, 23, 305-341. LAWRENCE, J. 1987. A Functional Biology of Echinoderms, The John Hopkins University Press, Baltimore, MD. LE GALL, B., BILLA, M., BOS, P., GARREAU, J., LE GOFFIC, M. & PARADIS, S. 1992. Notice explicative, carte gkologique de la France (1/50 000), feuille Gourin (311). Editions du Bureau de Recherches G6ologiques et Mini6res, Orl6ans, 81. LEFEBVRE, B. 2000a. Les 6chinodermes stylophores du Massif armoricain. Bulletin de la Soci~t~ des Sciences Naturelles de l'Ouest de la France, 22, 101-122. LEFEBVRE, B. 2000b. A new mitrate (Echinodermata, Stylophora) from the Tremadoc of Shropshire (England) and the origin of the Mitrocystitida. Journal of Paleontology, 74, 890-905. LEFEBVRE, B. 2003a. Functional morphology of stylophoran echinoderms. Palaeontology, 46, 511555. LEFEBVRE, B. 2003b. Stephen J. Gould, les mitrates et les monstres. Comptes Rendus Palevol, 2, 509-522. LEFEBVRE, B. 2006. Early Palaeozoic palaeobiogeography and palaeoecology of stylophoran echinoderms. Palaeogeography, Palaeoclimatology, Palaeoecology, doi: 10.1016/j.palaeo.2006.02.021. MATSUMOTO, H. 1915. A new classification of the Ophiuroidea; with descriptions of new genera and species. Proceedings of the Academy of Natural Sciences of Philadelphia, 67, 43-92. MEYER, D. L. 1997. Implications of research on living stalked crinoids for paleobiology. In: WATERS,J. A. & MAPLES,C. G. (eds) Geobiology of Echinoderms. Paleontological Society Papers, 3, 31-43. MIKULAS, R., PETR, V. & PROKOP, R. 1995. The first occurrence of a 'brittlestar bed' (Echinodermata, Ophiuroidea) in Bohemia (Ordovician, Czech Republic). Vestnik Cesk~ho geologick~ho ~stavu, 70, 17-24. MILLER, S. A. 1889. North American Geology and Palaeontology for the Use of Amateurs, Students and Scientists, Published by the author, 1-664. MORTENSEN, T. H. 1938. Ober die stratigraphische verwendbarkeit der mikropischen EchinodermenReste. Senckenbergiana, 20, 342-345. MORZADEC, P. & UBAGHS, G. 1969. Ophiurina armoricana n. sp., ophiuroide nouveau du D6vonien inf6rieur de la Bretagne. Annales de Palkontologie, 55, 179-186. NICHOLS, D. 1972. The water-vascular system in living and fossil echinoderms. Palaeontology, 15, 519-538.
86
A.W. HUNTER ETAL.
PARIS, F. & ROBARDET, M. 1977. Pal6og6ographie et relations ib&o-armoricaines au Pal6ozoique ant6Carbonif&e. Bulletin de la Sociktk gkologique de France, 19, 1121-1126. PARSLEY,R. L. 1988. Feeding and respiratory strategies in Stylophora. In: PAUL, C. R. C. & SMITH, A. B. (eds) Echinoderms Phylogeny and Evolutionary Biology. Clarendon Press, Oxford, 347-361. PARSLEY, R. L. 1994. Mitrocystitid functional morphology, evolution and their relationships with other primitive echinoderm classes. In: DAVID, B., GUILLE, A., FERAL, J.-P. & Roux, M. (eds) Echinoderms Through Time. Balkema, Rotterdam, 167172. PETR, V. 1989. Revision of morphology and ecology of Bohemura jahni JAEKEL, 1903 (Ophiuroidea, Protasteridae) from Bohemian Middle Ordovician. Sbornik Narodni Muzeum v Praze, 45, 1-20. PRUVOST, P. & LE MAITRE, D. 1949. Carte gkologique de la France ~ 1/80 000,feuille Chdteaulin, 26me 6dn. Institut G6ographique National, Paris. ROMANO, M. & HENRY, J. L. 1982. The trilobite genus Eoharpes from the Ordovician of Brittany and Portugal. Palaeontology, 25, 623-633. RUTA, M. & BARTELS, C. 1998. A redescription of the anomalocystitid mitrate Rhenocystis latipedunculata from the Lower Devonian of Germany. Palaeontology, 41, 771-806. SCHAFER,W. 1972. Ecology and Paleoecology of Marine Environments. University of Chicago Press, Chicago, IL. SHACKLETON, J. 2005. Skeletal homologies, phylogeny and classification of earliest asterozoan echinoderms. Journal of Systematic Palaeontology, 3, 29-114. SIMMS, M. J., GALEA. S., GILLILAND,P., ROSE, E. P. F. & SEVASTOPULO,G. D. 1993, Chapter 25, Echinodermata. In: BENTON, M. J. (ed.) The Fossil Record 2. Chapman & Hall, London, 491-528.
SPENCER, W. K. 1950. Asterozoa and the study of Palaeozoic faunas. Geological Magazine, 87, 393408.
SPENCER, W. K. 1951. Early Palaeozoic starfish. Philosophical Transactions of the Royal Society of London, Series B, 235, 87-129. THORAL, M. 1935. Contribution ~ l'ktude palkontologique de rOrdovicien infkrieur de la Montagne Noire et rkvision sommaire de lafaune cambrienne de la Montagne Noire. Imprimerie de la Charit6, Montpellier. TROMELIN, G. DE & LEBESCONTE, P. 1876. Essai d'un catalogue raisonn6 des fossiles siluriens des d6partements de Maine-et-Loire, de la Loireinf&ieure et du Morbihan, avec des observations sur les terrains pal6ozoiques de l'Ouest de la France. Comptes Rendus de l'Association franDaise pour l'Avancement des Sciences, 4kme session, Nantes, 1875, 601-661. UBAGHS, G. 1968. Stylophora. In: MOORE, R. C. (ed.) Treatise on Invertebrate Paleontology, Part S, Echinodermata 1, 2. Geological Society of America, Boulder, CO and University of Kansas Press, Lawrence, KS, 495-565. UBAGHS, G. & ROBISON, R. A. 1988. Homalozoan echinoderms of the Wheeler Formation (Middle Cambrian) of Western Utah. University of Kansas Paleontological Contributions, 120, 1-18. WARNER, G. F. 1971. On the ecology of a dense bed of the brittle-star Ophiotrix fragilis. Journal of the Marine Biological Association of the United Kingdom, 51,267-282. YOUNG, T. P. 1990. Ordovician sedimentary facies and faunas of southwest Europe: palaeogeographic and tectonic implications. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Palaeozoic Palaeogeography and Biogeography. Geological Society Memoir, 12, 421M30.
Micritic fabrics define sharp margins of Wenlock patch reefs (middle Silurian) in Gotland and England S T E V E K E R S H A W 1, Y U E LI 2 & L I G U O 3
1Department of Geography and Earth Sciences & Institute for the Environment, Halsbury Building, Brunel University, Uxbridge, Middlesex UB8 3PH, UK (e-mail." stephen, kershaw@brunel, ac. uk) 2Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 East Beijing Road, Nanjing, 210008, China (e-mail."
[email protected]) 3CASP, Department of Earth Sciences, University of Cambridge, West Building, 181a Huntington Road, Cambridge CB3 0DH, UK (e-mail."
[email protected]) Silurian reefs are well known to comprise frame-building corals, stromatoporoids and algae, but also a range of calcimicrobial components and micritic sediments of possible microbial origin. The margins of Wenlock patch reefs tend to have diffuse edges that grade into the surrounding bedded facies because of talus shed from the reefs. However, portions of patch reefs show sharp-bounded reef margins, with bedded limestones terminating abruptly against the reef edge. Examples of sharp boundaries where the reef comprises only carbonate mudstone-wackestone with poorly-defined gross fabric, and containing no metazoan framework, have been found in Wenlock patch reefs the UK and Gotland. Although in some cases the micrite may demonstrate a peloidal structure, in others there is no clear structure, broadly fitting the aphanitic (structureless) type of fabric found in leiolites (suspected microbial facies that show no structure). The fabrics are interpreted to have been formed by microbial mediation of micrite precipitation as part of reef construction, and are therefore automicrites. In all cases the sharp reef edges indicate coherence of the micritic fabric, interpreted as a lithified wall against which bedded limestones were deposited. This arrangement supports published interpretations of pronounced topography of Wenlock patch reefs, and shows the presence of steep, vertical and, possibly, overhanging reef margins, formed prior to bedded sediment accumulation. Thus, there is likely to have been a time interval between reef formation and deposition of bedded sediments, possibly caused by reef upward growth in transgressive settings, followed by catch-up of bedded limestone deposition. Abstract:
Silurian patch reefs occurring in argillaceous bedded limestones in Wenlock rocks were described by Scoffin (1971, 1972) to have formed as low-profile structures in relatively quiet waters, and produced some talus that is distributed only a short distance from the reef. Although the reefs are clearly identifiable as unbedded patches in well-bedded limestones (Fig. 1), margins of the reefs are mostly not sharply defined (Fig. 2a); debris of reef frames shed onto the surrounding sea floor, and in some cases decreasing grain size of crinoidal debris away from the reef leaves the reef boundaries somewhat diffuse in most cases. At many sites reef margins are partially interbedded with off-reef bedded limestones. However, in some portions of reefs, the boundaries are so well defined that the reef edge forms an abrupt border, against which bedded sediments terminate sharply (Fig. l b, c). Some small patch reefs are comprised of structure that forms a sharp From: LkLVARO,J.
boundary across the entire exposed margin. In some cases the appearance of reefs in vertical section is highlighted by differential compaction of the surrounding bedded lithofacies (Fig. lb). Descriptions of Wenlock reefs (e.g. Scoffin 1971, 1972; Abbott 1976; Watts 1988; Riding & Watts 1991; Ratcliffe & Thomas 1999; Watts & Riding 2000) have not highlighted the potential significance of sharp boundaries in reef development; this paper describes the essential features of sharp reef margins and presents evidence that microbial processes were responsible, using three examples: Coates and Lea Quarries, Wenlock Edge, U K (Figs la, c & 2); Hobbs Ridge, Longhope, Gloucestershire (Fig. lb), UK; and Millingsklint (Ireviken 2 site), Gotland, Sweden (Figs 4 & 5). Thus, the principal aim is to draw attention to the potential importance of such reef margins in palaeoenvironmental interpretations of reefs' settings, but this work does not attempt to comprehensively document the distribution of such features.
J., ARETZ, M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN, E. (eds) 2007.
PalaeozoicReefsandBioaccumulations:ClimaticandEvolutionaryControls.Geological Society, London, Special Publications, 275, 87-94.0305-8719107l$15.00 9 The Geological Society of London.
88
S. KERSHAW E T A L .
Fig. 1. Geometry of Wenlock patch reefs, UK. (A) Patch reef in Coates Quarry, showing unbedded structure in bedded limestones; the reef, although clearly identifiable, has a somewhat diffuse margin in some parts and a sharp margin in others. (B) Sharp margin of reef in Hobbs Quarry, near Longhope, Gloucestershire; note onlapping bedded limestone. See text for locality information. (C) Sharp margin of Patch reef in Lea Quarry.
General character of Silurian patch reefs Wenlock age patch reefs in England are small, on average 12 m in diameter and 4.5 m thickness (Fig. 1), forming a string of patch reefs, possibly a barrier structure. Reefs are limited to the Much Wenlock Limestone Formation (20-30 m thick at Wenlock Edge) and represent a narrow time-window of low clastic supply during the Wenlock highstand, the formation being underlain by shales and overlain by silts (Scoffin 1971, 1972). Ratcliffe & Thomas (1999) described two types of reefs: (A) those dominated by skeletal metazoans are the most abundant, represented here by those at Lea and Coates quarries; and (B) reefs dominated by intepreted microbial reef builders are rarer, represented here by those at Hobbs Ridge. Clay seams commonly pass through reef structures, and
interrupted growth, followed by recolonization, so that individual bioherms consist of a stack of lenses (Scoffin 1972). In contrast, Wenlock patch reefs of the H6gklint Formation on Gotland are mostly much larger structures, several hundreds of metres across (Riding 1981); these reefs are associated with a variety of non-reef facies of shallow-marine sediment. The bioherms are mostly developed on grainstone lenses; in many places they contain moderate-diversity skeletal frameworks of laminar-domical morphologies, with biotic zonation from corals upwards to stromatoporoids. In detail, H6gklint reefs have a five-stage zonation, terminating with shallow-water algal facies (Watts 1988; Riding & Watts 1991; Watts & Riding 2000). There are diverse associated shelly faunas, dominated by crinoids.
MICRITIC FABRICS OF WENLOCK PATCH REEFS
89
gross morphology often found constructed of leiolites. Hrgklint patch reefs on Gotland are more variable because of prominent vertical zonation from corals to stromatoporoids to algal-microbial communities. Small patch reefs at Hobbs Ridge are skeleton-poor micritic mounds surrounded by bedded limestones; the reef bases are not exposed, but no zonation of biota is apparent. The micritic structure of these reefs comprises patches of leiolite with sharp edges within the reefs and at reef edges. Ratcliffe & Thomas (1999, p. 197) noted that the lowest visible parts of the Hobbs Ridge reefs consist of crinoidal grainstone and that the reefs are steep sided.
Description of reef margins The sites selected to illustrate the micritic reef margin facies are outlined below. In none of the three sites were the sharp margins studied found to be loose blocks eroded from reefs; thus the true edge of the in situ reef mass was examined.
Lea and Coates quarries, Wenlock Edge Fig. 2. (A) Margin of a patch reef (Coates Quarry), showing diffuse margin. Field of view approximately 10 wide. (B) Margin of another patch reef (Coates Quarry) showing margin interfingering with adjacent bedded sediment. Field of view approximately 2 m wide. Microfacies of the U K reefs are relatively simple; laminar and domical coralstromatoporoid frames are subordinate to micritic sediment, which may have abundant skeletal debris and some microbial binding. Primary cavities contain micrite (at least two generations can be recognized), sparite and microbial growth of calcimicrobes dominated by genera such as Girvanella, Rothpletzella and Wetheredella. In places a significant calcimicrobial component is present. Of particular concern for this paper, are portions of bioherms poor in skeletal debris that appear composed of amalgamated peloids in some places; whilst in other places the fabric comprises structureless micritic sediment consistent with the aphanitic mesofabric micritic sediment of leiolites (Fliigel 2004, pp. 372-373) (see Figs 2 & 3). The term leiolite was introduced by Braga et al. (1995, p. 347) and described by Riding (2000, p. 195), from Greek leios (uniform or smooth) and lithos (stone) as a structureless macrofabric without lamination, clots or dendritic fabrics. However, the leiolite fabrics in the Wenlock material described here lack identifiable dome-shaped
Small patch reefs have been extensively described by Scoffin (1971, 1972) and are currently partly exposed in several quarries on Wenlock Edge, including Coates Quarry (Fig. l a) and Lea Quarry (which are now one large quarry), where the sharp margin of one reef is well exposed on the west quarry wall (Fig. lc). Thin sections of the latter reef (Fig. 3a, b) show the micritic construction of reef fabric with no framebuilding skeletal material; in contrast, adjacent bedded limestones are packstones-wackestones, rich in comminuted marine faunas of crinoids, brachiopods, corals, trilobites and others in micritic limestone matrix (Fig. 3c). Note that the micritic reef fabrics illustrated here do not show the laminated character of stromatolitic material, such as that figured by Scoffin (1971, p. 201). Although Scoffin (1971, p. 209) demonstrated that the reef frames contain a lot of micrite and that binding was by microbial carbonates (Scoffin 1971, p. 216), no direct mention was made of the sharp micrite-bounded margins of the reefs indicated here. In general, the micritic components that make up the sharp margins and distinct patches are consistent with the term automicrite, defined as in-place precipitation of micrite either by organic or inorganic processes (Neuweiler et al. 1999). The origin of other micrite sediment deposited around the bound areas is less certain, and could be allomicrite (allochthonous sediment carried into the reef area from surrounding facies). Whether or not this material is of polymud type (mixed origins) is
90
S. KERSHAW ET AL.
Fig. 3. Reef margins in Lea Quarry. (A) Mesofabric of reef margin showing inhomogenous micritic structure, consistent with leiolite fabric; Scale bar is 25 mm. (B) Microfabric enlargement from A, showing a compacted peloidal microstructure that exists in part of the area of A; scale bar is 2 mm. (C) Skeletal packstone-wackestone fabric of bedded sediment adjacent to the reef margin; scale bar is 4 mm.
not easily determined, but the high abundance of micrite in the carbonate sediments of both the reef and associated bedded limestones in these shallow-water facies suggests a significant component of local carbonate production, in keeping with the generally accepted views that carbonate sediment is generally formed within a basin of deposition, rather than being transported in (Calner et al. 2004).
Hobbs Ridge, Longhope, Gloucestershire Reefs are well exosed in a quarry on Hobbs Ridge (grid reference SO-695195), located on the south side of May Hill, Gloucestershire, as part of the May Hill inlier. Late Wenlock patch reefs are sharply defined in the associated bedded limestones (Fig. lb), and no blocks of reef material were found in the surrounding beds. The reefs are constructed of micrite with limited amounts of calcimicrobes, but lack a metazoan skeletal framework and may be better described as mud mounds rather than patch reefs. Bedded sediments onlap the mound margins at steep angles that were presumably enhanced by compaction (Fig. lb).
Ireviken 2, Gotland The locality is also known as Millingsklint (see Riding & Watts 1991 for locality details) and is a typical H6gklint patch reef (Fig. 4). Polished blocks (Fig. 4d) and acetate peels (Fig. 5) show the reef structure to be inhomogenously micritic; it contains some Girvanella, but otherwise the mesofabric is essentially leiolite (see also Flfigel 2004). In Figure 5, the leiolite structure forms a framework infilled with deposited micritic sediment. In contrast to the reef, the adjacent bedded limestone of the interreef beds (called the Irevik Member by Riding & Watts 1991) is wackestone of comminuted marine shell debris, sharply terminating against the reef margin (Fig. 4b, c).
Discussion
Origin of peloidal and leiolitic micritic sediments Reef margins composed of micritic fabrics are here interpreted as cemented into solid walls,
MICRITIC FABRICS OF WENLOCK PATCH REEFS
91
Fig. 4. Reef margin of H6gklint patch reef, Ireviken 2, Gotland. (A) & (B) Field views of lower portion of reef showing sharp contact between reef and bedded limestones. (C) Polished sample of wackestone fabric of bedded limestone adjacent to reef. (D) Polished sample of reef fabric at reef margin.
as automicrite. A microbial formation of such fabrics was reviewed by Flfigel (2004) and, although there is no proof of microbial mediation in the fabrics studied in this paper, we consider a microbial control to be the most probable reason for the fabrics. Note that the inhomogenous character of micritic fabrics illustrated in Figures 3-5 is not owing to bioturbation; they lack the typical diffuse concentric patterns of bioclastic material that accompany burrows in soft carbonate mudstone fabrics in these Silurian sediments. The best evidence that the sharp reef margins are synsedimentary is that they show a sharp facies
change. If instead, preferential cementation created the sharp margins diagenetically, the abundance of skeletal remains would be expected to be the same within the reef margin and the adjacent bedded limestone. Also, remains of bedding may be expected if the beds originally passed into the reef margin. However, these features are not seen, and the best explanation is an original sharp facies boundary.
Implications of lithified reef margins The formation of lithified micritic sediment as sharply defined reef margins, against which
92
S. KERSHAW E T AL.
Fig. 5. Acetate peel of Figure 4D, showing inhomogenous micritic texture. The bands of lighter micrite are automicrites of leiolite fabric, forming a lithified 'frame' of sediment; darker infilling of micrites of possible polymud composition is deposited in the spaces in the leiolite. Scale bar is 50 mm.
bedded sediment was deposited, suggest that reefs had prominent relief. Watts & Riding (2000) recognized that reef blocks up to 15 m high were found, lying on their sides in bedded sediment adjacent to reefs, and interpreted the H6gklint patch reefs to have had topography of at least 15 m as they grew. Scoffin (1971) considered Wenlock patch reefs to have had rounded tops, and gently sloping margins, but not the steep walls of a few metres height that are interpreted for the reef margins studied here. Although we agree that Wenlock patch reefs did have dome-shaped upper surfaces as described by Scoffin (1971), the steep walls envisaged here are an additional feature in some reefs and some parts of reefs. In all three examples described, reef tops have onlapping bedded limestones laid down on the sloping upper surface (e.g. Fig. la, d); however, all three also have bed-parallel sediments terminating against the reef sides. Interpretations of the growth rates of reefs and accumulation of the adjacent bedded limestones have suggested that the reefs kept pace
with sedimentation, and that the reefs did not have a topography of more than a few metres (e.g. Scoffin 1971; Jaanusson et al. 1979, p. 17). A characteristic of UK Wenlock reefs is that bedded sediments interfinger with reef facies, which may be interpreted as caused by variations, through time, of relative rate of reef growth compared to bedded sediment accumulation (Fig. 2b). Such intergrowth relationships between reef and bedded sediment demonstrate that the height of an individual reef top above surrounding sediment is likely to have varied throughout its growth, with periods when the reef top developed up to several metres height above the surrounding sediment, but then partly covered by sediment depostion at other times. Note, however, that much of the outcrop of reefs is in inaccessible quarry faces, preventing determination of the distribution of the micritic steep margins that suggest an increased topography. Nevertheless, more information is available from Gotland, considered next. For the H6gklint reefs, the excellent Vattenfallet section (Jaanusson et al. 1979) cuts through inter-reef beds. Jaanusson et al. (1979) showed that the lower parts of the H6gklint Formation, coeval with the lower parts of reefs, comprise richly bioclastic wackestones and packstones; a valuable review of literature concluded that the bioclastic debris was not derived from the nearby H6gklint reefs (Jaanusson et al. 1979, pp. 17-18). Upsection, the proportion of bioclastic debris in the Vattenfallet sequence decreases and is very sparse, coeval with the central portions of H6gklint reefs. Towards the top of the Vattenfallet section, bioclastic debris increases again, and this must be coeval with the upper portion of H6gklint reefs, but below the biostromal top phase identified by Watts & Riding (2000). That phase is found at higher levels in the sequence near Vattenfallet but above the topmost exposed beds at Vattenfallet. The upwards decrease and then increase of bioclastic debris is consistent with a process of reef growth during transgression. In initial transgression, rich faunas may be viewed as having exploited the shallow sea floor, but deepening water may be expected to have led to decreased colonization. Catch-up sedimentation could then have been responsible for an increase of bioclastic debris, and also explains the shallowing character of biotic zonation in the H6gklint reefs (Watts & Riding 2000) as catch-up reef growth. Thus, rather than the reefs growing to just keep ahead of depostion around (and presumably on) them, in many cases the reefs apparently formed in a setting of reduced sedimentation; bedded sediments were later laid
MICRITIC FABRICS OF WENLOCK PATCH REEFS on and around the structures. Almost all H6gklint patch reefs are underlain by crinoidal shoals, which indicate shallowing above the top of the underlying Upper Visby Formation; thus a transgression to generate accommodation space to form the reefs is consistent with the facies relationships. The sharp reef margins described here therefore corroborate at least 15 m of topography indicated for H6gklint patch reefs, and further implies that H6gklint patch reefs grew in transgressive settings. On Wenlock edge, there is evidence of a decrease and then increase of bioclastic debris in bedded sediments coeval with Wenlock reefs. Bassett (1989) outlined the key characters of sediments of the Longville Member and the succeeding Edgton Member of the Much Wenlock Limestone Formation in the so-called 'off-reef tract', located SW of the main reef outcrops in Wenlock Edge. The Longville Member comprises tabular-platy bedded wackestones with fine-grained bioclasts of principally pelmatozoan, bryozoan and brachiopod faunas. The lower half of the overlying Edgton Member is poorly fossiliferous nodular limestones, succeeded by coarse-grained pelmatozoan limestones, equivalent to the shallowing facies of the upper part of the Much Wenlock Limestone Formation. In contrast, the bedded limestones of the 'reef tract' area are more variable; however, there is a general increase in coarser-grained debris in pelmatozoan and coral limestones at the top of the sequence, overlying finer-grained limestones with sparser bioclasts. Consequently, therefore, there are similarities with the progression of the Vattenfallet section, suggesting water deepening associated with the timing of reef growth. Nevertheless, the review by Bassett (1989) indicates upwards shallowing through the portion of section containing reefs, and it is less clear that transgression was related to reef growth in this region. The Wenlock Edge reefs also lack large reef blocks that would provide the evidence of topography described for H6gklint reefs on Goland. Nevertheless, the formation of lithified steep reef edges against which bedded sediments were deposited is consistent with transgression for part of the time that the reefs grew. Sufficient accommodation space is needed to create the reef thicknesses observed in these outcrops, but whether that space was entirely provided by a static water depth that the reef system and associated sediments filled, or whether transgression partly provided space for reef and bedded sediment aggradation, is not clear at Wenlock Edge. Despite such uncertainty, interfingering of reef and bedded sediment indicates varying topographic height of any reef
93
top as the reef grew. Ratcliffe & Thomas (1999) argued that the Wenlock reefs of Wenlock Edge are diachronous, and would suggest a more complex pattern of relative sea-level change and reef growth if transgressive events influenced reef growth. At Hobbs Ridge, insufficient exposure is available to permit an attempt to relate the bioherm growth to relative sea-level change; an unknown part of the reefs is hidden beneath the quarry floor.
Conclusion Sharp margins of Wenlock patch reefs in England and Gotland have not been previously highlighted in research on those deposits, and provides further evidence that reefs had significant topography during growth. Sharp margins in the studied material are made of micrite, interpreted as automicrites of possible microbial origin. In the H6gklint reefs on Gotland, clear differences between structure and composition of bedded sediment and reef fabric, and the vertical changes in both reefs and bedded limestones, may be used to interpret that at least parts of reef growth took place in transgressive settings, broadly consistent with the interpreted sea-level change for the middle Silurian (Brunton et al. 1998). On Wenlock Edge, transgression during reef growth is less easy to interpret. We are grateful to the administrators of the Allekvia field station on Gotland for accommodation during many visits to Gotland since 1975, when the observations of sharp reef margins were first made. We are indebted to M. Ford for access to Lea Quarry for photography. S. Kershaw also thanks Brunel University for facilities. Fieldwork on Gotland in August 2005 was much enhanced by discussions with C. Brett, B. Cramer, P. McLaughlin and C. Forster, to whom the senior author is grateful for companionship. Finally, A. Thomas and R. Wood are thanked for their very helpful reviews of an earlier version of this paper.
References ABBOTT, B. M. 1976. Origin and evolution of bioherms
in Wenlock Limestone (Silurian) of Shropshire, England. AAPG Bulletin, 60, 2117-2127. BASSETT, M. G. 1989. The Wenlock Series in the Wenlock area. In." HOLLAND, C. H. & BASSETT, M. G. (eds) A Global Standard for the Silurian System.
National Museum of Wales Geological Series, 9, 51-72. BRAGA, J. C., MARTIN, J. M. & RIDING, R. 1995.
Controls on microbial dome fabric development along a carbonate-siliciclastic shelf-basin transect, Miocene, S.E. Spain. Palaios, 10, 347-361.
94
S. KERSHAW ETAL.
BRUNTON, F. R., SMITH, L., DIXON, O. A., COPPER, P., KERSHAW, S. & NESTOR, H. 1998. Silurian reef episodes, changing seascapes and palaeobiogeography. In: LANDING, E. & JOHNSON, M. (eds) Silurian Cycles," Linkages of Dynamic Strati-
graphy with Atmosphere, Oceanic and Tectonic Changes. New York State Museum Bulletin, 491, 265-282. CALNER, M., JEPPSSON, L. & MUNNECKE, A. 2004. The Silurian of Gotland - Part 1: Review of the stratigraphic framework, event stratigraphy, and stable carbon and oxygen isotope development. Erlanger geologische Abhandlungen, Sb5, 113-131. FLOGEL, E. 2004. Microfacies of Carbonate Rocks," Analysis, Interpretation and Application. Springer, Berlin. JAANUSSON, V., LAUFELD, S. & SKOGLUND, R. 1979. Lower Wenlock Faunal and Floral dynamics Vattenfallet section, Gotland. Sveriges Geologiska Unders6kning, Serie C, 762, Uppsala. NEUWEILER, F., GAUTRET, P., THIEL, V., LANGE, R., MICHAELIS, W. & REITNER, J. 1999. Petrology of Lower Cretaceous carbonate mud mounds (Albian, N-Spain): insights into organomineralic deposits of the geological record. Sedimentology, 46, 837-859. RATCLIFFE, .K. & THOMAS, A. 1999. Carbonate depositional environments in the late Wenlock
of England and Wales. Geological Magazine, 136, 189-204.
RIDING, R. 1981. Composition, structure and environmental setting of Silurian bioherms and biostromes in northern Europe. In: TOOMEY, D. F. (ed.) European Fossil Reef Models. Society of Economic Paleontologists and Mineralogist Special Publications, 30, 41-83. RIDING, R. 2000. Microbial carbonates: the geological record of calcified bacterial-algal mats and biofilms. Sedimentology, 47, (Suppl. 1), 179-214. RIDING, R. & WATTS, N. R. 1991. The lower Wenlock reef sequence of Gotland: facies and lithostratigraphy. Geologiska F6reningens i Stockholm F6rhandlingar, 113, 343-372. Sr T. P. 1971. The conditions of growth of the Wenlock reefs of Shropshire (England). Sedimentology, 17, 173-219. SCOFFIN, T. P. 1972. Cavities in the reefs of the Wenlock Limestone (mid-Silurian) of Shropshire, England. Geologisches Rundschau, 61,565-578. WATTS, N. R. 1988. Carbonate particulate sedimentation and facies within the Lower Silurian H6gklint patch reefs of Gotland, Sweden. Sedimentary Geology, 59, 93-113. WATTS, N. R. & RIDING, R. 2000. Growth of rigid high-relief patch reefs, Mid-Silurian, Gotland, Sweden. Sedimentology, 47, 979-994.
Siluro-Devonian Alpine reefs and pavements B E R N H A R D H U B M A N N 1& T H O M A S S U T T N E R 1'2
~University of Graz, Institute of Earth Sciences (Geology and Palaeontology), Heinrichstrasse 26, A-8010 Graz, Austria (e-maib thomas, suttner@uni-graz, at) 2Commission for the Palaeontological and Stratigraphical Research of Austria Abstract: Palaeozoic sediments of Austria are separated by the Periadriatic Fault into Eastern Alpine (Upper, Middle and Lower Austroalpine) and Southern Alpine units. We herein present six case studies showing up the different development of shallow-marine communities with special regard to carbonate factories and shell pavements occurring in both regions during the Siluro-Devonian time span. Upper Silurian-Upper Devonian deposits of the Eastern Alps comprise accumulations of serpulid tubes (Southern Burgenland) and Septatrypapavements, Amphipora mounds, coral-stromatoporoid-biostromes and Stachyodes-auloporoid beds regarded as pioneer reef communities (Graz Palaeozoic), respectively. Lower Silurian strata of the Southern Alps consist of pelagic sediments persisting to the Upper Silurian and therefore differ from contemporaneous successions in the Eastern Alps. Intercalated in Ludlow orthocerid limestone beds Cardiola pavements appear (Carnic Alps). Within the Lower Devonian sequence, mounds were built by baffling calcareous algae and tabulozoan communities. Coral-stromatoporoid patch reefs occur during the Pragian, Givetian and Frasnian stages.
Investigation of Silurian-Upper Devonian reefs and reef-related successions within Austria's borders has a long tradition. A survey of these developments is compiled in a stratigraphic scheme representing all non-metamorphic units of Austria. Generally all Palaeozoic strata suffered differing post-depositional destruction, i.e. tectonic splitting, dolomitization, metamorphic overprint, etc., during Variscan and Alpidic orogenesis thus complicating or limiting to some respect their investigation. From Southern Burgenland, the Graz Palaeozoic, the Carnic Alps and the Karawanken Mountains 'reefs', which cover a time-segment of some 55 Ma, are presented in overview.
Stratigraphical framework of dismembered units In Austria weakly and unmetamorphosed Palaeozoic successions are irregularly distributed (Fig. 1). Their incorporation as dismembered units into the Alpine Nappe System is rather complex, and their primary geographic positions and mutual bio(geo)graphic relations are only poorly understood. Generally two major regions of Palaeozoic developments separated by the Periadriatic Fault, the most prominent alpine fault system, can be distinguished: the Eastern Alpine Variscan sequences (Figs 2 & 3) (Upper
Austroalpine Unit: i.e. the Greywacke Zone, the Gurktal Nappe, the Graz Palaeozoic and some isolated outcrops in south Styria and Burgenland); and the Southern Alpine sequences (Fig. 4) (i.e. the Carnic Alps and the Karawanken Mountains). Developmental differences of both areas are of different depositional and organismic characters resulting from independent histories of subsidence rates, amounts of volcanic activity and climatic changes.
Eastern Alpine Units - exemplified on the succession of the Graz Palaeozoic The Graz Palaeozoic comprises an outcrop area of approximately 1250 km 2 resting on metamorphic basement. In the northern and western part it overthrusted the Middle Austroalpine (Gleinalm Crystalline), in the eastern part the Lower Austroalpine unit (Raabalpen Complex). In its western sector the Palaeozoic succession is unconformably overlain by the Upper Cretaceous Kainach Gosau and in the south it is onlapped by Neogene sediments of the Styrian Basin. The Graz Palaeozoic represents a pile of nappes, consisting of different depositional developments. Considering lithological similarities, tectonic position and metamorphic superimposition, the following nappe groups are discernible: the Basal Nappe System (Upper
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. PalaeozoicReefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 95-107. 0305-8719/07/$15.00 9 The Geological Society of London.
96
B. HUBMANN & T. SUTTNER
Fig. 1. Palaeozoic sediments of Austria. Abbreviations: carb. platform, carbonate platform; rew. Devonian Neog. boulders, reworked Devonian as Neogene boulders; allochth, accum., allochthonous accumulation.
Silurian-Lower Devonian) comprises the Sch6ckel Nappe and Anger Crystalline Complex. The Sch6ckel Nappe is made up of pre-Devonian rocks (Passail Group, Taschen Formation) and the Devonian Peggau Group. Generally, volcano-clastic sediments dominate the late Silurian-early Devonian, and carbonates the middle Devonian time span. The Intermediate Nappe System (Lower Silurian-Upper Devonian) includes the Laufnitzdorf Nappe and the Kalkschiefer Nappe (Lower-Upper Devonian). Both nappe groups occur in different structural levels. The former development contains pelagic limestones, shales and volcano-clastic rocks, and the latter limestone and siliciclastic deposits. The Upper Nappe System (Upper SilurianUpper Carboniferous) comprises the Rannachand Hochlantsch nappes. Both have a similar depositional development in common, especially in the Emsian-Givetian. Successions of the Rannach Nappe are composed of volcano-clastic
rocks (Silurian-early Devonian; Reinerspitz Group), fossiliferous siliciclastic and carbonate sediments (early-middle Devonian; Rannach Group) of a littoral environment followed by the pelagic Forstkogel Group (late GivetianBashkirian) and the shallow-marine Dult Group (Serpukovian, Bashkirian). According to a palaeogeographical interpretation of the entire Palaeozoic succession, the formations of the Rannach- and Hochlantsch nappes are interpreted as proximal nearshore deposits, while the 'Laufnitzdorf facies' represents distal offshore settings. Successions of the Sch6ckel Nappe occupy an intermediate position (Hubmann 1993; Hubmann & Messner 2005).
Southern Alpine Units - exemplified on the succession o f the Carnic Alps Silurian deposits range from shallow-water bioclastic limestones to nautiloid-bearing limestones, interbedded shales and limestones to
ALPINE REEFS AND PAVEMENTS black graptolite-bearing shales and cherts with an overall thickness not exceeding 60 m. The Silurian transgression (Loydell 1998) in the Carnic Alps started at the very base of the Llandovery (Akidograptus acuminatus Zone). Owing to a disconformity separating the Ordovician and the Silurian in many places a varying pile of sediments is locally missing, which corresponds to several conodont zones of Llandoverian-Ludlowian age. Even uppermost Pridolian strata may disconformably rest upon Upper Ordovician sediments. The shallow-marine Rauchkofel Boden section is regarded as one of the most fossiliferous Upper Silurian sections of the Carnic Alps. The contact with the underlying neritic Kok Formation is marked by concentrations of iron crusts. Development of scattered small-sized calcimicrobial structures (Histon 1999, and references therein) is also evident in the lower levels of the sequence. In the Wenlock-Ludlow transition thinly developed cyclic micritic limestone beds of bioclastic accumulations are separated by stylolites and horizons with iron-manganese crusts that may mark the end of depositional regimes. Concentrations of apparently juvenile and equidimensional articulate brachiopods, nautiloids and gastropods alternate with the dominantly nautiloid beds (the classic Orthoceras limestone) in the Lower Ludlow demonstrating the changing energy and oxygen levels of the formation, while the preservation and orientation of the fauna indicate many accumulated levels with intermittent changes in sea level particularly towards the top of the sequence. The Alticola Limestone, Pridoli in age, is a fine grey micritic limestone with abundant micritized bioclasts, frequent stylolites and an abundant, although less diverse, nautiloid fauna throughout the formation. The associated shallow-water fauna is similiar to the Kok Formation except for the presence of a few rugose corals near the top of the Silurian. A Scyphocrinites bed bearing complete specimens caps the formation and marks the Silurian-Devonian boundary and the shallowest level of the sequence (Ferretti et al. 1999). Additional to the shallow-marine development of the Rauchkofel Boden section the 'P16cken facies' represents a shallow to moderately deep marine carbonate series. Deposits were periodically affected by storm currents, with intervals of reduced depositional rates and non-sedimentation in an overall transgressive sequence (Flfigel et al. 1977). The pelagic Kok Formation consists of a transgressive carbonate series with alternating black shales and dark grey-slightly red micritic lenticular limestone turning into brown-red ferruginous limestone
97
with abundant nautiloids during the WenlockLower Ludlow. The overlying Cardiola Beds of Ludlow age are represented by a thinly developed dark limestone rich in cephalopods and bivalves indicating a slightly deeper offshore environment with probable contemporary non-deposition taking place. Albeit solitary rugose corals occur (Sch6nlaub pers. comm.) they are far-flunged distributed. A more stable pelagic environment is developed in the Alticola and Megaerella Limestone from the Upper Ludlow continuing into transgressive carbonate series of Pridolian age (Sch6nlaub 1997b). The boundary between the Silurian and Devonian is drawn based on graptolites and conodonts. Towards the Silurian-Devonian boundary the hydrodynamic regime was fairly dynamic and the sea level reached its lowermost magnitude at the base of the Rauchkofel Limestone (Lochkovian), which shows a distinct bryozoan fauna (Histon et al. 1999). A recent taphonomic study of the Silurian of the Cellon section has highlighted in more detail the faunal and environmental changes during this time interval (Priewalder 1997, 2000; Sch6nlaub & Histon 1999). The large oxygen isotope ratio excursion shown by Wenzel (1997) at the boundary may be explained by the more ventilated setting implied by the bryozoan fauna. The intermediate 'Findenig Facies' consists of interbedded black graptolitic shales, marls and blackish carbonates, which are locally underlain by a quartzose sandstone. The stagnant water graptolitic 'Bischofalm Facies' is represented by black siliceous shales, lydites and clayey alum shales. The evidence from the Silurian indicates faunal affinities, for example conodonts, trilobites, brachiopods, molluscs, chitinozoa and acritarchs with Baltica and Avalonia as opposed to loose relationships with Africa and southern Europe. In addition, first occurrences of rugose and tabulate corals, ooids and stromatolites indicate a moderate climate. An overall island setting may be inferred from a generally condensed and reduced sedimentary pattern without significant clastic imput. These data suggest an ongoing drift towards lower latitudes and, consequently, a palaeolatitudinal position between 30~ and 40~ In the central Alps rifting-related basic volcanism underpins these inferred plate movements (Sch6nlaub & Histon 1999). The Devonian Period (Kreutzer et al. 1997, 2000) is characterized by abundant shelly fossils, varying carbonate thickness, reef development and interfingering facies ranging from near-shore sediments to carbonate buildups, lagoonal and slope deposits, condensed pelagic cephalopod limestones to deep oceanic off-shore shales.
98
B. HUBMANN & T. SUTTNER
The neritic Rauchkofel Limestone (Lochkovian) characterizes a transgressional sequence (up to 180 m thick), which corresponds to some 20 m of pelagic limestones (Boden Limestone). During the Pragian and Emsian stages the differences even increased. Within short distances of less than 10 km a strongly varying facies pattern developed indicating a progressive but not uniform deepening of the basin. It was filled with thick reef and near-reef organodetritic limestones including different intertidal lagoonal deposits of more than 1000m thickness. According to Kreutzer (1992a, b) in the Carnic Alps five NNEto SSW-directed facies belts developed in the Devonian Period. During later orogenic events these belts were strongly deformed, being distributed in different nappes and tectonic slices, which from top to base can be subdivided into the following units: 9 southern shallow-water facies of the CellonKellerwand Nappe; 9 transitional facies of Cellon Nappe; 9 pelagic limestone facies of Rauchkofel Nappe; 9 pelagic offshore basinal facies of Bischofalm Nappe; 9 northern shallow-water facies. The reef development ended in the late
Porites-seagrass communities. Modern counterparts are more likely to be found in moderate water depth of approximately 2 0 4 0 m (Riegl & Piller 2000). A second type of very remarkable bioaccumulation is represented by in situ buildups of shelly fossils that are ecologically not capable of creating vertical bioconstructions which might be classified as reefs. Owing to their morphology and growth habit such organisms also are not able to construct a 'rigid' skeletal frame nor can they overgrow themselves. Such features as over/ inter/growths, cementation may be observable in biostromes. Organisms with exoskeletons of globular and ovate shape that occur in great numbers may leave a kind of fossil pavement. Therefore, we will call them 'colonized pavements' herein. These colonized pavements are normally flat accumulations showing a low-relief constructed by invertebrates that are dense enough to partially obscure the carbonate host rock. The individual shape of the organisms is oval, rounded or platy forming a cobbled or paved surface. These pavements are formed by bivalves or more commonly during Palaeozoic times by brachiopods. Modern (Cenozoicrecent) counterparts are, for instance, rhodolith pavements (e.g. Martin et al. 2004).
rhenana Zone of the Upper Frasnian. At
Southern Burgenland (Eastern Alps)
the Frasnian-Famennian boundary the reefs drowned and a uniform pelagic environment developed that lasted across the DevonianCarboniferous boundary.
Locality: Kirchfidisch Stratigraphic unit: Blumau Formation Age: Upper Silurian (A proper age could not
Case studies General opinion is that the question 'what is a reef?.' is a philosophical one. There is a great range of points of view in defining recent reefs but it is nearly impossible for Palaeozoic localities within the Alps, where large-scale outcrops are limited, thus impeding or precluding 'architectural' studies, i.e. geometrical features or internal variations of biota, etc., of reefal bodies. Therefore, it is wise to restrict the term 'reef' for the Alpine region to define in situ accumulations of sessile organisms that produced considerable amounts of biogenic sediments. Most of these 'reefs' are characterized by a biostromal rather than a biohermal geometry. They may genetically be compared with modern coral carpets (Riegl & Piller 2000) as their high mud contents and negligible cavity spaces suggest low to moderate hydrodynamic conditions of the depositional environment. The highly diversified organic content constructing these biostromes argue against a simple comparison with recent very shallow
be determined owing to lack of well-preserved and reliable numbers of distinctive conodont elements within that sedimentary interval.) The quarry area comprises a several tens of metres thick sequence of low-grade metamorphic sediments. The main part consists of laminated dolomite~tolomitic limestone (Pollack 1962; Sch6nlaub 1994). An interesting feature is that primary lamination is preserved with no trace of reworking nor of any ichnofabrics, although serpulid tube debris appears in the matrix in about the middle part of the laminated dolomite sequence. Microstructural investigations of the skeletons in thin section have indicated that tubes of that part were - although less densly packed much more recrystallized than in the type area (Fig. 5a) about 40 m further to the west. We suggest that the type area was cut off its original position by subsequent faulting, and that all the serpulid occurrences which are now distributed in different levels once belonged to a discrete horizon of approximately more than 20 m in lateral extension. Lithologically nearly planar limestone beds of about 7-15cm alternate with thin slate layers within a interval of 2 m.
ALPINE REEFS AND PAVEMENTS
99
Fig. 2. Stratigraphical column (based on Piller et al. 2004) with indication of reefs and pavements in Alpine units of the Silurian-Devonian timespan in Austria (Western and Eastern Greywacke Zone and Gurktal Nappe). Abbreviations: M. Auen Dolo., Middle Auen Dolomite; M. Dolomite, M61bling Dolomite; Reef Debt. Lst Althofen, Reef Debris Limestone Althofen; volc., volcanoclastic sediments.
100
B. HUBMANN & T. SUTTNER
Fig. 3. Stratigraphical column (based on Piller et al. 2004) with indication of reefs and pavements in Alpine units of the Silurian-Devonian time span in Austria (Graz Palaeozoic, Remschnigg/Sausal and Southern-Burgenland). Abbreviations: Hochl. Fm, Hochlantsch Formation; Os. Fm, Osser Formation; Parmas. Fm, Parmasegg Formation; P1. Fm, Plabutsch Formation; Sch6. Fm, Sch6ckel Formation; Schw. Fm, Schweinegg Formation; Tyrnauer. Fro, Tyrnaueralm Formation; Zachensp. Fm, Zachenspitz Formation.
ALPINE REEFS AND PAVEMENTS
101
Fig. 4. Stratigraphical column (based on Piller et al. 2004) with indication of reefs and pavements in Alpine units of the Silurian-Devonian time span of the Southern Alps (Austrian parts of the Carnic Alps and Karawanken Mountains). Abbreviations: Altico. Lst, Alticola Limestone; Cardio. Fm, Cardiola Formation; H. W. Lst, Hohe Warte Limestone; Kellerg. Reefal Lst, Kellergrat Reefal Limestone; K.W. Lst, Kellerwand Limestone; Lamb. Lst, Lambertenghi Limestone; M. U. Bischofalm S, Middle-Upper Bischofalm Slates; Seeberg C.-C. Lst, Seeberg Coral-Crinoid Limestone; Seew. Lst, Seewarte Limestone; S-Karawanken Mts, Southern-Karawanken Mountains.
102
B. HUBMANN & T. SUTTNER
Fig. 5. (a) Section at the Baron von Kottwitz quarry, Southern Burgenland. (b) Section of the Weisse Wand area, Graz Palaeozoic. (c) Lochkovian mound (Rauchkofel Limestone) at the Hohe Warte section, Carnic Alps. (d) Thick-unbedded limestone succession (Hohe Warte Limestone) at the Hohe Warte-Seewarte section (left to right), Carnic Alps.
Apparently only the limestone beds yield fossils (Suttner & Lukeneder 2004), especially small brachiopods, ostracods, poriferan spicules, some gastropods and few conodonts. These fossils occur together with huge amounts of accumulated serpulid tubes (Fig. 6a). Some of these can be found as solitary tubes and others may occur as fused clusters of 2-3 cm diameter. The average length of a single tube is about 5 mm and 0.8 mm in diameter. Towards the top of the section, the stabilizing siliciclastic input decreases and facies changes from thin laminated rocks to wellbedded limestone occur. Within the limestone beds a relatively abundant conodont fauna was obtained so that the upper part of the section could be referred to the Ozarkodina remscheidensis eosteinhornensis and Icriodus woschmidti Zones of uppermost Silurian-lowermost Devonian.
Graz Palaeozoic (Eastern Alps) Locality: Eggenfeld Stratigraphic unit: K6tschberg Thalwinkel Member
Formation;
Age: Upper Silurian-Lower Devonian At the base of the succession at Eggenberg (15 km north of Graz), approximately 2 m of greenish, massive diabase (Mensink 1953; Ebner 1976) is overlain by reddish-pink and greenish-grey tufts. From the latter originate the only so far known remains of a graptolite, Bohemograptus bohemicus tenuis. The volcanoclastic sediments are overlain by a succession of dark grey dolomite and grey-yellow limestone containing densely packed fossils (nautiloids, crinoids and brachiopods). Brachiopods referred to the species Septatrypa subsecreta were found in the second and third limestone horizons covering the entire bed surfaces (Fig. 6b). Nearly all individuals show articulated pedicle and brachial valves, suggesting an autochthonous burial. We therefore interpret them as brachiopod pavements. According to Ebner (1976) the profile covers conodonts of the siluricus (Upper Ludlow) to woschmidti Zones (Lochkovian). In addition, the identification of Bohemograptus bohemicus tenuis permits a referral to the leintwardinensis Zone (Lower Ludfordian).
ALPINE REEFS AND PAVEMENTS
103
Fig. 6. (a) Accumulation of serpnlid tubes, Kirchfidisch, Southern Burgenland, thin section, NHMW 2002z0166. (b) Septatrypa pavement, K6tschberg Formation, Eggenfeld, Graz Palaeozoic, rock surface. (c) Amphipora mounds, F16sserkogel Formation, Graz Palaeozoic. (d) Striatopora suessi, Plabutsch Formation, Graz Palaeozoic, polished section (collection: Fritz Messner). (e) Stachyodes Beds, Kollerkogel Formation, Weisse Wand, Graz Palaeozoic. (f) Stromatoporoid patches, Hohe Warte Limestone, Seewarte section, Carnic Alps, rock surface. (g) Renalcis sp., Hohe Warte Limestone, Seewarte section, Carnic Alps, thin section (IPUW-3803-4-27). (h) Calcareous green algae, Hohe Warte Limestone, Seewarte section, Carnic Alps, thin section (IPUW-3803-4-4). Abbreviation of collection numbers: IPUW, Institute of Palaeontology, University of Vienna (room number: 2B 191); NHMW, Museum of Nature History, Vienna.
104
B. HUBMANN & T. SUTTNER
Locality: Plabutsch Stratigraphic unit: Plabutsch Formation Age: Middle Devonian (Eifelian) The Plabutsch Formation represents a highly fossiliferous sequence whose stratigraphic boundaries are not clearly defined yet. Locally, the sequence may range from the Upper Emsian to the Lower Givetian. Deposition in restricted to open gently inclined platform environments is assumed (Hubmann 1993, 1995). Frequency distribution patterns of the fossil content show a clear dominance of biota belonging to the 'reefbuilding guild' or 'constructor guild' (Fig. 6d), i.e. stromatoporoids, favositids, various rugose corals. Representatives of the tabulate coral Thamnopora predominate and suggest a great importance for current reduction and 'baffling' of fine-grained sediment. Parts of the approximately 100 m-thick sequence may characterize an 'autoparabiostrome' following the classification schemes for organogenetically controlled successions lacking topographic relief (Kershaw 1994). In particular the distinct horizons of densely packed shells of the thick-valved pentamerid brachiopod Zdimir cf. hercynicus is a specific feature of the formation. Approximately half of the brachiopods are preserved with both valves suggesting an autochthonous-parautochthonous fossillagerstaette. According to the monospecific mass accumulation and the dense package we interpret it as brachiopod pavement.
Locality: Weil3e Wand Stratigraphic unit: Kollerkogel Formation Age: Middle Devonian (Givetian) Usually the Plabutsch Formation is overlain by dolostones passing into a sequence of massive limestone. At the Weil3e Wand area (Fig. 5b) (some 20 km north of Graz) a thin intercalating horizon of rauhwacke (cellular dolomite) marks the beginning of the Givetian succession starting with the Gaisbergsattel Member in this region. The latter comprises a basally developed 'reef pioneer settlement', some 10m in thickness dominated by densely packed Stachyodes (Fig. 6e) and auloporids in a black bituminous limestone matrix. This sequence passes into dark-grey fossil-rich limestones with special rugose coral consortia dominated by Mesophyllum and Stringophyllum. A thin horizon (approximately 30-50 cm) with small colonies of the phaceloid rugosan Thamnophyllum terminates the 'pioneer sequence', which is overlain by 20-30 m-thick, white and slightly dolomitized limestones. The latter contain accumulations: of various reef-building organisms (stromatoporoids, rugose and tabulate corals).
Carnic Alps (Southern Alps) Locality: Wolayer area, Rauchkofel Stratigraphic unit: Cardiola Formation Age: Upper Silurian (Ludlow) Overlying the Middle-Upper Silurian Kok Formation, the Cardiola Formation consists of interbedded dark micritic limestones and black marls yielding abundant cephalopods and bivalves. Cardiolid concentrations were found in thin layers on bedding planes. Although the outcrop area suffers from inferior exposure, two horizons can be observed, one at the base of the Cardiola Formation and a second one slightly above. Covering the nearly stable bed surface, fixed by parallel orientated orthocerids, bivalves settled in vast numbers. Owing to the in situ accumulations of abundant articulated valves of Cardiola docens and their monospecific occurrence totally covering the bedding surface, they are considered to represent a shell pavement.
Locality: Wolayer area, Hohe Warte Stratigraphic unit: Rauchkofel Limestone Age: Lower Devonian (Lochkovian) Microbial mounds (Fig. 5c) are documented in the Lower Devonian Rauchkofel Limestone. Baffling organisms, especially calcareous green algae (Hubmann 1994) and ramose-branching tabulate corals, predominate. In contrast to the well-bedded limestones at the base and the sedimentary cover, the mound features a higher organic content. The dimension of the lensshaped body is about 60-80 m in length and 8-10m in height. (Similar structures, even though stratigraphically much younger and much smaller in dimensions, can be recognized in the Amphipora mounds (Fig. 6c) of the Emsian F16sserkogel Formation of the Graz Palaeozoic.)
Locality: Wolayer area, Seewarte Stratigraphic unit: Hohe Warte Limestone Age: Lower Devonian (Pragian) The Hohe Warte Limestone reaches a maximum thickness of 300-350m. The Hohe WarteKellerwand Complex, together with some localities close to that area (Biegengebirge and Gamskofel Massif), are interpreted as relics of a shallow-water carbonate platform with reefs and back reefs. The proximal platform slope continues northwards and is found in the the Cellon Nappe. Lateral extensions from peritidal to pelagic settings were estimated with 8-9 km (Kreutzer 1992a). The area south of Lake Wolayer exhibits a thick massive limestone sequence (Fig. 5d) (Hohe
ALPINE REEFS AND PAVEMENTS Warte Lst) of Pragian age succeeding wellbedded crinoidal packstones. Its base consists of crinoidal grainstone grading into coralstromatoporoid-dominated framestones (Fig. 6f). Within the framework, cyanobacteria like Renalcis sp. (Fig. 6g) and calcareous green algae (Fig. 6h) become common constituents of the middle-upper part. At the boundary to the overlying Seewarte Formation facies changes from shallow-shelf to peritidal-lagoonal settings.
Discussion The depositional basins to which the 'ProtoAlps' belonged experienced from cold- to moderate-water conditions owing to a high-latitudinal position on a peri-Gondwanan terrane (Sch6nlaub 1992, 1997a) during the late Ordovician when the evolution and first prominent spreading of the mid-Palaeozoic reef community characterized by corals, stromatoporoid sponges and calcareous algae (Webby 1984; Poncet 1990) started. After the melting of the north Gondwanan ice sheet and the return to normal greenhouse climatic conditions the 'alpine terrane' - or several terranes on which the 'alpine' sequences were deposited - continued to shift from higher to lower latitudes during the Silurian Period. Apparently rather rapid northward plate rotations moved the Silurian Alpine depositional systems into a geographic position within a 300-40 ~ latitudinal interval (Sch6nlaub 1997b). Palaeomagnetic data from northern Gondwana areas, as well as increasing shallow-marine deposition during this time, support this inference. Although some regions remained in higher latitudes (Sch6nlaub 1992) others obviously reached the equator, for instance the Graz Palaeozoic during the Lower Devonian (Emsian) (Fenninger et al. 1997). Generally, the northward shift during time is apparent in changing facial patterns to progressive carbonate-rich sediments and increasing biotic diversity. The stratigraphic sequence of the Graz Palaeozoic (part of the Eastern Alpine Units) indicates a sedimentation area changing from a passive continental margin with intracontinental alkaline volcanism to shelf and platform geometries during the Silurian-Devonian periods (Fritz et al. 1992). Sea-level changes and probably synsedimentary tectonics were responsible for the lithological development (i.e. alternations of dolostones and limestones: Hubmann 1993) and the formation of stratigraphic gaps and mixed conodont faunas (Ebner 1978).
105
In contrast to this region north of the Periadriatic Fault the depositional history differs in the Southern Alpine Units. During the Silurian, the succession of the Carnie Alps is subdivided into four lithological facies representing different depths of deposition and hydraulic conditions suggesting a steadily subsiding basin and an overall transgressional regime from the Llandovery to Ludlow. The transitions at the Llandovery-Wenlock and Wenlock-Ludlow boundaries are marked by deepening events (Sch6nlaub 1997b). Uniform limestone sedimentation during the Pridoli suggests that more stable conditions were developed at this time. The Pragian, Givetian and, particularly, Frasnian stages climaxes for shallow-marine carbonate factories dawned; their products are still visible in the high cliffs of the Kellerwand and of the Hohe Warte in the central Carnic Alps. The ratio of thicknesses between shallowwater limestone and contemporary cephalopod limestone approximates to 12:1 and thus indicates differentially subsiding mobile basins affected by extensional tectonics. At the Frasnian-Famennian boundary the reefs drowned and the pelagic facies developed again. We kindly acknowledge support of the University of Vienna and the Austrian Academy of Sciences. We thank J. Pickett (retired, Geological Survey of New South Wales, Australia) for his help improving the style of the manuscript. F. Boulvain (University of Liege, Belgium) and H.P. Sch6nlaub (Geological Survey of Vienna, Austria) are gratefully acknowledged for helpful comments on the manuscript.
References EBNER, F. 1976. Das Silur/Devon-Vorkommen von Eggenfeld - ein Beitrag zur Biostratigraphie des Grazer Palfiozoikums. Mitteilungen der Gesellschaft Geologischer Bergbaustudenten Osterreichs, 37, 275-305. EBNER, F. 1978. Stratigraphie des Karbon der Rannachfazies im Palfiozoikum von Graz. Mitteilungen der Osterreichischen Geologischen Gesellschaft, 69, 163-196. FENNINGER, A., HUBMANN, B., MOSER, B. & SCHOLGER, R. 1997. Diskussion zur pal~iogeographischen Position des Grazer Terrane aufgrund neuer palfiomagnetischer Daten aus dem Unterdevon. Mitteilungen des naturwissenschaftlichen Vereins fiir Steiermark, 126, 33-43. FERRETTI, A., HISTON,K. & SCHONLAUB,H. P. 1999. The Silurian and Early Devonian of the Rauchkofel Boden Section, Southern Carnic Alps, Austria. In: HlSTON, K. (ed.) V International Symposium Cephalopods - Present and Past. Carnie Alps Excursion Guidebook. Berichte der Geologischen Bundesanstalt, 47, 55-62.
106
B. HUBMANN & T. SUTTNER
FLI3GEL, H. W., JAGER, H., SCHONLAUB,H. P. & VAI, G. B. 1977. Carnic Alps. In: MARTINSSON, A. (ed.) The Silurian-Devonian Boundary. lUGS Series A, 5, 126-142. FRITZ, H., EBNER, F. & NEEBAUER, F. 1992. The Graz Thrust-Complex (Paleozoic of Graz). In: NEUBAUER, F. (ed.) The Eastern Central Alps of Austria, ALCAPA-FieM Guide. IGP/KFU Graz, 83-92. HTSTON, K. 1999. V International Symposium Cephalopods - Present and Past. Carnic Alps Excursion Guidebook. Berichte der Geologischen Bundesanstalt, 47, 1-84. HISTON, K., FERRETTI, A. & SCHONLAUB,H. P. 1999. Silurian Cephalopod Limestone sequence of the Cellon Section, Carnic Alps, Austria. In: HTSTON, K. (ed.) V International Symposium CephalopodsPresent and Past. Carnic Alps Excursion Guidebook. Berichte der Geologischen Bundesanstalt, 47, 46-54. HUBMANN, B. 1993. Ablagerungsraum, Mikrofazies und Pal~io6kologie der Barrandeikalk-Formation (Eifelium) des Grazer Pal~iozoikums. Jahrbueh der Geologischen Bundesanstalt, 136(2), 393-461. HUBMANN, B. 1994. The oldest Udoteacean Green Algae known from Austria: Paralitanaia carnica n. sp. (Carnic Alps, Lower Devonian, Lochkovian). Neues Jahrbuch fiir Geologie und Paldontologie Monatshefte, 6, 329-338. HUBMANN, B. 1995. Middle Devonian shallow marine deposits of the Graz Paleozoic: fact and fiction for deposition under ecological stress. Beitrdge zur Paldontologie, 20, 107-112. HUBMANN, B. & MESSNER, F. 2005. Grazer Paldozoikum - 75. Jahrestagung der Palgiontologischen Gesellschaft in Graz. Berichte des Institutes ffir Erdwissenschaften, Universit~it Graz, 1-47. KERSHAW, S. 1994. Classification and geological significance of biostromes. Facies, 31, 81-92. KREUTZER, L. H. 1992a. Palinspastische Entzerrung und Neugliederung des Devons in den Zentralkarnischen Alpen aufgrund von neuen Untersuchungen. Jahrbuch der Geologischen Bundesanstalt, 135, 261-272. KREUTZER, L. H. 1992b. Photo-atlas of the variscian carbonate sequences in the Carnic Alps (Austria/ Italy). Abhandlungen der Geologischen Bundesanstalt, 47, 129. KREUTZER, U H., SCHONLAUB,H. P. & HUBMANN,B. 1997. The Devonian of Austria. In: SCHONLAUB,H. P. (ed.) IGCP 421 North Gondwanan Mid-Paleozoic Biodynamics, Guidebook. Berichte der Geologischen Bundesanstalt, 40, 42-60. KREUTZER, L. H., SCHONLAUB,H. P. & HUBMANN, B. 2000. The Devonian of Austria. Courier Forschungsinstitut Senckenberg, 225, 173-183. LOYDELL, D. K. 1998. Early Silurian sea-level changes. Geological Magazine, 135, 447-471. MARTIN, J. M., BRAGA,J. M., AGU1RRE,J. & BETZLER, C. 2004. Contrasting models of temperate carbonate sedimentation in a small Mediterranean embayment: the Pliocene Carboneras Basin, SE Spain.
Journal of the Geological Society, London, 161, 387-399. MENSINK, H. 1953. Eine tektonische Detailuntersuchung im Raum n6rdlich Gratkorn. Mitteilungen des naturwissenschaftlichen Vereins fiir Steiermark, 83, 123-129. PILLER, W. E., EGGER, H. ET AL. 2004. Die Stratigraphische Tabelle yon Osterreich 2004 (sedimentiire Schichtfolgen). Osterreichische stratigraphische Kommission und Kommission ffir die pal~iontologische und stratigraphische Erforschung {)sterreichs der Osterreichischen Akademie der Wissenschaften. POLLACK, W. 1962. Untersuchungen iiber Schichtfolge, Bau und tektonische Stellung des 6sterreichischen Anteils der Eisenberggruppe im siidlichen Burgenland. PhD thesis, University of Vienna. PONCET, J. 1990. Paleobiogeography of Ordovician calcareous algae. Palaeogeography, Palaeoclimatology, Palaeoecology, 81, 91-94. PRIEWALDER, H. 1997. The distribution of the Chitinozoans in the Cellon Section (HirnantianLower Lochkovian). - A preliminary report. In: SCHONLAUS, H. P. (ed.) IGCP 421 North Gondwanan Mid-Paleozoic Biodynamics, Guidebook. Berichte der Geologischen Bundesanstalt, 40, 74-85. PRIEWALDER, H. 2000. Die stratigraphische Verbreitung der Chitinozoen im Abschnitt CaradocLochkovium des Cellon-Profils, Karnische Alpen (K~irnten, Osterreich) - Ein vorl~ufiger Bericht. Mitteilungen der Osterreichischen Geologischen Gesellschaft, 91, 17-29. RIEGL, B. & PILLER, W. E. 2000. Reefs and coral carpets in the northern Red Sea as models for organism-environment feedback in coral communities and its reflection in growth fabric. In: INSALACO, E., SKELTON,P. W. & PALMER,T. J. (eds) Carbonate Platform Systems." Components and Interactions. Geological Society, London, Special Publications, 178, 71-88. SCHONLAUB, H. P. 1992. Stratigraphy, biogeography and paleoclimatology of the Alpine Paleozoic and its implications for plate movements. Jahrbuch der Geologischen Bundesanstalt, 135, 381-418. SCHONLAUB, H. P. 1994. Das Altpal~iozoikum im Sfidburgenland. In: LOBITZER, H., CsAszAR, G. & DAURER, A. (eds) Jubiliiumsschrift 20 Jahre Geologische Zusammenarbeit Osterreich-Ungarn. Geological Survey, Vienna, 365-377. SCHONLAUS, H. P. 1997a. The biogeographic relationships of Ordovician strata and fossils of Austria. In: SCHONLAUB, H. P. (ed.) IGCP 421 North Gondwanan Mid-Paleozoic Biodynamics, Guidebook. Berichte der Geologischen Bundesanstalt, 40, 6-19. SCHONLAUB, H. P. 1997b. The Silurian of Austria. In: SCHONLAUB, H. P. (ed.) IGCP 421 North Gondwanan Mid-Paleozoic Biodynamics, Guidebook. Berichte der Geologischen Bundesanstalt, 40, 20-42. SCHONLAUS,H. P. & HISTON, K. 1999. The Palaeozoic of the Southern Alps. In: HISTON, K. (ed.) Vlnternational Symposium Cephalopods - Present and Past.
ALPINE REEFS AND PAVEMENTS Carnic Alps Excursion Guidebook. Berichte der Geologischen Bundesanstalt, 47, 6-30. SUTTNER, T. & LU~ZENEDER,A. 2004. Accumulations of Late Silurian serpulid tubes and their palaeoecological implications (Blumau-Formation; Burgenland; Austria). Annalen des Naturhistorischen Museums in Wien, 105A, 175-187.
107
WENZEL, B. 1997. Isotopenstratigraphische Untersuchungen an silurischen Abfolgen und deren pal/iozeanographische Interpretation. Erlanger geologische Abhandlungen, 129, 1-117. WEBBu B. D. 1984. Ordovician reefs and climate: a review. Oslo University, Contributions to Paleontology, 295, 89-100.
Sedimentology and magnetic susceptibility of the Upper Eifelian-Lower Givetian (Middle Devonian) in SW Belgium: insights into carbonate platform initiation CI~DRIC MABILLE & FRI~DI~RIC BOULVAIN
P~trologie SOdimentaire, B20, Universitb de LiOge, Sart Tilman, B-4000 Libge, Belgium (e-mail: emabille@ulg, ac. be, fboulvain@ulg, ac. be) Abstract: The major part of the Hanonet Formation is deposited on a mixed siliciclasticcarbonate detrital ramp, whereas the top is dominated by carbonate-rimmed shelf-related sedimentation. The transition corresponds roughly to the Eifelian-Givetian boundary. This work is based on two stratigraphic sections located in the southern part of the Dinant Synclinorium. Petrographic study leads to the definition of 11 microfacies, which demonstrate important sedimentological differences existing between the sections. A curve showing microfacies evolution is interpreted in terms of changing bathymetry. An environmental model depicts the lateral transition from a multiclinal carbonate ramp (to the east) to a forereef setting (to the west). Magnetic susceptibility was used to establish accurate stratigraphic correlations between the two sections. It also leads to an appreciation of the relative importance of eustatic sea-level change and local sedimentation rate. The combined interpretation of the microfacies curves and the magnetic susceptibility provides a new view of the sedimentary dynamics of the studied sections and, in a more general way, a better understanding of the processes responsible for magnetic susceptibility variations in carbonate rocks.
Limestones in the Upper Eifelian-Lower Givetian of SW Belgium provide an instructive means to investigate the initiation of a carbonate platform. This leads to fundamental questions about the mechanisms responsible for the transition from a mixed siliciclastic-carbonate detrital ramp during the Eifelian to a rimmed carbonate shelf near the Eifelian-Givetian boundary, and thus insight into the start of the 'carbonate factory' to understand the parameters that influence the carbonate production. Although the La Couvinoise quarry has already been studied several times in the past, it has never been the subject of detailed sedimentological work, and nothing has ever been published on Les Monts de Baileux. This first use of magnetic susceptibility to obtain accurate stratigraphic correlations and information about the sedimentary dynamic of the sections is also a test for the use of magnetic susceptibility in multiclinal ramp, containing a series of cliniforms (La Couvinoise) and fore-reef settings (Les Monts de Baileux).
Location and geological context The Hanonet Formation is on both sides of the Eifelian-Givetian boundary. At this time, a large carbonate platform developed throughout northern Europe (Fig. l a) and overcame the mixed siliciclastic-carbonate ramp. The two studied sections are located along the southern flank of the Dinant Synclinorium,
in the area of Couvin (Fig. lb). The Dinant Synclinorium is part of the Rhenohercynian fold-and-thrust belt. Studying the Eifelian-Givetian in Belgium is of crucial importance to understand this rapid and dramatic transition. The Hanonet Formation, then, forms the link between the Jemelle Formation (the last ramp-related unit of the Eifelian) and the Trois-Fontaines Formation (the first carbonate platform-related unit of the Givetian) (Fig. 2). The first section (La Couvinoise) is located near the railway station in Couvin (400 m to the NW). Although the basal contact with the underlying Jemelle Formation is lacking, it exposes 85 m of the Hanonet Formation, up to the base of the biostromal unit of the Trois-Fontaines Formation. Les Monts de Baileux quarry is located to the NE of the 32nd kilometre marker along the N66 road between Couvin and Baileux. The first 6 m of outcrop are part of the Jemelle Formation. Data were collected over 113 m up to the base of the biostromal unit of the Trois-Fontaines Formation.
Previous work and historical context The biostratigraphy of the Eifelian at the southern flank of the Dinant Synclinorium is well established for the rugose corals (Coen-Aubert 1989, 1996, 1997, 1998), brachiopods (e.g.
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 109-123.0305-8719/07/$15.00 9 The Geological Society of London.
110
C. MABILLE & F. BOULVAIN
Fig. 1, (a) Palaeogeographic setting at the Eifelian (390 Ma), after Ziegler (1982) and McKerrow & Scotese (1990) showing the large carbonate platform that develops throughout northern Europe and overcomes the mixed siliciclastic-carbonate ramp, (b) Geological setting and location of the studied sections at the southern flank of the Dinant Synclinorium.
EIFELIAN CARBONATE PLATFORM INITIATION
111
Fig. 2. Generalized lithostratigraphic section of Middle Devonian formations at the southern border of the Dinant Synclinorium, after Bultynck & Dejonghe (2001). The studied interval corresponding to the Hanonet Formation in located at the boundary between the Eifelian (ramp-related sedimentation) and the Givetian (carbonate platform-related sedimentation). Same legend as in Figure 3.
Godefroid 1995) and conodonts (e.g. Bultynck 1970). The base of the Eifelian is now defined by the appearance of Polygnathus costatus partitus (Werner 1982). This species was found in Belgium above the basal limit of the old Couvinian stage. Moreover, the upper limit of the Eifelian, marked by the rising of Polygnathus hemiansatus, is below the upper limit of the Couvinian, which was placed at the top of the Hanonet Formation. Thus, the Eifelian and the Couvinian do not represent strictly the same stratigraphic interval, and the Eifelian is now regarded as the official name for the lower part of the Middle Devonian. For the Co2d interval, the formal name of the Hanonet Formation, the stratotype was established in La Couvinoise quarry (Bultynck 1970) and the actual name Hanonet Formation was introduced by Tsien (1973). The first sedimentological study of La Couvinoise quarry (Pr6at 1989) led to the definition of six microfacies. From these microfacies,
three distinct environments were defined: external ramp, middle ramp and inner ramp. The stratigraphic succession of these environments corresponds to a general shallowingupwards trend. A palaeoecological approach based on the ostracods (Casier et al. 1992) confirms this trend in La Couvinoise quarry from the top of the Hanonet Formation to the base of the Trois-Fontaines Formation. A more detailed study of the EifelianGivetian transition led to the definition of 10 major microfacies and several submicrofacies deposited on a mixed siliciclastic-carbonate detrital ramp (Pr6at & Kasimi 1995; Kasimi & Pr6at 1996).
Methods Bed-to-bed description and sampling were carried out in 2003 and 2004. From the samples, 550 thin sections were prepared. The textural classification used to characterize the microfacies follows Dunham (1962) and Embry & Klovan
112
C. MABILLE & F. BOULVAIN
(1972). The term 'coverstone' characterizes microfacies where laminar organisms cover mud and bioclastic debris (Tsien 1984). This led to the definition of 11 major microfacies and four submicrofacies, and to the establishment of a two-dimensional sedimentological model. These microfacies are compared to those defined for other Eifelian sections in Belgium (Pr6at 1989; Pr6at & Kasimi 1995). Then, each sample was submitted to magnetic susceptibility measurements with a KLY-3 (Kappabridge). Each sample was measured three times and weighed with a precision of 0.01 g. These operations allow the definition of the mass-calibrated magnetic susceptibility of each sample and the drawing of magnetic susceptibility curves for both sections.
Description of sections La CouvinoL~e quarry Five lithological units were defined (Fig. 3). The lowest consists of 39 m of dark very argillaceous limestone interbedded with subnodular beds of limestone. Brachiopods and crinoids that are often well preserved dominate the fauna. Some rugose corals, domical tabulate corals, Receptaculites and Orthoceras are also present near the top of the unit. The second unit (from 39 to 50 m) starts with the first development of laminar stromatoporoids and branching tabulate corals. This unit is also characterized by enrichment in rugose corals, whereas the brachiopods are less abundant. Some gastropods, laminar tabulate corals and Orthoceras are also found. This limestone is less argillaceous but some centimetre-thick argillaceous interbeds are present. The third unit (from 50 to 61 m) begins with pure limestone followed by slightly argillaceous limestone. The top of this unit is characterized by pure and very massive limestone in metre-thick beds. Crinoids, laminar stromatoporoids, rugose corals and branching tabulate corals dominate the fauna, but some domical stromatoporoids and tabulate corals are also present. The fourth unit (from 61 to 77 m) is similar to the second one, in terms of the presence of argillaceous limestone and joints. The faunal assemblage is also similar, except for the replacement of the laminar stromatoporoids by laminar tabulate corals about 2 m above the base of this unit. The uppermost unit (from 77 to 85 m) is composed of centimetre- to decimetre-thick beds of almost pure limestone interbedded with very argillaceous limestone. Crinoids dominate
the fauna, whereas brachiopods, tabulate corals, rugose corals, gastropods and laminar stromatoporoids are less abundant. This unit corresponds to the first part of the Trois-Fontaines Formation defined as 'bedded argillaceous crinoidal limestones' (Bultynck & Dejonghe 2001).
Les Monts de Baileux quarry Eight units are observed in this section. The lowest lithological unit (A) is 6 m thick, and is composed of very argillaceous limestone with a sparse fauna of crinoids and brachiopods. At the top of the unit some lenticular decimetresized beds of slightly argillaceous limestone are present. This unit corresponds to the top of the Jemelle Formation. A more abundant fauna and enrichment in gastropods characterize the B unit (from 6 to 26 m). This unit consists in an interbedding of several metre-thick sets of beds of purer limestone with more argillaceous limestone. Shaly interbeds are common. In the C unit (from 26 to 41 m) the fauna become more diverse with the first development of domical stromatoporoids (up to 50 cm in diameter), laminar stromatoporoids and branching, domical and laminar tabulate corals. Crinoids are again present whereas brachiopods and gastropods are less abundant. This third unit is composed of variably argillaceous limestone. The beginning of the D unit (from 41 to 57 m) is marked by an important faunal change. Although crinoids and domical tabulate corals are still present, domical stromatoporoids and branching tabulate corals disappear. As laminar skeletons of stromatoporoids and tabulate corals become less common, brachiopods and gastropods are more abundant. Finally, the first appearance of rugose corals and trilobites is observed. Lithologically, the limestone is less argillaceous even though some argillaceous interbeds are present. The E unit (from 57 to 67 m) starts with a distinctive metre-thick tempestite with brachiopods, crinoids and gastropods. The unit is composed of slightly argillaceous limestone. The fauna includes crinoids, rugose corals, and laminar and domical stromatoporoids and tabulate corals. The F unit (from 67 to 95 m) is characterized by argillaceous limestone becoming less argillaceous upwards. This unit resembles unit C. Some differences can be noted: rugose corals are present and some bioclastic decimetre-sized lenses are present. The beginning of the G unit (from 95 to 101 m) is marked by a 5 cm-thick shale bed. The lower half of this unit is more argillaceous than
EIFELIAN CARBONATE PLATFORM INITIATION
113
Fig. 3. Sedimentological log of the two studied sections showing the lithological units. These units are numbered from 1 to 5 for La Couvinoise section and from A to H for Les Monts de Baileux section. Both sections end at the base of the biostromal unit of the Trois-Fontaines Formation.
114
C. MABILLE & F. BOULVAIN
the upper half. The fauna consists of gastropods, crinoids, brachiopods, rugose corals and domical tabulate corals. Two decimetre-thick crinoidal grainstone beds are present; the first with an erosive base and the second lenticular. Finally, the H unit (from 101 to 113 m) consists of bioturbated, slightly argillaceous limestone. Crinoids dominate the fauna but some tabulate corals (laminar, domical and branching), rugose corals, stromatoporoids (laminar and domical) and gastropods are also present. This eighth unit corresponds to the base of the Trois-Fontaines Formation defined previously as locally coral-rich crinoidal limestone (Bultynck et al. 1991).
preserved (e.g. brachiopods, bryozoans, ostracods or trilobites). Detrital quartz reaches 7.5%, cubes of pyrite and micas are locally present.
Description of microfacies
MFC5." stromatoporoidal coverstone.
The marked differences between the two sections lead to the use of two sets of microfacies: seven microfacies for La Couvinoise (MFC1-MFC7) and four microfacies for Les Monts de Baileux (MFB 1-MFB4). Microfacies present in both sections (MFC5a and MFC6) are only described for La Couvinoise. For each section, the microfacies are described in order of increasing proximality (see Fig. 4).
- MFC5a." coverstone with reworked stromatoporoids in a slightly argillaceous matrix. This microfacies is dominated by laminar stromatoporoids. They are well preserved, and some reach more than 1 m in diameter and 20 cm in thickness. The matrix is slightly argillaceous, and textures range from packstone to mudstone. Crinoids, ostracods and brachiopods dominate the fauna. However, other reef-building organisms (like domical tabulate corals and stromatoporoids and branching tabulate corals) and algae are locally present. Peloids (spherical or ovoidal and from 0.2 to 0.5 mm in diameter) are only present in Les Monts de Baileux quarry.
L a Couvinoise quarry M F C I : slightly argillaceous mudstone with sparse fauna. Fossils are uncommon but characterize an open-marine environment: trilobites, crinoids, brachiopods, ostracods and bryozoans. These organisms are well preserved. Detrital quartz (up to 10%), framboidal pyrite and micas are present. Pressure-solution seams are also present. MFC2." slightly argillaceous wackestone with crinoids and brachiopods. The dominant bioclasts are trilobites, crinoids, brachiopods, ostracods and bryozoans with some reef-building organisms and algae. Moreover, well-preserved ostracods, brachiopods, bryozoans and crinoids are common. Detrital quartz, framboidal pyrite and pressure-solution seams are present whereas micas become less common than in MFC1. MFC3" slightly argillaceous packstone with crinoids and brachiopods. This microfacies is associated with MFC2 and MFC3 at the base of the section. At the top, it constitutes decimetreto metre-thick beds. The fauna is dominated by trilobites, crinoids, ostracods and bryozoans, but is more diversified: gastropods, stromatoporoids, tabulate corals and some algae (mainly palaeosiphonocladaleans). Some fossils are well
MFC4: slightly argillaceous floatstone and rudstone with stromatoporoids and tabulate corals. This facies corresponds to centimetre-sized fragments of rugose corals, tabulate corals (domical, laminar and branching) and stromatoporoids (laminar and domical) in a slightly argillaceous micritic matrix. Other organisms such as crinoids, brachiopods, bryozoans, ostracods and trilobites are present. The degree of preservation of reef-building organisms is higher than that of other fossils. The detrital quartz become less abundant (<1%) and cubes of pyrite are observed.
- MFC5b: coverstone with in situ stromatoporoids in a microsparitic matrix. These coverstones are similar to those described in MFC5a. The only difference is that they include in situ laminar stromatoporoids and a microsparitic nonargillaceous microsparitic matrix. This matrix, by its relative cohesiveness, is favourable to the preservation of shelter porosity corresponding to synsedimentary cavities under stromatoporoids (Boulvain 2001). MFC6: microsparitic packstone and poorly sorted peloidal grainstone rich in bioclasts. The peloids represent 20-30% of the thin-section surfaces. Two different types are observed: similar to those described above (for MFC5a) or larger (0.51 ram) and irregular. They can be related to the micritization of bioclasts, as suggested by local relics of the original fossil. This microfacies is rich in bioclasts: crinoids, brachiopods, bryozoans, ostracods, algae (Girvanella, dasycladaleans, udoteaceans and palaeosiphonocladaleans) and gastropods in order of decreasing abundance. Detrital quartz can reach up to 10% in the La Couvinoise quarry.
EIFELIAN CARBONATE PLATFORM INITIATION
115
o,-~
=:
_~ o='~
~ "~ "~ "~ ~ ~ ,~'~ ~11 ' ' ~ r ,.../ 0 ~ 0 ~.~ / - , , .,..~
"~ 0
0
.,--~
"~.= ~ o o 0
~ ~.~N 0~".~~ ' ~ ~'~
~
9~
,~ .,~
'-'~o o
,~
.~, o ..~
9
~:
"
c~
,
o
0
~ 0 ~ . ~
,~. ~ ~
~
0
'''~
rT
*-' -~' ~
0
0 . ~
o ~
.,~
~
~~.~
~ , o ~ o .~.-~ r.~ ~, .,_~
~.,,~
o ~ ~ ..o m ~u ~
~
0
~ ~D~
Cl
~ ~,~ ~ e ~ " ~
~
N ~ ~ ~ .~,.~ ~ o ~ .o ~ ~ ~.~ ~ o ~ rn " ~
o ~
~. o ~
o
~'~ ~
o o
o .~
~ r
~ ~ 0~ ~,~,~--.~
""~ 0
~
. ~ ~ ,-~ ~
116
C. MABILLE & F. BOULVAIN
MFC7: well-sorted peloidal grainstone. Small ovoidal peloids (cf. MFC5a) dominate this microfacies (up to 50%). The other elements are very uncommon: algae (mainly Girvanella), crinoids and ostracods. Les M o n t s de Baileux quarry MFBI: mudstone with high terrigeneous content. These mudstones, locally laminated, are particularly rich in detrital quartz and/or clay with local presence of micas flakes. Oxidized iron is often abundant, giving a yellowish-reddish colour to the rock. The fauna is very rare and poorly diversified (crinoids and brachiopods). Locally, it includes broken bryozoans, ostracods, tabulate corals, palaeosiphonocladaleans, trilobites and echinoid spines. Peloids range from rare to abundant. Millimetre-thick lenticular wackestones or packstones are also present. MFB2: peloidal wackestone with micritic matrix. Bioclasts of trilobites, brachiopods, crinoids, ostracods and bryozoans dominate the fauna. Small peloids (cf. MFC5a) are also present. Green algae are locally present (palaeosiphonocladaleans and dasycladaleans). These bioclasts are poorly preserved, except for those located in some packstone lenses. Note that dolomitized thin sections with replacement of matrix by euhedral dolomite crystals and less common preserved fossils (trilobites, brachiopods, crinoids, ostracods, bryozoans and palaeosiphonocladales) are considered as MFB2. MFB3: floatstone with stromatoporoids and tabulate coral bioclasts in a peloidal matrix. Between domical tabulate corals, solitary rugose corals, laminar stromatoporoids and tabulate corals, matrix is peloidal and rich in calcareous algae: dasycladaleans, palaeosiphonocladaleans, udoteaceans and Girvanella. MFC5a: coverstone with reworked stromatoporoids in a slightly argillaceous matrix. This is the same microfacies as that in La Couvinoise. MFB4: crinoidal grainstone and packstone. Crinoids dominate the fauna, whereas peloids and bioclasts, such as trilobites, ostracods, bryozoans, brachiopods and some calcareous algae (palaeosiphonocladaleans and dasycladaleans), are less common. The crinoids are well sorted and locally surrounded by syntaxic cement. MFC6: microsparitic packstone and poorly sorted peloidal grainstone. This is the same microfacies as that in La Couvinoise. However, two submicrofacies are distinguished in Les Monts de Baileux.
- MFC6a." microsparitic packstone and poorlysorted peloidal grainstone with trilobites. Calcareous algae and reef-derived debris are very rare, whereas the open-marine fauna (trilobites, brachiopods, crinoids, ostracods and bryozoans) is well represented. - MFC6b: microsparitic packstone and poorly sorted peloidal grainstone with calcareous algae and reef-building organisms. Whereas calcareous algae and reef-derived debris (solitary rugose corals, domical tabulate corals and stromatoporoids and laminar tabulate and stromatoporoids) corals are common, trilobite bioclasts are lacking.
Microfacies interpretation Different criteria are available to interpret the palaeoenvironmental setting of each microfacies. Faunal association and depositional texture directly reflect the level of energy and agitation. Two major sets of organisms have been described: open-marine fauna (trilobites, bryozoans, crinoids, brachiopods and ostracods); and reef-building organisms (rugose corals, tabulate corals and stromatoporoids). Calcareous algae (abundance and nature) and peloids are also significant constituents. Other criteria like sorting, terrigeneous content, nature of matrix and degree of preservation of bioclasts are also relevant. Moreover, a comparison with other microfacies defined in the literature for Eifelian rocks is made when possible. Every microfacies described here above may be interpreted in terms of degree of distality and relative bathymetry. La Couvinoise quarry The major difference between MFC1, MFC2 and MFC3 is texture, which ranges from mudstone to wackestone and packstone. However, they are similar in terms of faunal assemblage, nature of matrix, and detrital quartz and mica content. They were deposited in a similar environment and the faunal assemblage suggests an openmarine setting. The presence of packstone lenses within mudstones and wackestones can be interpreted as relatively distal storms deposits (Dott & Bourgeois 1982), such that MFC3 may represent storm deposits within MFC1 and MFC2. Thus, these microfacies correspond to an open-marine environment located below fair-weather wave base (FWWB) but above storm wave base (SWB). This is also the interpretation made for similar microfacies of the Eifelian-Givetian boundary interval in the Dinant Synclinorium (Pr6at & Kasimi 1995). A study of ostracod
EIFELIAN CARBONATE PLATFORM INITIATION fauna from La Couvinoise quarry (Casier et al. 1992) determined a dysaerobic environment for this microfacies. MFC4 includes debris coming from a reefal environment, but open-marine conditions still prevailed. Their more agitated nature, as shown by the floatstone/rudstone textures, points to an environment situated close to the FWWB. The same interpretation was made for similar microfacies (Pr6at 1989; Pr6at & Kasimi 1995). The development of laminar stromatoporoids characterizing MFC5 corresponds to favourable conditions in terms of bathymetry, substrate and sufficiently low detrital input (see, for example, Kershaw 1998). These favourable conditions may correspond to a lowering of detrital input from the nearby landmasses. In La Couvinoise, the first stromatoporoids are observed simultaneously with a lithological change from very argillaceous limestone to less argillaceous limestone, indicating a lowering in detrital input. In MFC5a some stromatoporoids are overturned, suggesting a higher influence of storms and a location near the FWWB (Kershaw 1980). This environment may be regarded as a potential source for part of the debris in MFC4 but another source, probably a biostromal unit, is needed for domical stromatoporoids and tabulate coral debris. The main characteristic of MFC6 is the abundance of peloids. They probably have a shallow-water, low-energy origin like a lagoon or a back-reef area (see, for example, Tucker & Wright 1990). Moreover, in other Eifelian sections studied in Belgium, the presence of peloids is also linked to the development of reefal settings (Pr6at & Kasimi 1995; Mamet & Pr6at 2005). This proximal environment might also be responsible for the production of calcareous algal debris. It is noticeable that there is a mixing between the two kinds of sediment (open-marine bioclasts and peloids + calcareous algae). This suggests that the proximal material (supplied by storm deposits or debris flow) and the openmarine bioclasts (supplied by storm deposits) are deposited in the same environment and then mixed by bioturbation. The grainstone texture suggests a location within the FWWB. The MFC7 shows a higher influence of the peloidal source than the MFC6 and the good sorting involves a more continuous degree of agitation. This microfacies is considered as the most proximal one. Les M o n t s de Baileux quarry MFB 1 represents the deepest microfacies of both sections. The primary sedimentation mechanism
117
process is slow accumulation of suspended mud and minute debris, but small wackestone and packstone lenses probably represent distal storm deposits. This suggests that this microfacies was located just above the SWB (Pr6at & Kasimi 1995). Again, two sources of debris must be considered to explain the nature of the MFB2 assemblage: an open-marine one (trilobites, bryozoans, crinoids, brachiopods and ostracods) and a transported but proximal one (peloids and calcareous algae, and possibly micrite). The proximal origin of the micrite is uncertain and we are not able to exclude a local production of this micrite. MFB2 was situated within the SWB, the packstone lenses representing storm deposits. MFB3 possess the same characteristics as MFC4, except that it is influenced by a proximal source supplying peloids and calcareous algae, and perhaps micrite. MFC5a is similar in both sections, except that in Les Monts de Baileux there was a greater supply of peloids. MFB4 is mainly characterized by well-sorted crinoidal grainstone and packstone. Such an accumulation of crinoids corresponds to storms deposits around the FWWB. The environment is largely influenced by an open-marine source while material originating from proximal areas is less abundant. MFC6 includes two submicrofacies. The main difference concerns the relative importance of the two sources of debris. MFC6a is more influenced by the open-marine source, whereas MFC6b is more influenced by the proximal area.
Palaeoenvironmental model Microfacies interpretation of each section leads to the conclusion that the two depositional environments are different (Fig. 5). La Couvinoise quarry is more influenced by a fine-grained detrital input, whereas Les Monts de Baileux quarry is characterized by a more proximal carbonate
La Couvinoise
Les Monts
de Baileux
Setting Reef Terrigenous input Carbonate input Total input
Multiclinal ramp
Fore-reef
None?
Present
High
Low
Low
High
Lower
Higher
Fig. 5. Summary of main palaeoenvironmental features.
118
C. MABILLE & F. BOULVAIN
influence. The palaeoenvironmental model proposed here is based on a synthesis of microfacies interpretation and explains the major differences between the two sections (Fig. 6). This model describes the lateral transition from a multiclinal carbonate ramp (La Couvinoise) to a fore-reef environment in a platform setting (Les Monts de Baileux). Without any barrier in La Couvinoise (this absence is perhaps caused by the terrigenous input itself) terrigenous material are allowed to spread, whereas a barrier would have created a more protected environment, potential source of peloids, calcareous algae and lime-mud. In the first case, however, the detrital input was not too great to prevent the development of isolated stromatoporoid bioconstructions in La Couvinoise, whereas in the second situation the input of carbonate detritus was so abundant that the development of stromatoporoids is very poor.
Microfacies curves interpretation The palaeoenvironmental evolution, highlighted by the microfacies curves, is described lithological unit by lithological unit for each section.
This allows a better understanding of the differences existing between them (Fig. 7).
La Couvino&e quarry The first unit is characterized by MFC1, MFC2 and MFC3 (note that in the second third of the unit, MFC1 is very dominant). Referring to the microfacies interpretation, this unit corresponds to open-marine setting, under the FWWB and above the SWB. A wide variety of microfacies (MFC2, MFC3, MFC4, MFC5a, MFC5b and MFC6) is present in the second unit. It marks: (1) an increase of storms energy and influence, shown by the apparition of MFC4; (2) the presence of MFC6, which shows that the FWWB is reached at the top of the unit; (3) the development of stromatoporoids (MFC5a and MFC5b); and (4) there is an evolution to more proximal microfacies along this unit. The microfacies present in the third unit are similar to those present in the second one (except MFC2). The evolution, however, is just the opposite and goes from proximal to distal settings.
Fig. 6. Proposed palaeoenvironmental model. Same legend as in Figure 3. This model shows the lateral transition from a multiclinal carbonate ramp (La Couvinoise) mainly influenced by a fine-grained detrital input to a fore-reef setting (Les Monts de Baileux) characterized by the major influence of a proximal source of carbonate (peloids and calcareous algae, and possibly micrite).
EIFELIAN CARBONATE PLATFORM INITIATION The fourth unit is characterized by MFC3, MFC4, MFC5a and MFC6 and can be divided in two. The first half is characterized by an oscillating microfacies curve showing a background sedimentation around the FWWB (MFC3, MFC4 and MFC5a) often flooded by peloids, calcareous algae and carbonate input originating from a more proximal area (MFC6). Nevertheless, the second half of the fourth unit shows an evolution from MFC3 to MFC6. The fifth and last unit is also characterized by an oscillating curve between MFC3 and MFC4, on the one hand, and MFC6 and MFC7, on the other. MFC3 and MFC4 involve a location under the FWWB with a high terrigeneous input (confirmed by the abundance of shale). However, proximal carbonate inputs are also well represented, as shown by MFC6 and MFC7.
Les Monts de Baileux quarry Except for the A unit, the microfacies curve is characterized by lots of oscillations in Les Monts de Baileux. These oscillations occur between two groups of microfacies. The first group (MFB2, MFB3, MFC5a and MFB4) represents the background sedimentation with a limited but present proximal influence, whereas the second (MFC6a and MFC6b) corresponds to high proximal inputs in carbonate (peloids and calcareous algae, and possibly micrite). A unit is divided into two parts. The first part is only composed of MFB1, whereas the top is composed of MFB4. This shows a rapid transition from the SWB to the FWWB. In the B unit microfacies are MFB2 and MFB4 (Group 1) and MFC6a and MFC6b (Group 2). Moreover, it is remarkable that the interbedding of several metre-thick sets of beds of purer limestone within more argillaceous limestone corresponds, respectively, to parts where group 2 and group 1 are dominant. C and D units are both characterized by a group 1 represented by MFB2 and MFB3 and a group 2 composed of MFC6a and MFC6b. The first group indicates energetic settings, but still under the FWWB. The difference between C and D consists of the fact that the first group of microfacies dominates the C unit, whereas the second dominates the D unit. This explains why the D unit is composed of more massive limestone. The E unit is very massive, this is related to the dominance of MFC6b and MFC6a (Group 2). MFB2 is poorly represented. Oscillations are again observed within the F unit between group 1 (MFB2 and MFB3) and group 2 (MFC6b and MFC6a). The last is still
119
dominant. It is also noticeable that laminar stromatoporoids (MFC5a) are present only in this unit. Group 1 (MFB2 and MFB3) dominates group 2 (MFC6b and MFC6a) in the G unit. This indicates a lower influence of the proximal carbonate input. The H unit is dominated by MFC6a at the base and MFC6b at the top. Group 1 (MFB2) is poorly represented.
Microfacies curves: conclusions There is also a large difference between the two sections that prevents any correlation based on sequence stratigraphy from being made (Fig. 7). In fact, it is possible to plot the bathymetric evolution in La Couvinoise by interpreting the microfacies curves; this interpretation suggests a general shallowing-upwards trend. For Les Monts de Baileux, however, the major process that defines the microfacies curve involves the pulses in the carbonate influx, which were independent of bathymetry. These two distinct sedimentary dynamics explain why no correlation based on sequence stratigraphy can be made.
Magnetic susceptibility Principles Magnetic susceptibility (MS) is a measure of the sample response to an external magnetic field first employed in the study of Palaeozoic rocks during the 1990s. For sedimentary rocks, the major influence on MS is the terrestrial fraction. This can be linked to eustasy because when the sea level falls, the erosion of exposed continental masses increases and this typically leads to higher MS values. On the other hand, when the sea level rises, MS shows lower values. Thus, MS can be used to obtain accurate correlations with higher resolution than that offered by biostratigraphy (Crick et al. 1997; da Silva & Boulvain 2002). It is important to note that other influences like climatic changes (precipitation, ice ages, pedogenesis), tectonism, diagenesis, volcanism, impact ejecta and so on may also influence MS values.
M S values and correlations When comparing the MS values of the two sections, similar trends and events are observed (Fig. 7). These are considered as isochronous and facies-independent, and thus correlatable (see also Ellwood et al. 1999). Moreover, the average value for each section (6.38 x 10-8 m 3 kg -1 for La Couvinoise and 2.92 x 10-8 m 3 kg -1 for Les Monts
120
C. M A B I L L E & F. B O U L V A I N
EIFELIAN CARBONATE PLATFORM INITIATION
Fig. 8. Average, minimal and maximal MS values plotted by microfacies for each location. These curves show a general decreasing in average values from distal to proximal microfacies in both sections.
de Baileux) confirms that the environment in La Couvinoise was subjected to a greater detrital input, accounting for its higher MS values. Available data in literature (e.g. Hladil et al. 2002; da Silva & Boulvain 2002) support the notion that, generally, proximal microfacies possess higher values of MS than distal ones. This is explained by the relative proximity to the terrestrial source. However, the average MS values of each microfacies (Fig. 8) shows just the opposite trend, with higher MS values for distal microfacies and lower values for proximal ones. This can be explained if the environment of deposition of the Hanonet Formation was located sufficiently far from the detrital source to homogenize the detrital supply. In this situation there is no great difference in the terrestrial input for each microfacies. Thus, the major influence on the MS value is the dilution by the carbonate production. So the greater the carbonate productivity, the greater the dilution of the MS signal.
M S interpretation If the MS response is related to sea-level change, it is surprising to observe such a divergence between it and the microfacies curves (Fig. 7).
121
Fig. 9. Average, minimal and maximal magnetic susceptibility values plotted by percentage of detrital quartz for each location. Although quartz is diamagnetic and has a weakly negative magnetic susceptibility signature, and thus does not affect the overall MS values, it is regarded as a good proxy for detrital content.
Published data (da Silva & Boulvain 2002) usually report strong correlation between MS and bathymetric interpretation based on microfacies. This is noticeably the case for La Couvinoise, where the microfacies curve suggests a general shallowing-upwards trend, whereas the MS curve corresponds grossly to a general deepening-upwards trend followed by a (slight) shallowing-upwards trend. Making a semi-quantitative estimate of detrital quartz content for each thin section, a strong correlation between the abundance of detrital quartz and MS values (Fig. 9) is apparent. While the detrital quartz does not carry the MS signal, it is a good indicator of the detrital input, at least for the two sections studied here. Therefore, it is certain that the MS is correlated to the detrital input. In other sections, for example Aywaille and Tailfer from the Frasnian of Belgium (da Silva & Boulvain 2003), the MS and microfacies curves are nearly parallel. Each transgressive or regressive trend is registered in both curves. It is quite different for La Couvinoise and Les Monts de Baileux. In fact, in La Couvinoise, the two curves seem to be opposed for the two first lithological units. Then they are more correlated, even if
Fig. 7. Schematic sedimentological log, microfacies curves and magnetic susceptibility curves. Same legend as in Figure 3. Arrows represent trends in curves and dashed lines the correlation lines mainly based on MS features. For microfacies curves, trends are defined lithological unit by lithological unit in La Couvinoise. Note also that the oscillating microfacies curve in Les Monts de Baileux, related to a different sedimentary dynamic, prevents from any reliable trend from being considered. This is why no correlation based on sequence stratigraphy can be made.
122
C. MABILLE & F. BOULVAIN
there is a kind of time lag in that the same trends are present somewhat later in the MS curve. To understand the different mechanisms or environmental parameters that constrain the different curves, it is necessary to focus on the contrasting depositional environments of the two sections: a carbonate platform for Aywaille and Tailfer; and a ramp and fore-reef setting for La Couvinoise and Les Monts de Baileux. At the very least, the microfacies curve gives information about the local bathymetric evolution. A hypothesis is that the MS variations, reflecting continental erosion, are correlated to global sea-level change. In this case it corresponds to a general deepening-upwards trend followed by a shallowing-upwards trend, allowing the development of the Trois-Fontaines Formation biostromal unit. It so happens that this is confirmed by the global sea-level curve available in the literature (Johnson et al. 1985). If the trends observed for local and global sea-level evolution are not the same, different sedimentation rates may explain these differences. Here, the shape of the microfacies curves is related to local bathymetry, dependent on both global sea-level fluctuation and the sedimentation rate. Considering local variations in the sedimentation rate, significant differences between the local and global sea-level evolution can be obtained. The different sedimentary dynamics shown by the microfacies analysis can explain the differences between the sedimentation rate in each section. For Les Monts de Baileux, microfacies analysis has already shows that the major influence on the microfacies evolution is the carbonate input, creating a relative independence between the global sea-level curve (shown by MS) and the local sea-level change (shown by the microfacies curve). However, the differences between the local and global sea-level evolution in La Couvinoise can be explained by considering major variations in sedimentation rate provided by strongly contrasting carbonate productivity. Thus, the 'carbonate factory' worked at different rates (owing to ecological parameters) to fill in the free space left by the global sea-level pattern. For example, if the 'carbonate factory' worked faster than the eustatic sea-level rise, it can produce a regressive event that has the appearance of a local sea fall.
Conclusions For this study of the Hanonet Formation, two sections were considered: La Couvinoise (the stratotype) and Les Monts de Baileux, whose marked differences are evident even in the field.
Petrographic analyses led to the definition of 11 microfacies and four submicrofacies, of which only three are observed in both locations. All these microfacies are integrated to a twodimensional palaeoenvironmental model depicting the lateral transition from a multiclinal carbonate ramp (La Couvinoise) to a fore-reef setting (Les Monts de Baileux). The former environment is mainly characterized by enhanced terrigenous input, whereas the latter is greatly influenced by back-reef-derived sediment deposition. This also implies a major divergence between both sections in terms of sedimentary dynamics that does not allow suitable highresolution stratigraphic correlations based on sequence stratigraphy. However, magnetic susceptibility analyses revealed itself to be a powerful tool to establish accurate stratigraphic correlations between the two sections. The combined interpretation of the microfacies and the magnetic susceptibility curves proved instructive. This interpretation, mostly based on the La Couvinoise quarry, explains the apparent divergence existing between the general shallowing-upwards trend recorded by the microfacies curve and the two deepening-upwards trends followed by a shallowing-upwards trend shown by magnetic susceptibility. This situation can be explained by a difference between the evolution of the local bathymetry and global sea level induced by differences in the rates of sedimentation. In this case, eustasy is reflected in the evolution of the magnetic susceptibility, whereas the microfacies curve records the local relative sea-level evolution. C. Mabille benefited from a F.R.I.A. grant from the Belgian Fond National de la Recherche Scientifique (F.N.R.S.). F. Boulvain acknowledges support through research grant FRFC 2-4501-02 (F.N.R.S.). The authors are especially grateful to M. Humblet for help during field study, and to E. Poty and M. Coen-Aubert for rugose corals determination and interesting discussions. We are deeply grateful to B. Pratt for critical reading of the manuscript, as well as to S. Kershaw and M. Whalen for their accurate and very helpful criticisms and recommendations.
References BOULVAIN,F. 2001. Facies architecture and diagenesis of Belgian Frasnian mounds (Petit-Mont Member). Sedimentary Geology, 145, 269-294. BVLTYNCK, P. 1970. ROvision stratigraphique et palOontologique (Braehiopodes et Conodontes) de la coupe type du Couvinien. M6moires de l'Institut
g6ologique de l'Universit6 de Louvain, 26.
EIFELIAN CARBONATE PLATFORM INITIATION BULTYNCK, P. & DEJONGHE, L. 2001. Devonian lithostratigraphic units (Belgium). Geologica Belgica, 4, 39-69. BULTYNCK, P., COEN-AUBERT, M. ET AL. 1991. Les formations du Ddvonien moyen de la Belgique. M6moires pour servir fi l'explication des cartes g6ologiques et mini6res de la Belgique, 30. CASIER, J.-G., PREAT, A. & KASIMI, R. 1992. Ostracodes et S6dimentologie du sommet de l'Eifelien et de la base du Givetien, ~ Couvin (bord sud du Bassin de Dinant, Belgique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, Sciences de la Terre, 62, 75-108. COEN-AUBERT, M. 1989. Repr6sentants des genres Sociophyllum BIRENHEIDE, 1962 et Beugniesastrea n. gen. fi la base du Calcaire de Givet de Pondr6me et de Resteigne (bord sud du Bassin de Dinant, Belgique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, 58, 5-31. COEN-AUBERT, M. 1996. Siphonophrentides et Cyathophyllides pr6s de la limite Eifelien-Givetien ~t Resteigne (Ardenne, Belgique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, 66, 19-36. COEN-AUBERT, M. 1997. Rugueux solitaires pros de la limite Eifelien-Givetien fi Pondr6me (Belgique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, 67, 5-24. COEN-AUBERT, M. 1 9 9 8 . Thamnophyllides et Acanthophyllides pros de la limite Eifelien-Givetien /t Wellin et Pondr6me (Belgique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, 68, 5-24. CRICK, R. E., ELLWOOD,B. B., EL HASSANI,A., FEIST, R. & HLADIL, J. 1997. Magnetosusceptibility event and cyclostratigraphy (MSEC) of Eifelian-Givetian GSSP and associated boundary sequences in north Africa and Europe. Episodes, 20, 167-175. DA SILVA,A.-C. & BOULVAIN,F. 2002. Sedimentology, magnetic susceptibility and isotopes of a Middle Frasnian carbonate platform: Tailfer section, Belgium. Facies, 46, 89-102. DA SILVA,A.-C. & BOULVAIN,F. 2003. Sedimentology, magnetic susceptibility and correlations of Middle frasnian platform limestone, Tailfer and Aywaille sections, Belgium. Geologica Belgica, 6, 81-96. DOTT, R. H., JR. & BOURGEOIS, J. 1982. Hummocky stratification: significance of its variable bedding sequences. Geological Society of America Bulletin, 93, 663-680. DUNHAM, R. J. 1962. Classification of carbonate rocks according to depositional texture. In: HAM, W. E. (ed.) Classification of carbonate rocks. AAPG Bulletin, 87, 108-121. ELLWOOD, B. B., CRICK, R. E. & EL HASSANI,A. 1999. The magneto-susceptibility event and cyclostratigraphy (MSCE) method used in geological correlation of Devonian rocks from Anti-Atlas Morocco. AAPG Bulletin, 1, 108-121. EMBRY, A. F. & KLOVAN, J. E. 1972. Absolute water depth limits of Late Devonian paleoecological zones. Geologische Rundschau, 61,672-686.
123
GODEFROID, J. 1995. Les brachiopodes (Pentamerida, Atrypa et Spiriferida) de la fin de l'Eifelien et du d6but du Givetien ~t Pondr6me (Belgique, bord sud du Synclinorium de Dinant). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique, 65, 69-116. HLADIL, J., PRUNER, P., VENHODOVA,D., HLADILOVA, T. & MAN, O. 2002. Toward an exact age of Middle Devonian Celechovice corals - Past problems in biostratigraphy and present solutions complemented by new magnetosusceptibility measurements. Coral Research Bulletin, 7, 65-91. JOHNSON, J. G., KLAPPER,G. & SANDBERG,C. A. 1985. Devonian eustatic fluctuations in Euramerica. Geological Society of America Bulletin, 96, 56%587. KASIMI, R. & PRI~AT,A. 1996. S6dimentation de rampe mixte silico-carbonat6e des couches de transition eifeliennes-givetiennes franco-belges. Deuxi~me partie: cyclostratigraphie et pal6ostructuration. Bulletin des Centres de Recherche ExplorationProduction Elf Aquitaine, 20, 61-90. KERSHAW, S. 1980. Cavities and cryptic faunas beneath non-reef stromatoporoids. Lethaia, 13, 327-338. KERSHAW, S. 1998. The application of stromatoporoids palaeobiology in palaeoenvironmental analysis. Palaeontology, 41,509-544. MAMET, P. & PRI~AT,A. 2005. Microfaci6s d'une lentille biohermale ~t la limite Eifelien/Givetien (Wellin, bord sud du synclinorium de Dinant). Geologica Belgica, 8, 85-112. MCKERROW, W. S. & SCOTESE, C. R. 1990. Palaeozoic Palaeogeography and Biogeography. Geological Society, London, Memoir, 12. PRI~AT,A. 1989. Sedimentology, facies and depositional environment of the Hanonet (Upper Eifelian) and Trois-Fontaines (Lower Givetian) Formations in Couvin (Dinant Basin, Belgium). Bulletin de la Soci&~ beige de Gdologie, 98, 14%154. PRI~AT,A. & KASIMI,R. 1995. S6dimentation de rampe mixte silico-carbonat6e des couches de transition eifeliennes-givetiennes franco-belges. Premi6re partie: microfaci6s et mod61e s6dimentaire. Bulletin des Centres de Recherche Exploration-Production Elf Aquitaine, 19, 329-375. TSlEN, H. H. 1973. Le Couvinien dans la rkgion de Couvin. Comitk H - Dkvonien. Service G6ologique de Belgique, Document, 8. TSIEN, H. H. 1984. R6cifs d6voniens des Ardennes Pal6o6cologie et structure. In: GLISTER, J. & HERB, R. (eds.) GOologie et palkoOcologie des rkcifs, Institut Geologique de l'Universit6 de Berne, Berne, 7.1-7.20. TUCKER, M. E. & WRIGHT, W. P. 1990. Carbonate Sedimentology. Blackwell Science, Oxford. WERNER, R. 1982. On Devonian stratigraphy and palaeontology of the Ardenno-Rhenisch Mountains and related Devonian matters. Senckenbergische Naturforschende Gesellschaft, Frankfurt. ZIEGLER, A. 1982. Geological Atlas of Western and Central Europe. Shell, Den Haag, the Netherlands.
Frasnian carbonate mounds from Belgium: sedimentology and palaeoceanography FR}~DI~RIC B O U L V A I N
Pktrologie skdimentaire, B20, Universit~ de Likge, Sart Tilman, B-4000 Liege, Belgium (e-mail." fboulvain@ulg, ac. be) Abstract: The facies architecture, sedimentary dynamics and palaeogeographic evolution were reconstructed for a number of middle-late Frasnian carbonate mounds from the south side of the Dinant Synclinorium (Belgium). Nine facies were recognized in the buildups, each characterized by a specific range of textures and assemblage of organisms: spiculitic wackestone with stromatactis (facies Pml), which becomes progressively enriched in crinoids and corals (Pm2); grey or pinkish limestone with stromatactis, corals and stromatoporoids (A3-L3, Pm3); grey limestone with corals, peloids and dasycladales (A4-L4, Pm4); grey, microbial limestone (A5-L5, Pm5); grey limestone with dendroid stromatoporoids (A6-L6); grey, laminar fenestral limestone, (A7-L7); and grey, bioturbated limestone (A8-L8). Sedimentological evidence suggests that facies Pml and Pm2 correspond to iron bacteriasponge-dominated communities, developing in a quiet aphotic and hypoxic environment. A3-L3 developed between storm and fair-weather wave base, in an oligophotic environment. Facies A5-L5 developed close to fair-weather wave base. Facies A6-L6 and the fenestral limestone A7-L7 correspond to an environment with slightly restricted water circulation. Facies AS-L8 developed at subtidal depths in a quiet, lagoonal environment. The main differences between the middle and late Frasnian mounds concern facies architecture, and are a consequence of different palaeoceanographic settings. The large flattened middle Frasnian Arche and Lion buildups show limited vertical differentiation, large-scale progradation features, extensive exportation of material towards off-reef environment and development of inner lagoonal facies. They grew offshore from a well-developed carbonate platform with a healthy carbonate factory. Middle Frasnian sea-level fluctuations were relatively mild, and sedimentation was able to keep up with sea-level rise. At the opposite extreme, during the late Frasnian, severe eustatic rises, together with rising oceanic hypoxic conditions, were responsible for frequent collapses of the carbonate factory, drowning of the middle Frasnian carbonate platform, and development of buildups with relatively limited lateral extension, high vertical facies differentiation, low potential for material exportation and high content in microaerophilic iron bacteria.
Among the various Palaeozoic carbonate mounds known throughout the world (e.g. Bosence & Bridges 1995; Monty 1995; Pratt 1995; Bourque 1997), the Frasnian carbonate mounds of Belgium are probably the earliest studied. This remarkable interest carried by generations of geologists derives from the number and quality of outcrops: currently 75 carbonate mounds are known, and the majority have been actively quarried since R o m a n time. The combination of extraordinary outcrop exposure and a welldocumented Devonian stratigraphy makes Frasnian carbonate mounds in Belgium ,of international significance (e.g. Tsien 1975). This paper is devoted to the middle Frasnian Arche and Lion buildups and the late Frasnian PetitMont Member. It illustrates various facies architectures, interprets mound palaeoenvironments, and assesses the possible relationship between evolution of carbonate mounds and changes in palaeoceanographic setting of the sedimentary basin.
Location and geological context The Belgian Frasnian lithostratigraphy has been revised recently (Boulvain et al. 1999). Three main levels of carbonate mounds are recognized in the Frasnian of the southern border of the Dinant Synclinorium (Figs 1 & 2), which is a large-scale unit of the West-European Variscan fold-and-thrust belt. These are, upsection, the Arche, Lion and Petit-Mont members. In the Philippeville Anticlinorium, mounds occur only in the Petit-Mont Member, the other moundbearing levels being replaced landwards by bedded limestone, locally with back-reef characters. At the northern border of the Dinant Synclinorium and in the Namur Synclinorium, the entire Frasnian consists of bedded limestone and argillaceous strata (Da Silva & Boulvain 2002, 2004). The best-known middle Frasnian Arche and Lion buildups are located in the immediate neighbourhood of Frasnes: the Nord quarry
From: •LVARO, J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 125-142. 0305-8719/07/$15.00 9 The Geological Society of London.
126
F. BOULVAIN
Fig. 1. Schematic geological map of southern Belgium, with location of main outcrops, l, La Boverie quarry and Humain section; 2, Lompret quarry; 3, Frasnes railway section; 4, Neuville railway section; 5, Aisemont quarry; 6, Lustin and Tailfer sections; 7, Crupet section; 8, Huccorgne section. See Figure 3 for a detailed map of the Frasnes-Philippeville area (framed).
Fig. 2. Schematic N-S cross-section and main lithostratigraphic subdivisions of the Frasnian sedimentary basin before the Variscan orogeny.
FRASNIAN CARBONATE MOUNDS
127
Fig. 3. Detailed geologicalmap of the Frasnes-Philippeville area with location of some Frasnian carbonate mounds. 1, Arche quarry; 2, Nord quarry; 3, Lion quarry; 4, Moulin Bayot sections; 5, Beauchfiteauquarry; 6, Tapoumont quarry; 7, Les Bulants quarry; 8, Les Croisettes quarry; 9, Les Wayons quarry; 10, Rochefontaine quarry; 11, Hautmont quarry.
(Lecompte 1954; Boulvain et al. 2004), the Lion quarry (Boulvain et al. 2004) and the Arche quarry (Lecompte 1954; Boulvain et al. 2004); the first two are in the Lion Member, the third is in the Arche Member (Fig. 3). Very recently, Boulvain et al. (2005) brought information about a set of outcrops located some distance from Frasnes: the La Boverie quarry close to Rochefort (Fig. 4) and the Moulin Bayot sections close to Vodel6e (Fig. 1). At both locations, it was possible to study a series of buildups, starting from the Arche Member and ending with the Lion Member. Among the 69 late Frasnian carbonate mounds currently listed (Fig. 3), seven buildups from the Philippeville Anticline were examined: Beauchfiteau, Les Bulants, Les Wayons (Fig. 4), Rochefontaine, Hautmont, Les Croisettes and Tapoumont (Boulvain 2001). Several sections in each buildup were described bed by bed, and more than 3000 thin sections were produced. Polished slabs were examined under a binocular microscope. Sections were also studied in peri- and off-mound environments at the southern border of the Dinant Synclinorium (Frasnes, Chimay:
Humblet & Boulvain 2001; Boulvain et al. 2004) and in the Philippeville Anticline (Neuville railway sections: Boulvain 2001). Coeval sections were also examined in the internal zones of the Frasnian platform, at the northern border of the Dinant Synclinorium (Lustin, Crupet, Tailfer: Boulvain 2001; Da Silva & Boulvain 2002), at the southern border of the Namur Syncline (Aisemont) and at the northern border of the Namur Syncline (Huccorgne: Boulvain 2001)
(Fig. 1). Previous work Dewalque (1868) first recognized the 'reefal' character of Belgian Frasnian mounds. After stratigraphic subdivision by Maillieux (1913 1926), Delhaye (1913) recognized a distinct vertical succession of lithofacies within these mounds, but it was Dumon (1932) who related this succession to bathymetric variation. Lecompte (1936, 1959) realized that the Frasnian buildups developed in relatively deep water under conditions of subsidence. His facies zonation (successively lower red zone with corals and stromatactis;
128
F. B O U L V A I N
FRASNIAN CARBONATE MOUNDS median pink or grey zone with corals, brachiopods, crinoids, stromatoporoids and algae; and upper red zone with corals) was focused on bathymetric and hydrodynamic criteria (below turbulent zone for the red zones and within turbulent zone for the grey or pink zone). According to Tsien (1975, 1980), reefal biofacies were built in the wave action zone (red and pink limestone with corals and cyanobacteria; red and pink limestone with corals and stromatactis; grey limestone with corals and cryptalgal structures), whereas others formed below the wave action zone (red limestone with stromatactis; pink limestone with Renalcis). Tsien (1980) also showed that the slopes of buildups were partly oversteepened by differential compaction between buildup limestone and argillaceous nonbiohermal sediments. Monty et al. (1982) and Monty & Van Laer (1988) asserted that the majority of cavity cements, like the micrite in buildups, were of bacterial origin, and that the presence of microbial gel allowed the development of steep slopes and significant relief. They concluded that the buildups developed below wave base and below the photic zone.
Facies models Six buildup facies can be defined in the Arche ('A') and Lion ('L') members (facies A3-L3 through to A8-L8) and five (facies Pml-Pm5) in the Petit-Mont Member, each facies being characterized by a specific range of textures and organic assemblages. Three other facies ('flank facies') can also be defined; these bedded bioclastic and lithoclastic facies being the lateral time-equivalents of the buildup facies. The components in the buildup facies are essentially autochthonous and directly reflect the influence of water parameters such as agitation and light intensity. By contrast, the flank facies contain large amounts of transported material, much of
129
it derived from nearby buildups (Humblet & Boulvain 2001). The organic assemblages found in the flank facies therefore do not necessarily reflect the environments in which those facies were deposited. The logic behind the coding scheme used here for designating the buildup facies is the same as that used by Boulvain et al. (2001): identical facies are given identical facies numbers, even when they are in buildups at different stratigraphic levels (e.g. A3, L3, respectively, for middle Frasnian Arche and Lion members). This scheme also facilitates comparison with the mounds of the late Frasnian Petit-Mont Member (facies Pml-Pm5), which have been described in this way (Boulvain 2001). The following descriptions of the buildup facies are organized bathymetrically, from deep to shallow, according to textures and fossil assemblage.
Buildup facies: late Frasnian Petit-Mont Member Red limestone with stromatactis (facies Pml). The intense red pigmentation of this facies is the consequence of a hematite content up to 5% Fe203 (Fig. 4C, E, F). The occurrence of stromatactis is variable. Stromatactis may be grouped in metre-scale beds forming a reticulate structure and exceeding 50% of the rock. Stromatactis may exceed 50 cm in length, but generally diminish in size towards more argillaceous zones. Stromatactis are cemented by inclusion-rich radiaxial calcite. The cement surmounts various types of internal sediment (laminar microspar; microspar with vermiform texture of Pratt 1982; peloids and pseudosparite; microbial mats; ooids with a microsparitic cortex). A strict geometrical relationship between spicular networks and stromatactis does not exist: spicules can overlie, penetrate or form concentrations below stromatactis. In addition to stromatactis,
Fig. 4. (A) Upper part of the Beauchgtteau mound, near Senzeilles, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. The height of the outcrop is 30 m. (B) Lower part of the Les Wayons mound, near Merlemont; late Frasnian Petit-Mont Member, Philippeville Anticlinorium. Log of the mound, see Figure 6. (C) Red limestone with stromatactis (facies Pml); Les Croisettes quarry, Vodec6e, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. (D) Stromatactoid fenestra surrounded by a spicular network (facies Pml); thin section HMC17, normal light; Hautmont quarry, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. (E) Iron bacteria in sparite; thin section RFX, normal light; Rochefontaine quarry, Villers-le-Gambon, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. (F) Pink limestone with corals, crinoids, brachiopods, stromatactis, fenestrae, stromatoporoids and nebuloids (grey horizontal beds) (facies Pro3); Les Bulants quarry, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. (G) Wackestone with stromatactoid fenestrae, crinoids, fenestellids and peloids (facies Pm3); thin section TP 44, normal light; Tapoumont quarry, late Frasnian Petit-Mont Member, Philippeville Anticlinorium. (H) Packstone with peloids and Trelonella (facies Pm4); thin section TPG2a, normal light; Tapoumont quarry, late Frasnian Petit-Mont Member, Philippeville Anticlinorium.
130
F. BOULVAIN
millimetre-scale fenestrae with granular sparitic cement are common (Fig. 4D). Sponges are abundant (e.g. tetractinellids, hexactinellids and other non-rigid demosponges with a simple monaxone assemblage of spicules: Termier et al. 1981). Matrix is microsparitic-pseudosparitic.
subspherical coral colonies (Hankaxis, Phillipsastrea, Alveolites), Thamnopora, brachiopods and
subordinate
(Amphipora).
dendroid
stromatoporoids
Renaleis is locally abundant.
former by the addition of crinoids and lamellar corals such as Alveolites and Phillipsastrea. Decimetre-scale stromatactis and centimetrescale stromatactoid fenestrae (see Neuweiler et al. 2001) with granular cement are abundant. Supported cavities filled with radiaxial cement typically occur below lamellar organisms.
Thrombolitic structures and microbial mats are present. Within thrombolites, Renalcis is associated with Palaeomicrocodium (Mamet & Boulvain 1992). All thrombolites and stromatolites appear as a canvas made up of irregular peloids set in a yellowish pseudosparitic cement (the 'structure grumeleuse' of Cayeux 1935). Irregular peloids and fragments of microbial mats are abundant. Most organisms (including Palaeomicrocodium) are strongly encrusted by Rothpletzella, Girvanella, Wetheredella and microbial mats.
Pink limestone with corals, crinoids, brachiopods, stromatactis and lamellar stromatoporoids (facies Pro3). This facies shows decimetre-scale beds
Buildup facies: middle Frasnian Arche and Lion members
Red limestone with stromatactis, lamellar corals, crinoids (facies Pro2). This facies differs from the
rich in millimetre-scale stromatactoid fenestrae, crinoids, brachiopods, and other bioclasts intercalated with beds containing sparse fenestrae, corals and subordinate stromatoporoids. Corals are generally tabular (Alveolites, Phillipsastrea, Thecostegites), branching (Thamnopora, Senceliaepora) or fasciculate (Thamnophyllum); solitary rugose corals are also present. Receptaculites is locally abundant. This facies is locally interbedded with more argillaceous bioclastic units rich in crinoids, coral fragments and brachiopods. Enigmatic structures consisting of decimetre-thick pockets or beds of dark grey radiaxial cement containing brachiopods and crinoids occur (Fig. 4F). These particular structures (called 'nebuloids', cf. Boulvain 2001) may pass laterally by reduction in the proportion of cement, into a network of centimetre-scale stromatactis or fenestrae. Girvanella and Rothpletzella (here interpreted as cyanobacteria) form partial coatings around particles, and peloids are common and irregular.
Grey limestone with algae, fenestrae, branching tabulate corals and brachiopods (facies Pro4). This facies is devoid of stromatactis. Microfacies are characterized by common peloids and encrusting cyanobacteria. Coatings are generally composite, consisting of an association of various algae and bryozoans. Green algae (Trelonella, Radiosphaeroporella) are abundant (Mamet & Boulvain 1992) (Fig. 4H).
Grey limestone with corals, stromatoporoids, microbial mats, thrombolites (facies Pro5). This facies forms massive limestone with stylolites. Decimetre- to metre-scale growth cavities cemented by granular spar are abundant. Breccia is locally present. The fauna is dominated by
Like that from late Frasnian mounds, the following descriptions of the middle Frasnian buildup facies are organized bathymetrically, from deeper to shallower environments.
Grey, pinkish or greenish limestone, with stromatactis, corals and stromatoporoids (facies A3 and L3). This facies is composed of wackestones and floatstones showing decimetre-scale stromatactis and centimetre-scale stromatactoid fenestrae, with abundant branching tabulate corals, brachiopods and crinoids (Fig. 5C, F). Locally, there are massive or tabular (rarely dendroid) stromatoporoids, bryozoans, peloids and fasciculate rugose corals. Some subordinate cricoconarids, palaeosiphonocladale algae and calcispheres are present. Coatings (Rothpletzella) are rarely developed. Many of the fenestrae correspond to cavities situated below a lamellar organism (umbrella effect) or to growth cavities. Local reworking and concentration of bioclasts by storm action might result in this facies evolving into a bioclastic rudstone. This facies resembles late Frasnian Pm3, but with less hematitic pigment.
Grey limestone with algae, fenestrae, branching tabulate corals, stromatoporoids and brachiopods (facies A4 and L4). This facies is composed of rudstones, grainstones and floatstones, with peloids, lithoclasts (fragments of coating), branching tabulate corals coated by Rothpletzella, brachiopods, some crinoids, dendroid stromatoporoids, radiospheres and calcispheres. Locally, some Udotaeacea occur. Stromatactoid fenestrae and stromatactis are present. This facies corresponds to the first occurrence of Udotaeacea, together with the development of
FRASNIAN CARBONATE MOUNDS very thick and symmetrical coatings. This facies is similar to Pm4.
Grey microbial limestone (facies A5 and L5). This facies is composed of thrombolitic and stromatolitic bindstones and baffiestones, with Renalcis, stromatoporoids, tabulate corals, some Udotaeacea, brachiopods, bryozoans and rugose corals (Fig. 5G). Thick coatings of Rothpletzella alternate with encrusting microbial laminae rich in peloids. This facies is commonly found associated with facies A3-L3 or A4-L4, as isolated or coalescent metre-scale lenses. This facies is similar to Pm5 Grey limestone with dendroid stromatoporoids (facies A6 and L6). This facies is composed of rudstones, floatstones or grainstones that are rich in peloids, lithoclasts and dendroid stromatoporoids (mainly Amphipora) (Fig. 5H). These latter components are thickly coated by Rothpletzella or microbial laminae; the coatings are symmetrical. Calcispheres, palaeosiphonocladales (Issinella, Proninella) and Udotaeacea (Trelonella?) occur. Branching tabulate corals, gastropods and crinoids are present in places. Irregular fenestrae occur in matrix-rich zones.
Grey laminar fenestral limestone (facies A7 and L7). This facies is composed of grainstones and wackestones, with peloids, lithoclasts, calcispheres and palaeosiphonocladales. There are abundant millimetre-length fenestrae scattered throughout the sediment, bedding-parallel (Fig. 5E, I). Locally, there are dendroid stromatoporoids, which are commonly thickly coated.
Bioturbated grey limestone (facies A8 and L8). This facies is composed of wackestones and mudstones with palaeosiphonocladales, calcispheres and peloids. There is commonly evidence of bioturbation: open, vertical burrows filled by pseudosparitic-sparitic cement. Branching tabulate corals, stromatoporoids, ostracods and gastropods are present.
Off-mound and flanking facies The following descriptions of the off-mound and flanking facies are organized according to content and to grain size.
Shale, nodular shale and argillaceous limestone. This facies includes green-brown shale with sporadic limestone nodules. Nodules are of centimetre-decimetre scale and commonly are irregular. In some cases, nodules coalesce to
131
produce irregular beds of greenish-grey argillaceous limestone containing brachiopods and crinoids. Sponges (lyssakine hexactinellids, and tetractinomorph demosponges) are locally abundant.
Microbioclastic packstones. This facies is composed of thin-bedded, dark, commonly argillaceous, fine-grained (c. 100 ~tm) bioclastic packstones, containing brachiopods, crinoids, rugose corals, tabulate corals, fenestellids, ostracods, trilobites, peloids and cricoconarids. Locally, there are laminar stromatoporoids. Deformative bioturbation is commonly intense. The development of microsparite in the matrix is typically more intense than in all other facies and seems to be related to a higher clay content. Crinoidal packstones and grainstones. This dark grey limestone with crinoids and bioclasts forms decimetre-thick lenses or beds, with a slightly undulating planar upper surface and an erosive lower surface. Upper surfaces locally were used as substrates by corals. Bioclastic packstones and grainstones. This facies is composed of dark, centimetre to decimetrethick beds of rudstones, packstones and grainstones. It forms isolated lenses within the microbioclastic facies or within shales. The bioclasts are the same as in the microbioclastic facies, but are coarser grained (c. 500 ~tm). Some lithoclasts, radiospheres and calcispheres are present. Hummocky cross-lamination is developed in places.
Packstones and grainstones with peloids and lithoclasts. This facies is composed of packstones and grainstones. It differs from the last facies in containing abundant and commonly sorted peloids and lithoclasts, with some subordinate crinoids, fragments of corals and stromatoporoids, brachiopods and bryozoans. The grain size is approximately 300 ~tm.
Facies interpretation Below all the carbonate mounds was a soft argillaceous bioturbated substrate that was colonized by sponges, corals, branching bryozoans and some crinoids (Figs 5D & 6). The environment was situated below the photic and wave zones. Sediment had an oxic character (Boulvain 1993). This type of substrate is rather surprising if compared with published reports in which crinoidal sands (Burchette 1981), hardgrounds (Walker & Alberstadt 1975), lamellar corals (Maurin et al. 1981) or breccia (Mountjoy & Riding 1981) are common mound substrates.
132
F. B O U L V A I N
FRASNIAN CARBONATE MOUNDS The deepest mound facies is red limestone with stromatactis (Pml). Bourque & Boulvain (1993) concluded that stromatactis formed from sponge degradation in a relatively coherent, gel-like sediment. Cavities left after degradation evolved within sediment by collapse of the upper part and internal sedimentation on the base (Wallace 1987). It seems noteworthy that, in spite of the presence of sponges in argillaceous limestone below the carbonate mounds, no stromatactis were observed. In the late Frasnian Petit-Mont Member, the intensity of pigmentation decreases gradually from the red limestone with stromatactis (Pml) to grey microbial limestone (Pm5). The surrounding argillaceous facies are devoid of ferruginous pigment and are low in pyrite. Two hypotheses are possible to account for the presence of this pigment: (1) trapping of hematite detrital particles on the carbonate mounds (Lecompte 1936); or (2) local production, possibly of microbial origin (Monty et al. 1982; Boulvain et al. 2001). In the red matrix, a micrometre-scale hematitic pigment occurs among crystals of microspar. In early cemented cavities, the pigment forms possibly organically precipitated coccoids 5-10 gm in diameter and threads 1-3 gm in diameter (Fig. 4E). These structures can be referred to Siderocapsa-like and Sphaerotilus-Leptothrix-like iron bacteria (Boulvain et al. 2001). It is likely that ironoxidizing bacteria also were present in the matrix, but that they were later partially destroyed during matrix neomorphism. The ecology of recent iron-oxidizing bacteria is wellknown (Van Veen et al. 1978; Ghiorse 1984). They develop in environments where iron is available in reduced form, but where the redox potential is sufficiently high so that oxidation can occur. This is the case for oxygen-poor sediment or water (Pringsheim 1952; Nealson 1983). It can also be related to microenvironments (e.g. within pellets) where iron-bacteria are associated with
133
other micro-organisms that maintain a low 0 2 concentration (Cowen & Silver 1984). No lateral zonation of buildups occurred during the development of Pml (Fig. 7). This suggests a certain isotropy of the sedimentary environment (few currents, no significant vertical gradient). Absence of algae and the prevalence of muddy facies indicates a quiet environment located below photic and wave action zones. The transition between the argillaceous limestone below the mounds and this facies is abrupt and is accompanied by a reduction in biological diversity. However, the establishment of the stromatactis facies was not associated with any change in laterally equivalent facies. The development of Pml facies could point to a local hypoxic environment, perhaps following a strong increase in organic productivity. This hypothesis is supported by decreasing bioturbation in red limestone with stromatactis. The elimination of endofauna suggests locally reducing sediment. The ubiquitous organisms present below the mounds would be replaced by a sponge-microbe community whose only fossilized representatives would be stromatactis and iron-oxidizing bacteria. Byers (1977) indicated that sponges can live hypoxically. The organic community in red limestone with stromatactis, corals and crinoids (Pm2) is more diverse than that in the underlying facies. Pm2 facies was primarily muddy and the presence of delicate branching forms and some partial encrustations indicates low energy. Sponges were abundant, but large stromatactis are rare. The less homogeneous character of the facies could explain the replacement of the large cavities by networks of small stromatactis in the zones richest in grains. Cemented cavities located under lamellar corals could have played the role of keystones, vertically limiting the collapse of the roof of the cavities left by the degradation of sponges. This facies does not show any lateral zonation. A low-energy environment below the photic zone is suggested.
Fig. 5. (A) Photo mosaic giving a complete NE-SW panorama of the middle Frasnian Lion mound (Lion quarry, Lion Member, Frasnes). The highest point of the quarry is nearly 40 m high. (B) Middle part of the middle Frasnian Arche mound (Arche quarry, Frasnes), showing grey algal and microbial bindstones and bafflestones (facies A4-A5). The stratification is nearly horizontal and the high of the quarry wall reaches 20 m. (C) Lower part of the middle Frasnian Arche carbonate mound (Arche quarry, Frasnes), characterized by red coverstones with stromatactis and shelter cavities, zebra, tabulate corals, crinoids, brachiopods and stromatoporoids (facies A3). (D) Large amounts of Disphyllum corallites in shale, forming the substrate of the Arche mound (middle Frasnian Arche Member, Frasnes). (E) Laminar fenestral limestone (facies L7); La Boverie quarry, Jemelle, middle Frasnian Lion Member. (F) Wackestone with stromatactoid fenestra, crinoids and brachiopods (facies A3); thin section B209, normal light; La Boverie quarry, Jemelle, middle Frasnian Arche Member. (G) Bafflestone with thrombolites and Renalcis (facies L5); thin section N46b, normal light; Nord quarry, middle Frasnian Lion Member. (H) Floatstone with dendroid stromatoporoid (facies L6); thin section B407b, normal light; La Boverie quarry, Jemelle, middle Frasnian Lion Member. (I) Fenestral intraclastic packstone (facies L7); thin section B46, normal light; La Boverie quarry, Jemelle, middle Frasnian Lion Member.
134
F. BOULVAIN
Fig. 6. Simplified logs of the middle Frasnian Arche and Lion members in the La Boverie quarry and of the late Frasnian Petit-Mont Member in the Les Wayons quarry.
FRASNIAN CARBONATE MOUNDS
135
Fig. 7. Sedimentological model of the late Frasnian Petit-Mont mounds in the Philippeville Anticlinorium, with third-order sequential canvas.
Fig. 8. Sedimentological model of the middle Frasnian Arche and Lion mounds along the south side of the Dinant Synclinorium, with third-order sequential canvas for the Lion Member.
136
F. BOULVAIN
The pink or grey limestone with corals, crinoids, brachiopods, stromatactis, fenestrae and stromatoporoids (Pm3 and A3-L3) is faunally more diverse than the surrounding lateral facies. The presence of cyanobacteria suggests deposition close to the photic zone. The sediment is essentially microsparitic, but locally packstone and rudstone pockets occur, indicating temporary increases in turbulence (Fig. 4F). Grey limestone with algae, fenestrae, branching tabulate corals and brachiopods (Pm4 and A4-L4) contains the most diverse organic community. Green algae, algal peloids and symmetrical coatings are common. Intraclasts indicate synsedimentary lithification. The coarse texture and presence of green algae indicate a shallow environment, in the photic and wave action zones. The total disappearance of hematite (Fig. 5B) could be related to better oxygenation of water, suppressing iron-oxidizing bacteria communities. Skeletal organisms in grey limestone with corals, stromatoporoids, thrombolites and microbial mats (Pm5 and A5-L5) display subspherical or encrusting morphologies. Algal and microbial coatings are complex, thick and symmetrical. This is best explained by a more agitated environment. A reduction in organic diversity is also observed. This facies is also characterized by the importance of 'structure grumeleuse'. In the absence of obvious sponges, these structure are generally interpreted as microbial (cf. Pratt 1982; Tsien 1985). Perforations and lithoclasts show that the mats were synsedimentarily cemented. The A6-L6 facies is characterized by its lithoclastic character, the abundance of dendroid stromatoporoids and the dominant grainstone texture, with possible graded bedding. This facies resembles the 'Amphipora floatstone and rudstone' from the subtidal facies association of the Miette and Ancient Wall buildups (Whalen et al. 2000). It corresponds to an environment located above fair-weather wave base, with possible evolution to restricted conditions marked by a relatively low faunal diversity. This Amphipora-rich facies is also observed in debris-flow beds deposited on the flanks of Lion Member mounds, especially in the fore-mound environment (Fig. 8). In the upper central parts of the mounds, facies A6-L6 shows a progressive transition to fenestral limestone rich in peloids, calcispheres and palaeosiphonocladales (A7L7). This facies is very similar to the 'laminite facies' from the reef-fiat interior of the classic Frasnian Golden Spike reef complex, Alberta (Mc Gillivray & Mountjoy 1975) or to the 'peloidal packstones and grainstones' from the
peritidal facies association of the Miette and Ancient Wall buildups (Whalen et al. 2000). For Hopkins (1972), this facies was deposited in a lagoonal environment. In the Belgian buildups, this very shallow facies develops in a moderately restricted intertidal area. The last facies (A8-LS) is very fine grained and is deposited in a relatively shallow quiet subtidal lagoonal environment. Flank beds of Petit-Mont Member mounds are rich in crinoidal packstone and grainstone. These crinoids were probably indigenous. This type of community has commonly been reported from the flanks of Palaeozoic mounds (Burchette 1981; Pratt 1982). These communities developed when the mounds had relief; i.e. starting from the base of the upper part of the Petit-Mont mounds. Crinoidal beds with planar tops and undulating bases resulted from the reworking of the bioclastic material and its transport downslope. The bedded bioclastic-lithoclastic facies of the Lion and Arche members result from the input of eroded material exported directly from the buildups by gravity flows (Stoakes 1980) or from the reworking and sorting of already deposited material by storm waves (Humblet & Boulvain 2001). Microbioclastic packstones are characterized by an open-marine facies with brachiopods, bryozoans and crinoids, whereas bioclastic rudstones and lithoclastic packstones and grainstones show a clear buildup influence through abundant input of bioclastic and lithoclastic material. These flank facies are similar to the bioclastic-lithoclastic fore-reef strata of the Frasnian carbonate buildups from the Leduc Formation (Mc Gillivray & Mountjoy 1975) or from the Miette and Ancient Wall buildups (Whalen et al. 2000).
Architecture and sediment dynamics of the buildups Middle Frasnian Arche and Lion members are relatively large buildups, more than 150 m thick and nearly 1 km in diameter (Fig. 5A). Late Frasnian Petit-Mont mounds are smaller limestone bodies, 60-80 m thick, with a diameter of 150-250 m (Fig. 4A). All the buildups are included in argillaceous limestone, nodular shale or shale. Contradictory inferences about the initial mechanical state of carbonate mound mud appear to derive from field observations. The persistence of dips as high as 35 ~ on the flanks of several mounds, the presence of lithoclasts in the grey limestone (Pro5, A5-L5) and the sharp distinct character of some fractures indicate early lithification. Conversely, plastic deformation
FRASNIAN CARBONATE MOUNDS of the sediment, presence of overturned coral colonies, formation of zebra structures by lateral compression, scarcity of hardgrounds and of sediment borings, and the irregular character of some synsedimentary fractures indicate an absence of early lithification. It appears that the sediment was initially sufficiently ductile to permit synsedimentary deformation, yet sufficiently coherent to have maintained open cavities and significant relief. It is likely that the sediment had a gel-like consistency, probably related to the presence of significant quantities of organic matter.
Late Frasnian Petit-Mont Member Interpretation of facies succession. The succession of facies in the Petit-Mont Member (Figs 6 & 7) mounds poses an interpretive problem. Two principal models have been proposed to explain the evolution of communities of constructing organisms: the autogenic model of Walker & Alberstadt (1975), and the allogenic model inspired from the work of Lecompte (1959) and Hoffman & Narkiewicz (1977). Lecompte considered that the succession observed in the carbonate mounds of the Petit-Mont Member corresponded to an adaptation of communities to decreasing depth, from intermittent agitation to turbulence. The more general model of Walker & Alberstadt (1975) distinguished a succession of pfiases in the evolution of mound communities from stabilization of the substrate until domination by some very specialized species. In the case of the Petit-Mont Member, the colonization of the substrate is related to sponges (Pml). The diversification phase corresponded to the establishment of an assemblage of corals, crinoids, brachiopods, stromatoporoids and cyanobacteria (Pm3), followed by branching tabulate corals, brachiopods and green algae (Pro4). The domination stage is marked by the appearance of grey limestone with corals, stromatoporoids, thrombolites and microbial mats (Pm5), with general encrusting morphology. In off-mound environments, however, no apparent change in facies and assemblages was observed. This appears to exclude the intervention of allogenic processes in the evolution of mound communities, but off-mound communities are mainly made up of generalists of which the sensitivity to variations of oceanic parameters is less than that of more specialized organisms (Walker & Alberstadt 1975). Furthermore, various sedimentological features indicate an increase in turbulence and luminosity when passing from the base of the mounds to grey limestone (Pm5). In the
137
internal zones of the platform, the shallowingupwards succession of facies was a response to a relative sea-level fall (Boulvain 1993, 2001). Pm5 development is also accompanied by an increase in the diameter of the mounds and by progradation of lateral facies Pm3 down flanks. These arguments suggest that at least the development of the grey algal-microbial core (Pm5) in late Frasnian Petit-Mont Member mounds was related to a relative sea-level fall. If ecological evolution (Pml-Pm5) of the Petit-Mont mounds is related to bathymetry, a similar interpretation for the opposite sequence (Pm5-Pm2), which caps, after a hardground, the largest mounds, could be suggested. This sequence, related to increasing depth, is accompanied by an upwards reduction in the diameter of the carbonate mounds (Fig. 7). A specialized community of corals, stromatoporoids, thrombolites and microbial mats (Pro5) was replaced by a community of sponges, corals and crinoids (Pm2). However, carbonate production was not able to compensate for the rise in sea level, and the last beds of the mounds, already very argillaceous, were covered by argillaceous limestone and nodular shale. This transgressive sequence is marked in the internal zones of the platform by the disappearance of oncoid shoals and overall deposition of argillaceous sediment.
Palaeobathymetry. Grey limestone with fenestrae, branching tabulate corals and brachiopods (Pm4) developed when mounds reached the wave action and photic zones. Depth of the wave action and photic zones may be related to the geometrical characteristics of the basin, and to climatic and other parameters. About 30 m is used here as a base (Fliigel 2004). Using this depth of development for facies Pm4, it is possible to estimate the depth of development of red limestone with stromatactis (Pml) by knowing the average thickness of decompacted sediments separating both facies. An average rate of compaction of 1.5 (Boulvain 1993,2001) was estimated by the method of Beaudoin et al. (1987). Calculation gives a value of about 100-150 m for water depth during deposition of the base of the mounds. In the Upper Devonian of Alberta, Stoakes (1980) related hypoxic to anoxic sediments that result from basin starvation at comparable depths. Middle Frasnian Arche and Lion members Lateral facies. Microbioclastic packstones mainly occur in off-mound facies. In this facies the influence of reefs on the sediment budget remains relatively low. On the other hand, the lithoclastic grainstones and bioclastic rudstones
138
F. BOULVAIN
are facies where extensive supply of reefal debris is significant. This reefal input consists of bioclastic-lithoclastic sediment reworked from the mound and deposited by debris flows showing decimetre-deep basal erosion structures (Humblet & Boulvain 2001). The base of the buildups. The buildups began with the development of large coral colonies (fasciculate rugose corals in the famous Arche quarry) on a muddy sea floor, followed by the progressive colonization of this substrate by sponges and, finally, more intense microbial (?) carbonate production in the form of centimetric
Whalen et al. 2000). Moreover, the general size of some Alberta mounds, although variable and dependant on the subsidence rate, is similar to that of the Belgian mounds; for example, the Golden Spike Leduc reef is 3.6 x 2.4 km in area and 182m in height, according to Mountjoy (1980). However, an important difference between Belgian and Alberta buildups lies in the fact that the latter are characterized by a welldeveloped rim of stromatoporoid biostromes (Whalen et al. 2000), whereas the Belgian buildups show only relatively mud-rich mound-type facies.
Mound types and palaeoceanography Late Frasnian carbonate mounds from the Petit-Mont Member offer an ideal case study of the architecture of carbonate mounds according to bathymetric evolution. Above a substrate of argillaceous limestone rich in sponges, corals, brachiopods and bryozoans, red limestone with stromatactis (Pml) initiated mound development. This facies was produced by a community of sponges and iron-bacteria, below the photic and wave action zones, in a hypoxic environment at a depth of 100-150 m. Above this, a transition towards red limestone with stromatactis, corals and crinoids (Pm2), then to pink limestone with corals, crinoids, brachiopods, stromatactis, fenestrae and stromatoporoids (Pm3) occurred. The photic zone was reached for cyanobacteria. Grey limestone with fenestrae, branching tabulate corals and brachiopods (Pm4) developed in the wave action zone. At the same time, grey limestone with corals, stromatoporoids, thrombolites and microbial mats was deposited in the mound core (Pm5), marking the domination of encrusting organisms. This general facies succession records a third-order sequence pattern: a highstand system tract (Pro 1-Pro4 facies succession), followed by a lowstand (Pro5) and a rapid transgressive system tracts (Pm3-Pml facies succession), responsible for the final drowning of the mounds (Fig. 7). Inshore time-equivalent facies belts are ramp-type sediments, dominated by argillaceous limestone during highstand systems (HST) and transgressive systems tracts (TST) and oncoidal shoals during a lowstand systems tract (LST) (Boulvain 1993, 2001). The facies of Petit-Mont Member mounds are indicators of palaeobathymetry. They also point to a particular palaeoceanography. They recorded hypoxic environments at relatively shallow depth preceding a large-scale anoxic event: the Lower Kellwasser event (Copper 2002a, b).
FRASNIAN CARBONATE MOUNDS
139
Fig. 9. Schematic log of Frasnian formations along the south side of the Dinant Synclinorium with sea level curve and conodont zonation (after Boulvain et al. 1999).
Can we consequently consider this type of carbonate mound as an indicator of hypoxic conditions on a large scale? Other case studies could test this hypothesis. Palaeotemperature history of the Frasnian shows that this very warm greenhouse climatic period is interrupted by two short-term cooling events, evidenced by positive excursions in 6~80 of conodont apatite in the Late rhenana Zone and at the Frasnian-Famennian boundary (Joachimski et al. 2004). The first cooling pulse coincides with the Lower Kellwasser and could be contemporaneous with the drowning of the Petit-Mont Member mounds. On the other hand, middle Frasnian carbonate mounds record different oceanographic conditions. They pass landwards to a 'normal' carbonate platform, with well-developed lagoonal complexes and barrier reefs (Da Silva & Boulvain 2002, 2004). More specifically, by comparison with recent models of atoll development in response to eustatic variations (Warrlich et al. 2002), a dynamic interpretation is suggested for the geometry and succession of sedimentary units in the middle Frasnian Lion and Arche members. After the growth of the lower part of the mounds during a transgression (Fig. 9), a clear progradation is recorded by fore-mound
sedimentation of reworked material. Lower sea level then restricted reef growth to downslope positions only, culminating in the development of a circular reef margin during the following transgressive stage. The presence of relatively restricted facies is therefore possibly the result of a balance between sea-level rise and reef growth. A third-order sequence subdivision of the Lion mounds and their lateral sediments is proposed, based on the geometry and bathymetry of the sedimentary bodies (Fig. 8). The lower and middle parts of the buildups correspond to the succession of a transgressive systems tract (TST1) and/or a highstand systems tract (HST1), with strong progradation associated with reduced accommodation occurring during the HST. The mound development during the lowstand systems tract (LST1) was restricted to the margin of the buildups, with possible emergence and synsedimentary lithification (Sandberg et al. 1992). This lowering of sea level was recorded in the internal platform by subsequent widespread development of palaeosols in the upper part of the Lustin Formation (Da Silva & Boulvain 2002). The development of an atoll-like margin corresponds to the TST2, with significant lateral facies differentiation between fore-mound and mound lagoon.
140
F. BOULVAIN
Conclusion The sequential canvas proposed here for middle and late Frasnian carbonate mounds gives rise to the following observations: TSTs may correspond to various unit types according to the developmental stage of the buildups and to different rates of sea-level rise. They may correspond to aggrading 'deep' mound facies (Lion TST1), to shallow facies developing in a lagoon (Lion TST2) or to give-up type sequences (PetitMont TST4) (Neumann & Macintyre 1985). Highstand systems tracts could correspond to catch-up aggrading sequences (Petit-Mont HST3) or to prograding stages with high rates of sediment exportation (Lion HST1). The LSTs could be poorly developed, implying low accommodation and a temporary emergence of the buildups (Lion LST1) or could correspond to the maximum development of the mounds (Petit-Mont LST3). This overall picture suggests that the main differences between the Arche or Lion members and the Petit-Mont Member are a consequence of different palaeoceanographic setting. The middle Frasnian Arche and Lion members are large flattened buildups showing limited vertical differentiation, large-scale progradation features, extensive exportation of material towards off-reef environment and development of inner lagoonal facies. They grew offshore from a welldeveloped carbonate platform with a healthy carbonate factory. Middle Frasnian sea-level fluctuations were relatively mild, and sedimentation was able to keep up with sea-level rise. At the opposite end of this spectrum, during the late Frasnian, severe eustatic rises (Johnson et al. 1985), together with rising oceanic hypoxic conditions (Copper 2002a, b), were responsible for collapse of the carbonate factory, drowning of the middle Frasnian carbonate platform and development of Petit-Mont type mounds: buildups with relatively limited lateral extension, vertical facies differentiation, low potential for material exportation and high content in microaerophilic iron bacteria. I am grateful to all those who shared their remarks and observations when visiting the Belgian Frasnian mounds. Suggestions from G. Webb and G. Racki improved this paper greatly.
References BEAUDOIN,B., COJAN, I. ETAL. 1987. Mesure directe de la compaction dans les s6diments. In: AISSAOUI, D. M. er Ar~.(eds) Genkse et Evolution des Bassins
skdimentaires. Notes et MOmoires Total CFP, 21,
235-247. BOULVArN, F. 1993. Skdimentologie et diagenkse des monticules micritiques 'F2j' du Frasnien de l'Ardenne.
Service g6ologique de Belgique, Professional Papers, 260. BOULVAIN,F. 2001. Facies architecture and diagenesis of Belgian late Frasnian carbonate mounds (Petit-Mont Member). Sedimentary Geology, 145, 269-294. BOULVAIN,F., BULTYNCK,P. Er AL. 1999. Les Formations du Frasnien de la Belgique. Memoirs of the Geological Survey of Belgium, 44. BOULVAIN,F., DE RIDDER, C., MAMET,B., PRI~AT,A. & GILLAN,D. 2001. Iron microbial communities in Belgian Frasnian carbonate mounds. Facies, 44, 47-60. BOULVAIN,F., CORNET,P. ET AL. 2004. Reconstructing atoll-like mounds from the Frasnian of Belgium. Facies, 50, 313-326. BOULVAIN,F., DEMANY,B. & COEN-AUBERT,M. 2005. Frasnian carbonate buildups of southern Belgium: the Arche and Lion members interpreted as atolls. Geologica Belgica, 8, 69-91. BOSENCE, D. W. J. & BRIDGES, P. 1995. Origin and evolution of carbonate mud-mounds. In: MONTY, C. L. V., BOSENCE, D. W. J., BRIDGES, P. H. & PRATT, B. R. (eds) Carbonate Mud-Mounds: Their Origin and Evolution. International Association of Sedimentologists, Special Publications, 23, 3-9. BOURQUE, P.-A. 1997. Paleozoic finely crystalline carbonate mounds: cryptic communities, petrogenesis and ecological zonation. Facies, 36, 250-253. BOURQUE,P.-A. & BOULVAIN,F. 1993. A model for the origin and petrogenesis of the red stromatactis limestone of Paleozoic carbonate mounds. Journal of Sedimentary Petrology, 63, 607-619. BURCHETTE, T. P. 1981. European Devonian reefs: a review of current concepts and models. In: TOOMEY, D. F. (ed.) European Fossil Reef Models. Society of Economic Paleontologists and Mineralogists, Special Publications, 30, 85-142. BYERS,C. W. 1977. Biofacies patterns in euxinic basins: a general model. In: CooK, H. E. & ENOS, P. (eds) Deep-water Carbonate Environments. Society of Economic Paleontologists and Mineralogists, Special Publications, 25, 5-17. CAYEUX, L. 1935. Les Roches Skdimentaires de France. Roches
Carbonat~es
(Calcium
et
Dolomies ) .
Masson, Paris. COPPER, P. 2002a. Reef development at the Frasniard Famennian mass extinction boundary. Palaeogeography, Palaeoclimatology, Palaeoeeology, 181, 27-65. COPPER, P. 2002b. Silurian and Devonian reefs: 80 million years of global greenhouse between two ice ages. In: KIESSLING,W., FLOGEL,E. & GOLONKA,J. (eds) Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 181-238. COWEN, J. P. & SILVER, M. 1984. The association of iron and manganese with bacteria on marine macroparticulate material. Science, 224, 1340-1342.
FRASNIAN CARBONATE MOUNDS DA SILVA,A.-C. & BOULVAIN,F. 2002. Sedimentology, magnetic susceptibility and isotopes of a middle Frasnian carbonate platform: Tailfer section, Belgium. Facies, 46, 89-102. DA SILVAA.-C. & BOULVAIN,F. 2004. From palaeosols to carbonate mounds: facies and environments of the middle Frasnian platform in Belgium. Geological Quarterly, 48, 253-266. DELHAYE, F. 1913. Etude de la formation des r6cifs de calcaire rouge ~ Acervularia et Hypothyris cuboides. Annales de la SociSt~ gkologique de Belgique, XL, B469-481. DEWALQUE, G. 1868. Prodrome d'une description gkologique de la Belgique. Carmanne, Liege. DUMON, P. 1932. Compte rendu de l'excursion du 11 juin 1932 aux carri6res de marbre rouge de Vodel6e et de Soulme. Bulletin de la SociktO beige de Gkologie, XLII, 118-128. FLOGEL, E. 2004. Microfacies of Carbonate Rocks. Analysis, Interpretation and Application. Springer, Berlin. GHIORSE, W. C. 1984. Biology of iron- and manganesedepositing bacteria. Annual Reviews of Microbiology, 38, 515-550. HOFFMAN, A. & NARKIEWICZ,M. 1977. Developmental pattern of Lower to Middle Paleozoic banks and reefs. Neues Jahrbuch fiir Geologie und Paliiontologie Abhandlungen, 5, 272-283. HOPKINS, J. C. 1972. Petrography and distribution and diagenesis of foreslope and basin sediments, Miette and Ancient Wall carbonate complexes', (Devonian), Alberta. Doctoral thesis, McGill University, Montreal. HUMBLET, M. & BOULVAIN, F. 2001. Sedimentology of the Bieumont Member: influence of the Lion Member carbonate mounds (Frasnian, Belgium) on their sedimentary environment. Geologica Belgica, 3,97-118. JOACHIMSKI, M. M., VAN GELDERN, R., BREISIG, S., BUGGISH, W. & DAY, J. 2004. Oxygen isotope evolution of biogenic calcite and apatite during the Middle and Late Devonian. International Journal of Earth Sciences, 93, 542-553. JOHNSON, J. G., KLAPPER,G. & SANDBERG,C. A. 1985. Devonian eustatic fluctuations in Euramerica. Geological Society of America Bulletin, 96, 567-587. LECOMPTE, M. 1936. Contribution ~t la connaissance des 'r&ifs' du Frasnien de l'Ardenne. MOmoires de l'Institut gbologique de l'Universitk catholique de Louvain, X, 30-113. LECOMPTE, M. 1954. Quelques donn6es relatives h la gen6se et aux caract+res 6cologiques des 'r6cifs' du Frasnien de l'Ardenne. In: Volume jubilaire Victor van Straelen, I, 153-181. LECOMPTE, M. 1959. Certain data on the genesis and ecologic character of Frasnian reefs of the Ardennes (transl. by P.F. Moore). International Geological Review, 1, 1-23. ME GILLIVRAY,J. G. & MOUNTJOY,E. W. 1975. Facies and related reservoir characteristics Golden Spike reef complex, Alberta. Bulletin of Canadian Petroleum Geology, 23, 753-809 MAILLIEUX, E. 1913. Nouvelles observations sur le Frasnien et en particulier sur les pal6or6cifs de la
141
plaine des Fagnes. Bulletins de la SociktO beige de G~ologie, XXVII, 6%104. MAILLIEUX,E. 1926. Contribution/~ l'6tude du 'Massif' de Philippeville. Bulletin de la Socibtb beige de Gkologie, Palbontologie et Hydrologie, 36, 86-112. MAMET, B. & BOULVAIN, F. 1992. Microflore des monticules frasniens 'F2j' de Belgique: Revue de Micropalkontologie, 35, 283-302. MAURIN, A. F., PHILIP, J. & BRUNEL,P. 1981. Possible microbial accretions in Cenomanian mounds, S.E. France. In: MONTY, C. L. V. (ed.) Phanerozoic Stromatolites. Springer, Berlin, 121-133. MONTY, C. L. V. 1995. The rise and nature of carbonate mud-mounds: an introductory actualistic approach. In: MONTY, C. L. V., BOSENCE,D. W. J., BRIDGES, P. H. & PRATT, B. R. (eds) Carbonate Mud-mounds: Their Origin and Evolution. International Association of Sedimentologists, Special Publication, 23, 11-48.
MONTY, C. L. V. & VAN LAER, P. 1988. The Upper Devonian mud mounds from the south western Dinant Synclinorium. In: HERBOSCH, A. (ed.) 9th IAS European Regional Meeting on Sedimentology, Excursion Guidebook, Ministry of Economic Affairs, Brussels, 157-176. MONTY, C. L. V., BERNET-ROLLANDE, M. C. & MAURIN, A. F. 1982. Re-interpretation of the Frasnian classical 'reefs' of the southern Ardennes, Belgium. (Extended abstract). Annales de la Sociktk gOologique de BeIgique, 105, 339-341. MOUNTJOY, E. W. 1980. Some questions about the development of Upper Devonian carbonate buildups (reefs), Western Canada. Bulletin of Canadian Petroleum Geology, 28, 315-344. MOUNTJOY, E. W. & RIDING, R. 1981. Foreslope stromatoporoid-renalcid bioherm with evidence of early cementation, Devonian Ancient Wall reef complex, Rocky Mountains. Sedimentology, 28, 299-320. NEALSON, K. H. 1983. The microbial iron cycle. In: KRUMBEIN, W. E. (ed.) Microbial Geoekemistry. Blackwell Science, Oxford, 159-191. NEUMANN,A. C. & MACINTYRE,I. 1985. Reef response to sea-level rise: keep-up, catch-up or give-up. Proceedings of the Fifth International Coral Reef Congress, 3, 105-110. NEUWEILER, F., BOURQUE, P.-A. & BOULVAIN, F., 2001. Why is stromatactis so rare in Mesozoic carbonate mud mounds? Terra Nova, 13, 338-346. PRATT, B. R. 1982. Stromatolitic framework of carbonate mud-mounds. Journal of Sedimentary Petrology, 52, 1203-1227. PRATT, B. R. 1995. The origin, biota and evolution of deep-water mud-mounds. In: MONTY, C. L. V., BOSENCE, D. W. J., BRIDGES,P. H. & PRATT, B. R. (eds) Carbonate Mud-mounds: Their Origin and Evolution. International Association of Sedimentologists, Special Publications, 23, 49-123. PRINGSHEIM, E. G. 1952. Organismes ferrugineux. Endeavour, XI, 208-214. SANDBERG, C. A., ZIEGLER, W., DREESEN, R. & BUTLER, J. L. 1992. Conodont biochronology, biofacies, taxonomy and event stratigraphy around
142
F. BOULVAIN
Middle Frasnian Lion mudmound (F2h), Frasnes, Belgium. Courier Forschungsinstitut Senckenberg, 150, 1-87. STOAKES, F. A. 1980. Nature and control of shale basin fill and its effect on reef growth and termination: Upper Devonian Duverney and Ireton Formations of Alberta, Canada. Bulletin of Canadian Petroleum Geologists, 28, 345-410. TERMIER, H., TERMIER, G. & TSIEN, H. H. 1981. Spongiaires des calcaires rrcifaux du Frasnien de l'Ardenne. Bulletin de la Sociktk beige de Gkologie, 90, 287-298. TSIEN, H. H. 1975. Introduction to the Devonian reef development in Belgium. In: 2nd Symposium International sur les Coraux et rdcifs coralliens fossiles, Paris, livret-guide exc. C, Service Grologique de Belgique, Bruxelles, 3-43. TSIEN, H. H. 1980. Les rrgimes rrcifaux d6voniens en Ardenne. Bulletin de la Sociktk belge de Gdologie, 89, 71-102. TSIEN, H. H. 1985. Algal-bacterial origin of micrites in mud mounds. In: TOOMEY, D. F. & NITECKI, M. H. (eds) Paleoalgology. Contemporary Research and Applications. Springer, Berlin, 291-296.
VAN VEEN, W. L., MULDER, E. G. & DEINEMA, M. H. 1978. The Sphaerotilus-Leptothrix group of bacteria. Microbiological Review, 42, 329-356. WALKER, K. R. & ALBERSTADT,L. P. 1975. Ecological succession as an aspect of structure in fossil communities. Paleobiology, 1,238-257. WALLACE,M. W. 1987. The role of internal erosion and sedimentation in the formation of Stromatactis mudstones and associated lithologies. Journal of Sedimentary Petrology, 57, 695-700. WARRLICH, G. M. D., WALTHAM, D. A. & BOSENCE, D. J. W. 2002. Quantifying the sequence stratigraphy and drowning mechanisms of atolls using a new 3-D forward stratigraphic modelling program (CARBONATE 3D). Basin Research, 14, 379-400. WHALEN, M. T., EBERLI, G. P., VAN BUCHEM, F. S. P. & MOUNTJOY, E. W. 2000. Facies models and architecture of Upper Devonian carbonate platforms (Miette and Ancient Wall), Alberta, Canada). In: HOMEWOOD, P. W. & EBERLI, G. P. (eds) Genetic Stratigraphy on the Exploration and Production Scales- Case Studies From the Pennsylvanian of the Paradox Basin and the Upper Devonian of Alberta. Bulletins des Centres de Recherche Elf ExplorationProduction, Mrmoire, 24, 139-178.
Late Frasnian phillipsastreid biostromes in Belgium EDOUARD
POTY l & EMMANUEL
CHEVALIER 2
~Palkontologie animale et humaine, Universitk de Likge, Bdt. B18, Sart Tilman, B-4000 Likge, Belgium (e-mail:
[email protected]) 2Carmeuse s.a., Rue du Chgtteau, 13A, B-5300 Seilles, Belgium (e-mail: emmanuel, chevalier@carmeuse, be)
Abstract: In the Belgian Namur-Dinant Basin the boundary between the Lustin Formation and the Aisemont Formation (in the Lower rhenana conodont Biozone) corresponds to a fall followed by a rise in sea level, leading to the first recorded late Frasnian coral crisis. The Aisemont Formation records a transgressive-regressive cycle. Prior to the crisis most of the colonial rugose corals were members of the Family Disphyllidae, but these were largely replaced by corals belonging to the Phillipsastraeidae. Among these Frechastraeacolonized all environments of the basin and was the main constructor of a biostromal reef in its northernmost proximal area, in the fair-weather wave zone. Corals did not encrust each other and therefore were not firmly attached, but they hug tightly the substrate (a dead coral colony) and rest closely on it to resist to the turbulence of waves. During the Silurian and Devonian, up until the late Frasnian crisis, shallow-water reefs in turbulent water were usually built by encrusting stromatoporoids, whereas rugose corals were restricted to waters of lower energy. Indeed, they were unable to encrust substrates, unlike stromatoporoids and post-Palaeozoic scleractinians, and to live in turbulent habitats. In Belgium argillaceous sedimentation prevented the development of stromatoporoids and provided an opportunity for the corals to colonize empty niches and to construct biostromes in relatively high-energy environments. At the same time Alveolites and stromatoporoids were dominant in a mid-proximal environment below the fair-weather wave base, but within the storm wave zone, where they also constructed biostromes.
Rugosa were usually not able to encrust a substrate, therefore, in contrast to Scleractinia which are major contributors to reef building, rugose corals do not occur commonly in high waterenergy environments (Scrutton 1998) and their contribution to bioconstructions was usually limited as debris bound by encrusting organisms (stromatoporoids, sponges, algae, microbial communities, etc.) or to minor frame building in biostromal reefs developed in relatively lowenergy environments (Rodriguez 1996; Aretz 2002; Aretz & Chevalier 2007). Like Rugosa, most Tabulata were unable to encrust, except for some of them such as alveolitids, which were common binders in Middle Devonian and Frasnian bioconstructions. During the middle Frasnian, large bioherms developed within the argillaceous-dominant facies in distal parts of the N a m u r - D i n a n t Basin (see, for example, Boulvain et al. 1999, 2004) whereas a carbonate platform developed in proximal areas. These facies contain often rich coral and stromatoporoid faunas. In Belgium the first Frasnian ecological crisis occurred during the earliest late Frasnian, parallel to a shift in the sedimentary pattern on the platform where argillaceous-dominant facies
progressively developed. This crisis is marked by the extinction of the disphyllids, the main stock of colonial rugose corals, and the decline of most tabulate corals and stromatoporoids. The newcoming phillipsastreids, especially Frechastraea replaced the extinct disphyllids and colonized widely distal-proximal environments (CoenAubert 1994; Boulvain et al. 1999). Moreover, they were able to become constructors in environments where usual builders (stromatoporoids, Alveolites) were unable to live. Thus, at the beginning of the late Frasnian, they constructed biostromal reefs (lower unit of the Aisemont Formation). These biostromes and their relationships with other contemporaneous biota developed on the same ramp are studied here.
Geological setting and palaeontology The Belgian Frasnian (Fig. 1) is now exposed in the Dinant and N a m u r synclinoria and in the Vesdre area ('Vesdre Massif', 'Vesdre Nappe', 'Vesdre Synclinorium' in the papers, reflecting different tectonic conceptions). These Variscan structural units were part of the N a m u r - D i n a n t Basin during the time of deposition, in which the
fikLVARO,J. J., ARETZ, M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special
From:
Publications, 275, 143-161. 0305-8719107l$15.00 9 The Geological Society of London.
144
E. POTY & E. CHEVALIER
Fig. 1. Location map and geological setting of the studied sections.
recognized sedimentation areas and the succession of facies reflect a ramp setting, with breaks of slope (Bultynck & Dejonghe 2002; Boulvain et al. 2004; Da Silva & Boulvain 2004). The Aisemont Formation extends in the Namur Synclinorium, the northern part of the Dinant Synclinorium and the Vesdre area, which probably all belong to the same part of the ramp. It comprises two limestone units, more or less argillaceous, at the base (lower limestone member; 'first biostrome with Phillipsastrea' of Coen-Aubert & Lacroix 1979) and at the top (upper limestone member; 'second biostrome with Phillipsastrea' of Coen-Aubert & Lacroix 1979), and an intercalated shale unit (middle shale member). It is separated from the underlying Lustin Limestone Formation and from the overlying Lambermont Shale Formation by two disconformities corresponding to erosional transgressive surfaces (see Fig. 3b, c later). The Aisemont Formation records a transgressive-regressive cycle considered here as corresponding to a third-order sequence. The maximum of relative sea level is marked by the development of the argillaceous deposits of the middle member, which comprises a unit with dysaerobic-anaerobic facies characterized by
azoic or poorly fossiliferous levels. The top of the Lustin Formation and the lower limestone member of the Aisemont Formation are correlated with the Lower rhenana Conodont Biozone, and the upper member and the lower part of the Lambermont Formation with the Upper rhenana Biozone (Bultynck et al. 2000; Gouwy & Bultynck 2000). The intercalated shale member is not dated but, by correlation with the Philippeville Anticlinorium (see hereafter), its dysaerobic-anaerobic unit could be the record of the Lower Kellwasser Event (LKW; lower part of the Upper rhenana Biozone), which is situated in the lowest part of the Valisettes Formation (Bultynck et al. 1998). The LKW corresponds to a rise of anoxic waters on the platform (Schindler 1993), which consequently would be correlated with the maximum flooding surface of the 'Aisemont sequence'. Southwards, in the Philippeville Anticlinorium, the Lustin Formation and the lower limestone member of the Aisemont Formation pass laterally, respectively, into the carbonates of the Philippeville Formation and the shales of the Neuville Formation (Fig. 2). These are also separated by a disconformity corresponding to a erosion surface (Fig. 3a). The middle-upper
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
145
Fig. 2. Stratigraphy and correlation of the considered interval. Fm, Formation; A., Ancyrognathus; C.,
Cyrtospirifer; N., Navalacria; T., Tiocyrspis.
Fig. 3. Thin sections showing the sharp contact corresponding to a transgressive erosion surface, (a) between the Philippeville Formation and the Neuville Formation in the Neuville new railway section, and (b) & (c) between the Lustin Formation and the Aisemont Formation, respectively, in the Prayon and La Mallieue sections. Note in (a), a Hexagonaria sp. (a) fixed and growing on the surface, and in (c) a thamnoporid rugose coral (a) sharply cutted by the surface.
146
E. POTY & E. CHEVALIER
Frasnian boundary is considered as corresponding to this limit (Boulvain et al. 1999; Bultynck & Dejonghe 2002). The middle shale member and the upper limestone member correspond more or less to the Valisettes Shale Formation (Upper rhenana Biozone), which overlies the Neuville Formation. There is a diachronism between the bases of the Neuville and Aisemont formations as demonstrated by:
9 Ancyrognathus triangularis, which appears about 4 m above the base of the Neuville Formation at Neuville, but at the base of the Aisemont Formation (Boulvain et al. 1999); 9 the base of the Tiocyrspis bironensisCyrtospirifer condrusorum interval Brachiopod Zone, which is at 8.6 m above the base of the Neuville Formation, but corresponds to the base of the Aisemont Formation in the Condroz and in the Namur areas (Mottequin 2005); 9 the last representatives of the colonial Disphyllidae (Disphyllum, Hexagonaria) are present in the top of the Lustin Formation, but are not known in the Aisemont Formation where they are replaced by phillipsastraeids (Frechastraea, Phillipsastrea) from the base, whereas Hexagonaria, the last disphyllid, is found with the first Frechastraea, about 5 m above the base of the Neuville Formation (Coen & Coen-Aubert 1974). This diachronism between the bases of the Neuville Formation and of the lower limestone member of the Aisemont Formation, and the resulting gap, demonstrate a retrograding pattern of deposition and that the two units belong to a transgressive system tract (TST). Such a pattern is also recorded in the Lambermont Formation, whose lowest part is missing in the Vesdre area by comparison with the northern area of the Dinant Synclinorium (Mottequin 2005), and characterizes the transgression of the following sequence. The base of the Aisemont Formation is also considered here as diachronous and the lower limestone member as retrograding northwards (Fig. 2). The extinction of the Disphyllidae in the lowest part of the Neuville Formation marks the first Frasnian coral crisis in Belgium and surrounding areas (Poty 1999). The recovering coral fauna is dominated by newcomers not previously known in Belgium: Frechastraea (including F. pentagona and F. limitata), Phillipsastrea (P. ananas) and Hankaxis (H. insignis). It comprises also some uncommon species belonging to Peneckiella, Macgeea and Tabulophyllum, genera
which were previously present. The newcoming taxa appeared just after or together with the last representatives of the previous family, as recorded by Coen & Coen-Aubert (1974). But further investigations will be necessary to confirm and to precise the overlap of these two faunas. No extinction at the generic level has been recorded in tabulate corals. Alveolites remains very common; however, as with the stromatoporoids, the rest suffered a dramatic decline and became relatively uncommon compared to the Middle Frasnian Philippeville and Lustin formations. Indeed, the tabulate coral family Pachyporidae (mainly Thamnopora), as well as branching and bulbous stromatoporoids which were dominant in older limestone facies, became uncommon and never recovered a dominant position Therefore, this coral turnover is correlated with the beginning of the rise in sea level triggering the TST of the 'Aisemont sequence' the top of the Philippeville and Lustin formations (which followed the fall in sea-level marking). But it was not a result of the Lower Kellwasser Event sensu stricto, which occurred later (in the middle member of the Aisemont Formation) and induced strictly no extinction in corals, those found in the lower limestone member being still present in the upper member. Note that the beginning of the Aisemont and the Neuville formations can be correlated with the beginning of the argillaceous input on the platform, whereas argillaceous deposits were previously characteristic of the deep parts of the basin. Biostromes developed in the Namur Synclinorium and in the Vesdre area during the onset of the TST of the 'Aisemont sequence', but not in the other areas of the basin (Dinant Synclinorium). There are no biostrome units in the upper member, only coral beds, despite the fact that it has been known in the literature as the 'second biostrome' since the work of CoenAubert & Lacroix (1979). Similarly, the 'third biostrome' of the same authors is a limestone unit rich in corals within the Lambermont Formation.
Description of key sections Several key sections as representatives of the different facies of the lower limestone member of the Aisemont Formation are described herein (Figs 1 & 2). They are situated in the northern part of the Dinant Synclinorium (Baugn6e), in the eastern part of the Namur Synclinorium (Engis, La Mallieue, see Fig. 6 later) and in the Vesdre area (Fonds-des-Cris, Prayon, see Fig. 5 later). Sections in the Philippeville Anticlinorium
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
147
(mainly Neuville) are discussed but not described here (see among others Coen & Coen-Aubert 1974; Boulvain et al. 1993, 1999; Bultynck et al. 1998). They evolve from distal to proximal facies northwards. The Baugn6e section is 10 km SSE of Engis and La Mallieue, but during Frasnian times it may have been more than 45 km as a result of the intense folding and horizontal displacement of about 25 km along the Eifel Thrust. The Prayon section is 19 km east of Engis and 16 km NE of Baugn6e. During the Frasnian times it had to be situated SE of Engis and NE of Baugn6e (the direction of the folding is W S W ENE), but on a intermediate facies line between these two.
Baugnke section (northern Dinant Synclinorium) : distal facies of the lower limestone member The Baugn6e section (Figs 1 & 2) is situated along the road from Nandrin to Esneux, north of the Tavier stream. It exposes the top of the Lustin Formation, the Aisemont Formation and the base of the Lambermont Formation (Mottequin 2005). The top of the Lustin Formation is composed of 4 m of thick-bedded bioclastic limestones with bulbous stromatoporoids. The lower limestone member of the Aisemont Formation rests in disconformity on an erosional transgressive surface capping the Lustin Formation. It comprises two lithological units. The lower unit is composed of 3 m of decimetre- to pluridecimetre-thick limestone, more or less argillaceous and nodular, with intercalated shale layers. The limestone is a bioturbated mudstone/wackestone-packstone with numerous brachiopods - well preserved or in debris (Fig. 4) -pelmatozoans, gastropods, some bryozoans, rare siliceous sponge spicules and a few Frechastraea. The base of the unit contains some endoclasts and is sandy. The upper unit comprises 4 m of alternating pluridecimetre- to metre-thick argillaceous limestone beds and decimetre- to pluridecimetrethick shale beds, with numerous brachiopods. The base of the lower member is interpreted as relatively shallow, but a relatively deep environment with argillaceous inputs below the fair weather wave zone developed rapidly. It was unfavourable for the development of stromatoporoids, siliceous sponges and Alveolites. The middle shale member is composed of 22.6 m of green shale with some brachiopods, and the upper limestone member of about 17 m of thick-bedded limestones, more or less nodular, with some brachiopods and phillipsastreids.
Fig. 4. Mudstone-wackestone with brachiopod shells; Baugn6e section, basal unit of the Aisemont Formation. Note the stylo-nodular fabrics in the lower part of the thin section.
Its uppermost bed is a 60 cm-thick argillaceous limestone with numerous phillipsastreids. The base of the Lambermont Formation is composed of shales, sometimes calcareous and nodular, with brachiopods.
Prayon and Fond-des- Cris sections ( Vesdre area): mid-ramp facies of the lower limestone member Considering the geometry of the basin and the evolution of facies, these sections were in a medium position between the Baugn6e section and the Engis and La Mallieue sections. As in the other sections, the Aisemont Formation rests in disconformity on an erosional transgressive surface capping the Lustin Formation (Fig. 3b). The Fond-des-Cris section is composed of a succession of small disused quarries crossing an anticline exposing the Lustin and the Aisemont formations. The latter, being highly tectonized and not easily accessible, was sampled but not logged. The Prayon section crops out along a footpass and exposes 3 m of bioclastic limestones rich in corals and stromatoporoids of the top of the Lustin Formation, and the lower limestone member of the Aisemont Formation (10.94 m thick), which can be divided into three lithological units (Fig. 5).
148 Litho.
E. POTY & E. CHEVALIER Simplified log
Relative sea-level
Facies
/R.
T..
m
Z
~P---.---~ . ~ . . . . (t)
< O r
~E
"~LL
_d
5
...... ..
6
~
7
Fig. 5. Simplified log and facies evolution of the lower limestone member of the Aisemont Formation at Prayon (Vesdre area). Facies: A, coralstromatoporoids limestones of open-marine carbonate platform; B, laminar stromatoporoid-coral meadow; C, Frechastraea-Alveolites meadow; D, laminar Alveolites-stromatoporoid biostromal reef. Relative sea-level: R, regressive; T, transgressive; D (ETS), disconformity (transgressive erosion surface). Lithology: 1, limestone; 2, slightly nodular limestone with stromatoporoids and corals; 3, Alveolitesstromatoporoid biostrome; 4, limestone with corals; 5, laminar stromatoporoid; 6, laminar Alveolites; 7, phillipsastreid. Litho., lithostratigraphy; Fm, Formation.
The lower unit comprises 2.7 m of decimetreto pluridecimetre-thick bedded limestone (packstone), locally bioturbed and slightly argillaceous and dolomitized, with numerous laminar stromatoporoids and common laminar-tabular Alveolites and Frechastraea. This latter may be reworked and overturned, but the stromatoporoids and Alveolites are in living position, encrusting and stabilizing the substrate. Other allochems comprise crinoids, fragments of brachiopods, siliceous sponge spicules (pseudomorphosed in calcite), bryozoa and gastropods. The base of the unit is sandy. The matrix of the packstone is a microspar (from a former micrite),
locally slightly argillaceous. Compaction gives a stylonodular fabric, which is enhanced when bioturbation occurs. This unit was deposited in a shallow-water environment, under a moderate influence of the fair-weather wave zone, but subjected to storm waves, as indicated by the reworked fossils; the argillaceous input was low, allowing the growth of stromatoporoids, Alveolites and sponges, which stabilized the substrate. The middle unit is composed of 1.54 m of massive limestone (packstone), bioturbated and more or less dolomitized, with numerous laminar-tabular Frechastraea and common laminar Alveolites (both from 20 up to 60% of the volume of the rock). The latter often bind the sediment. Some Phillipsastrea are present. Laminar stromatoporoids are uncommon and are only present at the base of this unit. Corals are partly reworked and overturned. The composition of the packstone is similar to that of the lower unit. Peloidal grainstone may occur inside bioturbations. The 1.54m-thick massive limestone was deposited in an environment similar to, but possibly slightly deeper than, the previous one, as suggested by the general evolution of the section still subjected to storm waves. Frechastraea and Alveolites formed a coral meadow. A high sedimentation rate could have prevented the growth of stromatoporoids. The upper unit is a 6.7 m-thick biostrome composed of laminar Alveolites-stromatoporoid bindstone with some Frechastraea (see Fig. 12 later). The matrix varies from packstone to mudstone/wackestone, sometimes slightly argillaceous and dolomitized. When slightly argillaceous, a stylonodular fabric develops. The biostrome formed under or near the storm wave base and was subjected to an input of limestone sediments, perhaps triggered by storms (it will be discussed later in the section devoted to the biostromes). The middle shale member is not exposed at Prayon. The upper limestone member is composed of at least 17 m of stylonodular limestones with oncolites and some corals; it has yet to be studied.
Engis and La Mallieue sections: proximal facies of the lower limestone member The Engis section is situated in a disused quarry redeveloped as a public geological park (Parc des Tchafornis; see Fig. 7b) in the town of Engis. It exposes the upper part of the Lustin Formation, the lower limestone member and the base of the middle shale member of the Aisemont
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
149
Fig. 7. (a) La Mallieue road section; and (b) Engis disused quarry section, low. lim. mbr., lower limestone member of the Aisemont Formation; mid. sh. mbr., middle shale member; up. lira. mbr, upper limestone member.
Fig. 6. Simplified log and facies evolution of the Aisemont Formation in Engis and La Mallieue sections (SE Namur Synclinorium). Facies: A, coralstromatoporoids/amphiporids/stromatoliticlimestones of open-marine to restricted carbonate platform; B, Frechastraea-Alveolites biostrome in the fair-weather wave zone with argillaceous input; C, coral meadow on soft carbonated-argillaceous substrate in the storm wave zone; D, brachiopod association in a aerobic-dysaerobic carbonated-argillaceous-argillaceousenvironment under the storm wave base; E, dysaerobic-anaerobic argillaceous environment with rare cyrtospiriferids; F, coral/oncolite open-marine argillaceous limestones; G, carbonated argillaceous environment with some corals. Relative sea-level: R, regressive; T, transgressive; D (ETS), disconformity (erosion transgressive surface); FWWB, fair-weather wave base; AD-DAn, aerobic-dysaerobic/dysaerobic-anoxic transition; SWB, storm wave base; Lithology: 1, limestone; 2, biostrome; 3, dolomitic shale; 4, shale; 5, calcareous shale; 6, argillaceous limestone; 7, colonial coral; 8, oncolite. Litho., lithostratigraphy; Lamb., Lambermont Formation.
Formation. The La Mallieue section (CoenAubert & Lacroix 1979; Mottequin 2005) is composed of a disused quarry and an outcrop along the road from Engis to Amay (Fig. 7a) on the left bank of the River Meuse. It exposes the upper part of the Lustin Formation and the whole Aisemont Formation. The two sections are 2 km apart on the reversed southern limb of an anticline, on the SE edge of the Namur Synclinorium. They show the same lithology and therefore their description will be combined (Fig. 6). The Lustin Formation comprises bioclastic limestones rich in rugose, tabulate corals and stromatoporoids, and mudstones and stromatolitic boundstones. Some nodular argillaceous levels corresponding to palaeosols are present. These facies are distributed in plurimetre-thick shallowing upwards parasequences. The lower limestone member ('first biostrome' sensu Coen-Aubert & Lacroix 1974) of the Aisemont Formation can be divided in four lithological units. The lowest unit is composed of 40 (Engis) to 50 cm (La Mallieue) thin-bedded, more or less argillaceous, strongly bioturbated wackestonegrainstone rich in siliceous sponge spicules and
150
E. POTY & E. CHEVALIER
Fig. 8. (a) Colonies of Frechastraea in the lowest part of the lower limestone member of the Aisemont Formation (view of the base of a reversed bed). (b) Wackestone with siliceous sponge spicules (S, a tetraxon spicule) and bryozoa (B), in the upper part of the thin section; the lower part of the thin section is a grainstone. Engis section, base of the Aisemont Formation. crinoids, alternating with some layers of calcareous shales (Fig. 8b). Colonies of Frechastraea and Alveolites are common, being in living position or more or less disturbed (Fig. 8a). Other bioclasts are debris of bryozoans, trilobites, brachiopods and gastropods. The base of this first unit is weakly sandy and rests in disconformity on an erosional surface at the top of the Lustin Limestone (Fig. 3c). This unit is considered as being deposited in a shallow-water environment, within the fairweather wave zone, with episodes of clay deposition. The carbonate episodes allowed the growth of numerous siliceous sponges, but not of stromatoporids. Note that these shallow-water facies rich in siliceous sponges, but with a low biodiversity and with quartz grains developing at the base of the unit, are similar to those described by Geldsetzer
Fig. 9. (a) View of the Frechastraea-Alveolites biostrome at Engis (the photograph is presented in a stratigraphic position). (b) Close-up view of the biostrome. et al. (1993)just after the Frasnian-Famennian
extinction event in platform carbonates along the Trout River (Northwest Territories, Canada). These authors consider this low biodiversity and the relative abundance of primitive organisms as the first colonization stage after a global extinction event, which, in a lesser extent, is the case here. The second unit is a 3.7 (Engis) to 4 m-thick (La Mallieue) coral biostrome sensu stricto (Figs 9 & 10). The rugose coral colonies are laminar, discoid, of various sizes, from a few centimetres to 70-80 cm in diameter (and possibly more), and
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
151
Fig. 10. (a) Framestone constructed by Frechastraea (at the base of the photograph) and Alveolites. (b) Frechastraeaframestone; note the contact between two colonies (arrows). Frechastraea-Alveolitesbiostrome at La Mallieue.
from some millimetres to 6-7 cm thick; most are between 10 and 20 cm in diameter and from 1 to 3 cm in height. They were not firmly attached to the substrate (a dead colony), but they utilize its shape and rest intimately on it. Alveolites are laminar-conical, encrusting the substrate (a fossil). Uncommon laminar stromatoporoids and Sphaerocodium are present. Colonies of rugose corals are often overturned and are used as substrate for other ones (Fig. 11 a). The density of the frame builder can reach 90% of the rock (Fig. 9b); the matrix is a shale or a slightly argillaceous packstone more or less dolomitized. Compaction was high and lamellar colonies are often broken, fissures being filled up with sparry calcite. The biostromal unit developed in an argillaceous environment with uncommon limestone inputs, in the fair-weather wave zone and under the influence of storm waves, usually not allowing the growth of stromatoporoids. Argillaceous limestone (Fig. 11c), 1.5 m thick, passing vertically to calcareous shale composes
the third unit. It is more or less dolomitized, with numerous Frechastraea, Alveolites and Hankaxis, and some Phillipsastrea. Corals are separated from each other, often tipped or overturned (Fig. 1 lb). Some of them show rejuvenescence after tilting. The shape of the massive Rugosa varies from tabular to domal, with margins more or less ragged; their base can be conical, flat or even concave. Some ragged domal colonies show a deflected growth indicating a response to a unidirectional current (see Fig. 19 later). Alveolites are domalcolumnar with ragged margin, from some centimetres up to 12-15 cm in height. The unit deposited in an environment probably below the fair-weather wave zone, but subjected to bottom currents and affected by storm waves, with a relatively high rate of slightly carbonated argillaceous sedimentation sometimes disturbing the coral growth and preventing the growth of stromatoporoids and bryozoans. The last unit is a 35 cm-thick argillaceous dolomite with some rugose and tabulate corals.
152
E. POTY & E. CHEVALIER
Fig. 11. (a) Two Frechastraea (a, b) growing on the basal epitheca of an overturned colony of another Frechastraea (c); Frechastraea-Alveolites biostrome at La Mallieue. (b) A colony of Frechastraea in normal position (a) near a overturned colony (b); third unit of the lower limestone member at Engis. (c) Coral bed at the base of the same unit at Engis showing tipped colonies.
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES It suggests a similar environment, but slightly deeper than the previous one. Dolomitization is diagenetic. The palaeoecology of the lower limestone member will be discussed later. The middle shale member is divided in four lithological units. The lowest is a 70 cm-thick calcareous shale, more or less dolomitized, with brachiopods (mainly spiriferids); the second is composed of about 4 m of shales almost devoid of fossils (very rare spiriferids); the third comprises about 7.3 m of shales, sometimes slightly calcarous, with some brachiopods (comprising spiriferids, productids and lingulids), 'paper pectens', fenestellids, gastropods, remains of crustaceans and orthocerids; and the last one is a 1 m-thick bioturbed calcareous shale passing into argillaceous limestone, with some brachiopods and gastropods. The middle member deposited below the wave zone. A deepening-upwards trend (upper part of a TST) reaches its maximum in the second unit (maximum flooding surface correlated with the Lower Kellwasser Event), and is followed then by a shallowing-upwards (beginning of the highstand system tract (HST)). The upper limestone member comprises 10 m of more or less nodular and slightly argillaceous limestone, partly dolomitized, rich in oncolites, with sparse colonies of Frechastraea and Alveolites. It is interpreted as deposited in a shallow-water environment and corresponding to the upper part of the highstand of the 'Aisemont sequence'.
153
Fig. 12. Thin section in the Alveolites (a)-stromatoporoid (s) biostrome; Fonds-des-Cris section. stromatoporoids, often alternately encrusted (Fig. 12). Their size varies usually from a few centimetres to 20 cm in diameter and from a few millimetres to 2 cm in thickness. Some Alveolites are domal, and may show ragged margins, and can reach a height of 5 cm. Low tabular-laminar colonies of Frechastraea and Phillipsastrea are present. Some Frechastraea reach up to 50-60 cm in diameter and 7-15 cm in height, indicating growth for a long period without disturbance, but most are of relatively small size, with a diameter of less than 15 cm. Laminar cavities (probably corresponding to decayed sponges) intercalated between Alveolites colonies, stromatoporoid or Alveolitesstromatoporid colonies are not uncommon (Fig. 13). Preserved desmosponges are sometimes
The biostromes The biostromes of the lower member correspond to two different associations of builders: the first one, developed in the Vesdre area and well exposed in the Prayon section, is composed of laminar Alveolites and stromatoporoids; the second, developed in the eastern part of the Namur Synclinorium and well exposed in the Engis and La Mallieue sections, comprises mainly Frechastraea and secondarily Alveolites and Phillipsastrea. The latter type of biostrome is unusual because it is almost entirely built by rugose corals and contains only a few matrix.
The laminar Alveolites-stromatoporoid association The association comprises mainly laminar colonies of Alveolites tenuissimus and laminar
Fig 13. Laminar stromatoporoid bindstone (s) with cavities (c) probably corresponding to decayed sponges as suggesting by the rests of some of these (sp); Aulopora (a) encrusting a stromatoporoid; a debris of Scoliopora (sc) partly encrusted by an underlying stromatoporoid; Alveolites-stromatoporoid biostrome at Prayon.
154
E. POTY & E. CHEVALIER
present (Fig. 4a). Encrusting Sphaerocodium and bryozoans (Fig. 4b) are minor bioconstructors. Auloporids commonly encrust surfaces of dead colonies (Fig. 13). Frechastraea and Phillipsastraea are in living position and sometimes overturned, such as in the following Frechastraea-Alveolites biostrome, suggesting an environment exposed to storm waves. The matrix varies from mudstone to packstone and includes mainly small-sized bioclasts from crinoids, debris of brachiopod shells and bryozoans, and, secondarily, debris of Scoliopora, Aulopora and other corals. It is slightly argillaceous and more or less compacted, stylonodular and dolomitized. The ratio of builders to matrix varies widely from 1/3 to 3 with an average of 2/3. It indicates
that the builders are mainly binders (Flfigel 2004), grading occasionally into a framestone. Between the layers of constructors, the matrix is sometimes clearly graded, from a lower coarse-grained packstone layer to an upper slightly argillaceous wackestone/mudstone. These plurimillimetric- to centimetric-thick layers suggest that the growth of the Alveolitesstromatoporoids association could be disrupted by sudden inputs of sediments (possibly tempestites) and that the later colonization started in a relatively quiet environment. Some colonies of Alveolites and Frechastraea are ragged - the result of a competition with a high rate of sedimentation. The biostrome had to develop in an environment situated below the fair-weather wave base, as indicated by the lithofacies as well as the laminar shape of stromatoporoids and Alveolites (Fig. 14). Indeed, laminar shapes are more easily overturned than domal forms, which are typically developed in relatively turbulent environments (Scrutton 1998). The graded layers
Fig. 15. (a) Alveolites (a) encrusting the surface of a Fig. 14. (a) Laminar stromatoporoid (s)-sponge (sp) bindstone (F, Frechastraea). (b) Laminar Alveolites (A)-stromatoporoid (s) bindstone with an encrusted bryozoa (b) and an encrusting annelid Spirorbis omphalotes (Sp). Fonds-des-Cris section.
dead colony ofFrechastraea (b); FrechastraeaAlveolites biostrome at La Mallieue. (b) Aulopora encrusting the surface of a dead colony of Frechastraea, Frechastraea-Alveolitesbiostrome at Engis.
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES of sediments and the overturned colonies of rugose corals indicate occasional exposure to storm waves and sediment input.
The F r e c h a s t r a e a - A l v e o l i t e s biostromal association Frechastraea is the dominant rugose coral in the biostrome and belongs to only two species:
F. limitata, the most common, and F pentagona. They are always dominant, composing up to 80-90% of the whole builders and of the volumetry of the biostrome. Other Rugosa present are: Hankaxis insignis, relatively common in
155
some levels, Phillipsastrea ananas, uncommon small Tabulophyllum sp. and Peneckiella sp., only known from a single large colony on the top of the biostrome. Tabulates comprise Alveolites tenuissimus and A. suborbicularis, composing up to 10-15% of the association, two species of Aulopora (A. serpens and A. repens) encrusting dead parts of, or dead colonies of, philipsastreids (Fig. 15b) and Alveolites, and uncommon fragments of ramose colonies (Thamnopora sp. and Scoliopora sp.). Lamellar and domal stromatoporoids and Sphaerocodium are other uncommon builders. Accessory organisms are some athrypids and spiriferids. The matrix is only
Fig. 16. (a) Frechastraea showing tabular surfaces corresponding to successive deaths of parts of the colony; Frechastraea-Alveolites biostrome, Engis area. (b) Longitudinal thin section in a Frechastraea showing seasonal growth lines, some of them marked by an argillaceous input and the death of parts of the colony. (c) Longitudinal thin section in a low-domal ragged colony of Frechastraea showing successive lateral deaths following by rejuvenescences; seasonal growth lines are well marked; third unit of the lower limestone member at Engis.
156
E. POTY & E. CHEVALIER
present in some levels. It varies from a shale to a packstone comprising numerous siliceous sponge spicules, crinoids, bryozoans and debris of brachiopod shells, such as in the underlying limestones. The weak amount of deposited matrix is probably a result of to the permanent washing owing to the waves. This buildup is unusual because it is developed in the fair-weather wave zone and almost entirely built by Frechastraea with no or few matrix in-between the coralla, although they did not encrust, and therefore were not firmly attached to, the substrate. But, in order to resist the agitation, they hug closely the shape of the substrate (a dead rugose or tabulate coral colony) and rest intimately on it. Consequently, their growth pattern was laminar, which is not the best colony shape in which to live in turbulent waters (Scrutton 1998). Thus, however, they were able to resist to the fair-weather wave agitation and were reworked by storms, as shown by the numerous overturned colonies (which are used as substrate for others). Note that in the field aggregated colonies can often be separated from each other with a knife. In contrast however, Alveolites encrusted the substrate - a fossil (Fig. 15a). Thus, their growth patterns were laminar-conical, the latter pattern being better adapted to turbulent water. Both phillipsastreids and Alveolites can occupy up to 80-90% of the volume of the biostrome, and sometimes substantially more when colonies rest upon each other, and therefore they actually act as framestone constructors. The amount of matrix is low, sometimes less than 10% and never exceeding 20%. It consists mainly of clay infill between colonies and, secondarily, of more or less argillaceous packstone. The input of clay could be seasonal and/or triggered by storms, as suggested by their correlation with deceased parts of colonies, whereas the carbonated sediments probably came from a more proximal area during storms (Fig. 16) By lateral growth, rugose coral colonies can get in touch with one another (Fig 17a). This was observed between corals belonging to the same species of Frechastraea, between two species of this genus (Fig. 17b), and between Frechastraea and Phillipsastraea (Fig. 17c). In all cases, the individuals stop their lateral increase at the point of contact and remained separated by their coral epitheca. Lateral growth continues on both sides of the point of contact along a line separating the two colonies (Fig. 17a). The vertical growth of the colonies follows the same pattern (Fig. 17b). Neither integration between two colonies of the same species nor the dominance of a colony on another was observed. With this pattern of
Fig. 17. (a) View of the basal epitheca of two touching colonies of Frechastraea continuing their lateral growth parallel from their point of contact. (b) Longitudinal thin section showing the vertical contiguous growth between two touching species of Frechastraea. (c) Transversal thin section showing the contact between a Freehastraea (on the left) and a Phillipsastrea (on the right). All samples from the Frechastraea-Alveolites biostrome at La Mallieue.
colonization of space, contemporary colonies sometimes covered the complete bottom surface, forming a continuous framestone. The basal epitheca of Frechastraea and Alveolites (when developed on a soft substrate) can show regular growth bands that are interpreted as seasonal (Fig. 18). Their width
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
157
Fig. 18. (a) Seasonal growth bands on the basal epitheca of a colony of Frechastraea; La Mallieue. (b) Similar growth bands in Alveolites; Engis. (c) Longitudinal thin section in Frechastraea showing seasonal lines (arrows) corresponding to densely packed dissepiments.
varies from 5 to 8 ram, with an average of 6 mm in Frechastraea (Fig. 18a), and from 5 to 12 mm in laminar Alveolites (Fig. 18b), with an average of about 7-8 mm. These results show a pattern that is consistent with the data collected by Scrutton (1998). Inside the skeleton of Frechastraea, when colonies become low tabular, seasonal growth variations are also recorded (Fig. 18c). They are marked by a regular development of thin bands (about 1 mm thick) of alternating layers of normally packed dissepiments (0.7-1.1 mm thick) and densely packed dissepiments (about 0.2 mm thick), sometimes correlatable with the decease of parts of colonies (Fig. 16) and/or with an argillaceous line resulting from an argillaceous input. The latter could be correlated with seasonal storms. Seasonal variations are also indicated by constrictions and rejuvenescences in the solitary rugose coral Hankaxis. This seasonal variation is better marked in specimens from the shaly limestone overlying the biostrome, where the shape of
phillipsastreid colonies are commonly domal or tabular and Alveolites columnar, with ragged margins. Thus, in the laminar Frechastraea the growth is dominantly lateral: from 5 to 8 mm in radius per year, with 6 mm as average, based on seasonal growth bands; the vertical increase varies from almost 0 to about 1.5 mm year -1. In the biostrome there are colonies of all sizes, the most common ones being 15-20 cm in diameter and 2 cm thick; the largest colonies vary from 30 to 40 cm in diameter and from 3 to 4 cm in thickness in the lower 2.3 m of the biostrome, i.e. about 2533 years old, and reach up to 60-80 cm in diameter and 5-8 cm in thickness in the upper 1.7 m, i.e. possibly about 50-65 years old. Moreover, many colonies were post-mortally encrusted by Aulopora, which did not contribute to develop the biostrome. So, considering the growth rate of Frechastraea, the duration for the formation of the entire biostrome can be estimated to be in the range of 4 (maximum growth rate) to 5 ka
158
E. POTY & E. CHEVALIER
Fig. 19. (a) Ragged colony of Frechastraeawith a deflected growth (arrow) owing to a constant bottom current. (b) Stromatoporoid (s) encrusting the top and the edge of the Frechastraea(F), suggesting that the latter was dressed above the sediment. Third unit of the lower limestone member at Engis.
(minimum growth rate with stops in the construction). The overlying 1.5m coral bed contains smaller colonies of all sizes, indicating high mortality affecting all the ages in the population. The maximum diameters decrease progressively upwards, from about 30 cm at the base of the unit to 12-15 cm in the upper part, indicating that colony ages dropped from a maximum of 20 years to about 10-12 years based on their annual growth rate. Coral growth probably stopped as a result of argillaceous inputs and/or storms that tipped or overturned many of them. This suggests that the development of this deposit was relatively shorter, possibly about 1 ka. Therefore, the rate of the relative sea-level rise, from some metres deep at the base of the
Aisemont Formation to the substorm wave zone at the base of the shale member, seems to have been relatively fast and suggests that it could be owing to an icecap melt, following a short glaciation and the former emersion of the carbonated platforms.
Comparison between the two types o f biostrome The Alveolites-stromatoporoids biostrome formed in a more distal position than the Frechastraea-Alveolites biostrome (Fig. 20). The lesser number or the lack of stromatoporoids in the latter is probably linked to the higher rate of argillaceous input. Frechastraea (and Phillipsastrea) were present everywhere in the
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES
159
Fig 20. Composite reconstruction of the biostromes and distal environment of the lower part of the Aisemont Formation. A, Frechastraea-Alveolitesbiostrome; B, Alveolites-stromatoporoid biostrome; C, deeper distal brachiopod association. FWWB, fair-weather wave base; SWB, storm wave base. 1, laminar Alveolites; 2, laminar stromatoporoid; 3, laminar-tabular phillipsastreid, with the dead part showing calices (a) and the living part with polyps (b); 4, overturned phiUipsastreid; 5, conical-columnar Alveolites; 6, sediment; 7, brachiopods (spiriferids and atrypids); 8, cavity with coarse-grained sparry calcite.
basin, colonizing a wider range of environments than stromatoporoids, from proximal to distal, carbonate-dominant to argillaceous-dominant environments, submitted or not to the wave action. They became dominant when the other competitors (especially stromatoporoids and, to a lesser extent, Alveolites) were absent. This is the case in the proximal La Mallieue-Engis area, where they constructed a biostrome, and in the distal Philippeville areas where they formed rugose coral beds.
Conclusions The Alveolites-stromatoporoid biostrome is a bindstone (grading sometimes into a framestone)
composed of encrusting organisms and belonging to a type commonly known from the Silurian to the Frasnian. It developed in a medium distal environment, relatively quiet (under the fair-weather wave zone) and protected from constant argillaceous inputs. In contrast, the Frechastraea-Alveolites biostrome is an unusual bioconstruction in the sense that rugose corals constructed the major part of the buildup, but did not encrust each other, thus only laying flat on dead colonies. This biostrome was a structure relatively resistant to fair-weather waves, but not to storm waves. Upper Frasnian phillipsastreids and, more especially, Frechastraea species, which are the most common, were opportunists that lived everywhere on the platform and took over
160
E. POTY & E. CHEVALIER
the positions of organisms whose development was prevented or limited by ecological factors. They particularly proliferated in proximal shallow-water environments not favourable for the development of the stromatoporoids because of a permanent argillaceous input (which was constantly removed and almost not deposited because of the permanent agitation). This situation was triggered by the shift in the depositional pattern, from a carbonatedominant (Philippeville and Lustin formations) to a mixed carbonate-argillaceous (Aisemont Formation) or argillaceous-dominant (Neuville Formation) sedimentation on the platform, at the middle-upper Frasnian boundary, and the correlated ecological crisis probably resulted from a cooling (Joachimski et al. 2004). The fall and then a rapid rise in the sea level at the middle-upper Frasnian transition could correspond to a ice cap freezing-melting cycle. Despite their large number, the generical and specifical diversity of rugose, tabulate corals and stromatoporoids was very low after the first Frasnian crisis. The latter and the development of unfavourable environments favoured the development of pre-adapted taxa, whereas others were decimated. For Joachimski et al. (2004), on the basis of 5180 analysis of conodont apatite, the Kellwasser horizons were triggered by two short-term cooling events responsible for the late Frasnian crises during the very warm greenhouse climatic period characterizing the late Frasnian and early-middle Famennian. However, in the N a m u r - D i n a n t Basin, the Lower Kellwasser Event corresponds to the maximum flooding surface of the first third-order sequence of the late Frasnian ('Aisemont sequence'), i.e. a rise in sea level, and no direct influence on the distribution of corals and brachiopods is observed. Therefore, it seems to have a time-lag and no relationship between the first Frasnian coral extinction in Belgium (extinction of the disphyllids) and the Lower Kellwasser Event, which had strictly no influence on the stratigraphic distribution of corals. We wish to thank the two referees, F. Boulvain and S. Schr6der, for their helpful comments, suggestions and corrections, and M. Aretz for his constructive remarks and corrections.
References ARETZ, M. 2002. Habitatanalyse und Riffbildungspotential kolonialer rugoser Korallen im Unterkarbon (Mississippium) von Westeuropa. Kdlner Forum fiir Geologie und Paliiontologie, 10, 1-155.
ARETZ, M. & CHEVALIER,E. 2007. After the collapse of stromatoporid - coral reefs - the Famennian and Dinantian reefs of Belgium: much more than Waulsortian mounds. In: ALVARO,J. J., ARETZ, M., BOULVAIN, F., MUNNECKE, A., VACHARD, D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 163-188. BOULVAIN, F., BULTYNCK, P. ET AL. 1999. Les formations du Frasnien de la Belgique. Memoirs of the Geological Survey of Belgium, 44, 1-125. BOULVAIN, F., COEN, M., COEN-AUBERT, BULTYNCK, P., CASIER, J.-G., DEJONGHE, L. & TOURNEUR, F. 1993. Les formations frasniennes du Massif de Philippeville. Service g6ologique de Belgique, Professional Paper, 1993ll, 1-37. BOULVAIN,F., CORNET,P. E T AL. 2004. Reconstructing atoll-like mounds from the Frasnian of Belgium. Facies, 50, 313-326. BULTVNCK, P. & DEJONGHE, L. 2002. Devonian lithostratigraphic units (Belgium). In: BULTYNCK, P. & DEJONGHE, L. (eds) Guide to a Revised Lithostratigraphic Scale of Belgium. Geologica Belgica, 4, 3%69. BULTYNCK, P., COEN-AUBERT, M. & GODEFROID,J. 2000. Summary of the state of correlation in the Devonian of the Ardennes (Belgium-NE France) resulting from the decisions of the SDS. Courier Forschungsinstitut Senckenberg, 225, 91-114. BULTYNCK,P., HELSEN,S. & HAYDUCKIEWICH,J. 1998. Conodont succession and biofacies in upper Frasnian formations (Devonian) from the southern and central parts of the Dinant Synclinorium (Belgium) - Timing of facies shifting and correlation with late Frasnian events. Bulletin de l'Institut royal des Sciences naturelles de Belgique, Sciences de la Terre, 68, 25-75. COEN, M. & COEN-AUBERT,M. 1974. Conodontes et coraux de la partie sup6rieure du Frasnien dans la tranch6e du chemin de fer de Neuville (Massif de Philippeville, Belgique). Bulletin de l'Institut royal des Sciences naturelles de Belgique, Sciences de la Terre, 50, 1-8. COEN, M., COEN-AUBERT, M. & CORNET, P. 1977. Distribution et extension stratigraphique des r6cifs/~ ~ Phillipsastrea ~ dans le Frasnien de l'Ardenne. Annales de la SocidtO gdologique du Nord, 96, 325-331. COEN-AUBERT, M. 1994. Stratigraphie et syst6matique des Rugueux de la partie moyenne du Frasnien de Frasnes-lez-Couvin (Belgique). Bulletin de l'Institut royal des Sciences naturelles de Belgique, Sciences de la Terre, 64, 21-56. COEN-AUBERT, M. & LACROIX, D. 1979. Le Frasnien dans la partie orientale du bord sud du synclinorium de Namur. Annales de la Socikt~ Gkologique de Belgique, 101,269-279. DA SILVA,A.-C. 2004. Skdimentologie de laplate-forme carbonatkefrasnienne belge. PhD thesis, University of Li6ge. DA SILVA,A.-C. & BOULVAIN,F. 2004. From paleosols to carbonate mounds: facies and environments of the Middle Frasnian platform in Belgium. Geological Quarterly, 48, 253-266.
LATE FRASNIAN PHILLIPSASTREID BIOSTROMES FLt3GEL, E. 2004. Microfacies of Carbonate Rocks; Analysis, Interpretation and Application. Springer, Berlin. GELDSETZER, H. H. J., GOODFELLOW, W. D. & MCLAREN, D. J. 1993. The Frasnian-Famennian extinction event in a stable cratonic shelf setting: Trout River, Northwest Territories, Canada. Palaeogeography, Palaeoclimatology, Palaeoecology, 104, 81-95. GouwY, S. & BULtYNCK, P. 2000. Graphic correlation of Frasnian sections (Upper Devonian) in the Ardennes, Belgium. Bulletin de l'Institut royal des Sciences naturelles de Belgique, Sciences de la Terre, 70, 25-52. JOACHIMSKI, M. M., VAN GELDERN, R., BREISIG, S., BUGG~SCH,W. & DAY, J. 2004. Oxygen isotope evolution of biogenic calcite and apatite during the Middle and Late Devonian. International Journal of Earth Sciences ( Geologische Rundschau) , 93, 542553. MOTTEQUIN, B. 2005. Les brachiopodes de la transition FrasnienlFamennien dans le Bassin de NamurDinant (Belgique). Systkmatique-palkokcologie-
161
biostratigraphie-extinctions. PhD thesis, University of Li6ge. POTY, E. 1999. Famennian and Tournaisian recoveries of shallow water Rugosa following late Frasnian and late Strunian major crises, southern Belgium and surrounding areas, Hunan (South China) and the Omolon region (NE Siberia). Palaeogeography, Palaeoclimatology, Palaeoecology, 154, 11-26. RODRIGUEZ, S. 1996. Development of coral reef-facies during the Vis6an at Los Santos de Maimona Lower Carboniferous Basin (Badajoz, SW Spain). In: STROGEN, P., SOMMERVILLE,I. & JONES, G. I. (eds) Recent Advances in Lower Carboniferous Geology. Geological Society, London, Special Publications, 107, 145-152. SCHINDLER, E. 1993. Event-stratigraphic markers within the Kellwasser Crisis near the Frasnian/ Famennian boundary (Upper Devonian) in Germany. Palaeogeography, Palaeoclimatology, Palaeoecology, 104, 115-125. SCRUTTON, C. T. 1998. The Palaeozoic corals, II: structure, variation and palaeoecology. Proceedings of the Yorkshire Geological Society, 52, 1-57.
After the collapse of stromatoporid-coral reefs m the Famennian and Dinantian reefs of Belgium: much more than Waulsortian mounds MARKUS ARETZ 1& EMMANUEL
CHEVALIER 2
llnstitut fiir Geologie und Mineralogie, Universitiit zu K61n, Ziilpicher Strasse 49a, 50674 Kgln, Germany (e-mai# markus, aretz@uni-koeln, de) 2Carmeuse s.a., Rue du Chateau, 13a, 5300 Seilles, Belgium Abstract: Reef development in the Famennian and Carboniferous successions of Belgium is more common than previously thought, and 10 broad time intervals of reef development can be differentiated. Reef formation is due to a variety of reef fabrics. Microbial communities are important for most reef frameworks, and often crucial for formation and stabilization of frameworks. Larger skeletal frameworks are rare. However, the interaction of skeletal bioconstructors and microbial communities is common, and results in successful reef building. However, microbial communities are still the backbone of these reefs. The majority of reefs are small, and a significant number formed in environments of restricted marine facies. Large reefs developed only in the late Tournaisian and late Vis6an. Their initiation and formation was controlled by the geometry of the shelf. Three hierarchical levels, discussed below under the headings palaeobiology, local environment, and regional and global environment, controlled reef formation. Important limiting factors were relative water depth, sea-level oscillations, climate, shelf geometry and the needs of the individual bioconstructor. In general, Belgian reef diversity reflects the global picture, but significant differences can be recognized in the different time slices. In particular, the abundance of middle Vis6an reefs is a unique feature. The onset of the Variscian orogeny terminated all reef development in Belgium, and reefs younger than late Vis6an are unknown.
The late Frasnian Kellwasser events are generally equated with the collapse of the middle Palaeozoic reef community of stromatoporids, rugosa and tabulates. Copper (2002) demonstrated that the demise and collapse of most Frasnian reefs started and ended earlier. Therefore, the major change in the reefal environment at that time must have been decoupled from the event interval. The demise seems to have been gradually throughout the entire Frasnian period. Good indications are the gradual reduction in biodiversity of stromatoporids throughout the Frasnian following the first prominent loss at the end of the Givetian time (Stearn et al. 1999; Stearn 2001), and the existence of very different types of reefs throughout the Frasnian (e.g. see Boulvain 2007; Poty & Chevalier 2007). The Famennian and most of the Carboniferous are commonly characterized as a period of arrested reef development (Copper 1988). In this context, the dominance of bioconstructions that are devoid of a metazoan framework has often been highlighted (Copper 1988; West 1988), and Waulsortian (mud-)mound has become a synonym for Early Carboniferous reef. However, the global spectrum of the Famennian and Carboniferous reefs is much wider, and numerous reef types in very different tectono-sedimentary
environments have been identified (e.g. Webb 1994, 2002; Aretz 2002). The Belgian Famennian and Carboniferous successions around the Brabant Massif (Fig. 1) have been intensively studied, and biostratigraphical and sedimentological data are mainly well constrained (Streel 1985; Thorez & Dreesen 1986; Conil et al. 1991; Bultynck & Dejonge 2002; Poty et al. 2002). During the time interval studied Belgium remained in tropical latitudes. However, during the cooler climate of the Famennian time, siliciclastic material was dominant in the marine settings of the Dinant and Namur synclinoria and the Vesdre Massif (Figs 1 & 2). A major facies change occurred during latest Famennian time (Strunian) when carbonate facies re-appeared widely on the shelf. This mixed siliciclastic-carbonate ramp system south of the Caledonian Brabant Massif of that time was the precursor of the Dinantian carbonate ramps and platforms (Fig. 2). The Brabant Massif retained its position as a structural high during Dinantian time, but was then fully surrounded by full-marine facies, including the Campine Basin towards the north (Figs 1 & 2). The onset of the Variscan orogeny in Belgium (Fig. 2) terminated the development of carbonate facies, resulting in the exposure of former shelf
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs'and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 163-188.0305-8719107l$15.009 The Geological Society of London.
164
M. ARETZ & E. CHEVALIER
Fig. 1. Main Palaeozoic structures of Belgium. The Brabant Massif is a structural high, which was emerged during the time interval studied, and is surrounded by marine shelf facies. Only the Carboniferous succession of the Campine Basin contains considerable marine facies, although often very discontinuous. The most complete successions for the studied interval are in the Dinant Synclinorium.
carbonates around the Vis6an-Namurian boundary, and led to an important reconfiguration of the basins. Later (Namurian-Westphalian time) paralic coal basins developed only in the vicinity of the Brabant Massif. The important changes in the shelf geometry and basin configuration around the Brabant Massif throughout the time interval studied offer the possibility of studying a wide range of very different environments and ecological niches, which may have been suitable for very different bioconstructors and reef types. Although not all possible positions and environments were developed during all the individual time slices, the small geographical extension of the studied area is relatively unique. All reefs can be placed in their local tectono-sedimentary environment and, in combination with biostratigraphical data, the limiting factors can be identified at several hierarchical levels.
Ever since the Waulsortian mounds were named after the village of Waulsort in southernmost Belgium, the equation of Wauslortian and Carboniferous reefs has been very common in Belgium. Various types of reefs in a broad sense have been described from Famennian, Tournaisian and Vis6an rocks in Belgium (for details see sections below). The present study tries to compile the reef localities from the Famennian-Vis6an interval (younger strata are devoid of reefs.). With the exception of the Waulsortian reef facies, all potential reef localities known to us have been evaluated and examined. This has lead to significant changes in the number of known reef localities, and partly to the reinterpretation of formerly described reefs. A consensus on reef terminology does not exist. Based on Aretz (2002) and Chevalier & Aretz (2005), the following terminology is employed herein. Bioherm and biostrome are
BELGIAN FAMENNIAN AND DINANTIAN REEFS
165
Fig. 2. Overview of the biostratigraphical and lithostratigraphical column with indications of reef intervals, shelf geometry and climate. The abbreviations in the lithostratigraphical column are from Bultynck & Dejonghe (2002) and Poty et al. (2002).
166
M. ARETZ & E. CHEVALIER
general terms to describe only the morphology (see discussion in Aretz 2002). Biostromes can be subdivided into allo-, para-, autopara-, and autobiostromes based on the percentage of organisms/bioconstructors in place (Kershaw 1994). Bioconstruction is used as a term that describes a carbonate body that is the result of constructional activity of organisms, without respect to its geometry. Reefs are rigid, biologically induced and biologically controlled structures with topographic relief. Framework is the meshwork that leads to rigidity; it is a result of the skeletal organisms in place as well as a broad range of non-skeletal organisms and processes (Webb 1996). The absence of skeletal organisms does not imply the lack of rigidity and vice versa. The term mound is herein restricted to carbonate mud-dominated bioherms that lack a clear framework. The implication of microbial communities in the formation of these mounds and/or the carbonate mud is favoured here (see Lees & Miller 1995; Pickard 1996; Webb 1996). The term complex (here mound complex) is used for carbonate bodies of large size, supposedly formed by several bioherms but which are not clearly separable. It is the aim of this paper: (1) to analyse the occurrence and distribution of reefs in the Belgian Famennian-Dinantian successions; (2) to describe the reef fabrics; (3) to evaluate the factors controlling reef development; and (4) to place the Belgian reefs in a global reef context. Reef intervals Each interval of reef development, which can clearly be separated on lithostratigraphical and
biostratigraphical grounds in the FamennianVis6an succession, has been labelled with a letter (Fig. 2). The durations of the individual reef intervals vary considerably. The geographical extent of an individual reef development is commonly bound to a particular facies realm resulting in varying abundance from a single occurrence to numerous occurrences on a basinal scale.
Famennian reefs Interval A- microbial reefs (Baelen type). The oldest Famennian reefs formed during the deposition of the Souvrain-Pr6 Formation (upper middle Famennian; late marginifera Zone), the first major re-appearance of marine carbonate facies in the Famennian succession. Reefs are rare in that formation, but a local concentration of reefs is observed in the Limbourg area in the Vesdre Massif (Dreesen et al. 1985) (Fig. 3). There the typical nodular limestone with shaly intercalations of the Souvrain-Pr6 Formation is replaced by the Baelen facies. The Baelen facies is decribed as 'reef mud mound forming a heterogeneous limestone complex' (Bultynck & Dejonghe 2002, p. 61). Recent field observations indicate that the Baelen facies consists mainly of two lithotypes: (1) fine-grained, red or pale limestone partly with abundant stromatactoid cavities; and (2) dark to reddish shales with numerous intercalations of often coarse-grained crinoidal limestone. The later may be locally intercalated with micaceous sandstone. Based on these field observations, the extension of the Baelen facies is stratigraphically considerably lower than previously thought (Laloux et al. 1996). It seems to be possible to distinguish at least five carbonate
Fig. 3. Distribution of Famennian reefs. @ Microbial reefs of the Baelen-type; | microconchid reefs; 9 Strunian bioconstructions.
BELGIAN FAMENNIAN AND DINANTIAN REEFS bodies (Fig. 4a), whose cores consist of lithotype 1, which are surrounded by lithotype 2, which itself grades laterally and vertically into the nodular limestone with shaly intercalations of the Souvrain-Pr6 Formation. The classical 'La Forge' quarry (Dreesen et al. 1985) represents the most northern carbonate body and shows at its entry the transition from the massive core facies to the surrounding facies. Although of massive appearance, all supposed core areas of the carbonate bodies show some kind of bedding on a metre-sized scale. The size of the carbonate bodies is difficult to establish, but estimates are in the range of up to some tens of metres in thickness and up to 200 m in width. The composition of the two lithotypes is microscopically heterogeneous. Although macroscopically relatively homogeneous, lithotype 1 at the level of microfacies revealed important variations in the abundance of carbonate mud, allochems and cavity systems (Fig. 4b, c). Levels of dominance of carbonate mud, sponge spicules and bryozoans, and abundant stromatactoid cavities, alternate with levels of spar-dominance, and abundance of crinoids, palaeoberesellid algae, foraminifers and ostracods. This alternation may be totally random, but a significant number of thin sections indicate vertical variations of horizontally distinctive horizons. Cavity systems developed in the mudstone-wackestone facies as well as in the packstone-grainstone facies are complex and of different origin. They range from classical Stromatactis and zebra structures (Fig. 4b) to dissolution cavities. Crinoids and algal filaments are found in coarse blocky cements of some cavities (Fig. 4c). The sediment fill of the cavities consists of dark peloidal carbonate mud and fine- to mediumgrained bioclastic sediments. Fenestellid bryozoans in some examples form the roof of the cavities. In some cases the cavities are surrounded by a thin layer of brownish fibrous cements before the sediment infill starts. Although the distribution of pale and red coloured limestone in lithotype 1 is random, areas with packstone-grainstone textures tend to be paler, possibly owing to the abundance of bioclasts and blocky cements. The complex diagenetic evolution of the Baelen facies (Dreesen et al. 1985) is also well documented in lithotype 2, where dissolution phenomena at the margins of crinoids are conspicuous (Fig. 4d). The abundance, sizes and density of packing of crinoid fragments varies considerably in the individual limestone layers. In addition to the pelmatozoan fragments, foraminifers and ostracods are common in some intercalations.
167
Layers of both lithotypes were irregularly stacked upon each other at the edges of the massive core facies (Fig. 4e). However, the abundance and size of lithofacies 1 layers decreases significantly with increasing distance from the centre of the carbonate body. The classification and interpretation of these bioherms are difficult and various terms from mud-mound to reef have been applied to them (e.g. Belli~re 1953; Dreesen & Flajs 1984; Dreesen et al. 1985; Bultynck & Dejonghe 2002). This may be partly due to the complex diagenetic history, the poor outcrop quality in some parts of the carbonate bodies and the lack of further studies. However, the term microbial reef is applied herein. The clotted nature of some peloidal textures observed in the lithotype 1, cavity roofs formed by carbonate mud, and the co-occurrence of fine- and coarse-grained textures, implicate the incorporation of microbial communities in the formation of the framework, although calcimicrobes are rarely preserved. The framework was secondarily enhanced by the incorporation, of other biotic components, as seen in the fenestellid sheets forming cavity roofs, and the formation of early cements. The reefs are intensively colonized by crinoids and palaeoberesellid algae. They accumulated on the flanks and the reef surface, and thus were repeatedly incorporated into the reef. This hydrodynamically controlled accumulation of biota and their incorporation into the reef formation is a further part of the nomenclatural problem. In addition, the carbonate bodies contain slumps, which point to areas with reduced or no rigidity. The reefs were surrounded by mainly finegrained siliciclastic sediments, but the relatively continuous export of bioclasts resulted in the mixed siliciclastic-carbonate facies, which surrounds the carbonate bodies. It is so far not fully understood if lithotype 2 generally or only partly formed flanks. Some of the bedded areas of lithotype 1 may also be related to flank sediments. The intensive folding of the area and the differences in compaction of the lithotypes hamper an evaluation of the palaeotopography, but flank dips may have been in the range of 15~ ~ The formation of the reefs is believed to be between the base of wave activity and the base of storm activity, as indicated by the co-existence of grainstone textures and the large amount of fragmented bioclasts and mudstone textures. The position of the reefs may be related to tilt-blocks and faults (J.M. Marion pers. comm.). The factors controlling the initiation of the reefs are unknown; the termination of reef growth was possibly owing to reduced accommodation
168
M. ARETZ & E. CHEVALIER
,...,
0
o
o
,._~o
"'~
0.~
o
"'~
~.~ ~ ~.~ ~
"0.,~0
0
~
~
~
~
o
~,-~
~~'~ ~
.
~ . -
" ~ ~ ~
.
~
0~.~
~
0
~
~
~
~
~-~
o0 . ~ 0 0 o,'~ .~_~,.~-~ ..~,
~ ' ~
~.~~ .~.=~ ~ ~ ' ~ ~
0
~'
~'~
9
~
BELGIAN FAMENNIAN AND DINANTIAN REEFS space, as a consequence of a regression and/or vertical reef growth. The dimensions estimated for the carbonate bodies do not support the interpretation of Dreesen & Flajs (1984) that the Baelen facies formed a barrier, which protected the depositional environment of the Souvrain-Pr6 facies. Further reports of bioconstructional activity in late middle Famennian time are from the southern and eastern Dinant Synclinorium (Fig. 3). Stainier (1893) mentioned the temporary exposure of Baelen-type facies during the construction of a railway tunnel in the Lesse Valley. Dusar & Dreesen (1984) described up to 3 m-thick, domal-shaped accumulations of crinoid fragments from the Hamoir area. Interval B: microconchid reefs. The peritidallagoonal deposits (Fontin Member) of the Evieux Formation (upper Famennian, eastern Dinant Synclinorium) contain small reefs (Dreesen & Jux 1995). The framework of these small domal-shaped reefs was owing to microconchids and calcimicrobes (Garwoodia, Mitcheldeania). The term microconchid coined by Weedon (1990) describes tube-like fossils, which have commonly been termed 'vermetid gastropods' in Carboniferous deposits (e.g. Burchette & Riding 1977; Hance & Hennebert 1980; Wright & Wright 1981; Aretz 2001). The bioconstructional activity of microconchids resulted from sediment trapping in consequence of their erect growth form and their abundance (Dreesen & Jux 1995). However, framework building and sediment trapping may also result from the interaction of calcimicrobes and microconchids, which resulted in digitated laminar fabrics (oncoids of Dreesen & Jux). The reef growth took place in harsh environments of a marginal marine setting. Dreesen & Jux (1995) associated the reefs with a supratidal setting of a lagoonal pond; however, reef growth in the intertidal setting of a restricted lagoon cannot be excluded. Reef initiation might be related to cryptalgal laminations of the underlying mudstones. Reef growth stopped with further shallowing of the depositional environment, which resulted in the occasional development of palaeosols. Interval C: strunian bioconstructions. The youngest Devonian reef interval occurred in the latest Famennian ('Strunian substage'). It provides the last glimpse of bioconstructional activity of stromatoporids, the typical bioconstructors of the Middle Palaeozoic. The reappearance of carbonate facies on the shelf south of the Brabant Massif gave rise to a widespread colonization of
169
shallow-marine environments by fully marine organisms such as brachiopods, foraminifers, stromatoporids and corals. Van Steenwinkel (1990) and Weber (2000) recognized small-scaled sedimentary cycles in the Strunian deposits. According to Weber (2000), the older cycles were dominated by siliciclastics of marginal-intratidal zones, whereas younger cycles were carbonate-dominated and reflected fully marine settings. The carbonate increase towards the top of the Strunian may be related to a combination of transgression, and lower erosion rates and reduced siliciclastic influx. Conil (1964) described the Strunian localities of Belgium and named some horizons that contained stromatoporids biostromes. However, the abundance of the stromatoporids in most localities was rather low, and the distances between individuals were rather large. The term biostromes seems to be inappropriate, as a horizontal-extended accumulation of the sponges is not seen. In fact, the occurrences of stromatoporids and other associated marine biota only reflected the appearance of fullmarine conditions and carbonate facies. The term level-bottom community fits much better. These level-bottom communities may be developed in all Strunian cycles that contained the appropriate facies (e.g. zones C and D of Weber 2000). The Dolhain area in the Vesdre Massif has had much attention since Conil et al. (1961) described three Strunian biostromes. The third and thickest biostrome ( = main biostrome of the Vesdre) is of most interest; the lower biostromes fall into the category of level-bottom communities. Conil et al. (1961) indicated a thickness of 12.5 m for the third biostrome in the railway section north of Dolhain. Laloux et al. (1996) described the biostrome as an assemblage of sandy limestone, nodular limestone, crinoidal limestone and shaly intercalations, which is rich in stromatoporids and corals. The same level is exposed in the Dison section (Weber 2000). Based on the distribution of stromatoporids, Weber (2000) described in that interval a small lenticular structure, which he termed either a mound or patch reef. A facies analysis of the particular interval in the Dison section indicated a similar heterogeneous composition to that described by Laloux et al. (1996) for the Dolhain section (Fig. 5a). The horizontal and vertical extension of a particular facies was generally low. Even within single beds, the changes occurred rapidly and are numerous. The limestones were generally impure; in thin section, dispersed detritical quartz grains can be seen in some samples forming layers. Rapid facies changes,
170
M. ARETZ & E. CHEVALIER
Fig. 5. Strunian bioconstructions. (a) Intercalation of limestone and argillaceous beds. Characteristic heterogeneous composition of the Strunian succession in the Dison section. (b) Laminar stromatoporid (s) stabilizing a bioclastic substrate. The white dotted line indicates its lower surface. A fistulioporid bryozoan (b) started sediment stabilization, but was later encrusted. Dison section. Scale bar is 3 mm.
seen in the composition and texture of single thin sections, occurred. Stromatoporids, and tabulate and rugose corals, may be abundant locally. Colonies of encrusting growth mode acted mainly as sediment stabilizers (Fig. 5b). The encrustation of other colonies is only rarely observed. The development of a larger scale framework by stromatoporids and/or corals has not been observed. The dimensions of stromatoporid and coral boundstones are often limited, and do not exceed the dimension of a specimen and its immediate surroundings. However, isolated boundstone patches attain up to 1 m in thickness. The principal biostrome is possibly best described as a carbonate unit that consists of many different facies types, and has partly a higher abundance and biodiversity of stromatoporids and corals than other Strunian deposits. Few individual horizons of the carbonate unit may be classified as autoparabiostrome, but most parts are too heterogeneous to be classified as biostrome. The occurrence of the lenticular structure in the Dison section (Weber 2000) may point to the development of a reef, but our field observations do not show any uniform facies at this horizon. The origin of the distribution pattern of the stromatoporids remains enigmatic. Therefore, the general term bioconstruction is preferred for the Strunian interval, until better outcrops reveal the geometry of supposed framework. The development of a substantial framework cannot be excluded for the Belgian successions, but the rapid facies changes in connection with the numerous small sedimentary cycles would have hampered the development of any
large-scale framework. Therefore, reefs may only be found in more stable and/or protected environments. The main biostrome of the Vesdre Massif was the last occurrence of bioconstructional activity of stromatoporids in Belgium. Younger horizons in the Dolhain and Etroeungt formations containing stromatoporids were level-bottom communities. This ecosystem finally collapsed in the D - C boundary Hangenberg Event, and the stromatoporids became extinct.
Tournaisian reefs In the early Tournaisian (Hastarian) reefs are not known. The absence of reefal fabrics correlates with the scarcity of potential bioconstructors in the aftermath of D - C boundary extinctions and unfavourable facies, for example black shale environments of the Pont d'Arcole Formation. Microbial communities, successful reef builders in Eastern Australia during that time interval (Webb 1998), are not known from Belgium. Most rugose corals were solitary - colonial forms developed in later radiation events mainly during the Vis6an (Poty 1981, 1984). The first bioconstructions of Carboniferous age appeared with the Waulsortian mounds in the Ivorian (late Tournaisian).
Interval D: Waulsortian mounds. The Waulsortian mounds have attracted much attention since the late 19th century (see Lees 1988 for an historical review). The interpretation and nomenclature of these massive, mud-dominated carbonate bodies, up to several hundreds of metres thick, has changed several times (see Lees 1997). As the
BELGIAN FAMENNIAN AND DINANTIAN REEFS
171
Fig. 6. Distribution of Tournaisian reefs. (~) Waulsortian mounds. Note that the circles only indicate spatial distribution and do not correspond to individual mounds.
Waulsortian mounds are by far the most abundant bioconstructions in the Belgian Carboniferous, they have commonly been regarded as the only Carboniferous 'reefs'. The geographical distribution of the Waulsortian mounds is restricted to the western and central Dinant Synclinorium; no Waulsortian mound is known east of the town of Ciney (Fig. 6). It is not the aim of the present study to discuss the Waulsortian mounds in any detail; for this purpose the reader is referred to Lees & Miller (1995) and Lees (1997). Lees & Miller's definition (1995, p. 259) of a Waulsortian mound (their bank) is based on the dominance of an association of polymuds, the presence of at least one grain type assemblage (Lees et al. 1985) and the stratigraphical position. Cavities, including stromatactoid cavities, may or may not exist (Lees & Miller 1995). The origin of the carbonate mud is believed to be microbial, as indicated by the clotted textures (see Lees & Miller 1995; Devuyst & Lees 2001). In general, the locally abundant macrofauna within a Waulsortian mound (e.g. Sosoye Mound: Demanet 1923) did not contribute to any framework-building processes. The Belgian Waulsortian mounds consist macroscopically of three main lithofacies (Lees et al. 1977): (A) the so-called 'facies ~ veines bleues' comprises fine-grained pale limestones (biomicrites) with abundant spar-filled cavities (Stromatactis), which are often associated with fenestellid sheets (Fig. 7d, e); (B) the crinoid facies with mainly wackestone textures; and (C) the pale, homogenous limestones of the biomicrite facies. Polymuds are abundant in lithofacies A, but may also be present in the other facies. The carbonate bodies are surrounded by
peri-Waulsortian facies (Fig. 7b), which consist of well-bedded limestones, commonly with numerous chert nodules. Laterally equivalent facies corresponding to the two growth intervals of the mounds have been termed Bayard and Leffe facies. The Molign6e facies is always separated by the Leffe facies from the Waulsortian carbonate bodies. The lateral shallower equivalents of the Waulsortian facies are found in the Yvoir, Ourthe, Martinrive and Longpr6 formations (Fig. 2). According to Hance et al. (2001), the mounds formed in the distal areas of the upper Tournaisian carbonate ramp in the Dinant and locally in the Condroz (Ciney area) sedimentation areas (Fig. 6). Waulsortian mounds are of very different sizes and various geometries. They range from small metre-sized isolated lenticular carbonate bodies to aggregates ( = m o u n d complexes) of several hundreds of metres in thickness. This marked differences in size and geometry can be correlated to the abundance of Waulsortian mounds and the geometry of the shelf system on a north-south transect (Fig. 7a, b). In the northern area (e.g. Sosoye) mounds are relatively small and the distance between individual mounds is relatively large (up to some kilometres). Mound sizes and abundances increased in a southern direction, whereas the distance between mounds decreased. This development culminates in the southern mound complexes (Dehantschutter & Lees 1996). The Waulsortian mounds are commonly dolomitized in Belgium, making the study of primary fabrics rather difficult. However, Lees et al. (1985) proposed a bathymetric model for the Waulsortian mounds based on four grain-type
172
M. ARETZ & E. CHEVALIER
Fig. 7. Waulsortian mounds. (a) Well-known outcrops of Late Tournaisian age in the Dinant area. (b) Model for the distribution of Waulsortian mounds on a S-dipping ramp system, and for the geometry and thickness of the individual mound shape. (c) Sketch of the palaeobathymetric distribution of the grain type assemblage of Lees et al. (1985). (d) Polished slab of the core facies showing the digitated system of stromatactoid cavities (s) in polymuds (p). Slightly dolomitized. Moniat mound. Scale bar is 3 cm. (e) Thin section with polymud fabrics and stromatactoid cavity. Moniat mound. Scale bar is 12 mm. (a) and (b) are modified from Lees (1997)
assemblages (Fig. 7c). Phase A may represent palaeodepth of up to 300 m, whereas phase D represents deposition within the photic zone. Mound initiation could start in all four phases and mound growth could cease well before the photic zone. However, the ramp geometry in Belgium suggests that the southern-most mounds and complexes should reflect the deepest member of the suite. The stratigraphical occurrence of the Waulsortian mounds in Belgium is now limited to the Ivorian (see Poty et al. 2006). Waulsortian mounds formed during the third and fourth sedimentary cycles (interpreted as third-order sequences) of Hance et al. (2001), thus enabling the differentiation of two growth intervals. Waulsortian mounds are not reported from early Vis6an strata, possibly related to the major sea-level drop around the Tournaisian-Vis6an boundary. Most of the shallower environments in middle and proximal positions of the Tournaisian ramp
are devoid of any bioconstructional activity. Only the Tournai region may contain some bioconstructions. Mortelmans (1976, p. 149) reported local accumulations of the massive tabulate corals Michelina favosa and M. tenuispeta in the Tournai Formation (Vaulx Member). Possibly they are identical with the Michelina biostromes from the top of the Vaulx Member of Legrand et al. (1967, p. 167). Field observations in the Lemay Quarry showed some concentrations of michelinid corals at this stratigraphical level, but the formation of a bioherm or biostrome cannot be confirmed. The distribution of the colonies in the horizons immediately below the 'Gras D61it' (local stratigraphical marker) shares some similarities with coral meadows described from the late Vis6an of N W Ireland (Aretz 2002). The development of a substantial framework was not observed. A lenticular structure of some 10 m in width and about 15 m in height is exposed in the SW part of the Lemay Quarry (the Gras D61it passes
BELGIAN FAMENNIAN AND DINANTIAN REEFS
173
Fig. 8. Facies types within the Terwagne Formation of the Northern Namur Syncline, which may be related to reefal facies. (a) Boundstone formed by the heterogeneous meshwork of fistuliporid bryozoans (b), Aphralysia (A), laminated stromatolitic (1) and micritic (m) layers. Scale bar is 5 mm. (b) Accumulations of microconchids in partly laminated peloidal limestone. Scale bar is 5 ram.
above its top). It consists of dark, fine-grained bioclastic wackestone with thin layers of coarser crinoidal debris. This facies is the same as that of the surrounding sediments. This structure is a part of the Vignoble Member, which has been interpreted as large slump by Mortlemans (1969). Therefore, this structure is probably owing to sedimentological and not bioconstructional phenomena, although it resembles a mud-mound in some aspects.
Vis4an reefs A major change in basin configuration characterized the early Vis6an time. The former Late Tournaisian ramp was transformed into a carbonate platform, resulting from southward progradation of shallow-marine facies (Hance et al. 2001). Early Vis6an reefs are so far not known. Their absence is enigmatic, as suitable environments were widely available. However, the Terwagne Formation in a borehole in the Northern Namur Syncline contains two facies types, which may be related to bioconstructional activity. Type A (Fig. 8a) represents a boundstone formed by the heterogeneous meshwork of crusts, which consisted of fistuliporid bryozoans, Aphralysia, calcimicrobes, microconchids, and stromatolitic and micritic layers. Small spar-filled cavities of very irregular shape occur. Numerous bioclasts are incorporated through encrustation into this meshwork. Type B (Fig. 8b) comprises accumulations of microconchids in laminated, peloidal limestones, which may be found in reefs, but are not restricted to them (see Dreesen & Jux 1995; Aretz 2001). As the geometries of the horizons that contain these two facies types are unknown, and at least for type A are rather thin ( < 20 cm), only their
presence is recorded here. Better exposures may help to demonstrate the development of reefal fabrics in the early Vis6an.
Interval E: microbe-fenestellid bryozoan reefs. The oldest recorded Vis~an reefs in Belgium were small patch reefs, which formed in the Corphalie Member (sequence 0) of the Lives Formation (middle Vis6an). These reefs developed in the eastern Namur Syncline (Fig. 9). Lauwers (1992) described five small and one unusual large reef (Fig. 10d) from a disused quarry in a suburb of Namur (Bomel), Chevalier & Aretz (2005) recorded a single reef from the active Engihoul Quarry (Fig. 10a). However, during the present study at least two more reefs were found in the Engihoul Quarry, and disused quarries north of Engis contain possibly three more reefs (Fig. 10b). The reef framework consisted of a meshwork of fenestellid bryozoans and microbial communities. The evidence for these microbial communities is the presence of 'skeletal' calcimicrobes, numerous thin dark micritic layers, which encompassed the fenestellid bryozoans, and clotted peloidal textures (Fig. 10d, e). The framework was locally further enhanced by abundant brachiopods (Fig. 10c), and aggregates of microbial coatings and mats interbedded with layers of early fibrous cements, and Aphralysia crusts (Chevalier & Aretz 2005). The primary framework contained open pores of various sizes. This primary porosity was reduced by cryptic communities (e.g. columnar Aphralysia with downward growth orientation, Fig. 10e), and the infill of fine-grained carbonate mud (often peloidal). Finally, porosity was occluded by blocky calcite cements.
174
M. ARETZ & E. CHEVALIER
Fig. 9. Distribution of Vis6an reefs. | Microbes-fenestellid bryozoan reefs; Q stromatolite reefs; ~ microbesbryozoan biocemenstone reefs; @ stromatolite-microconchid reefs; (~) microbial reefs (Campine type); | late Vis6an coral bioconstructions.
All reefs developed in the transition interval from full-marine to restricted-marine facies of the sequence 0 (Chevalier & Aretz 2005). The macroscopically sharp change from bioclastic to stromatolitic limestone (bed to bed contact) is microscopically more complicated. The installation of the restricted-marine facies was gradual and was characterized by the patchy occurrence of bioclastic and peloidal facies, in an up to 2 mthick interval, before the general establishment of restricted marine facies. Reef initiation is correlated with the deposition of a bed remarkably rich in brachiopods (Composita bed). The composition of this widespread lithostratigraphic marker bed varies between the two end members: brachiopod rudstone and oncolitic grainstone. The stratigraphically lowest microbial coatings were observed some decimetres below the Composita bed. However, successful reef initiation and growth seems to be related to a well-balanced abundance pattern of fenestellids and microbial communities. This process is as yet not well understood. The succession of sequence 0 in the Engihoul Quarry may or may not contain reefs, although the composition of the Composita bed does not vary significantly. In contrast, the cessation of reef growth was related to decreasing water depth, which enabled the installation of the restricted-marine stromatolitic facies. This sea-level fall is well documented in the diagenetic
history of the larger Bomel reef (Lauwers 1992). However, this reef, which is 10 times larger than all other reefs, is enigmatic. Although it started to form at the same level as the other reefs, it still persisted when reef growth was stopped elsewhere and stromatolitic facies dominated. Possibly more accommodation space may have been locally available (subsidence, synsedimentary fault?). Further localities in the Dinant Synclinorium possibly contain reefs in the Lives Formation. Among the Vis6an reef localities reported by Dupont (1883) was Assesses, where below the police station some massive limestone blocks occur that are rich in brachiopods. Microscopically, the limestone consists of peloidal limestone with grainstone-wackestone textures with abundant brachiopods. A framework has so far not been an observed constraint, but the very poor outcrop conditions hamper interpretation. The Lives Formation east of Modave contains two particular localities. The succession in the former railway cut contains levels with peloidal boundstone facies, and one thin section shows the development of a Multithecopora boundstone (tabulate coral). Encrustation around fenestellid bryozoans in the peloidal boundstone occurs. The boundstone layers occur within a bioclastic-dominated facies. As the geometries and distributions of the boundstone levels are not well understood, the term reef is
B E L G I A N F A M E N N I A N A N D D I N A N T I A N REEFS
175
O.-~
O
,,i.o
O
q3
"~ az ..
;..~ r/~ +.o
~0 ,.c~ 0 ,.~
I=~ ~ ~ ~.~
o
Oc~
o
~
,,,
~e
N~
~ .~ ~~ ~~ ~Nu~ o~ x~ ~
~
~'~
~
.'.~
~
176
M. ARETZ & E. CHEVALIER
Fig. 11. Stromatolites reefs in Transcar Quarry. (a) Outcrop photograph indicating the alignment of small reefs (R). (b) Onlap (arrow) on a reef. (e) Peloidal boundstone showing some thrombolitic textures, which form a larger lamination. Scale bar is 5 mm
not used herein, although framework-building processes may be found here. A locality about 100 m to the west shows many similarities in the texture of the limestones and the abundance of brachiopods with the Assesses locality.
Interval F: stromatolite reefs. The Maizeret Member of the Grands-Malades Formation (middle Vis6an) exposed in the disused Transcar quarry contains a series of patch reefs. Access to the reef level is now very difficult owing to the configuration of the quarry. The description given here is based on the field data, a photo mosaic and a measured section, which cuts through the only accessible reef. The reefs developed in the lower part of the Maizeret Member, which is formed by bedded limestones of laminated peloidal grainstonewackestone. These stromatolitic limestone beds showed many gently wavy surfaces, but up to 15 well-delineated lenticular-domal bodies have been identified in at least three horizons (Fig. 1 l a). The individual bodies are less than 2 m in height and 5-15 m in width. An onlap on these bodies has been documented locally (Fig. l lb). The carbonate bodies consisted of laminated peloidal boundstones. Some small patches of thrombolitic fabrics occur within this mainly stromatolitic fabric (Fig. 1 l c). Bioclasts have not been found so far.
The carbonate bodies are interpreted as stromatolitic reefs that developed in the restricted marine environment of the Maizeret Member. The limiting factors of these reefs are not understood. The development of stromatolite reefs in a stromatolitic limestone is somewhat paradoxical. The thrombolitic patches are the only difference between reef and off-reef facies, and possibly enhanced framework-building processes, but their relative rareness does not explain the initiation of reef growth. Although there is no obvious reason for the cessation of reef growth, the development of reef horizons may point to the existence of small-scale sedimentary cycles. These cycles could be coupled to similar processes as in the older Lives Formation, and thus repeated shallowing would limit the accommodation space and so terminate reef growth.
Interval G: microbes-biocementstone-bryozoan reefs. Dissolution breccias are common phenomena in the upper middle Vis6an succession of the D i n a n t - N a m u r Basin. The term 'Grande Br6che' has commonly been applied to these deposits, but different names are in use for specific regions (e.g. Charleroi area: Delcambre & Pingot 2000). Reefs of time interval G are either encompassed by these breccias, included in the breccias or the position in respect to the breccias is poorly understood owing to poor outcrops. Therefore,
BELGIAN FAMENNIAN AND DINANTIAN REEFS
177
Fig. 12. Microbes-bryozoan-biocemenstone reefs. (a) Boundstone facies of the Bouvigne reef, polished section showing the loose meshwork of bryozoans and early fibrous cements and peloidal sediments. Note the abundant stromatactoid cavities. Polished slab. Scale bar is 2.5 cm. (b) Brecciation within the reef fabric at Bouvigne. Note the 'lamination' of fibrous cements around the fenestellid sheets. Polished slab. Scale bar is 2.5 cm. (c) Detail of (b) showing the several generations of fibrous cements formed around the fenestellid sheets. Scale bar is 5 mm. (d) Thrombolitic fabric developing into a stromatolite surrounded by peloidal and automicritic fabrics, Bouffioulx reef. Note the small cavities with an outer fibrous and inner blocky cement generation. Scale bar is 4mm.
the biostratigraphic control of this interval of reef development is not always good. Parts of the reefs may be age-equivalent of the reefs in intervals F and H. Reefs in interval G developed west of a line drawn between Dinant and Namur (Fig. 9). Typical examples are found in a field west of Bouvignes (location in Delcambre & Pingot 1993) and the cliff at Bouffioulx (location in Pirlet 1967b). The dimensions of the Bouvignes reef are difficult to establish, as the outcrop is poor, but the reef facies has been traced for up to 100 m with a maximum thickness of 10 m which indicate, respectively, minimum diameter and height. The reef consists of a complex fabric (Fig. 12a) of fenestellid bryozoans in a peloidal matrix, and multigenerations of cements. Abundant thin-shelled brachiopods are incorporated into the reef fabric; sponge spicules occur locally.
The fuzzy shapes of the small peloids indicate that their formation was related to microbial communities. However, the peloids and micritic filaments do not show structures and alignments typical of stromatolitic or thrombolitic fabrics. The fenestellid sheets are commonly surrounded by thin, fibrous, inclusion-rich, brownish cements (Fig. 12a). Cavities in the primary reef framework were irregularly distributed and filled by fine-grained peloidal mud and/or cement. Some cavity roofs only consisted of peloidal mud, thus indicating rigidity of the primary microbial structure, as otherwise the cavities would have collapsed. The cavities are always filled by several generations of cement. The first generation is mainly fibrous to radiaxial fibrous and inclusion-rich; later generations tend to be equigranular and blocky. Peloidal layers may occur between individual cement generations, thus pointing to an early cementation processes.
178
M. ARETZ & E. CHEVALIER
However, dissolution cavities were always filled by clear blocky cements. The relation of rare bioclastic wackestones and the reef fabric is not clear. These fine-grained wackestones could be interpreted as intrareef facies. There is no direct contact between the reef and the supposed surrounding Lives Formation (Delcambre & Pingot 1993). Therefore the depositional environment of the reef remains uncertain, but the abundance of bryozoans and brachiopods indicate almost full-marine conditions. The depositional environment may be comparable to those of time interval E, thus middle-upper parts of the platform might be favoured. The Bouffioulx reef developed in the Bouffioulx Member (Delcambre & Pingot 2000). This member formed on top of the Lives Formation and below the Hoyoux Group, which contained all formations of the Warnantian. Therefore, the member is time-equivalent to the Grands-Malades Formation. The depositional environment of the Bouffioulx Member was restricted marine, as evidenced by evaporitic phenomena and the dominance of algal limestones (Delcambre & Pingot 2000). The Bouffioulx Member exposed next to the 'Grotte de Mantreau' consists macroscopically of homogeneous, massive limestone, which is locally rich in brachiopods and bryozoans. This homogeneity is not seen in thin section. The limestones include bioclastic wackestonespackstones, peloidal grainstones, peloidal boundstones (possibly non-skeletal microbialites) and microbreccias. All lithotypes have been strongly overprinted diagenetically, mainly as a result of dissolution and/or brecciation. The 'veines bleues' ( = stromatctoid cavities) of Delcambre & Pingot (2000) were found in all lithotypes, and most seem to represent dissolution cavities. The complex diagenetic overprint makes the reconstruction of the primary sedimentary structures very difficult. Some parts of the section, especially on top of the cliff, contained primary reef fabrics (Fig. 12b). These reef fabrics consist of stromatolites and thrombolites. The microbial-dominated frame is enhanced through the formation of biocemenstones and locally the incorporation of bryozoans. However, brachiopods and bryozoans show abundance peaks in the bioclastic sediments. The internal structure of the reef fabric is rather chaotic, and possibly best described as patchy. The growth direction of individual microbial mats may vary by as much as 45 ~ (Fig. 12c). These steep dips implied rigidity for the microbial mats. The spatial distribution of the reef facies is not well established. Alternations with off-reef bioclastic facies
occurred frequently. The microbreccias may represent fore-reef breccias, but their brecciation could also be related to later dissolution. The interpretation of the peloidal grainstones is difficult, as they may be associated with the reef fabrics or the off-reef sediments. Possibly, the Bouffioulx reef was not a single large bioconstruction. The alternation of bioclastic and algal limestones may point to sedimentary cycles, and then small patch reefs (Fig. 12d) of the microbial fabrics may have formed repeatedly. The patchy occurrence of the reefal fabrics may point to that spatial distribution. If the entire Bouffioulx Member is interpreted as a breccia, it has to be stated that this breccia would contain clasts that were clearly derived from reefal material described above. Further reefs of time interval G were found south of Salet (Blanc Cailloux), north of Arbre (old railway cut), and at Landelies. Pirlet (in Paproth et al. 1983) mentioned reefs in the Dinant area (Chession, Citadelle), which might be attributed to this reef interval, but their presence could not be confirmed in the present study. Interval H: stromatolite-microconchid reefs. The youngest middle Vis6an reefs are found in the Bay Bonnet Member of the Grands-Malades Formation. The reefs are small, commonly metre-sized, and developed locally in the Namur and Condroz sedimentation areas. In the northern Namur Syncline, Seilles Quarry the reefs formed well-delineated lenticular structures of peloidal boundstone. Within the Andenne area 10 reefs could be identified. Field observations showed a prominent horizontal lamination of the boundstone, with dips of up to 35 ~ at the margin of the reefs. Microscopically, the stacking pattern of the layers was very irregular. Lamination is the result of the spacing of peloids, and the preservation of thin elongated micritic 'filaments'. Individual layers were often rather short, and the upper boundaries of individual layers are commonly irregular wavy structures (Fig. 13a). Locally, small domed stromatolites developed from such a layer. Lamination is not developed throughout an entire reef. Non-laminated areas of carbonate mud with only few peloids occur. However, this facies type is in strong contrast to the mud- and wackestones below the reefs (Fig. 13a), which contain bioclasts. The patch reefs of the Bay-Bonnet quarry (Vesdre Massif) contain abundant microconchids. The framework of these reefs (Fig. 13b) is a result of the interaction of the microbial communities, forming stromatolitic reef fabrics, and the erect growth mode of the
BELGIAN FAMENNIAN AND DINANTIAN REEFS
179
Fig. 13. Bay Bonnet reefs. (a) Base of a stromatolitic reef in the northern Namur Syncline. The reef formed on top of a fine-grained bioclastic wackestones. Scale bar is 5 mm. (b) Stromatolite-microconchid reef of Bay Bonnet Quarry. Top to the left. Scale bar is 7 mm. microconchids. However, the stromatolitic fabrics seem to have dominated. The depositional environment of the reefs was restricted-marine, as indicated by low biodiversity and widespread formation of stromatolitic limestone. Reef formation possibly started when a critical number of microbial communities forming the first stromatolitic reef fabrics had been reached. Termination of reef growth might have been owing to reduced accommodation space. The compositional differences of the reefs may be related to the harshness of the environments. The total absence of bioclasts in the Bay Bonnet section may indicate a restricted lagoon. Less restricted periods occurred, at least periodically, in the Andenne area.
Phase I: microbial reefs (Campine type). Upper Vis6an reefs have been described under the terms 'cryptalgal reef or microbial buildups' from the Campine Basin based on surface and subsurface data (Muchez & Peeters 1986; Muchez et al. 1987, 1990). Surface data for the Campine Basin are restricted to the Vis6 area. Intensive block-faulting characterizes the Vis6 area and very different tectonic blocks have been recognized (Poty 1991). This highly influenced the deposition of Vis6an sediments, and thus the Vis6 Formation comprises very different lithotypes, and varied greatly in thickness (zero to some hundreds of metres). Muchez & Peeters (1986) described a microbial reef from the stratotype of the Vis6an (quarries F, G and H: Pirlet 1967a). This section is now not accessible, but two more reef localities can be recorded for the Vis6 area. Blocks recovered by E. Poty west of Richelle consist of microbial boundstone. This boundstone facies (Fig. 14a) contained: (1) a thick crust of
calcimicrobes and Aphralysia around small solitary corals (Amplexocarinia sp.), and bryozoans; (2) peloidal packstones, with thrombolitic outline; and (3) various stromatactoid cavities. These cavities are either filled with carbonate mud or cemented with at least two generations of cement. Coarser bioclastic and intraclastic sediment may be accumulated in open spaces between large crusts. The geometry of the reef to which this reef fabric belonged is unknown. In an outcrop north of Dalhem, in the Berwinne Valley, the thick-bedded Vis6 Formation contains sedimentary breccias, which included various clasts of reefal origin. Some clasts represent the same facies as that recovered at Richelle; others consist of peloidal limestone with sponge spicules, or bioclastic wackestones with abundant microbial crusts of several mm in widths and sizes. The source for the breccias must have been nearby; a more accurate provenance would require the study of transport directions, which is hampered by poor outcrops and fast lithological changes. Seismic reflection surveys revealed the existence of domal structures in the subsurface of the Campine Basin. The Heibaart and Poederlee boreholes drilled in these structures contained thick packages of microbial boundstones and peloidal packstones-wackestones (Fig. 14b). This reef fabric is characterized by peloidal mud commonly with clotted textures, thrombolites and thick crusts of Aphralysia. The crusts originated on clotted peloidal limestone (indicating its rigidity), cements, bryozoans and corals. Automicrites formed locally. Early cementation played a vital role in the formation of the microbially initiated framework, as the meshes of the framework were, in places, rather large and the formation of step and overhanging structures
180
M. ARETZ & E. CHEVALIER
Fig. 14. Microbial reefs of the Campine type. (a) Cracoean-typed reef facies of the Richelle area. Amplexocarinia sp. surrounded by crusts of Aphralysia?; scale bar is 9 mm. (b) Heibaart borehole. Dominance of early cements and micritic crusts of supposed microbial origin. Polished slab. Scale bar is 3 cm.
were supported by cementation (Fig. 14b). The cementation prevented the collapse of the framework meshes, and reduced the open pore space. Smaller stromatactoid cavities were completely cemented by brown, fibrous cements, but the larger cavities contain numerous generations of cement (Fig. 14b). However, the biodiversity of some microbial boundstones was high. Almost the entire spectrum of the Vis6an marine fauna can be recognized within the various bioclasts. A remarkable feature is the local abundance and excellent preservation of Koninckopora. Other calcareous algae are mainly present as unidentified tubes and filaments. However, none of these organisms played an active role in framework-building processes. The passive dwellers were incorporated into the reef fabric through incrustation, and therefore passively enhanced rigidity. The reef fabric facies intercalates with bioclastic wackestones that contained a rich fauna and algal flora. The Poederlee borehole contained several bioclastic horizons, whereas only a single 20 m-thick bioclastic unit was reported in the Heibaart borehole. If the domal structure can be correlated for the most part with the distribution of the microbial boundstone facies, then reef dimensions would have been several hundreds of metres in width and up to about 150 m in height. Muchez et al. (1987) reported depositional surfaces with a dip of 28 ~ for their lower subunit 4 and 10~ for their upper subunit 4. If the bioclastic units form distinctive horizons that could be traced laterally throughout the domal structure, the domal structures are aggregates or complexes of microbial reefs. If the bioclastic units do not form laterally persistent horizons, the domal
structures correspond to a single reef which is partly covered by off-reef facies several times, possibly connected to sea-level fluctuations. Since both borehole were drilled in the central positions of the domal structures, Muchez et al. (1990) rejected the possibility that the Poederlee borehole crossed a more marginal position of a single reef, and supported the idea of a deeper depositional environment for the Heibaart development. Initiation and termination of reef growth are poorly understood. Initiation might have been linked to the increase in the abundance of the microbial communities, and thus implies an improvement for factors controlling these communities. Reef termination was possibly related to the major sea-level drop around the Vis6anNamurian boundary. The only biohermal structures of Warnantian age south of the Brabant Massif are bioherms, up to 3 m high, in the Haut-le-Wastia area reported by Pirlet (in Paproth et al. 1983). Their composition, abundance and precise stratigraphical position are unknown. Reef type K: Viskan coral biostromes. The bioconstructional potential of coral biostromes was very varied (see Aretz 2002). Hydrodynamically controlled accumulations of coral debris ( = Allobiostromes) do not possess any, whereas autobiostromes may represent failed reef developments. Both types can be found in the Vis6an succession. Siphonodendron martini biostromes are a common feature in the middle Vis6an Lives Formation. These biostromes consist mainly of coral fragments of various sizes. Intact colonies are
BELGIAN FAMENNIAN AND DINANTIAN REEFS rarely documented. There are no frameworkbuilding processes known, as encrustation and thus enhancement of a potential framework is absent. These low-height, monospecific biostromes developed repeatedly in the same position of several parasequences. Biostrome formation took place in the short interval where wackestone textures changed into grainstones. Biostromes formed the top of the lower mud-dominated unit. The vertical growth of the fragile phaceloid colonies was hampered by higher energy levels, but the horizontal growth of the biostrome was nearly unlimited because the vast middle Vis6an platform could be colonized. Thus, the biostromes can be traced horizontally for kilometres, although rarely exceed the height of two colonies. Biostrome formation stopped when higher energetic levels were attained across the entire platform. Hence, the development, distribution and preservation of the biostromes were mainly controlled by the hydrodynamic regime and the geometry of the shelf. A different type of biostrome is preserved at Royseux in the late Vis6an Anh6e Formation (Aretz 2001, 2002). The development of parasequences characterizes the sedimentation pattern of the Royseux sections. A polyspecific autobiostrome developed at the base of sequence + 2. It consists of phaceloid and cerioid colonial rugose corals. Following the stabilization of a crinoidal grainstone, corals colonized the area intensively and grew upon each other. Thus, coral boundstones formed at least locally, but this facies did not persist very long. The absence of encrusting bioconstructors and a rapid sea-level rise terminated the development of a substantial coral-dominated reef framework. A threefold second coral biostrome developed on top of sequence + 2 (Aretz 2001, 2002). It consists of a thin lower Siphonodendronjunceum autoparabiostome, which is succeeded by a conglomerate containing a rich coral fauna. The upper unit is formed by a coral meadow of large hemispherical Siphonodendron martini colonies. The development of a substantial framework could not be seen in this second biostrome.
Discussion
Controlling factors Reefs are complex ecosystems that are controlled by numerous intrinsic and extrinsic factors. Most of theses factors are connected to each other, and it may be difficult to distinguish a single controlling factor (Aretz 2002, p. 98, Fig. 57)
181
Fig. 15. Hierarchical model for the factors controlling reef development. The factors directly controlling the reef organisms are summarized under palaeobiology; mainly abiotic factors controlling the local tectonosedimentary environment are encircled and related to global extrinsic factors.
Famennian-Vis6an reef building and destruction was controlled by a variety of factors. They comprise a spectrum with its end members: (1) factors that controlled the development of the individual reef; and (2) factors of global importance, which influenced reef development by coupled processes. In this study controlling factors were grouped into the three hierachical levels: palaeobiology; local environment; and regional and global environments (Fig. 15).
Palaeobiology. As reefs were biologically controlled carbonate bodies, the understanding of the biology of a potential reef builder and its biological requirements is essential. This task is often very problematic owing to the incompleteness of the geological record for certain information, especially where soft tissues and metabolic processes were involved. However, some crucial information on possible controlling factors can be gathered from the skeletal remains. The development of reef framework depends among other factors, on growth modes, growth rates, fragility and abundance of bioconstructots. Superstratal and constratal growth modes were recognized as basic types for scleractinian corals (Insalaco 1998). Thus, ideal frameworks should contain superstratal bioconstructors for developing topographic relief and constratal bioconstructors for its stabilization. The transfer of this concept to
182
M. ARETZ & E. CHEVALIER
Palaeozoic reef builders is not without problems and is not well understood (Aretz 2002). Flat, laminar stromatoporids within the Strunian bioconstructions showed the typical sedimentbounded constratal mode. Especially large coral colonies of domal and branching growth forms indicate superstratal growth. However, some larger specimens (stromatoporids in the Dison section, Siphonodendron and Lithostrotion in the Vis6an) show recovery growth styles in consequence of the partial cover by sediments. This pattern is evidence for low elevation above the sediment surface and therefore implies constratal growth. Constratal growth may also support the development of dendroid colonies because of reduced fragility, but, by analogy with modern corals, would not have been the general growth mode for this colony shape. High growth rates would have favoured the development of a superstratal growth mode and, according to the available data (Scrutton 1998), some fasciculate corals show the highest growth rates of Palaeozoic corals. This would point to a superstratal growth mode for at least some corals. The low growth rates of stromatoporids (Scrutton 1998) may suggest a constratal growth mode. This suggestion is backed by the aforementioned growth patterns in the Dison section. However, growth rates of Palaeozoic corals have only rarely been measured, and Scrutton (1998) reported significant differences in a single Silurian species depending on growth in reef or non-reef positions. The unusually high growth rates of Acropora, which forms the bulk of modern Caribbean reefs, have not been reported for Palaeozoic corals. Therefore, slow growth rates of Palaeozoic reef builders in combination with sedimentation rate were an important controlling factor for reef-building processes. Thus, the failure of Strunian bioconstructors to form large frameworks was owing to their inability to keep up with the periodic sediment influx of the sedimentary cycles. The middle Visdan biostromes provide an example of the hydrodynamic control of coral growth, and show the fragility of the fasciculate Siphonodendron colonies. In addition, the weak or non-existing fixation to the sediment of Palaeozoic corals further reduced the ability for successful framework building, at least for higher energy conditions. The absence of reefs from the high-energy facies within the Belgian successions may be based, at least partly, on these palaeobiological limitations. However, the limitations do not generally exclude corals or stromatoporids from successful reef building in these facies.
Some other factors are difficult to evaluate. If Palaeozoic corals lived in symbiosis with photosymbionts, light would have been a controlling factor for their distribution. To what extent so-called calcareous algae were lightdependent during the time interval studied is unknown, especially as the taxonomic position of most taxa is highly questionable. Therefore, reconstructions of the depth of deposition are somewhat problematic. Nutrients would have strongly controlled the distribution and abundance of organisms. If the present day situation were to be used as an analogue, DevonianCarboniferous corals would prefer oligotrophic conditions, whereas more eutrophic conditions would be favoured by algae and (?) microbial communities. The influx of nutrients is commonly interpreted based on the amount of siliciclastic input. Therefore, the Famennian ocean would have been nutrient rich, and thus the development of successful reef construction by corals may have been strongly hampered by this ocean chemistry. The dominance of microbial communities in the Vis6an reefs may indicate eutrophic conditions, and coral-dominated reefs did not exist in consequence of nutrient richness. However, as this factor is difficult to establish, and reefs with significant coral contribution existed elsewhere (Aretz & Herbig 2003a, b), the absence of coral-dominated reefs in Belgium seems to be related to other factors. Palaeobiological factors such as competition for space, nutrients, light, etc., on intra- and interspecific scales are likely to have affected and controlled reef formation, but their extent is poorly understood. The dominance of a species may point to optimum conditions for the particular species, or its success in competition with other species.
Local environment. This hierachical level comprises factors that control the reef directly, but also affect the immediately surrounding environments. Again, the geological record of the specific factors differed significantly. However, water depth is one, if not the most important, controlling factor for bioconstructions of the Famennian-Vis6an interval. Almost all bioconstructions reacted sensitively to changes in water depth. These changes, although not exclusively, influenced the initiation of reef growth (interval E), resulted in faunal and fabric changes (intervals D, E, G and I), and terminated reef development (almost all intervals). The water depth also controlled the composition of the reef fabrics, and the overall biodiversity of the reefs, e.g. interval D, in comparison with intervals E and F.
BELGIAN FAMENNIAN AND DINANTIAN REEFS Salinity and temperature are partly connected to water depth. Stromatolite reefs and especially stromatolite-microconchid reefs were adapted to restricted marginal marine settings in very shallow water. These environments certainly would have experienced anomalies in salinity and temperature. Turbulence, also strongly connected to water depth and surface geometry, controlled the distribution of the middle Vis6an biostromes. Short-termed perturbation in the energetic regime, as a consequence of storms or hurricanes, influenced the distribution of bioclasts, resulting in the winnowing of carbonate mud, the death of reef builders and dwellers, and possibly the reworking of reef areas. Apart from the last consequence, the influence of these perturbations is recorded in all reefs of moderate-high biodiversity. Reefs of low biodiversity, mainly developed in restricted environments, were rarely affected, possibly as reef development was directly stopped. The disintegration of reef limestone and the accumulation of reef debris on the flanks or next to the carbonate bodies are both partly due to the same perturbations (e.g. type A).
183
Siliciclastic input into the depositional environment strongly controlled reef formation and growth. Its relation to nutrients has been discussed above, but the complete cover of reef fabrics by siliciclastic sediment diminished or terminated reef growth as seen in the Strunian bioconstructions.
Regional and global environments. The highest hierachical level comprises all factors of regional and global importance, which mainly influence reefs indirectly. It is evident that most of factors of lower hierarchical levels are triggered by these factors, but connections are also abundant at this level. The shelf geometry as a consequence of subsidence rate, tectonics and geodynamic evolution strongly controlled reef formation. The shelf geometry is one important factor for the number of potential reef positions (Fig. 16), although not all potential positions were available at all times, and not all positions during a given time slice were necessarily occupied. The development of the Waulsortian mounds (phase D) was connected to the distal area of the upper Tournaisian ramp. In the Vis6an, this position was no longer
Fig. 16. Position of the reefs in a single time interval of reef development. Distribution on an idealized shelf transect indicating potential reef positions.
184
M. ARETZ & E. CHEVALIER
available because of the southward progradation of shallow-marine facies and the transformation into a platform. Thus, deep-water bioconstructions were not developed in the Vis6an. However, the important block-faulting in the Vis6 area, and possibly in the middle Campine Basin, allowed the formation of large microbial reefs (Campine type). The Strunian bioconstructions matched the depositional environments of the Strunian mixed siliciclastic-carbonate ramp, and were restricted to the time of this particular basin configuration. Catastrophic events influenced all of biological evolution. Reef builders became extinct, for example stromatoporids in the Hangenberg Event, or suffered heavy losses. Depending on the number of losses and perturbation of the reef fabrics, recovery took time and available reef positions could not be filled owing to the absence of reef builders, for example during Early Tournaisian time. Climate and sea-level oscillations are indirectly coupled. The time interval studied encompassed the transition from the Devonian greenhouse to the Carboniferous icehouse climates. The development of the Gondwana glaciations in the Namurian succeeded a time with repeated warming and cooling (e.g. Bruckschen et al. 1999). Weber (2000) correlated reefbuilding intervals in the Famennian with warmer climates. The numerous middle-upper Vis6an sedimentary cycles were of glacio-eustaic origin, and thus controlled the development of the small reefs that developed in some of these cycles. Note that Mistiaen (1984) speculated about the connection of ocean water chemistry and the presence of stromatoporids. Biomineralization processes may have been connected to ocean chemistry, but Stanley & Hardie (1998) showed that the change from a calcitic to an aragonitic sea took place after the interval studied.
Comparison and global trends The abundance of Famennian reefs is generally considered to be low. However, it has to be stressed out that many large complexes of reef carbonates are only known from subsurface data or have been insufficiently described (Webb 2002). When potential reef positions became available, however, reef development took place at least locally, as seen in the Belgian Famennian. Webb (2002) reported major Famennian reef systems, with platform-edge complexes, mounds and reefs of deeper water setting, stromatoporiddominated reef mounds, and Strunian-Late Famennian biostromes. The Belgian succession
contains only parts of this spectrum. The development of large complexes was totally absent. According to Webb (2002), the Baelen reef represents a deeper water mud mound, but the abundance of the biota there may point to a shallower depositional environment. The microconchid reefs of the Evieux Formation are so far unique in the Famennian reef pattern. The Carboniferous examples show that the development of such reef types is typical for marginal marine settings. A detailed study of comparable facies in Famennian succession elsewhere may prove their presence. Thus, this recorded rarity is seen as an artefact The Strunian bioconstructions of Belgium with the abundance of stromatoporids and some corals mirror the presence of biostromes in the Strunian successions of southern China (Shen et al. 1997). Reef development was completely absent from Belgium during early and middle Tournaisian time. Although not very common, reefs of these ages are known from Australia (Webb 1998; Aretz & Webb 2007), the Russian platform (for a summary see Webb 2002) and North America (Waulsortian of Montana). Common to all is the dominance of calcimicrobes and nonskeletal microbialites in the reef fabrics. The late Tournaisian was the first peak time for the development of mounds in deeper water settings. The formation of Waulsortian mounds is excellently documented in Belgium, but the development of Waulsortian-like mounds/buildups, as widely reported from the United States (see Lees & Miller 1995), is unknown although potential suitable depositional environments have been available in Belgium. The carbonate body in the Lemay Quarry might point to this type of buildup. The drop in the sea level around the Tournaisian-Vis6an boundary stopped the development of Waulsortian mounds in Belgian, although their development continued in other regions into the early Vis6an. The absence of reefs in the lower Vis6an succession of Belgium correlates with the global trend. The first reefs with a significant contribution of corals to framework-building processes are known from the Arundian of NW England (Adams 1984). Kelly & Somerville (1992) described mounds from Ireland, which may be intermediates between Waulsortian mounds and younger shallow-water reefs. Possibly the almost time-equivalent reefal facies found in the northern Namur Syncline point to this type of bioconstruction, and shows that reefs were more abundant than previously thought. However, their rareness on a global scale is evident. The high diversity of reefs in the Belgian middle
BELGIAN FAMENNIAN AND DINANTIAN REEFS Vis6an is unique and antedate the development observed elsewhere in the late Vis6an. The general importance of microbial framework building in all Belgian Vis6an reefs is also seen in most reefs developed elsewhere. The facies of Belgian Vis6an reefs were of relatively low diversity compared to the global pattern, where microbial frameworks were supported by very different bioconstructors, such as corals, sponges, calcareous algae, crinoids, brachiopods and bryozoans. However, the major diversification event in the late Vis6an is not well represented in Belgium, as strata of that age are only rarely preserved. However, the microbial reef (Campine type) in the Vis6 area and the Heibaart and Poederlee boreholes are very similar to the Cracoean buildups from northern England (Mundy 1994), and thus point to the development of a barrier of large microbial reefs along the platform edges north of the Brabant Massif. As the development of these reefs is possibly directly related to fault blocks, this barrier might have been an alignment of reefs with very different distances between the individual reefs. Because neither the Dinant-Namur nor the Campine Basin contained deeper water facies of late Vis6an age, the second global peak of deeper-water mounds was not established in Belgium.
Age
Reef type Microbial reefs A
(Baelen type)
B
Microconchid reefs
Occurrence and abundance of reefs
rare; Limbourg area, Souvrain-Pre Formation Vesdre Massif; and middle Famennian questionable Dinant area few localities in the Ourthe Evieux Formation; valley; eastern Dinant middle Famennian Synclinorium
C
Strunian bioconstructions
Etroeungt Formation; Strunian
D
Waulsodian mounds
Bayard und Leffe Formation; Ivodan
E
Microbe-fenestellid bryozoan reefs
Lives Formation, Sequence 0; Livian
F
Stromatolite reefs
Grands Malades Formation, Maizret Mb; Livian
G
Microbe-biocementLivian stone reefs
H
Stromatolitemicroconchid reefs
I
Microbial reefs (Campine type)
K
Coral biostromes
185
Owing to its geotectonic setting, Carboniferous reefs younger than Vis6an are unknown in Belgium, although marine facies existed in the Namurian and Westphalian, but the input of siliciclastic material prevented any reef growth. Conclusions 9 The Famennian and Carboniferous succession of Belgium contains various levels of reef development (Fig. 17), which have been termed reef intervals in this study. The individual reef intervals are often bound to a distinctive facies or position on the shelf. The lengths and durations of the 10 reef intervals varied considerably. 9 The abundance of reefs in the Famennian and Carboniferous successions of Belgium is higher than previously observed. In particular, the record of numerous reefs in restricted marine facies, often small patch reefs dominated by microbial frameworks, is a significant addition to the global picture. 9 Microbial communities contributed to the formation of most reefs. Their functional importance is high, as framework formation was often initiated and controlled by microbial communities. Skeletal frameworks
Dimensions
Framework
Biodiversity
up to several 100 metres in widths and some 10"s m in height
microbial communities, early variable cements; subordinate: sponges, bryozoans, crinoids
Small lenses, 35 cm in height; > 3 m in width
microconchids, calcimicrobes
low
stromatoporid sponges
high
microbial communities, fenestellid bryozoans, sponges
low at the base, high on top
isolated occurrences; several metres thick, some Vesdre Massif and eastem ten's metres wide Dinant Synclinorium up to several hundred metres in width and height: common in the southern aggregation to bank Dinant Synclinorium complexes
microbial communities, height: <3 m; width : < 10 m; fenestellid bryozoans, a single reef is 10 times Namur Syncline; Dinant sponges, subordinate: Synclinorium questionable larger brachiopods common in the eastern
one locality, Namur Syncline; up to 10 reefs
height: < 2 m; width: 5 - 20 m
several localities in Dinant unknown Synclinorium and Namur Syncline
medium
microbial communities
low
microbial communities, brachiopods, bryozoans, early cements
variable
two localities; more than 10 Grands Malades microbial communities; less low Formation, Bay-Bonnet; reefs (?), Namur Syncline height: <4 m; width : <10 m; abundant microconchids and Vesdre Massif Livian microbial communities, 6 localities including 2 height: up to 200 m; width: fenestellid bryozoans, Vise Formation, high boreholes; Campine Basin sponges, subordinate: several hundred metres Wamantian and Vis6 area brachiopods, corals, cnnoids Lives and Anh~e Generally absent, only Common, Namur Syncline Up to several km in width; variable formations, Livian, locally coral boundstones and Dinant Synclinorium height < 1,2 m Wamantian
Fig. 17. The Belgian Famennian-Vis6an reef intervals. Short compilation indicating reef type, age, occurrence and abundance, dimensions, framework and biodiversity.
186
M. ARETZ & E. CHEVALIER
are rare, but skeletal bioconstructors contributed to a variety of reefs, resulting in mixed skeletal and microbial frameworks. Possibly the formation of reefs in restricted environments favoured the abundance of microbial communities. Reef formation was controlled on three hierarchical levels: palaeobiology; local environment; and regional and global environment. Palaeobiological limitation is commonly difficult to decipher, but bioconstructional activity is commonly controlled by these limitations. Relative water depth was probably the most crucial factor for reef formation. This factor is triggered by a multitude of factors of higher hierarchy, thus pointing to the complex and manifold influence on reef formation. However, at least from the middle Vis6an onwards, rapid sea-level oscillations as a consequence of the onset of the Gondwana ice sheet drastically influenced sedimentation patterns and thus reef formation. Therefore, climate is, at least partly, a very important factor. Reefs of large dimensions occur only in two reef intervals, Waulsortian mounds in the upper Tournaisian and microbial reefs of the Campine type in the upper Vis6an. The formation of these reefs required certain shelf configurations, and they formed at some distance from elevated areas. This situation was only developed in these specific time slices. Small, often isolated reefs and patch reefs developed in more nearshore positions. The comparison of the reef development in Belgium with the global picture reveals that the Belgian Famennian reefs show only a part of the global reef diversity. During the early Tournaisian, Belgium was cut off from the global development, but with the formation of Waulsortian mounds regained its attachment to global reef diversity. The general rareness of early Vis6an reefs is well documented in Belgium, although the first record of reef facies of this age is reported here. The abundance of middle Vis6an reefs predates the global development, but as many Belgian reefs of that age formed in restricted environments the picture may be biased. The upper Vis6an microbial reefs of the Campine type resemble the Cracoean reefs of northern England, and possibly formed in comparable shelf positions. However, the marked increase in abundance and diversity of upper Vis6an and Serpukhovian reefs is not recorded in Belgium, because the onset of the Variscian orogeny suppressed further reef formation.
E. Poty, J.-M. Marion, B. Delcambre, F. Boulvain and L. Barchy are thanked for indication, guidance to outcrops and discussion about reefal facies in southern Belgium. Carmeuse S.A. is thanked for the permission to work on reef facies on their properties. G. Sevastopulo and E. Poty are acknowledged for their constructive reviews. R. B~iumler and F. Noebert prepared the numerous thin sections for this study. References ADAMS, A. 1984. Development of algal-foraminiferalcoral reefs in the Lower Carboniferous of Furness, northwest England. Lethaia, 17, 233-249. ARETZ, M. 2001. The upper Vis6an coral-horizons of Royseux - The development of an unusual facies in Belgian Early Carboniferous. Tohoku University Museum, Bulletin, 1, 86-95. ARETZ, M. 2002. Habitatanalyse und Riffbildungspotential kolonialer rugoser Korallen aus dem Unterkarbon (Mississippium) yon Westeuropa. K6lner Forum fiir Geologie und Paliiontologie, 9, 1-155. ARETZ, M. & HERBIG,H.-G. 2003a. Coral-rich bioconstructions in the Vis6an (Late Mississippian) of southern Wales (Gower Peninsula, UK). Facies, 49, 223-249. ARETZ, M. & HERSIG, H.-G. 2003b. Contribution of rugose corals to Late Vis6an and Serpukhovian bioconstructions in the Montagne Noire (Southern France). SEP M Society of Economic Paleontologists and Mineralogists, Special Publications, 78,119-132. ARETZ, M. & WEB8, G. E. 2007. Western European and eastern Australian Mississippian shallow-water reefs: A comparison. In: WONG,TH.E (ed.) Proceedings of the XVth International Congress on Carboniferous and Permian Stratigraphy. Utrecht, 10-16 August 2003. Royal Dutch Academy of Arts and Sciences, Amsterdam. BELLIERE, J. 1953. Note sur le calcaire famennien de Baelen et ses stromatactis. Annales de la Sociktb GOologique de Belgique, 76, 115-128. BOULVAIN,F. 2007. Frasnian carbonate mounds from Belgium: Sedimentology and palaeoceanography. In: ALVARD, J. J., ARETZ, M., BOUVAIN, F., MUNNECKE, A., VACHARD,D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations." Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 125-141. BRUCKSCHEN, P., OESMANN, S. & VEIZER, J. 1999. Isotope stratigraphy of the European Carboniferous: proxy signals for ocean chemistry, climate and tectonics. Chemical Geology, 161,127-163. BULTYNCK, P. & DEJONGHE, L. 2002. Devonian lithostratigraphicunits (Belgium). In: BULTYNCK,P. & DEJONGHE, L. (eds) Guide to a Revised Stratigraphic Scale for Belgium. Geologica Belgica, 4, 39-68 BURCHETTE,T. P. & RIDING, R. 1977. Attached vermiform gastropods in Carboniferous marginal marine stromatolites and biostromes. Lethaia, 10, 17-28. CHEVALIER, E. & ARETZ, M. 2005. A microbebryozoan reef from the middle Vis6an of the Namur Syncline (Engihoul Quarry). Geologica Belgica, 8, 109-119.
BELGIAN FAMENNIAN AND DINANTIAN REEFS CONIL, R. 1964. Localit6s et coupes types pour l'6tude du Tournaisien inf6rieur (r6vision des limites sous l'aspect micropal6ontologique). Mdmoires de l'Acaddmie Royal de Belgique, Classe des Sciences, 4 skrie, 15, 1-87. CONIL, R., DICKENSTEIN, J. & DRICOT, E. 1961. Le biostrome strunien du massif de la Vesdre. Bulletin de la Sociktk beige de Gkologie, 70, 28-34. CONIL, R., GROESSENS, E., LALOUX, M., POTY, E. & TOURNEUR, F. 1991. Carboniferous guide foraminifera, corals and conodonts in the Franco-Belgian and Campine basins: their potential for widespread correlation. Courier Forschungsinstitut Senckenberg, 130, 15-30. COPPER, P. 1988. Ecological succession in Phanerozoic reef ecosystems: is it real? Palaios, 3, 136-151. COPPER, P. 2002. Reef development at the Frasnian/ Famennian mass extinction boundary. Palaeogeography, Palaeoclimatology, Palaeoecology, 181, 27-65. DEHANTSCHUTTER, J. A. E. & LEES, A. 1996. Waulsortian buildups of Waulsort, Belgium. Geological Journal, 31, 123-142. DELCAMBRE,B. & PINGOT, J. L. 1993. Dinant-Hasti~re. Carte gOologique de la Wallonie 1/25 000 (+ notice explicative). Minist6re de la R6gion Wallonne. DELCAMBRE, B. & PINGOT, J. L. 2000. Fontainel'Evdque Charleroi. Carte gkologique de la Wallonie 1/25 000 (+ notice explicative). Minist6re de la R6gion Wallonne. DEMANET, F. 1923. Le Waulsortien de Sosoye et ses rapports fauniques avec le Waulsortien d'fige Tournaisien sup6rieur. Mkmoires de l'Institut Gkologique de l' Universitk de Louvain, 2, 37-283. DEVUYST, F. X. & LEES, A. 2001. The initiation of Waulsortian buildups in Western Ireland. Sedimentology, 48, 1121-1148. DREESEN, R. & FLAJS, G. 1984. Le marbre rouge de Baelen, une barri6re carbonat6e importante Crinoides, Spongiaires et Algues dans le D6vonien sup6rieur du Massif de la Vesdre (Belgique orientale). Comptes Rendus des Sdances de l'Acaddmie des Sciences, Skrie 2, 299, 639-644. DREESEN, R. & Jux, U. 1995. Microconchid buildups from Late Famennian peritidal-lagoonal settings (Evieux Formation, Ourthe Valley, Belgium). Neues Jahrbuch fiir Geologie und Paliiontologie, Abhandlungen, 198, 107-121. DREESEN, R., BLESS, M. J., CONIL, R., FLAJS, G. & LASCHET, C. 1985. Depositional environment, paleoecology and diagenetic history of the 'marbre rouge fi crinoides de Baelen' (late Upper Devonian, Verviers Synclinorium, eastern Belgium). Annales de la Sociktk Gdologique de Belgique, 108, 311-359. DUPON7, E. 1883. Sur les origines du Calcaire Carbonif6re de la Belgique. Bulletin de l'Acaddmie Royale des Science Belges, 3e skrie, 5, 211-229. DUSAR, M. & DREESEN, R. 1984. Stratigraphy of the Upper Frasnian and Famennian Deposits in the Region of Hamoir-sur- Ourthe (Dinant Synclinorium, Belgium). Service G6ologique de Belgique, Professional Paper, 209, 1-52. HANCE, L. & HENNEBERT, M. 1980. On some Lower and Middle Vis6an carbonate deposits of
187
the Namur Basin, Belgium. Mededelingen Rijks Geologische Dienst, 32, 66-68. HANCE, L., POTY, E. & DEVUYST, F. X. 2001. Stratigraphie s6quentielle du Dinantien type (Belgique) et corr61ation avec le Nord de la France (Boulonnais, Avesnois). Bulletin de la Socidtk Gkologique de France, 172, 411-426. INSALACO, E. 1998. The descriptive nomenclature and classification of growth fabrics in fossil scleractinian reefs. Sedimentary Geology, 118, 159-186. KELLY, J. G. & SOMERVILLE, I. D. 1992. Arundian (Dinantian) carbonate mudbanks in north-west Ireland. Geological Journal, 27, 221-241. KERSHAW, N. 1994. Classification and geological significance of biostromes. Facies, 31, 81-92. LALOUX, M., DEJONGHE, L., GEUKENS, F., GHYSEL,P. & HANCE, L. 1996. Carte gdologique de Wallonie 1/25 000. FlOron-Verviers 42/7~ (+ notice explicative). Minist~re de la R6gion Wallonne. LAUWERS, A. S. 1992. Growth and diagenesis of cryptalgal-bryozoan buildups within a mid-Vis6an (Dinantian) cyclic sequence, Belgium. Annales de la Sociktk gOologique de Belgique, 115, 187-213. LEES, A. 1988. Waulsortian 'reefs': the history of the concept. Mkmoires de l'Institut Gkologique de l'Universitd de Louvain, 34, 43-55. LEES, A. 1997. Biostratigraphy, sedimentology and palaeobathymetry of Waulsortian buildups and peri-Waulsortian rocks during the late Tournaisian regression, Dinant area, Belgium. Geological Journal, 32, 1-36. LEES, A. & MILLER, J. 1995. Waulsortian Banks. In: MONTY, C. L. V., BOSENCE, D. W. J., BRIDGES, P. H. & PRATT, B. R. (eds.) Carbonate Mud-mounds, Their Origin and Evolution. International Association of Sedimentologists, Special Publications, 23, 191-271. LEES, A., HALLET,V. & HIBO, D. 1985. Facies variation in Waulsortian buildups. Part 1. A model from Belgium. Geological Journal, 20, 133-158. LEES, A., NOEL, B. & BOUW, P. 1977. The Waulsortian (Lower Carboniferous) 'reefs' of Belgium: a progress report. Mdmoires de l'Institut Gdologique de l'Universitk de Louvain, 29, 289-315. LEGRAND, R., MAMET, B. & MORTELMANS, G. 1967. Sur la stratigraphie du Tournaisien de Tournai et de Leuze. Probl6mes de l'6tage Tournaisien dans sa localitd-type. Bulletin de la Sociktk belge de Gdologie, 74, 140-188. MORTELMANS, G. 1969. L'&age Tournaisien dans sa localit6 type. Compte Rendus 6e Congrks International de Stratigraphie et Gkologie du Carbonifdre, Sheffield 1967, 1, 19- 44. MORTELMANS, G. 1976. Evolution pal6o6cologique et s6dimentologique du calcaire de Tournai. Quelques lignes directrices. Bulletin de la Sociktk belge de Gkologie, 82, 141-180. MISTIAEN, B. 1984. Disparition des stromatopores paldozoiques ou survie du groupe: hypoth~se et discussion. Bulletin de la Sociktd Gdologique de France, skrie 7, 26, 1245-1250. MUCHEZ, P. & PEETERS, C. 1986. The occurrence of a cryptalgal reef structure in the upper Vis6an of the Vis6 area. Annales de la Socidtk Gkologique de Belgique, 109, 573-577.
188
M. ARETZ & E. CHEVALIER
MUCHEZ, P., CONIL, R., VIAENE,W., BOUCKAERT,J. & POTY, E. 1987. Sedimentology and biostratigraphy of the Vis6an carbonates of the Heibaart (DzH1) borehole (Northern Belgium). Annales de la Sociktb Gkologique de Belgique, 110, 199-208. MUCHEZ, P., VIAENE, W., BOUCKAERT,J., CONIL, R., DUSAR, M., POTY, E., SOILLE, P. & VANDENBERGHE, N. 1990. The occurrence of a microbial buildup at Poederlee (Campine Basin, Belgium): biostratigraphy, sedimentology, early diagenesis and significance for early Warnantian paleogeography. Annales de la Sociktk Gkologique de Belgique, 113, 329-339. MUNDY, D. J. C. 1994. Microbialite-spongebryozoan-coral framestones in Lower Carboniferous (Late Vis6an) buildups of Northern England (UK). Canadian Society of Petroleum Geologists, Memoir, 17, 713-729. PAPROTH, E., CONIL, R. ET AL. 1983. Bio- and lithostratigraphic subdivisions of the Dinantian in Belgium, a review. Annales de la SociOtk Gkologique de Belgique, 106,185-239. PICKARD, N. A. H. 1996. Evidence for microbial influence on the development of Lower Carboniferous buildups. In: STROGEN, P., SOMERVILLE, I. D. & JONES, G. L1. (eds) Recent Advances in Lower Carboniferous Geology. Geological Society, London, Special Publications, 107, 65-82. PIRLET, H. 1967a. Nouvelle interpr6tation des carri6res de Richelle; Le Vis~en de Vis6. Annales de la Socikt~ Gkologique de Belgique, 90, 298-329. PIRLET, H. 1967b. Sur l'age et la signification tectonique de la br6che de Bouffioulx. Annales de la Socikt~ Gkologique de Belgique, 92, 123-130. POTY, E. 1981. Recherche sur les T6tracoralliaires et les H6terocoralliaires du Vis6en de la Belgique. Medelingen Rijks Geologische Dienst, 36, 1-161. POTY, E. 1984. An evolutionary pattern for the Western European Lithostrotionidae. Palaeontographica Americana, 54, 465-467. POTY, E. 1991. Tectonique de blocs dans le prolongement oriental du Massif de Brabant. Annales de la Sociktb Gkologique de Belgique, 114, 265-275. POTY, E. & CHEVAHER, E. 2007. Development of a rugose coral biostrome at the beginning of the late Frasnian in Belgium. In: ,h,LVARO,J. J., ARETZ, M., BOULVAIN, F., MUNNECKE, A., VACHARD, D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations." Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 143-161. POTY, E., HANCE, L., LEES, A. & HENNEBERT,M. 2002. Dinantian lithostratigraphic units (Belgium). In: BULTYNCK, P. & DEJONGHE, L. (eds) Guide to a Revised Stratigraphic Scale for Belgium. Geologica Belgica, 4, 69-93. POTY, E., DEVUYST, F.-X. & HANCE, L. 2006. Upper Devonian and Mississipian foraminiferal and rugose coral zonations of Belgium and northern France: a tool for Eurasian correlations. Geological Magazine, 143, 829-857. SCRUTTON, C. T. 1998. The Palaeozoic corals, II: structure, variation and palaeoecology. Proceedings of the Yorkshire Geological Society, 52, 1-57.
SHEN, J. W., Yu, C. M. & BAO, H. 1997. A LateDevonian (Famennian) Renalcis-Epiphyton reef at Zhaijiang, Guilin, South China. Facies, 37, 195-209. STAINIER, X. 1893. Marbre rouge ~t crinoides darts le Famennien de la Lesse. Bulletin de la Sociktk belge de Gkologie, 7, 177. STANLEY, S. M. & HARDIE, L. A. 1998. Secular oscillations in the carbonate mineralogy of reef-building and sediment-producing organisms driven by tectonically forced shifts in seawater chemistry. Paleogeography, Paleoclimatology, Palaeoecology, 144, 3-19. STEARN, C. W. 2001. Biostratigraphy of Devonian stromatoporid faunas of arctic and western Canada. Journal of Paleontology, 75, 9-23. STEARN, C. W., WEBBY, B. D., NESTOR, H. & STOCK, C. W. 1999. Revised classification and terminology of Palaeozoic stromatoporoids. Acta Palaeontologica Polonica, 44, 1-70. STREEL, M. 1985. Biostratigraphie par spores du D6vonien ardenno-rhenan. Annales de la Sociktg Gkologique du Nord, 105, 85-95 THOREZ, J. & DREESEN, R. 1986. A model of a regressive depositional system around the Old Red Continent as exemplified by a field trip in the upper Famennian 'Psammites du Condroz' in Belgium. Annales de la Soci~tk Gkologique de Belgique, 109, 285-323 VAN STEENWINKEL, M. 1990. Sequence stratigraphy from 'spot' outcrops; example from a carbonatedominated setting; Devonian-Carboniferous transition, Dinant Synclinorium. Sedimentary Geology, 69, 259-280. WEBB, G. E. 1994. Non-Waulsortian Mississippian bioherms: a comparative analysis. Canadian Society of Petroleum Geologists, Memoir,17, 701-712. WEBB, G. E. 1996. Was Phanerozoic reef history controlled by the distribution of non-enzymatically secreted reef carbonates (microbial carbonate and biologically induced cement?). Sedimentology, 43, 947-971. WEBB, G. E. 1998. Earliest known Carboniferous shallow-water reefs, Gudman Formation (Tnlb), Queensland, Australia: implications for Late Devonian reef collapse and recovery. Geology, 26, 951-954. WEBB, G. E. 2002. Latest Devonian and Early Carboniferous reefs: depressed reef building after the Middle Paleozoic collapse. In: KIESSLING, W., FLU'GEL, E. & GOLONKA,J. (eds) Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 239-269. WEBER, H. M. 2000. Die karbonatisehen Flachwasserschelfe im europfiischen Oberfamennium ( Strunium ) : Fazies, Mikrobiota und Stromatoporen-Faunen. PhD thesis, Universitfit zu K61n. WEEDON, M. J. 1990. Shell structure and affinity of vermiform 'gastropods'. Lethaia, 23, 297-309. WEST, R. R. 1988. Temporal changes in Carboniferous reef mound communities. Palaios, 3, 152-169. WRIGHT, V. P. • WRIGHT, E. V. G. 1981. The paleontology of some algal-gastropod bioherms in the Lower Carboniferous of South Wales. Neues Jahrbuch fiir Geologie und Palfiontologie, Monatshefte, 1981, 546-558.
The late Atokan (Moscovian, Pennsylvanian) chaetetid accumulations of Sierra Agua Verde, Sonora (NW Mexico): composition, facies and palaeoenvironmental signals EMILIO ALMAZAN-VAZQUEZ
1, B L A N C A E. B U I T R O N - S A N C H E Z
D A N I E L V A C H A R D 3, C Y N T H I A M E N D O Z A - M A D E R A
2,
1&
CATALINA GOMEZ-ESPINOSA 1
1Universidad de Sonora, Departamento de Geologia, Boulevard Luis Encinas y Rosales, 83000 Hermosillo, Sonora, Mexico (e-mail:
[email protected]) 2Universidad Nacional Aut6noma de Mdxico, Instituto de Geologia, Departamento de Geologia, Ciudad Universitaria, Delegaci6n Coyoacdn, 04510 M~xico, D.F, Mexico 3Universitd de Lille 1, Sciences de la Terre, U M R 8014 du CNRS, Laboratoire LP3, Bfitiment SN5, 59655 Villeneuve d'Ascq Cedex, France (e-mail: daniel,
[email protected])
Abstract: The La Joya Formation of the Sierra Agua Verde, Sonora (NW Mexico) is late Atokan in age, equivalent to the early late Moscovian (Podolskian) and fusulinid biozone A3. In this alternance of cherty limestones and thin shaly beds, fusulinellid or anchicodiacean ('phylloid algae') wackestones-packstones and crinoidal rudstones-grainstones are the predominant microfacies, but chaetetid boundstones are conspicuous. These chaetetid occurrences of the Sierra Agua Verde are compared with the accumulations of Arizona, Texas, Kansas and Nevada (USA), and the Cantabric Cordillera (Spain). In Sonora, the environments with chaetetids were quiet, and located below wave base. Shallower facies with staffellids and Komia generally top the chaetetids. Because of the associated micritic deposits, the chaetetids have inhabited probably a soft or firm substrate. As a result of the disphotic-aphotic reconstructed environments, the possible symbionts of the chaetetids are more probably heterotrophic bacteria than autotrophic algae. The most comparable ecological conditions exist in the Atokan Marble Falls Formation of central Texas (USA). Chaetetids are not mentioned in the southern suspect terranes of Mexico, but were possibly present because these regions were located along the probable migration way to Peru, the southernmost area where Pennsylvanian chaetetids are known. As noted by many authors, chaetetids were one of the rare reef-mound builders during the Middle Pennsylvanian. They were responsible for unique framestones during this period, whereas other bioconstructions consist of algal bindstones or baffiestones with Donezella, beresellaceans, archaeolithophyllaceans and anchicodiaceans, i.e. 'phylloid algae' (e.g. Roux 1985; Mamet et al. 1987; Minwegen 2001; Della Porta et al. 2003; Mamet & Villa 2004). Occurrences of chaetetids are generally well described in the literature, and the aim of this paper is preferentially to describe a newly investigated outcrop in Sonora and some of its biosedimentological characteristics (see also Buitr6n-S/mchez et al. 2007).
Geological setting In the state of Sonora, in the N W comer of Mexico, the deposits of the Pennsylvanian
subsystem are generally poorly described. They are listed in some general handbooks and stratigraphic compilations (L6pez-Ramos 1985; Peiffer-Rangin 1987; Sfinchez-Zavala et al. 1999; Vachard et al. 2000b). First, the Horquilla Formation, defined in the southern USA, has been studied the northern part of Sonora in the Cerros de Tule, and in the Sierra de Palomas in the adjacent state of Chihuahua (Wilson et al. 1969; Tdllez-Gir6n 1979; Gonz/tlez-Le6n 1986). In central Sonora, the Pennsylvanian is well exposed in the Sierra Agua Verde, an hill located 120 km N E of Hermosillo City, the capital of Sonora (Fig. 1). This hill is formed by a series (2900 m thick) of platform rocks from Cambrian to Permian in age. In its southern flank (29~ 29~ and 109~176 the Pennsylvanian is referred as to the La Joya Formation (Ochoa-Granillo & Sosa-Le6n 1993). It is composed of a series (e. 100 m thick) of micriticbioclastic limestone, extensively silicified with
From: •LVARO, J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007.
Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 189-200.0305-87191071515.009 The Geological Society of London.
190
E. ALMAZAN-VAZQUEZ E T AL.
Fig. 1. Maps indicating the locations of (a) Sonora State in Mexico and (b) Sierra Agua Verde in Sonora.
nodular cherts and interstratified with siltstone (Fig. 2). The fossiliferous facies are generally decimetre-thick beds of cherty pale grey limestone. They contain mainly calcareous algae, fusulinids, chaetetids, gastropods, brachiopods, crinoids and conodonts (Stewart et al. 1997, 1999; Buitr6n-Sfinchez et al. 2004; MendozaMadera et al. 2004).
Biostratigraphy In the United States, the subsystem Pennsylvanian is divided into five stages and 16 zones of
fusulinids (Wilde 1990): Morrowan (biozones M 1-2); Atokan (A 1-4), Desmoinesian (DS 1-5); Missourian (MC 1-4) and Virgilian (VC 1-3). The equivalent international subdivisions (Heckel 2004) are the early-middle Bashkirian for the Morrowan; the late Bashkirian-early Moscovian for the Atokan; the late Moscovian for the Desmoinesian; the Kasimovian for the Missourian; and the Gzhelian for the Virgilian. In the Sierra Agua Verde, several beds ('Fusulinids' levels of Fig. 2) permit the datation of the series thanks to their relatively diversified assemblages of carbonate microfossils. The
ATOKAN CHAETETIDS OF SONORA (MEXICO) It
Me
Brachiopods Fusulinids
Crinoids Solitary corals Brachiopods
80-
~ _ ' - ~
70-
60-
Syringopora
~
" Fusulinids
~
Crinoids Chaetetid reef mound
~
Solitary corals Fusulinids
5o--
Crinoids
Syringopora Brachiopods Solitary corals
"7 20--
I
,, I ~
I
I Chaetetids
J
- " - T - - ~ - - Syringopora . n ~ m ~ n u i - - , - -
~,,
lo--
.,
' I
~ , ~ ,Brachiopods
'
'
-7
-i I
n -
I t o m - "
j _ _ l _ ]:_ _ l _ _ ~ O-- wJ ~! ~' Covered base ~ Limestone with ~
J Crinoids j Fusulinids chert nodules
Siltstone
Fig. 2. Lithostratigraphic column of the Pennsylvanian La Joya Formation in Sierra Agua Verde, Sonora, Mexico.
191
problematic calcareous algae are represented by sporadic Eugonophyllum sp., Zidella (?) sp., Kamaena (?) sp. and Komia eganensis. Taxa of associated smaller foraminifers are Pachysphaerina pachysphaerica, Eotuberitina reitlingerae, Insolentitheca horrida, Endothyra ex gr. bowmani, Globivalvulina bulloides, Climacammina ex gr. moelleri, Deckerella sp., Calcivertella sp, Baryshnikovia sp. and Syzrania sp. The following fusulinids have been identified: Mediocris breviscula, Eostaffella grozdilovae, Millerella sp., Pseudostaffella sp., Staffella powwowensis, Eoschubertella texana, Fusulinella thompsoni, F. llanoensis, F. aff. llanoensis and Nipperella (?) sp. indet. This fusulinid assemblage indicates a late Atokan age (biozone A3: Wilde 1990). Similar species of fusulinids are known in the Upper Marble Falls Limestone of central Texas (Groves 1991), which also contains 'impressive chaetetid colonies' (West 1992, p. 165). The first results of our analyses of biostratigraphy (C. G6mez-Espinosa, B. Buitr6n-S/mchez and D. Vachard) and sequence stratigraphy (E. Almaz/m-V/tzquez and D. Vachard) suggest that the 100 m of the La Joya Formation are entirely late Atokan in age, and were accumulated in the form of a transgressive systems tract (TST). This TST is probably coeval with the upper part of the TST of the Podolskian stage identified in the Ukrainian Donets Basin (Izart et al. 1996, text-fig. 9; sequences SM9 and/or SM10). The dominant lithologies of the Atokan TST of the Sierra Agua Verde are: (a) fusulinellid, especially Fusulinella llanoensis, wackestone, packstone and floatstone; (b) crinoidal rudstone; (c) Chaetetes bafflestone; and (d) Eugonophyllum bindstone, alternating with shales. The chaetetid organisms
Chaetetids are coralline (i.e. hypercalcified) demosponges (see discussion in Connolly et al. 1989, table 1), with unusual pseudomorphs of spicules (e.g. Cremer 1995). They are possibly capable of heterotrophic symbiosis (West 1994; Stanton et al. 1997), not necessarily attached to a hard substrate (Stanton et al. 1997), and essentially present during the Early and Middle Pennsylvanian (Morrowan, Atokan and Desmoinesian stages), but not exclusively a guide for this period (West 1992; Fliigel 2004). The palaeocological and taphonomical elements of description of the chaetetids are clearly summarized in Connolly et al. (1989), Stanton et al. (1994) and Minwegen (2001); and the vocabulary for the description of the skeletons can be borrowed from Cremer (1995) and Stanton et al. (1997). According to West (1992, 1994), the seven
192
E. ALMAZ/~N-V/i~ZQUEZ E T AL.
Fig. 3. Field photograph in Sierra Agua Verde, Mexico, showing some silicified chaetetids in life position, numerous cherts and the thickness of the associated limestone beds. The scale bar is 20 cm.
Fig. 4. Silicified Chaetetes showing the clinogonal growth of the tubes (Sierra Agua Verde, Sonora, Mexico). The scale bar is 30 mm.
Fig. 5. Polished longitudinal section of a chaetetid. The scale bar is 1 cm.
species of Carboniferous Chaetetes in North America could be synonyms. This preliminary report does not describe the detailed taxonomy of the chaetetids collected in the Sierra Agua Verde by our team. Chaetetids of the La Joya Formation are present in two stratigraphic levels, 19 and 60 m above the base of the section, respectively (Fig. 2). They are represented by many silicified massive structures with a hemisphericalcolumnar shape (Figs 3-5). The chaetetid skeletons are comprised of a series of thin and long tubes, quadrate-hexagonal in cross-section with an average maximal dimension of 0.5001.000 mm, with a clinogonal microstructure and increasing by longitudinal parting (Figs 3-7). The primary walls, not perforated, show at
regular intervals some quasi-tabulae, but no growth bands are distinct. Monaxon megascleres are entirely missing in our specimens. Often, the chaetetids of the Sierra Agua Verde are in life position and exhibit columnar structures up to 50 cm in height and approximately 35 cm in diameter (Fig. 5). Accumulations of chaetetids grew by vertical development of cloned individuals and trapping of micritic sediment, i.e. constituting baffiestones (Figs 3 & 4). Limestones either with chaetetids or devoid of them do not differ significantly, and the lithological and palaeotopographic differences between bafflestones and wackestone-packstone deposit areas are generally faint, as we do not observe preserved palaeoslopes such as those illustrated by Krainer et al. (2003) in the Anthracoporella
ATOKAN CHAETETIDS OF SONORA (MEXICO)
Fig. 6. Longitudinal section of a chaetetid showing the growth of the tubes by longitudinal parting. The scale bar is 1 mm.
mounds of the Carnic Alps (Austria). Hence, the chaetetids in the Sierra Agua Verde constitute cluster reefs in the sense of Riding (2002, p. 190) or reef mounds such as those of Arizona (Connolly 1990). The columnar forms are dominant among these reef mounds, as in Kansas outcrops described by Suchy & West (2001), or the Hueco Mountains (west Texas) and Whetstone Mountains of southern Arizona (Connolly et al. 1989; Connolly 1990). Among the environments inhabited by chaetetids summarized by Connolly et al. (1989, p. 147), the Sonora occurrence is similar to that described for the outcrops of the Mason County (central Texas) located in the Atokan part of the Marble Falls Formation. The common characteristics are the association with foraminiferal wackestones and packstones, the Chaetetes reef mounds situated below normal wave base, the low faunal diversity, an estimated relief of 1-2 m, and the evidences of episodic storms.
Environments and facies Diverse accumulations of fusulinids, phylloid, and other algae, sponges and corals are visible in
193
Fig. 7. Transverse section of a chaetetid showing the tubes and their quadrate to hexagonal cross section. The scale bar is 1 mm.
the field (Fig. 2) and in thin sections; for example, fusulinellids Fusulinella llanoensis, phylloid algae Eugonophyllum sp., problematic ungdarellacean algae Komia eganensis (sometimes difficult to distinguish from Pseudokomia), corals (Michelinia, Syringopora), chaetetids and crinoids (Buitr6nSfinchez et al. 2007). The micritic matrix is homogeneous, and no siliciclastic input is obvious in the bioclastic wackestone, packstone or floatstone microfacies. Some beds are diagenetically dolomitized and silicified with poorly preserved fusulinellids. Indicators of shallow-water environments (i.e. within the zone of normal water activity), such as oolites, microbialites, current figures, granular carbonate cements, palaeosols or other sequence boundaries, are totally lacking along the stratigraphic column, which remains from base to top of late Atokan age (see above) and exhibits consequently the characters of a TST deposited in a strongly subsident platform. The biota of the chaetetid beds is characterized by an epifaunal macrofauna (crinoids and fusulinellids that probably lived upon these crinoids, in view of the absence of algae and the consequent impossibility of an epiphytic way of
194
E. ALMAZAN-VAZQUEZ E T AL.
life) and infaunal small foraminifers, living in the superficial centimetres of the bottom carbonate mud. In the sediments associated with the Chaetetes the tests of fossils are well preserved, rarely encrusted by microbial biofilms and preferentially composed of an exclusive species of Fusulinella. A sorting by buoyancy is probable because of the absence of juvenile stages and uniform size of the adult Fusulinella tests entirely filled by microsparite, i.e. transported empty to the terminal deposit. They are believed to have been relatively deeply deposited, i.e. at the limit of the photic zone, because green algae are not present in the assemblages. The absence of staffelloid foraminifers and other indicators of shallow waters, and the abundance of crinoids, brachiopods and bryozoa, are other characteristics of relative depth. Nevertheless, in the succession, shallower environments can be identified when: (a) Fusulinella are associated with Staffella; and (b) phylloid algae form bindstones or platestones with abundant micrite. Chaetetids are generally preserved in life position, showing a clinogonal increase (Fig. 5), with the tubes directed upwards and outward. No specimens grew downwards, contrary to the specimens of the Hueco Mountains described by Stanton et al. (1997). Preservation in situ appears to be common in these quiet environments, as our taphonomic studies in progress indicate that the disarticulation of crinoids was the result of mass transport accelerating the muscular and articular post-mortem decay (Buitr6n-S~nchez et al. 2007).
Discussion As indicated above, in the Sierra Agua Verde, the chaetetid cluster reefs were constructed during the deposition of an Atokan transgressive systems tract (TST). Inversely, the Kansas chaetetid buildups were developed during a Desmoinesian HST (Suchy & West 2001) and exhibit associated syringoporid tabulate corals Multithecopora or algal crusts and storm events, all these phenomena are not recorded in Sierra Agua Verde and we propose to interpret the chaetetid biotopes to have been located below storm wave base and photic zone; i.e. to approximately - 1 0 0 m of depth (according to the synthetic data of Bosscher & Schlager 1992; Wright & Burchette 1996; Biju-Duval & Savoye 2001; Cojan & Renard 2003). This proposed palaeobathymetry is consistent with the lateral equivalence to Fusulinella wackestones having a matrix of homogeneous micrite, and with the absence of photophile algae in the
microfacies, absence of current or storm sedimentary figures, absence of oolites and absence of stromatolites, and because the Recent coralline sponges are known from 10 to 185 m (Fallon et al. 2005). The life habit of Chaetetes here was relatively restricted, as it can occupy areas 'from intertidal to below wave base' (Connolly et al. 1989), and in some cases quoted by these authors the phylloid algal mounds can be in turn capped by Chaetetes constructions (Connolly et al. 1989, text-fig. 5). Algae appear with Eugonophyllum and Komia in the overlying beds of La Joya Formation, thus suggesting deposition in water depths of more or less 25 m (see discussion in Connolly et al. 1989, p. 153). This important variation of the deposit depths, from -100 to - 2 5 m (Fig. 8a), indicates a strong local subsidence, responsible for the creation of a regional palaeoslope and rapid variations of the accommodation on the carbonate platform (Fig. 8b). A bathymetry of 'much less than 100 feet (-30 m)' is indicated for the Chaetetes of Kansas by Suchy & West (2001, p. 432), and shallower than that of the underlying phylloid algal beds. Consequently, the bathymetrical relations between Chaetetes bioconstructions and phylloid algal bioconstructions vary according to the series and the outcrops. In the Sierra Agua Verde, we have supposed that the Chaetetes grew deeper (c. -100 m) than the phylloid algae Eugonophyllum (c. -25 m) (Fig. 8a). In the literature exists a discussion about the types of substrates of chaetetids (Connolly et al. 1989; Stanton et al. 1997). The associated carbonate microfacies in the Sierra Agua Verde are devoid of pressure-solution features, deformation or beacking of fusulinellids and sutured contacts between grains. That seems to indicate an early cementation of the overlying beds, and consequently a firm-hard substrate for the local chaetetids. Moreover, Chaetetes can support agitated waters (Connolly et al. 1989). Generally, a strong hydrodynamism permits a rapid cementation of the granular sediments. Consequently, in agitated environments, the chaetetids have still more probably a hard substrate. Nevertheless, Suchy (pers. comm. 2006) suggests that the chaetetids attached themselves to softgrounds held together by covalent forces within the muds or by algal or bacterial mats that left no visible trace. Bromley & Heinberg (2006) admit also that in many cases 'there is an organic membrane or layer of glue between the skeleton and the substrate'. However, owing to the large dimensions of some individuals, their base must be strengthened by rapid sedimentary inputs to stop the chaetetids from taking a tumble and permit them to hold in vertical position (Fig. 8a).
ATOKAN CHAETETIDS OF SONORA (MEXICO)
195
Om
0 0o~~0o0
8
25 m
d
o
.
d
do
V~~ lOOm ~m 8
|
| sea
~ ~ ~ _ . . ~ ~
b
SWArizona
l
USA/Mexlco
W Texas
border
,
1,
level
_ l
'~ Sonora
100m Der~h
Fig. 8. Reconstructions of environments of deposits. (a) Idealized parasequence showing the succession of the deposits, biotopes and corresponding bathymetries, l, dasyclads; 2, fusulinids; 3, phylloid algae; 4, Chaetetes; 5, crinoids; 6, brachiopods; 7, solitary corals; 8, limestone; 9, shales. (b) Palaeogeographical reconstruction from Arizona to Sonora during the Atokan.
196
E. ALMAZAN-VAZQUEZ E T AL.
Suchy & West (2001) used growth rates of extant coralline sponges (more or less 0.151.2 mm year -1, according to Fallon et al. 2005) to make inferences about the longevity of the chaetetids preserved in the rocks and thus the duration of deposition of the beds. We have obtained more or less similar data (life time of the biggest chaetetid up to 3 ka (3000 years) and time of deposit of a chaetetid bed (10-20 ka)). Consequently, using those as a standard, we propose to discuss and compare these data with the growth rates of fusulinids (probable average longevity of 3 years; compare with Vachard & Bouyx 2002) and crinoids in order to estimate, with precise biochronometres, the sedimentation rate all around and within the bioconstructions of chaetetids (G6mez-Espinosa's PhD). A life time of the chaetetids of up to 3000 years, compared with probable average longevity of 3 years for the fusellinellids and an intermediary time for the crinoids, allows a precise correlation of the growth of the chaetetids and their successive trapping of fusulinellid wackestone-packstone and/or crinoid rudstone to be made (see also Buitr6n-S~mchez et al. 2007, fig. 4), and eventually allows a calculation of the frequency of storms during the Atokan in Sonora. Symbiosis with heterotrophic bacteria and not autotrophic algae, as in many modern sponges, may be implied by the probable growth of our chaetetids in the disphotic or aphotic zone but we lack precise arguments. The Chaetetes might be also proposed, as well as the Recent coralline sponges (Fallon et al. 2005), as a recorder of water temperature in order to verify the eventual episodes of cold waters that can occur in some tropical environments (Weidlich 2007). Nevertheless, owing to the large spectrum of recorded habitats, it is difficult to accurately express nowadays if chatetids were warm or cold, stenotherms or stenotopes. A control by studies of sequence stratigraphy might allow to the true biotopes of chaetetids to be accurately expressed. Finally, the most comparable ecological conditions exist in the Atokan Marble Falls Formation of central Texas. According to the characters summarized by Connolly et al. (1989, text-fig. 3), the chaetetid bioconstructions in SW Arizona and east Texas seem to correspond to shallower environments; consequently, during the Atokan, a slight slope towards the south existed near the current boundary between Mexico (Sonora) and the United States (Arizona) (Fig. 8b), and another one, or the prolongation of the latter one, in central Texas.
Associated fusulinellids in the Sierra Agua Verde are classical species of the southern USA, e.g. Fusulinella llanoensis (Skinner & Wilde 1954; Groves 1991). Smaller foraminifers are similar to the assemblages of Marble Falls (Groves 1992). Among the algae, the incertae sedis K o m i a is also widespread in North America (Mamet et al. 1987). In the two geodynamic domains of Sonora, Craton and Caborca Terrane (Gonz~tlez-Le6n 1989; Sedlock et al. 1993), the microfossil assemblages are similar. Apparently, these two domains were not isolated during the Carboniferous-Permian. Continuity with the central Mexican Mixteco Terrane is evident at least as early as the Missourian (Late Pennsylvanian), based on the identity of species of Triticites (Vachard et al. 2000b). The continuity of a carbonate platform as early as the Middle Pennsylvanian, between Sonora and South America, via Oaxaquia and Maya Blocks, is probable (Fig. 9), as some species of Fusulinella from Sonora are very similar to some Fusulinella from South America (for example F. peruana (Meyer) described from central Peru by Dunbar & Newell 1946). The platform reconstructed here (Fig. 9) seems to be the unique possible means of migration for these foraminifers, strictly benthic, living in rather shallow waters and devoid of pelagic stages (see discussion in Vachard & Bouyx 2002). However, the Middle Pennsylvanian Chaetetes are also known into South America (Suchy & West 2001). These data validate the previous palaeobiogeographical reconstructions of our team (Vachard et al. 2000 a-c). The North American Craton was separated from the Gondwanan continent of southern America by a remnant of the Rheic Ocean, where some individualized tectonostratigraphic terranes were present, such as Mixteco and Oaxaquia, exhibiting a carbonate platform (Fig. 9). Flysch basins are developed in the intervening parts of the Pennsylvanian of Mexico (Fig. 9). In the Mixteco and Oaxaquia terranes, Chaetetes have not been found owing to the weak development of limestone in this area, which was probably the bottom of an ocean at the time of deposition. However, a carbonate platform is probably present elsewhere as the chaetetids are known in the Carboniferous of South America (Suchy & West 2001, p. 431), as are fusulinellids (Dunbar & Newell 1946; Newell et al. 1953).
Conclusions 9 The Atokan Chaetetes bioconstructions of the Sierra Agua Verde (Sonora) can be correlated
ATOKAN CHAETETIDS OF SONORA (MEXICO)
NORTH
AMERICAN
CRATON e
. ""
197
s
,'
/
RHEIC ,,
OCEAN
J
/
/
m-~ 1
I
I
baG O N D WAN A
".<. ~ j ' ,
Peru
Fig. 9. Atokan palaeogeography from Sonora to Peru showing an hypothetical carbonate platform as a possible migration seaway for the chaetetids and fusulinellids. 1, carbonate platform; 2, flysch basins; 3, mainlands. Not to scale.
with coeval buildups of central Texas; the presence of the fusulinellid group Fusulinella llanoensis confirms the biogeographical affinity with the Texas assemblages; the algae Komia and Eugonophyllum are abundant in both areas. The Sierra Agua Verde and central Texas environments are relatively deeper and were less similar to the environments and constructions of Kansas, Arizona and eastern Texas. Contrary to the Kansas reefs, no associated syringoporid tabulate corals Multithecopora
or algal crusts have been observed, confirming the deeper palaeobathymetry of the Sonora outcrop. 9 In the Sierra Agua Verde, the colonies of Chaetetes were located below wave base (generally located by the authors to be c. -100 m). Algae appear with Komia and Eugonophyllum in the topping beds (whose deposit depth decreased probably to 25 m). 9 The chaetetid reef mounds of the Sierra Agua Verde were accumulated within a TST of the late Atokan.
198
E. ALMAZ,~N-V,/~ZQUEZ ET AL.
9 The growth of chaetetids in firm, lithified substrate is inferred. 9 Symbiosis with heterotrophic bacteria and non-phototrophic algae is suggested by the probable growth of the Sierra Verde chaetetids in the disphotic or aphotic zone. 9 Life time of the chaetetids up to 3000 years, and a probable average longevity of 3 years for the fusellinellids, allow the to precise correlation of the growth of the chaetetids and their successive trapping of fusulinellid limestone. The crinoids, which have an intermediary longevity, provide some regional encrinites partially enveloping the chaetetids and must also be studied from this point of view. 9 Although the biotopes are slightly different (shallower), in Sonora, Arizona and Texas the three provinces were connected in only one tectono-stratigraphic domain with similar fossil assemblages. This domain was necessarily related to the Mexican terranes of Oaxaquia and Mixteco, as the unique way through the Rheic Ocean, for the chaetetids and fusulinellids migrating to South America. The project M00U01 'Un estudio sedimentol6gico, micropalentol6gico y geoquimico del Paleozoico de M6xico' of ECOS-Nord (Evaluation-orientation de la Coop6ration Scientifique), ANUIES (Asociaci6n Nacional de Instituciones de Educaci6n Superior), CONACYT (Consejo Nacional de Ciencia y Tecnologia) and UNAM (Universidad Nacional Autonoma de Mexico), and the project UNAM-PAPIIT No. IN 104103-3 'Bioestratigrafia de rocas de plataforma del Pensilvfinico y P6rmico de Sonora, M6xico', financially supported this research.We thank L. Pille and J. Gaillot for technical help. The criticisms of reviewers R. West, D. Suchy and M. Aretz were pertinent and, in a word, constructive.
References BIJU-DUVAL,B. & SAVOYE,B. 2001. Ockanologie, 2nd edn. Dunod, Paris. BOSSCHER, H. • SCHLAGER, W. 1992. Computer simulation of reef growth. Sedimentology, 39, 503-512. BROMLEY, R. G. & HEINBERG, C. 2006. Attachment strategies of organisms on hard substrates: A palaeontological view. Palaeogeography, Palaeoclimatology, Palaeoecology, 232, 429-453. BUITRON-SANCHEZ, B., GOMEZ-ESPINOSA, C., ALMAZAN-VAZQUEZ, E. & VACHARD, D. 2007. A late Atokan regional encrinite (early late Moscovian, Middle Pennsylvanian) in the Sierra Agua Verde, Sonora state, NW Mexico. In: ~kLVARO, J. J., ARETZ, M., BOULVAIN, F., MUNNECKE, A., VACHARD,D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations: Climatic
and Evolutionary Controls. Geological Society, London, Special Publications, 275, 201-209. BUITRON-SANCHEZ, B., ALMAZAN-VAZQUEZ, E. & VACHARD, D. 2004. Benthic invertebrates of Carboniferous-Permian age from Sonora: their paleogeographic implications. In: 32nd International Geological Congress, Florence, Italy, Abstracts, 202. COJAN, I. 8~ RENARD, M. 2003. SOdimentologie. 2nd edn. Dunod, Paris. CONNOLLY, M. 1990. Detailed measured section of Atokan and Desmoinesian strata, Horquilla Limestone, Dry Canyon, Whetstone Mountains, Cochise County, Arizona. In: SCHEIBER,J. F. (ed.) Upper Paleozoic Stratigraphy of the Whetstone Mountains, Cochise and Pima Counties, Arizona. Geological Society of America, 86th Cordilleran Section Meeting, Tucson, Supplement Volume, 19-44. CONNOLLY,W. M., LAMBERT,L. L. & STANTON,R. J. 1989. Paleoecology of Lower and Middle Pennsylvanian (Middle Carboniferous) Chaetetes in North America. Facies, 20, 139-168. CREMER, H. 1995. Spicule pseudomorphs in Upper Triassic (Norian) chaetetid sponges from the western Taurids (Antalya-Region, SW Turkey). Gkobios, 28, 163-174. DELLA PORTA,G., KENTER, J. A. M., BAHAMONDE,J. R., IMMENHAUSER,A. 8z VILLA,E. 2003. Microbial boundstone dominated carbonate slope (Upper Carboniferous, N Spain): microfacies, lithofacies distribution and stratal geometry. Facies, 49, 175-208. DUNBAR, C. O. & NEWELL,N. D. 1946. Marine Early Permian of the central Andes and its fusuline faunas. American Journal of Sciences, 244, 377-402. FALLON, S. J., MCCULLOCH,M. T. & GUILDERSON,T. P. 2005. Interpreting environmental signals from the coralline sponge Asterosclera willeyana. Palaeogeography, Palaeoclimatology, Palaeoecology, 228, 58-69. FLUGEL, E. 2004. Microfacies of Carbonate Rocks, Analysis, Interpretation and Application. Springer, Berlin. GOMEZ-ESPINOSA, C. 2007. Palaeobiology and taphonomy of the biota of Sierra Agua Verde (late Moscovian - late Atokan) Sonora, Mexico. Unpublished PhD thesis, Universidad de Sonora, Mexico. GONZALEZ-LEt3N,C. 1986. Estratigrafia del Paleozoico de la Sierra del Tule, noreste de Sonora. Universidad Nacional Aut6noma de Mkxico, Instituto de Geologia, Revista, 6, 117-135. GONZALEZ-LEON, C. 1989. Evoluci6n de terrenos mesozoicos en el noroeste de M6xico, Boletin del Departamento de Geologia de la Universidad de Sonora, 6, 39-54. GROVES, J. R. 1991. Fusulinacean biostratigraphy of the Marble Falls Limestone (Pennsylvanian) western Llano Region, Central Texas. Journal of Foraminiferal Research, 21, 67-95. GROVES, J. R. 1992. Stratigraphic distribution of non-fusulinacean foraminifers in the Marble Falls Limestone (Lower-Middle Pennsylvanian), western Llano region, central Texas. In: SUTHERLAND,
ATOKAN CHAETETIDS OF SONORA (MEXICO) P. K. & MANGER, W. L. (eds) Recent Advances in Middle-Carboniferous Biostratigraphy. A Symposium. Okhahoma Geological Survey, Circular, 94, 145-161. HECKEL, P. 2004. Chairman's column. Newsletter on Carboniferous Stratigraphy, 22, 1-3. IZART, A., BRIAND, C., VASLET, D., VACHARD, D., COQUEL, R. & MASLO, A. 1996. Stratigraphy and sequence stratigraphy of the Moscovian in the Donets Basin. Tectonophysics, 268, 189-209. KRAINER, K., FL(JGEL, E., VACHARD, O. & JOACHIMSKI,M. 2003. A close look at Late Carboniferous algal mounds: Schulterkofel, Carnic Alps, Austria. Facies, 49, 325-350. LOPEZ-RAMOS, E. 1985. Geologia de Mkxico, Volume II, 4th edn, primera reimpresi6n. Publicaciones Secretariga de Educaci6n Pfiblica, M6xico D. F. MAMET, B. & VILLA, E. 2004. Calcareous marine algae from the Carboniferous (Moscovian-Gzhelian) of the Cantabrian zone (NW Spain). Revista EspaBola de Paleontologia, 19, 151-190. MAMET, B. L., ROUX, A. & NASSICHUK,W. W. 1987. Algues carbonif6res et permiennes de l'Arctique canadien. Geological Survey of Canada, Bulletin, 342, 1-143. MENDOZA-MADERA, C., ALMAZA.N-VAZQUEZ, E., BUITRON-SANCHEZ, B. E. t~ VACHARD, D. 2004. Bioestratigrafia de la secuencia del Pensilvfinico en la Sierra Agua Verde, en la porci6n central del Estado de Sonora. Universidad de Sonora, Divisi6n de Ciencias Exactas y Naturales, Departamento de Geologia, Res~menes, Semana Cultural, XXIX, 9-10. MINWEGEN, E. 2001. Die Biokonstruktionen im Pennsylvanium des Kantabrischen Gebirges (Nor&panien). K61ner Forum fiir Geologie und Pal~iontologie, 912001, 1-139. NEWELL, N. D., CHRONIC, J. & ROBERTS, T. G. 1953. Upper Paleozoic of Peru. Part 3, Fusulinidae. Geological Society of America, Memoirs, 58, 167-226. OCHOA-GRANILLO, J. A. & SOSA-LEON, J. P. 1993. Geologia y estratigrafia de la Sierra Agua Verde con dnfasis en el Paleozoico. Departamento de Geologia Tesis, Universidad de Sonora. PEIFFER-RANGIN, F. 1987. Biostratigraphic study of Paleozoic rocks of northeastern and central Sonora. PhD thesis, Universit6 de Paris. RIDING, R. 2002. Structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth-Science Reviews, 58, 163-231. Roux, A. 1985. Introduction fi l'6tude des algues fossiles pal6ozoiques (de la bact6rie/t la tectonique des plaques). Bulletin Centres Recherche Exploration-Production Elf-Aquitaine, 9,465-699. S/kNCHEZ-ZAVALA, J. L., CENTENO-GARCIA, E. & ORTEGA-GUTII~RREZ, F. 1999. Review of the Paleozoic stratigraphy of Mexico and its role in the Gondwana-Laurentia connections. Geological Society of America, Special Paper, 336, 211-226. SEDLOCK, R. L., ORTEGA-GUTII~RREZ,F. 8r SPEED, R. C. 1993. Tectonostatigraphic terranes and tectonic evolution of Mexico. Geological Society of America, Special Paper, 278, 1-153.
199
SKINNER, J. W. & WILDE, G. L. 1954. New early Pennsylvanian fusulinids from Texas, Journal of Paleontology, 28, 796-803. STANTON, R. J., CONNOLLY,W. M. • LAMBERT, L. L. 1994. Palaeoecology of Upper Carboniferous Chaetetes - morphology, growth style, and spatial distribution. Courier Forschungsinstitut Senckenberg, 172, 365-372. STANTON, R. J., LAMBERT, L. L. & WEBB, G. E. 1997. Positive geotropic growth in Chaetetes. Boletin de la Real Sociedad Espa1~ola de Historia Natural, 92, 197-207. STEWART,J. H., AMAYA-MARTINEZ,R., STAMM,R. G., WARDLAW,B. R. ~r STANLEY,C. D. 1997. Stratigraphy and regional significance of Mississippian to Jurassic rocks in Sierra Santa Teresa, Sonora, Mexico. Revista Mexicana de Ciencias Geoldgicas, 14, 115-135. STEWART, J. H., POOLE, F. G., HARRIS, A. G., REPETSKI, J. E., WARDLAW,B. R., MAMET, B. L. & MORALES-RAMIREZ,J. M. 1999. Neoproterozoic (?) to Pennsylvanian inner-Shelf, miogeosynclinal strata in Sierra Agua Verde, Sonora, M6xico. Revista Mexicana de Cieneias Geoldgicas, 16, 35-62. SUCHY, D. R., & WEST, R. R. 2001. Chaetetid buildups in a Westphalian (Desmoinesian) cyclothem in southeastern Kansas. Palaios, 16, 425433. TI~LLEZ-GIRON,C. 1979. Microfacies y microfdsiles de la Formaci6n Horquilla, Norte de Mexico. Instituto Mexicano del Petr61eo, Proyecto, C-3044. VACHARD, D. t~r BOUYX, E. 2002. Les Eopolydiexodina g6antes (Foraminiferida, Fusulinina) du Permien moyen d'Afghanistan, remarques pr61iminaires. Annales de la Soci~tk G@ologiquedu Nord, 2e s6rie, 9, 163-189. VACHARD, D., FLORES DE DIOS, A., BUITRON, B. E. & GRAJALES-NISHIMURA, M. 2000a. Biostratigraphie par fusulines des calcaires carbonif6res et permiens de San Salvador Patlanoaya (Puebla, Mexique). G@obios, 33, 5-33. VACHARD, O., FLORES DE DIOS, A., PANTOJA, J., BUITRON, B., ARELLANO,J. & GRAJALESM. 2000b. Les fusulines du Mexique, une revue biostratigraphique et pal6og6ographique. G@obios, 33, 655-679. VACHARD,D., V1DAURRE-LEMUS,M., FOURCADE,E. & REQUENA, J. 2000c. New Early Permian fusulinid assemblage from Guatemala. Comptes Rendus de l'AcadOmie des Sciences de Paris, 33, 789-796. WEIDLICH, O. 2007. Permian reef and shelf carbonates of the Arabian platform and Neo-Tethys as recorders of climatic and oceanographic changes. In: flILVARO, J. J., ARETZ, M., BOULVAIN, F., MUNNECKE, A., VACHARD,D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations. Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 229-253. WEST, R. R. 1992. Chaetetes (Demospongiae): its occurrence and biostratigraphic utility. Oklahoma Gkological Survey, Circular, 94, 163-169. WEST, R. R. 1994. Species in coralline demosponges Chaetetida. In." OEKENTORP~ P. (ed.) Proceedings of the 4th International Symposium on Fossil Cnidaria and Porifera held in Munster,
200
E. ALMAZJ~N-V,/~ZQUEZ ET AL.
Germany, 9-14 September 1991. Courier Forschungsinstitut Senckenberg, 172, 399409. WILDE, G. L. 1990. Practical Fusulinid zonation" the species concept, with Permian Basin. West Texas Geological Society Bulletin, 29, 5-34, Midland Texas. WILSON, J. L., MADRID-SOLIS, A. & MALPICA-CRUZ, R. 1969. Microfacies of Pennsylvanian and Wolfcampian strata in southwestern U.S.A.
and Chihuahua, Mexico. In: CORDOBA, D. A., WENGERD, S. A. & SHOMAKER,J. (eds) Guidebook of the Border Region. New Mexico Geological Society, Twentieth Field Conference, 80-90. WRIGHT, V. P. • BURCHETTE, T. P. 1996. Shallow water carbonate environments. In." READING, H. G. (ed.) Sedimentary Environments: Processes, Facies and Stratigraphy, 3rd edn. Blackwe11 Science, Oxford.
A late Atokan regional encrinite (early late Moscovian, Middle Pennsylvanian) in the Sierra Agua Verde, Sonora state, NW Mexico B. E. B U I T R O N - S , / ~ N C H E Z l, C. G O M E Z - E S P I N O S A 1, E. A L M A Z A N - V J k Z Q U E Z 2 & D. V A C H A R D 3
' Universidad Nacional Aut6noma de MOxico, Instituto de Geologia, Departamento de Paleontologia, Ciudad Universitaria, Delegaci6n Coyoacdn, 14510 MOxico, D. F., Mexico (e-mail: blancab@servidor, unam. mx) 2Universidad de Sonora, Departamento de Geologia, Boulevard Luis Encinas y Rosales 83000 Hermosillo, Sonora, Mexico 3UniversitO de Lille 1, Sciences de la Terre, UMR 8014 du CNRS, Laboratoire LP3, Bdtiment SN5, 59655 Villeneuve d'Ascq COdex, France (e-mail: daniel,
[email protected]) Abstract: In the Sierra Agua Verde, central Sonora state, NW Mexico, the La Joya Formation exhibits an alternation (100 m thick) of calcareous siltstone and fossiliferous limestone with nodular cherts. This latter contains an abundant and diverse late Atokan (i.e. Podolskian =early late Moscovian, Middle Pennsylvanian) fossil assemblage composed of phylloid algae, fusulinids, chaetetids, tabulate corals, gastropods, fenestellid bryozoans, spiriferid and productid brachiopods, crinoids and conodonts. The crinoidal beds constitute a good example of a regional encrinite. They include several species of the parataxonomic stem form-genera Cyclocaudex, Cyclocrista, Heterosteleschus, Mooreanteris, Pentagonopterix, Preptopremnum, Cycloscapus and Pentaridica. Their preservation indicates the combination of preburial decay on the sea floor and post-burial decay within the sediment. The high degree of silicification of the crinoids indicates that they were possibly associated with siliceous organisms (Porifera?), not preserved in the assemblages. The studied thanatocoenosis is typical of tropical shallow seas, and reveals strong biogeographical affinities with the assemblages of the midcontinental and southern regions of the USA. Particularly, the Atokan crinoids of central Sonora are similar to those from Kansas and Texas, confirming the close palaeogeographic connection of southern USA and northern Mexico during the Middle Pennsylvanian.
The Sierra Agua Verde is located 120 km eastward of Hermosillo City, Sonora state, in the 'Sierras and Valles del Norte' Mexican subprovince. This range covers a surface area of 255 km 2 and its centre is at 109~176 and 29~176 (Fig. 1). It is composed of sedimentary, igneous and metamorphic rocks of Palaeozoic, Mesozoic and Cenozoic age. Geology and palaeontology of the Sierra Agua Verde were described by Poole et al. (1984), Stewart et al. (1988, 1999), Minjarez-Sosa et al. (1993), Ochoa-Granillo & Sosa-Leon (1993), Stewart & Poole (2002), Mendoza-Madera et al. (2004) and Buitr6n-Sfinchez et al. (2005a, b). The Palaeozoic series includes the following formations: Puerto Blanco, Cuarcita Proveedora, Buelna and Arrojos (Cambrian); E1 Boquinete (Ordovician); E1 Pollo (Devonian); Santiago (Mississippian); La Joya (Pennsylvanian); and Tuntunud6 (Permian).
Material and methods The La Joya Formation was logged and sampled bed by bed. All the beds marked as 'crinoids' in Figure 2 were studied, as well as those of other Pennsylvanian outcrops from Sonora in order to investigate several 'regional encrinites' in the sense of Ausich (1997) and Ginsburg (2005). Our study of sequence stratigraphy has determined that the La Joya Formation appears as a transgressive systems tract (TST), whose main lithologies are Chaetetes reef mounds (Almaz/mV~zquen et al. 2007), fusulinellid floatstones, phylloid algae framestones and crinoidal rudstones. These crinoidal rudstones are particularly abundant and well known in the Mississippian and Pennsylvanian of the USA, but only a very few have been studied in Mexico. Our investigation is essentially based on the extraction of the best-preserved specimens of crinoids for
~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special
From:
Publications, 275, 201-209.0305-8719/07l$15.00 9 The Geological Society of London.
202
B.E. BUITRON-S_ANCHEZ E T AL.
Fig. 1. Locations of (a) Sonora in Mexico and (b) Sierra Agua Verde in Sonora.
taxonomic purposes and a microfacies analysis from a taphonomic point of view, i.e. the taphofacies method of Speyer & Brett (1988) and C6zar (2003). Quantitative analyses based on the techniques summarized by Holterhoff (1996) concerning the crinoid biofacies will be applied to the material (G6mez-Espinosa's PhD 2007); this preliminary study presents only the systematic and taphonomic data obtained in hand specimens and thin sections.
Regional encrinites Regional encrinites were defined as crinoidal grainstones and packstones composed of more than 50% by volume of pelmatozoan debris that are at least 5-10 m thick and 500 km 2 in extension (Ausich 1997; Ginsburg 2005). They are an example of taphonomic feedback (Kidwell & Jablonski 1983) by the predominance of one special skeletal grain type. Regional encrinites
ATOKAN REGIONAL ENCRINITE OF NW MEXICO
Metres ~ l t / ~ I J I Brachiopods 10o_.~[ ~_": =1= =. _=j_== Fusulinids II , I , IjlCrinoids I L ' .... , " ~Solitary corals
/ F---i--i-.==- t
8o.-ll I
.
~
,o-lL ',,'
I/
',
Synngopora
','J
| Jr- ['_~_T_--_:.~Fus~,in'ds
/
eO--i ~ / ~
2 - c ; -i - i C.noJ ,
l e,~ I IChaetetid reef mound
!~-_-=--=~_
"_-_~-_-__~-= Solitary corals so-I ,, I w l
~
i ".T-'i--iT l l.,.. I
1
I
I
I Fusulinids
t l
'--'-~ --'~'l 40i ";U--'-----L4
' i ,~
3o-
~
I z /Crinoids
:---
Synngopora
,-i
Brachiopods Solitary corals Chaetetids
Syringopora Brachiopods
203
occur from the Ordovician to the Jurassic, but the acme of their development in North America is in the early Mississippian (Osagean) (Ausich 1997; Ginsburg 2005). At this time and during the Pennsylvanian individual regional encrinites occur with other carbonate lithofacies in different formations of the USA and northern Mexico. Three main factors related to the evolutionary taphonomical and diagenetical history of crinoids and their skeletal remains are important for the formation of regional encrinites: (1) the long geological history of crinoids is marked by episodes of exceptional abundance, which produce encrinites (Ausich et al. 1999); (2) the stalks and calyces of crinoids consist of separate plates and segments held together with ligamental tissue in a labyrinthine cavity system, the stereome; and (3) the post-mortem decay of this tissue provides plates of the calyx and more numerous segments of the stem, which are highly porous (Ginsburg 2005). These hydraulically light fragments can be entrained, transported and redeposited by currents and orbital wave velocities as low as 16 cm s-1 using the results from flume experiments with crinoid segments from living specimens (Ginsburg 2005). As a result of this ease of movement, bottom sediments of abundant crinoidal debris were probably frequently mobile resulting in the common occurrence of current bedding and graded bedding of encrinites (Ausich 1997; Ginsburg 2005). Furthermore, as highly porous crinoid ossicles are entrained at lower current velocities than solid carbonate particle of the same size (e.g. Fliigel 2004), accumulation rates in the encrinite are generally very high (more than 50% or even 60% of the rock volume), and the regional encrinites are generally deposited below normal wave base but within storm wave base (Fliige12004).
Fossil assemblage in Sierra Agua Verde
Crinoids
Fusulinids ~
Covered base Limestonewith chert nodules -~ Siltstone
Fig. 2. Lithological column of the La Joya Formation in Sierra Agua Verde (Sonora, Mexico). The lower contact is not exposed in the field. The upper contact is tectonic (normal fault). Fifteen field samples were collected in the three levels mentioned as 'Crinoids'. Each one contains about 100 crinoids, taxonomically studied.
The studied crinoid fauna has been sampled from marine fossiliferous limestone of the La Joya Formation (100 m thick), which is intercalated with nodular cherts and calcareous siltstone (Fig. 2). The associated biota includes: calcareous algae (Eugonophyllum sp., Zidella? sp., Kamaena? sp., Komia eganensis); problematic foraminifers (Pachysphaerina pachysphaerica,
Eotuberitina reitlingerae, Insolentitheca horrida); smaller foraminifers (Endothyra ex gr. bowmani, Climacammina ex gr. moelleri, Globivalvulina bulloides); fusulinids (Eostaffella grozdilovae (=E. acuta of the authors), Millerella sp., Pseudostaffella sp., Staffella powwowensis,
204
B.E. BUITR6N-SANCHEZ ETAL.
Eoschubertella texana, Fusulinella thompsoni, F. llanoensis, Nipperella? sp.); chaetetids; tabulate corals (Michelinia sp.) and zaphrentid solitary corals (Lophophyllidium sp.); gastropods (Euomphalus sp., cf. Donaldina robusta); brachiopods (Spirifer sp.); and conodonts. These fossils indicate an Atokan age (=Podolskian=early late Moscovian = Middle Pennsylvanian). The crinoid fauna (Fig. 3) includes 11 parataxonomic stem species: Cyclocaudex insaturatus Moore & Jeffords, Cyclocrista martini Miller, Heterosteleschus keithi Miller, Lamprosterigma erathense Moore & Jeffords, Lamprosterigma mirificum Moore & Jeffords, Mooreanteris waylandensis Miller, Pentagonopterix insculptus Moore & Jeffords, Preptopremnum laeve Moore & Jeffords, Preptopremnum rugosum Moore & Jeffords, Cycloscapus laevis Moore & Jeffords and Pentaridica simplicis Moore & Jeffords. This fauna appears similar to those described in the USA by Moore & Jeffords (1968). The thanatocoenosis is typical of tropical shallow seas and shows a strong affinity with the characteristic species of the midcontinental and southern regions of the USA. In particular, the Sonora Pennsylvanian crinoids show similarities to the assemblages of Kansas and Texas (Buitr6nSfinchez et al. 2004, 2005a, b). All these regions are located in tropical and subtropical latitudes during the Pennsylvanian epoch (Heckel 2002). They also have much in common with Eurasiatic-Arctic faunas corresponding to the same climatic zones during this epoch (see, for example, the reconstructions of Scotese & Langford 1995; Ross 1995; Vachard et al. 1997, 2000b, c). Similarly, fusulinids, calcareous algae and smaller foraminifers represented in the La Joya Formation show affinities with Arizona, New Mexico, Texas and California microfaunas. The Permian fusulinids (P6rez-Ramos 1992) share the same affinities. These new data confirm the palaeogeographical reconstructions of Vachard et al. (1997, 2000a-c) and Buitr6n-Sfinchez et al. (2004) in Mexico and Guatemala.
Taphonomical implications The taphonomical study of crinoids was not initiated until the 1960s (Ausich et al. 1999). Crinoids are rarely represented as complete fossils, commonly they are disarticulated after death and are preserved as crown parts, cups, arm ossicles, fragments and stem fragments. A few days after death the muscles and ligaments decay, which leads to the disintegration of the skeleton into isolated elements. In order to be
preserved completely, a crinoid must be deeply and quickly buried to prevent the reexcavation by currents or disruption by scavengers and burrowers (Donovan 1991; Ausich et al. 1999). In normal marine conditions the echinoderms usually disarticulate into individual ossicles within a period of 1-2 weeks, depending on their construction and environmental factors, but specifically the arms and cirri of recent crinoids begin to disarticulate within the 3 first days after their death, whereas calyx and certain segments of the arms may be disarticulated after 6 days (Dornbos & Bottjer 2001). In crinoids the articulations between multicolumnal segments of the column consist only of intercolumnal ligament, which are zones of structural weakness after death. This was demonstrated in experiments on recent isocrinids. A similar ligamentary organization of the column was observed in crinoids of the Early Mississippian. When a column begins to disarticulate the articulation between segments breaks first, producing fragments of column of almost equal length. Finally, when all the through-going ligaments decay, the whole column disarticulates into individual columnals (Baumiller & Ausich 1992; Ausich & Baumiller 1993). The effects of physical disturbance on echinoderms were studied by Kidwell & Baumiller (1990). The results obtained from laboratory experiments with living crinoids indicate that recently dead crinoids remain articulated through hours of physical disturbance, whereas dead echinoids disarticulate quickly after physical disturbance. In the La Joya Formation the majority of crinoids are represented by fragments of columns and a very low percentage by isolated plates (three specimens in 10 representative samples, each containing approximately 100 characteristic elements of skeleton). Thus, it can be inferred that decay allowed disarticulation of the arms and calyces, and a breaking of the intercolumnar ligaments, but not complete disarticulation of the columns, are evidenced by columns fragments. The preservation of crinoids in La Joya Formation probably shows a combination of decay of the crinoids preburial on the sea floor and decay of crinoids post-burial within the sediments. The limestones rich in crinoids exhibit various taphonomic processes and difference in the depositional environment. Coarse-grained crinoidal limestones are occasionally interpreted as having originated as a result of gravity flows (Cook & Mullins 1983; Martin 1999; Fltigel 2004) but more frequently as tempestites (e.g. Fliige12004). This interpretation is confirmed by the Sierra Agua Verde deposits, owing to the absence of turbidite figures of turbidites and microfacies
ATOKAN REGIONAL ENCRINITE OF NW MEXICO
205
Fig. 3. Late Atokan crinoid fauna from Sierra Agua Verde (Sonora, Mexico). All scale bars are 0.700 mm. (a) Cyclocaudex insaturatus Moore & Jeffords. (b) Cyclocrista martini Miller. (e) Heterosteleschus keithi Miller. (d) Lamprosterigma mirificum Moore & Jeffords. (e) Lamprosterigma erathense Moore & Jeffords. (f) Cycloscapus laevis Moore & Jeffords. (g) Preptopremnum laeve Moore & Jeffords. (h) Preptopremnum rugosum Moore & Jeffords. (i) Mooreanteris waylandensis Miller. (j) Pentagonopterix insculptus Moore & Jeffords and Pentaridica simplicis Moore & Jeffords. (k) Field picture of the crinoidal rudstones in the Sierra Agua Verde.
206
B.E. BUITR6N-SANCHEZ ETAL.
analyses that infer tempestites, and probably distal tempestites (according to the criteria summarized by Flfigel 2004, p. 596). The problem is to understand how and when favourable weak currents for suspension feeders like crinoids (see detailed data in Holterhoff 1997) become strong storm currents, i.e. the interaction of organisms with moving fluids. An average accumulation consists of 50% columnal plates that are totally disarticulated, 5% Fusulinella and very rare smaller foraminifers, 5% fenestellid bryozoa (and rare other remains of metazoans) and 40% syntaxial sparite cement. Nevertheless, the associated fusulinellids are poorly preserved, with many truncation facets indicating their allochthonous origin (e.g. Fernfindez-L6pez 2000). The cementation of the distal tempestites with crinoids is very rapid and took place very early, as the ossicles are not affected by compaction or other threedimensional finite strain (compare with Rowan 1991). The character of distal tempestite is also obvious in associated deposits, namely fusulinellid wackestone-packstone, because of the homogeneous micritic matrix and the absence of corrosion features on the fusulinellids, their oligotipy and the absence of associated photophile green algae, all characteristics indicating a rapid burial after transport and sorting. Ginsburg (2005) has underlined that fusulinids offer many opportunities for taphonomic feedbacks that are comparable to that of crinoidal debris. Chaetetid reef mounds, which developed in coeval beds, are located in the lower part of the photic zone, just below wave base (Almaz~mVfi.zquez et al. 2007) - thus supporting the interpretation of the depositional environments of the crinoidal beds. The absence of encrusting algae or foraminifers (e.g. Claracrusta, Calcivertella) on the fusulinellids corroborates the rapid burial of the biota within the microfacies, whereas the absence of ichnofossils in the outcrop indicates an absence of conditions for the establishment of an infrabenthic community. Consequently, in the Atokan Sierra Agua Verde carbonate platform, the crinoid biotopes have been very productive in supplying carbonate fragments for the distal tempestites, first by the ossicles and, secondly, because the stalked crinoids probably acted as substrates for the fusulinellids, another important component of the tempestites (Fig. 4). The chaetetid reef mounds were located in a lower part of the platform and trapped the sediments transported by the storm currents, i.e. the fusulinellids floatstones and the crinoid rudstones. Between the crinoid biotopes and these chaetetid mounds, infaunal organisms were probably inhabiting the micritic substrate (Fig. 4).
Fig. 4. Reconstruction of the biotopes of crinoids (1) and fusulinellids (2), with their first accumulations (1.1 and 2.1)and the final deposits (2.1.1 and 1.1.1) accumulated around the chaetetid constructions (3) in the Sierra Agua Verde (Sonora, Mexico). Not to scale.
The accumulation of crinoids may also correspond to colder water inputs, similar to those reconstructed in the Permian deposits of Oman (Weidlich 2007). Nevertheless, as indicated above, these crinoids are interpreted as tropical, and the possible mixing of cold and warm waters warrants further discussion in various outcrops and/or areas (see also the 'Auernig paradox' of Samankassou 2002 in the Carnic Alps). Other arguments for possible cold episodes are: (a) that the skeletal composition of the benthic fauna in these levels of the Sierra Agua Verde was principally calcium carbonate (Table 1), and poorly aragonitic, with a dominance of the 'heterozoan association' on the 'photozoan association' (e.g. Samankassou 2002); and (b) the existence of biogenic chert formation (see Beauchamp & Baud 2002) during the Atokan of Sonora, where most of microfossils are preserved by silicification and show internal diagenesis as molds with ornamental features. The source of silica is organic in origin because there is no evidence of volcanism or evaporation zones in the studied area, and because this type of silicification is common in all the investigated encrinites, although their provenances and deposit environments vary. The high degree of silicification suggests the presence of siliceous organisms (Porifera?); consequently, the apparently crinoid-dominated accumulations might correspond to crinoid-sponge associations. The physico-chemical conditions caused the dissolution of these skeletons of associated sponges, and this dissolved silica penetrated and preserved the dead organisms. In the microfacies the first silicifications appear as small points within the crinoid network. The lithostatic compression can be inferred preferentially to the sedimentation, and the silica could also have been mobilized from the interbedded siltstone from burial
ATOKAN REGIONAL ENCRINITE OF NW MEXICO
207
Table 1. Original mineralogy composition of the benthicfauna present in Agua Verde range C, common; X, less common (after Martin 1999) Taxon Gastropods Brachiopods Sponges Bryozoans Echinoderms Benthic foraminifers
Aragonite
Calcite low Mg
Calcite high Mg
C C
X C C C X
Aragonite + Calcite
Silica
C C C X
diagenesis. Therefore, more geochemical investigations are necessary in order to elucidate the fossil diagenesis of these crinoid accumulations.
Conclusions 9 The crinoid assemblage from the La Joya Formation is diversified and composed of 11 stem-based species. 9 The crinoids are typical forms of Pennsylvanian age; more precisely, they are late Atokan in age, as indicated by the associated Fusulinella. 9 Chaetetids, corals, fusulinellids, calcareous algae and brachiopods indicate tropical seas with normal salinity. Therefore, these conditions can be assumed for the crinoids; nevertheless, their possible value as local proxies of colder waters input is debatable. 9 The biotopes of each group are probably located in different environments along a carbonate platform and largely depending on the current system. 9 Crinoids indicate the presence of a hard substrate where they colonized, and were probably in turn the substrate for many epifaunal organisms, the fusulinellids for instance. 9 The crinoids of the La Joya Formation, represented by fragments of columns and a very low percentage of isolated plates, indicate that decay allowed disarticulation of the arms and calyx and the breaking of the intercolumnar ligaments, but not the through-going ligament of columnal plates. These crinoids show a typical combination of a preburial decay on the sea bottom and a post-burial decay within the sediments. 9 Regional encrinites of the La Joya Formation are the result of in situ accumulations rapidly reworked by storm waves and resedimented as distal tempestites. 9 These distal tempestites were finally accumulated near chaetetid bioconstructions developed in another part of the platform. Between the crinoid biotopes and these constructions
some infaunal organisms probably inhabited the micritic substrate. The high degree of silicification of the crinoids indicates that they were possibly associated with siliceous organisms (Porifera?), not preserved in the assemblages. The thanatocoenoses suggest a strong affinity with mid-continental and southern regions in the USA, especially with Kansas and Texas. The seaways of Sonora, southern USA and the midcontinent were apparently well connected during the Atokan, as the ecological conditions were remarkably similar. The Evaluation-orientation de la Coop6ration Scientifique, Asociaci6n Nacional de Instituciones de Educaci6n Superior, Consejo Nacional de Ciencia y Tecnologia, Universidad Nacional Autonoma de Mexico project No. MOOU01 'Un estudio sedimentol6gico, micropalentol6gico y geoquimico del Paleozoico de M6xico', and UNAM-PAPIIT project No. IN104103-3 'Bioestratigrafia de rocas de plataforma del Pensilvfinico y P~rmico de Sonora, Mdxico', financially supported this research. We are grateful to L. Pille for her technical help. We thank the reviewers T. Kammer, E. Gluchowski and M. Aretz for their constructive remarks and corrections.
References ALMAZAN-VAZQUEZ, E., BUITRON-SANCHEZ, B. E., VACHARD, D., MENDOZA-MADERA, C. ~ GOMEZESPINOSA, C. 2007. The late Atokan (Moscovian, Pennsylvanian) chaetetid accumulations of Sierra Agua Verde, Sonora (NW Mexico): composition, facies and palaeoenvironmental signals. In: ALVARO, J. J., ARETZ, M., BOULVAIN,F., MUNNECKE, A., VACHARD,D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 2"/5, 189-200. AusIcH, W. I. 1997. Regional encrinites: a vanished lithofacies. In: BRETT, C. E. & BAIRD,G. C. (eds) Paleontological Events, Stratigraphic, Ecologicaland Evolutionary Implications. Columbia University Press, New York, 509-520.
208
B.E. BUITRON-SANCHEZ ET AL.
AUSICH, W. I. & BAUMILLER,T. K. 1993. Taphonomic method for determining muscular articulations in fossil crinoids. Palaios, 8, 477484. AusIcH, W. I., BRETT, C. E. & HESS, H. 1999. Chapter 4 Taphonomy. In" HESS, H., AuSICH, W. I., BRETT, C. E. & SIMMS, M. J. (eds) Fossil Crinoids. Cambridge University Press, Cambridge. BAUMILLER, T. K. & AUSICH, W. I. 1992. The brokenstick model as a null hypothesis for crinoid stalk taphonomy and as a guide to the distribution of connective tissue in fossils. Paleobiology, l l , 105-119. BEAUCHAMP, B. & BAUD, A. 2002. Growth and demise of Permian biogenic chert along northwest Pangea: evidence for end-Permian collapse of thermocline circulation. Palaeogeography, Palaeoclimatology, Palaeoecology, 184, 37-63. BUITRON-SANCHEZ, B. E., ALMAZAN-VAZQUEZ,E. & VACHARD, D. 2004. Benthic invertebrates, of Carboniferous-Permian age, from Sonora: their paleogeographic implications. In: 32nd International Geological Congress, Florence Italy, Abstract, 202. BUITRON-SANCHEZ, B. E., ALMAZA,N-VA,ZQUEZ, E. & VACHARD, D. 2005a. Pennsylvanian crinoids close to chaetetid 'reef, Sonora, northwestern Mexico. In: Climatic and Evolutionary Controls on Paleozoic Reefs and Bioaccumulations, Colloquium 7-9 Sept 2005, Museum national d'Histoire naturelle, Paris, Abstracts, 16-18. BUITRON-SANCHEZ, B. E., ALMAZ./~N-V/~ZQUEZ, E., VACHARD, D., GOMEZ-ESPINOSA, C. & MENDOZAMADERA, C. 2005b. Crinoides pensilvfinicos asociados a facies 'arrecifales' de chaetetidos en Sierra Agua Verde, Estado de Sonora, Mexico. Geos, Uni6n Geofisica Mexicana, Boletin Informativo, Epoca II, Reunidn annual 2005, 25, 1. COOK, H. E. & MULLINS, H. T. 1983. Basin Margin Environment. In: SCHOLLE, P. A., BE~OUT, D. G. & MOORE, C. H. (eds) Carbonate Depositional Environments. AAPG Memoir, 33, 539-617. C0ZAR, P. 2003. Foraminiferal taphofacies in the Mississippian rocks of the Guadiato area, SW Spain. Facies, 49, 1-18. DONOVAN, S. K. 1991. The Processes of Fossilization. Columbia University Press, New York. DORNBOS, S. Q. & BOTTJER, D. J. 2001. Taphonomy and environmental distribution of helicoplacoid echinoderms. Palaios, 16, 197-204. FERNANDEZ-LOPEZ, S. R. 2000. Temas de Tafonomia. Universidad Complutense de Madrid. FLOGEL, E. 2004. Microfacies of Carbonate Rocks, Analysis, Interpretation and Application. Springer, Berlin. GINSBURG, R. N. 2005. Disobedient sediments can feedback on their transportation, deposition and geomorphology. Sedimentary Geology, 175, 9-18. GOMEZ-ESPINOSA, C. 2007. Palaeobiology and taphonomy of the biota of Sierra Agua Verde (late Moscovian-late Atokan) Sonora, Mexico. Unpublished PhD thesis, Universidad de Sonora, Mexico. HECKEL, P. H. 2002. Overview of Pennsylvanian cyclothems in Midcontinent North America and brief summary of those elsewhere in the world. In: HILLS, L. V., HENDERSON,C. M. & BAMBER, E. W.
(eds) Carboniferous and Permian of the WorM. Canadian Society of Petroleum Geologists, Memoir, 19, 79-98. HOLTERHOFF, P. F. 1996. Crinoid biofacies in Upper Carboniferous cyclothems, mid-continent North America; faunal tracking and the role of regional processes in biofacies recurrence. Palaeogeography, Palaeoclimatology, Palaeoecology, 127, 47-81. HOLTERHOFF, P. F. 1997. Filtration models, guilds, and biofacies: Crinoid paleoecology of the Stanton Formation (Upper Pennsylvanian), mid-continent, North America. Palaeogeography, Palaeoclimatology, Palaeoecology, 130, 177-208. KIDWELL, S. M. & BAUMILLER, T. K. 1990. Experimental disintegration of regular echinoids: roles of temperature, oxygen, and decay thresholds. Paleobiology, 16, 247-271. KIDWELL, S. M. & JABLONSKI,D. 1983. Taphonomic feedback: ecological consequences of shell accumulations. In: TEVESZ,M. J. S. & MCCALL, P. L. (eds) Biotic Interactions in Recent and Fossil Benthic Communities. Topics in Geobiology, 3, 195-248. MARTIN, R. E. 1999. Taphonomy, a Process Approach. Cambridge Paleobiology Series, 4. Cambridge University Press, Cambridge. MENDOZA-MADERA, C., ALMAZA.N-VA, ZQUEZ, E., BUITRON-SANCHEZ, B. E. & VACHARD, D. 2004. Bioestratigrafia de la secuencia del Pensilvfinico en la Sierra Agua Verde, en la porci6n central del Estado de Sonora. Universidad de Sonora, Divisi6n de Ciencias Exactas y Naturales, Semana Cultural XXIX, Resfimenes, 9. MINJAREZ-SOSA, I., OCHOA, J. G. & SOSA, L. P. 1993. Geologia de la Sierra Agua Verde, NE de Villa Pesqueira (Matape). In: GONZALEZ-LE6N, C. & VEGA-GRANILLO, E. L. (eds) Res•menes, Tercer simposio de la geologia de Sonora y (treas adyacentes. Hermosillo, Sonora, MGxico, Universidad de Sonora y Universidad Nacional Aut6noma de MGxico, Instituto de Geologia, 83-85. MOORE, R. C. & JEFEORDS, R. M. 1968. Classification and nomenclature of fossil crinoids based on studies of dissociated parts of their columns. University of Kansas Paleontological Contributions, 46, Echinodermata, Article 9, 1-86. OCHOA-GRANILLO, J. A & SOSA-LEON, J. P. 1993. Geologia y estratigrafia de la Sierra Agua Verde con Onfasis en el Paleozoico. Professional dissertation, Universidad de Sonora, MGxico. PI~REZ-RAMOS, O. 1992. Permian biostratigraphy and correlation between Southeast Arizona and Sonora. Boletin del Departamento de Geologla de la Universidad de Sonora, 9, 1-74. POOLE, F. G., STEWART, J. H. & ARMSTRONG, A. K. 1984. Newly Discovered Paleozoic Section in Central Sonora, in Geological Survey Research 1982. US Geological Survey, Professional Paper, 1375, 1-66. Ross, C. A. 1995. Permian fusulinaceans. In: SCHOLLE, P. A., PERYT, T. M. & ULMER-SCHOLLE,D. S. (eds) The Permian of Northern Pangea. Volume 1. Springer, Berlin, 167-185. ROWAN, M. G. 1991. Three-dimensional finite strain from crinoid ossicles. Journal of Structural Geology, 13, 1049-1059.
ATOKAN REGIONAL ENCRINITE OF NW MEXICO SAMANKASSOU, E. 2002. Cool-water carbonates in a palaeoequatorial shallow-water environment: The paradox of the Auernig cyclic sediments (Upper Pennsylvanian, Carnic Alps, Austria-Italy) and its implications. Geology, 30, 655-658. SCOTESE, C. R. ~; LANGFORD, R. P. 1995. Pangea and the Paleogeography of the Permian. In: SCHOLLE,P. A., PERYT, T. M. & ULMER-SCHOLLE, D. S. (eds) The Permian of Northern Pangea, Volume 1: Paleogeography, Paleoclimates, Stratigraphy. Springer, Berlin, 3-19. SPEYER, S. E. & BRETT, C. E. 1988. Taphofacies models for epiric sea environments: Middle Paleozoic examples. Palaeogeography, Palaeoclimatology, Palaeoecology, 63, 225-262. STEWART, J. H. & POOLE, F. G. 2002. Inventory of Neoproterozoic and Paleozoic strata in Sonora, Mexico. US Geological Survey, Open-file Report 02-97. http://geopubs.wr.usgs.gov/open-file/of02-97 STEWART, J. H., MADRID, R. J., POOLE, F. G. & KETNER, K. B. 1988. Studies of Late Proterozoic, Paleozoic, and Triassic rocks in Sonora, Mexico (ABS.). In: ALMAZ,kN-Vfi~ZQUEZ,E. & FERNANDEZ, A. M. A. (eds) Restimenes, Segundo simposio sobre geologia y mineria de Sonora. Universidad Nacional Aut6noma de M6xico, Instituto de Geologia, 60-62. STEWART, J. H., POOLE, F. G., HARRIS, A. G., REPETSKI, J. E., WARDLAW,B. R., MAMET, B. L. & MORALES, R. J. M. 1999. Neoproterozoic (?) to
209
Pennsylvanian inner-shelf, miogeoclinal strata in Sierra Agua Verde, Sonora, Mexico. Revista Mexicana de Ciencias Geol6gicas, 16, 35-42. VACHARD, D., FOURCADE, E. ET AL. 1997. Foraminif&es et algues du Permien du Guatemala. GOobios, 30, 745-784. VACHARD,D., FLORES DE DKOS,A., BUITRON, B. E. & GRAJALES-NISHIMURA, M. 2000a. Biostratigraphie par fusulines des calcaires carbonif~res et permiens de San Salvador Patlanoaya (Puebla, Mexique). Gkobios, 33, 5-33. VACHARD, D., FLORES DE DIOS, A., PANTOJA, J., BUITRON, B., ARELLANO,J. & GRAJALES,M. 2000b. Les fusulinides du Mexique, une revue biostratigraphique et pal6og6ographique. Gbobios, 33, 655-679. VACHARD,D., VIDAURRE-LEMUS,M., FOURCADE,E. 8r REQUENA, J. 2000c. New Early Permian fusulinid assemblage from Guatemala. Comptes Rendus de l'Acadkmie des Sciences de Paris, 33, 789-796. WEIDLICH, O. 2007. Permian reef and shelf carbonates of the Arabian platform and Neo-Tethys as recorders of climatic and oceanographic changes. In: /~EVARO, J. J., ARETZ, M., BOULVAIN, F., MUNNECKE, A., VACHARD,D. & VENNIN, E. (eds) Palaeozoic Reefs and Bioaccumulations: Climatic and Evolutionary Controls. Geological Society, London, Special Publications, 275, 229-253.
Coelobiontic communities in neptunian fissures of synsedimentary tectonic origin in Permian reef, southern Urals, Russia EMMANUELLE
VENNIN
BiogOosciences, UniversitO de Bourgogne, 21000 Dijon, France (e-mail." emmanuelle,
[email protected]) Sedimentary dykes in the Permian reef complexes of the Russian platform are well preserved and important in providing information about reef growth, the reef biota and, particularly, cavity-dwelling organisms and sediment sources. Two main fissure assemblages are recognized with N80 ~ and N 170~ (late Asselian-early Sakmarian) and N 130~176 and N60 ~ (Sakmarian-Artinskian) orientations. These contemporaneous orthogonal dyke sets present orientations corresponding to the regional tectonic fabric and a tectonic origin for fracturing associated with the foreland basin development. The largest dykes record eight lithofacies and several stages of fracture opening. Stromatoids and centimetre-thick deposits of peloidal grainstones-packstones (thromboids), which form in situ within microbial laminae on the fissure walls, preceded filling by skeletal and terrigenous sediments. The fissures contain a well-preserved biota similar to the Lower Permian cavity-dwelling organisms observed within the reef. The coelobiontic habitat was episodically enlarged by successive synsedimentary fracturing episodes reflecting several phases of encrustation and infill of recurrent lithofacies. The pioneer microbialites grew when nothing else was deposited in the fissures immediately after fracturing. Gastropods and ostracods occur as dense clusters, probably indicating in situ growth, or are considered as reworked material from the subsurface. Others organisms, such as crinoids, bryozoans, Tubiphytes, conodonts, brachiopods, ammonoids and spiculate sponges, are also found within fractures as reworked shells and skeletons, indicating that they were washed in from the overlying sea floor. The biota in the dykes is mainly represented by the dweller guilds living on the reef surface, whereas members of the binding and framebuilding guilds are poorly represented. The dykes formed during reef development and the last stage of argillaceous wackestone sedimentation within the fractures is related to a major flooding event corresponding to upper Artinskian drowning of the Russian platform. Biostratigraphically significant fossils in the dykes date the ages of the host rock and of the fissure formation and filling from late Asselian to upper Artinskian. The Tratau reef thus provides an instructive example of the interaction between carbonate accumulation and tectonic events. Abstract:
Carbonate buildups are characteristic components of the Carboniferous-Permian strata cropping out in the southern Urals, or present in the adjacent subsurface. They proliferated in a transitional area between the eastern margin of the Russian platform and the preduralian foredeep basin (Fig. 1a, b). This platform, flanking an eastern subduction complex, is characterized by the common reef occurrence during late Carboniferous-early Permian time. Biostratigraphic data, mainly based on the study of fusulinids and ammonoids (RauserChernoussova 1951; Vissarionova 1975; Chuvashov et al. 1996; Lys & Moreau 1989 unpublished) and the location of reef belts, indicate a westward migration of the reef depocentres (Chuvashov 1983) away from the megasuture, with a maximum development during the Asselian-Sakmarian. The Sterlitamak area in the southern Urals (Russia) records the largest reefs. These appear
as isolated reefs in scattered localities across a north-south-trending area (Fig. lb). The overall facies and thickness change along linear trends (Rauser-Chernoussova 1951), and their spotty distribution, in conjunction with evidence of synsedimentary faulting, suggest that their origin and growth were dictated by synsedimentary tectonism that also controlled local bathymetry and hydrodynamic conditions (Vennin et al. 2002). Neptunian dykes provide a significant key to the regional history of vertical movement relative to sea level (tectonic and/or eustatic). Neptunian dykes are sedimentary dykes in which the filling sediments are derived from above (i.e. the surface), in contrast to some sedimentary dykes with fill injected from below (Winterer & Sarti 1994). Neptunian dykes exhibit a complex infilling including loose skeletal material, peloidal packstones-grainstones, mudstones and wackestones, microbialites, breccias and argillaceous wackestones. They provide a site
From:/~kLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climatic"and Evolutionary Controls. Geological Society, London, Special Publications, 275, 211-227. 0305-8719107l$15.00 9 The Geological Society of London.
212
E. VENNIN
Fig. 1. (a) Regional map depicting main geological characteristics of the Urals Mountains. (b) Map showing the westward migration of reefs in the Sterlitamak area (central Urals). (c) Schematic sedimentary architecture and sequence framework of the Tratau reef. It shows a major transgressive-regressive sequence. (d) Tratau reef near Sterlitamak. The differential weathering of Permian basin-fill that covered the reef reveals the depositional architecture and a palaeo-relief of 250 m; Scale = 100 m.
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX where organisms could be preserved away from the destructive taphonomic processes active on the sea floor and the removal of sediments by erosion. Neptunian dykes give information about platform growth, sediment sources and the biota that were present on the platform. Early Lower Permian fracture-dwelling organisms are recorded within the fissure in Permian reefs in the Sterlitamak area. These fractures offer a record of new-eroded surface sedimentation. Fractures were episodically enlarged by synsedimentary tectonic fracturing reflecting several phases of infill of recurrent facies, and intraclasts. The knowledge of the diagenetic processes undergone by the reef is essential for an understanding of: (i) the timing of mud lithification and the preservation of high angles in the flanks; (ii) the origin of distinct fabrics, particularly related to cavity systems and synsedimentary fissures; (iii) the absence of both compaction and plastic deformation processes; and (iv) the marine origin of calcite cements (according to their isotopic signature). In addition, geochemical data from void-filling cements and documentation of the burial history of the rocks provide details regarding their depositional environments. These reefs have received considerable attention because subsurface ones form important reservoirs in the Ishimbaevo fields. The purpose of this paper is to: (1) demonstrate the synsedimentary tectonic signature reflected in one of the most significant reefs of the Sterlitamak area; (2) analyse the stratigraphic and facies distribution, geometry and diagenetic features of this reef; (3) determine how, where and when neptunian dykes developed; and (4) illustrate the stepwise colonization displayed by Early Permian benthic organisms (mainly microbial communities and gastropods) in sedimentary fissures.
Geological setting Tratau Hill is located 10 km SE of Sterlitamak village and is one of the better exposed Upper Carboniferous-Lower Permian carbonate reefs cropping out in the Sterlitamak-Ishimbaevo area of the southern Urals. During this time Laurasia and the Pangean supercontinent were assembled and reefal accumulations are restricted to intracontinental land masses and neighbouring foredeeps. The Uralian foreland basin is an asymmetric structure formed by flexure of the European craton in response to tectonic loading by the Uralian Orogen (Zonenshain et al. 1984). The Tratau reef belongs to the sedimentary cover of the east European craton, and is located at the
213
margin between the western Russian platform and the eastern preduralian foreland basin. It crops out in the external tectonostratigraphic unit of the Urals, separated from the internal one by the main Uralian fault (Fig. l a). In the preduralian basin, Uralian orogenesis produced a thick flysch that prograded across the axis of the basin rapidly to the west. During the Carboniferous-Permian the Russian platform recorded several episodes of backstepping that resulted in a shrinking of the carbonate shelf area and in gradual shifting of the shelf margin westwards (Fig. lb) (Chuvashov 1983; Vennin et al. 2002). Following margin backstepping, reefs formed a major belt mainly owing to differential subsidence in the whole region. A regional study of reefs shows the continuity of reef formation through the Sakmarian and the end of the carbonate productivity in the Burservskian (early Artinskian). During Asselian times there was widespread development of isolated reefs (Vennin et al. 2002). The uppermost Asselian is a time of subaerial exposure evidenced throughout the East European platform (Aisenberg et al. 1983). Whereas only three stratigraphic units have been observed in the Tratau reef for the Lower Permian, five units are identified in the Shaktau reef located 15 km to the north (Vennin et al. 2002). The two missing units in the Tratau reef correspond, respectively, to the lower Sakmarian deposits composed mainly of Palaeoaplysina and coral biostromes embedded in bioclastic packstones-grainstones, and to the upper Sakmarian deposits with fusulinid grainstones and small Palaeoaplysina biostromes and bryozoan bafflestones. The top of the Sakmarian is characterized by a significant karst surface on the platform (Vermin et al. 2002; Zempolich et al. 2002). Finally, the platform margin where reefs developed was progressively drowned, culminating with a flooding of the platform and the deposition of upper Artinskian argillaceous wackestones (dated by Gerasimov 1940; Ruzhencev 1956; Chuvashov 1984) overlaying an important erosional surface of Early Artinskian age (Rauser-Chernoussova 1951). This erosion corresponds to the Burservskian erosion recognized by Chuvashov et al. (1996). The late Artinskian Cladoconus-argillaceous wackestones coincide with a widespread transgressive trend of eustatic origin (Sloss 1972; Ross & Ross 1990). Differences in regional sedimentation between reefal areas during late AsselianArtinskian are related to a breakdown of the platform in the area of the isolated palaeohighs where reefs were initiated (Korolyuk 1985; Vennin et al. 2002).
214
E. VENNIN
The Tratau reef is classified as a Duvansktype buildup (Nalivkin 1950). The exposed dimensions of the reef are 270 m in height and about 1 km in diameter (Fig. lc, d), but the real thickness of this buildup is about 400 m according to borehole data. In the southern Urals exposed reefs retain their palaeodepositional relationship (i.e. relief) with respect to the basin floor. The Tratau Hill shows a hemisphericconical morphology owing to the preservation of flanks inclined up to 40 ~ (Fig. lc, d). It overlies Gzhelian carbonate strata (Shamov 1984), and is covered by upper Artinskian argillaceous wackestones as seen on the northern part of the reef. Biostratigraphically significant fossils (conodonts, ammonoids, foraminifers) in the dykes date the age of the host rock and fissure formation and infill. Sediments infilling fractures are late Asselian-upper Artinskian (Chuvashov et al. 1996).
Reef complex anatomy Three distinct stratigraphic units can be differentiated in the Tratau reef (Fig. lc). The two first units are separated by a sharp facies boundary identified as a maximum flooding surface, just on the top of the basal unit. This is a thin and continuous layer, 20 cm thick, made up of wackestones and packstones rich in ammonoids, orthoceratids, bryozoans, brachiopods, bivalves, crinoids, gastropods, foraminifers and problematic algae. The third unit is separated from the second one by an important erosive surface (sequence boundary). (i) The basal unit (Sphaeroschwagerina moelleri-Schwagerina fecunda Zone; lower Schwagerina Horizon) is about 100 m thick. Its core consists of massive bryozoan baffiestones and coral and bryozoan wackestones. Complete bryozoans (Fenestellidae and encrusting forms), Tubiphytes and corals are preserved in life position, locally encrusted by Archaeolithoporella, and associated with crinoids, brachiopods, Diplosphaerina, foraminifers and ostracods. The flanks are composed of bedded bryozoan and coral floatstones. Bryozoans, corals, Tubiphytes, sponge spicules and ammonoids are the predominant bioclasts in these flanks; foraminifers, crinoids, brachiopods, Palaeoaplysina, bivalves and ostracods are disseminated in the matrix. (ii) The second unit (Sphaeroschwagerina sphaerica-Schwagerina firma Zone) also shows lateral facies changes from core to flanking areas, although the core is not well exposed. Equivalent
core facies are observed in the nearly Shaktau reef situated 15 km to the south where it consists of Tubiphytes and Archaeolithoporella-rich bindstones. The flanking beds, 50 m thick, show a mosaic of facies: (a) the NW flank is composed of a vertical succession of Tubiphytes and bryozoans bindstones and floatstones, intraclast, coral and crinoidal floatstones, bryozoan cementstones and common Archaeolithoporella bindstones at the top; (b) the southern flank is characterized by Tubiphytes floatstones alternating with bryozoan cementstones, while its uppermost part shows Archaeolithoporella bindstones; (c) the SE flank consists of Palaeoaplysina floatstones, Tubiphytes and Archaeolithoporella bindstones; and (d) the northern flank is composed of Archaeolithoporella bindstones overlain by Artinskian argillaceous wackestones and cherts. (iii) The third unit consists of Artinskian argillaceous wackestones onlapping Asselian and, locally, Sakmarian deposits. Argillaceous wackestones crop out on the northern flank of the reef and fill synsedimentary fractures. In summary, the Tratau reef is composed mainly of non-framework facies characterized by: (i) dominantly baffling organisms (bryozoans); (ii) the occurrence of binding and encrusting organisms ( Tubiphytes, Archaeolithoporella and foraminifers); and (iii) a diverse dwelling fauna including corals, palaeoaplysinids, gastropods, bivalves, etc. Binding processes by the microbial organisms and various encrusting organisms, in conjunction with sediment baffling by bryozoans, are considered to have played a sediment-stabilizing and binding role rather than a frame-building one. Whereas the core is envisaged as the site of carbonate production and entrapment, the bedded nature of the flank facies indicates local downslope transport and lateral accretion of sediment derived from the core. The intraclasts are composed of material identical to that of the core, relocated by storm- and/or gravity-induced transport. The high dipping angle (up to 40 ~) suggests that the flanks have somehow been stabilized during reef growth by cementation and encrustation.
Diagenetic processes A complex diagenetic history is observed in the Tratau reef and comprises seven successive episodes: (i) deposition of homogeneous and peloidal-clotted micrite; (ii) dissolution; (iii) precipitation of multiple phases of radial fibrous cements; (iv) precipitation of cements alternating
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX with peloidal matrix; (v) second phase of dissolution; (vi) precipitation of a second generation of radial fibrous cements; and (vii) infill of fractures. The occurrence of massive cementstones in reef-slope and platform-margin settings suggests that cements were a fairly common component in reef flanks. These Lower Permian reefs contain fibrous magnesium calcite cements (high Mg contents) low in Sr (Wahlman & Konovalova 2002; Zempolich et al. 2002). The carbonate mud displays two textures that are present at all stages of reef growth: (i) homogeneous micrite is present in the reef core strata, in cavities and fractures; (ii) peloids are spherical ('structure grumeleuse'), up to 50-100 gm in diameter, and present in the flanking strata, in cavities and fractures. In cavities and fractures, peloids are interstratificated with bands of radial fibrous calcite cements, forming planar (0.55 mm thick) to dome-shaped structures (2-5 mm high). Under cathodoluminescence, the micrite presents a homogeneous dull luminescence. Peloidal sediment is thought to represent microbially induced marine cementation that contributed significantly to the cohesiveness of many reefs. Micrites are a quantitatively important component of the Tratau reef core (up to 40%). The recognition of producing organisms is difficult but the presence of micritic laminae may be assigned to microbialites (Kirkland et al. 1998; Weidlich & Fagerstrom 1999; Weidlich 2002). The isolated position of the Tratau reef flanked by terrigenous sediment (argillaceous wackestones) prevented mud input from the platform, whereas the presence of clotted and peloidal micrites, and fine microbial laminae, argue for microbial activity (automicrite: Reitner et al. 1995). The micrite in phylloid algal mounds as well as in reefs is interpreted as the result of biological in situ production (Guo & Riding 1992; Kirkland et al. 1998). Microbes played a significant role in reef building (Riding, 2000, 2002) for sediment stabilization, sediment binding and precipitation. In the Tratau reef, carbonate mud seems to have resulted from the trapping of fine particles, in situ calcification of microbial cells, and disintegration of Archaeolithoporella and bioerosion of bioclasts. Early-late dissolution episodes correspond, respectively, to primary pore types and secondary ones. Early dissolution created mouldic and inter- and intraparticle pores, fenestrae and stromatactis-like systems. Late dissolution episodes enlarged primary voids, which acted as access routes for the aggressive fluids, and produced large voids, vugs and discontinuities in fissures and fractures.
215
Stromatactis cavities identical to the classical Belgian forms (Bourque & Boulvain 1993) occur in both core and fank facies of the first two units (Figs 2 & 3). The size of these cavities ranges from centimetres to decimetres, and their length/height ratio ranges from 1 to 2.5. The orientation of stromatactis varies from horizontal in the core (Fig. 3a) to up to 40 ~ inclined in flanking beds (Fig. 3b). Stromatactis cavities can be interconnected by synsedimentary fissures (up to 0.5 cm wide) filled by fibrous calcite cements (Fig. 3c). The resulting cavity networks exhibit a more complex filling by internal sediments and centripetally oriented radial fibrous calcite (of marine origin) and botryoidal calcite (Fig. 3d). The various attributes of stromatactis show a direct relationship between complex filling of cavities and type of enclosing beds, and size and shape of cavities related to enclosing beds (Fig. 2). The largest cavities display four types ofinfill (Fig. 2): (i) cements (botryoidal and radial fibrous calcite cements) frequently associated with microbialites and peloidal grainstones; (ii) micrite with bioclasts; (iii) microdolomite; and (iv) argillaceous wackestones. Contacts between upper walls and cavity-filling cement are invariably sharp. Stromatactis in the Tratau reef has not affected surrounding bioclasts, and radial fibrous calcite has not replaced matrix indicating that they were true cavities. These obervations suggest formation by fluid escape after sediment collapse. Cavities may have formed as well as a result of a lateral compression resulting from gravity acting on the reef slope, a phenomenon that induced opening fractures and cavities. Stromatactis parallel to the inclined bedding of reef flank is thought to have been formed, in part, due to gravitational processes, which have been triggered by seismic activity, given the instable shelf position of reefs (Lees 1964). Perhaps decay of soft-bodied multicellular organisms (spicular sponges) could have left cavities (e.g. Bourque & Boulvain 1993). Marine phreatic cements (Vennin 1997) comprise 50-80% of the rock fabric and are botryoidal and isopachous radial fibrous calcite in primary porosity, stromatactis cavities, solution-modified pore space and locally in the open fractures. Banding is recognized in radial fibrous cements, with five alternating dark and light layers, while eight luminescent bands are visible by cathodoluminescence. The alternating luminescent radial fibrous calcite and peloidal laminae are well expressed in stromactatis cavities. The marine origin of these cements is evidenced by stable isotope values for the radial fibrous calcite crystals that show a limited range
216
E. VENNIN Size
Host facies
Infilling
width (cm)
thick (cm)
1- 10
1-2
Bryozoan and
Tubiphytes cementstones
Tubiphytes, bryozoan and coral bafflestones Algal, bryozoan and Tubiphytes bindstones
1-100
1-50
1-50
1-15
Tubiphytes, bryozoan and crinoid bafflestones
1-100
1-100
Dip of bases
Position in reef flank
cements (radiaxial, fibrous calcite), micrite, and argillaceous wackestones cements (botryoidal and radial fibrous calcite) - micritedolomite, and argillaceous wackestones cements (botryoidal and radial fibrous calcite): dolomite, and argillaceous wackestones cements (rare botryoidal and radial fibrous calcite): dolomite, and argillaceous wackestones
core
20-45 ~
X
0-3~
0_35 ~
X
X
0-20~
X
X
Fig. 2. Table showing a classification of stromatactis cavities in the Tratau reef according to their respective shape, size, infilling, measurements of the dip of bases and setting.
of variation in 13C and 180 values (5180 from-0.5 to -2.1%o and 613C from 4.7 to 6.2%0). Isotopic and trace-element analyses in the Shaktau reef conducted by Zempolich et al. (2002) gave similar values. Radial fibrous calcite cements are non-luminescent and inclusion-rich with elevated Mg contents and low Mn and Fe contents. They are interpreted therefore to have precipitated from marine waters as primary high-Mg calcite. Early marine cementation and the binding action of encrusting microbes are the most likely factors that led to an early stabilization of the flank slopes, and early internal filling of fractures. The origin of extensive cementstones correlated with the biota associated is suggested by Grotzinger & Knoll (1995). This association may be related to the delivery of trophic resources to slope and platform-margin settings. There is a clear correlation between the abundance of suspension-feeding and nutrient-tolerant organisms and the occurrence of marine cements (Vennin et al. 2002; Zempolich et al. 2002). The biotic and abiotic constituents need substantial quantities of organic matter and HCO3- to sustain growth and precipitation. These components are common in waters upwelling along continental margins (Zempolich 1993). Upwelling processes have
been proposed in the Uralian foredeep margin and related to the circulation of polar currents southwards along the margin of the Russian platform, and to progressive infilling of the Urals with clastics and resedimented carbonate during the early Permian (Chuvashov 1983).
Evidence for syndepositional tectonic processes Synsedimentary fissures and fractures, seen as neptunian dykes, are distinctive sedimentary structures of the Tratau reef. They are related to synsedimentary tectonism rather than gravitational phenomena on the reef flank.
Dyke sediments and biota Sediment-filled fractures are common (Figs 4 & 5). The fractures sizes range from a few centimetres up to 10 m in width, and up to 500 m in length. Most are subvertical to steeply inclined. The fracture opening post-dates reef formation, and thus has the latest Asselian as maximum age. Stromatactis cavity formation must have a pre-late Asselian age because they are cut by the complex synsedimentary fractures.
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX
217
Fig. 3. (a) Stromatactis characterized by a complex filling composed of micritic, peloidal and clotted matrix and radial fibrous calcite cements (hammer for scale). (b) Horizontal stromatactis-like cavity oblique, respectively, to bedding and characterized by a complex filling: micritic matrix and botryoid developed on the cavity roof. (e) Stromatactis parallel to bedding (white dotted line) and fracture network (white arrow) connecting stromatactis cavities filled by radial fibrous calcite cements. (d) Large botryoid (rfc, radial fibrous cement) alternating with microbial laminae (st, up to 5 cm thick) covered by peloidal packstone (tr) with millimetre-thick microbial layers; scale = 250 lam.
Correctness of the age determinations depends on whether the fossils lived contemporaneously with filling or had been reworked from older sediments. Preservation of diagnostic fossils as individual specimens not micritized and without taphonomic and diagenetic transformations indicates that reworking was insignificant, as opposed to specimens found in lithoclasts. Two successive assemblages of dykes can be identified. There orientations are: (i) N80 ~ and N170 ~ (late Asselian-early Sakmarian; Figs 5 & 6); and (ii)N130~ ~ and N60 ~ (SakmarianArtinskian; Figs 5 and 7). Their infills exhibit a fairly uniform composition of eight lithofacies. (i) A multiple phase of sediment layering, perpendicular or inclined to the fracture walls, dominates in the earlier assemblage (Fig. 5c, d). At least four distinct neptunian lithofacies separated by erosion surfaces are recognized in the more complete fractures, from inner to outer. The vertical lithofacies succession is represented
by: (a) oolithic, peloidal and lumpy (irregular aggregates of carbonate mud) grainstonespackstones (lithofacies 4a; Fig. 6c). Laminae are variably oblique to the fissure walls and show soft-sediment folding and convolution related to water escape and liquefaction; (b) intraclastic breccias (lithofacies 7) composed of mudstone. They are assigned to early lithification fracture of muddy sediment (in agreement with syndepositional seismic activity); (c) alternating wackestones and packstones rich in ostracods and gastropods (lithofacies 4b; Fig. 6a); (d) fenestrae mudstones (lithofacies 4c; Fig. 6b). Fenestrae are parallel to the inclined stratification. Silty dolomite fills the late porosity in fenestrae below the erosive surface observed at the top of these mud-rich deposits (lithofacies 4c). Its stratification within the fenestrae is concordant with the horizontal, respectively inclined to the stratification plane; (e) the erosion surface is covered by a bioclastic packstone composed mainly of crinoids, brachiopods, conodonts and rare
218
E. V E N N I N
,b'n~
-~.~:
~
~.~. 9~.~
~
~
0 "~
-~ N ~ ~,.~ O.N
i
.o
.~
~
~
Xeq .~ ~
o
m
.~=
02
~
=E
i ~
.
~.~
~ 0
~..~o~
.~ 9
r
cJ 0
o 0
8
~
~'~
-
._.9. , ~
2 ~
9~ ~
0
.
=o
N'~
0
r~
-~ ,--M ~ - ~
'~
;> . ~ " >,--
r
o ,.o ~
~.~.~ "~
oo~
=o '~
9 ~
o~ ~
_~u
""d ~"
.
~b "0
-d
9~ . ~
~ ~ ~',~
~.~
o
9
I:~
0"~
0
0
0
c~ ;>
..= o
9. ~
~
"0
0 "~
00",~
2:;o
~bz o~
c~
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX
219
Fig. 5. (a) Representation of fractures cross-cutting the Tratau reef complex mainly observed in the well-exposed south flank of the reef. (b) Tectonically induced evolution of fractures: two-stage timing of dyke formation and stages of filling based on biostratigraphic data. (e) Schematic drawing illustrating a fracture opening and filling (stage 1, N80~ so, stratification). Lithofacies 4a - ooidal-peloidal packstones; lithofacies 4b - wackestonespackstones rich in gastropods; lithofacies 4c - micritic and peloidal wackestones with fenestral fabric; lithofacies 5 - packstones rich in echinoderms, underlying an erosion surface. (d) Simplified diagram showing Schmidt stereographic projections of planes of fissure walls. Two separate fracture assemblages have N80 ~ and N 130~ 140~ orientations.
220
E. VENNIN
Fig. 6. (a) Neptunian dyke infilled with different lithofacies; ferrigeneous (fe), phosphatic crusts (lithofacies 1); stromatoid developed on dyke wall; gastropod-rich wackestone-packstone (lithofacies 4b). (b) Neptunian dyke (N80 ~ with complex infill: stromatoid encrusting the fracture wall (st); mudstone with fenestrae filled by fine dolomite (lithofacies 4c); erosion of the mudstone by crinoid-rich packstone (c), ammonoids (a) and other bioclasts (lithofacies 5). (c) Grainstone-packstone with ooids (oo) and lumps (lu); scale = 500 pm. (d) Neptunian dyke (fracture N140 ~ infilled with crinoid-rich packstone (lithofacies 5) and stromatoid (st) lining on the fracture wall; scale = 10 cm. (e) Dome-shaped microbial laminae (st, stromatoid) with depressions filled by radial fibrous cements (rfc); scale = 250 pm. (f) Neptunian dyke (N80 ~ infilled with encrinitic packstone (lithofacies 5); wall is encrusted by stromatoid crusts (st). (g) Radial fibrous cement layers (rfc) separated by thin dark microbial laminae (m); scale = 250 pin, top to the right. (h) Alternating grainstone layers with fibrous cements and peloidal packstones (tr); micritic dark layers seem microbial laminae (m, lithofacies 3); scale = 500 pm.
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX
221
Fig. 7. (a) Three-dimensional diagram of fracture infill (fracture with N 140~ orientation) with opening and three dated episodes of synsedimentary filling: level A, crinoid-rich packstone: lower Tastubian (Sakmarian); level B, crinoid-rich packstone: upper Tastubian (Sakmarian); level C, bioclastic packstone: lower Sterlitamakian (Sakmarian). (b) Transect illustrating the main lithofacies distribution. (c) Stromatoid composed of dome-shaped microbial crusts (Archeolithoporella-like); scale = 500 pin. (d) Complex filling showing alternating bioclastic facies (lithofacies 6a), microbial laminae and peloids; scale = 500 gm. (e) Layers of peloids regular and homogeneous in size alternating with microbial laminae; scale = 500 gm. (f) Fine-grained argillaceous wackestones, laminae are perpendicular to the wall of dyke (lithofacies 8); scale = 500 gm. (g) Intraclastic sediments composed of angular lithoclasts embedded in a micritic-peloidal matrix (lithofacies 7); scale = 500 gm. (h) Crinoid-rich packstones (lithofacies 5); scale = 1 mm. (i) Bioclastic packstones with gastropods and fragmented bivalves and ostracods (lithofacies 6a); scale = 1 mm. (j) Contact between flanking bed and fracture wall, which is encrusted by microbial laminae.
222
E. VENNIN
ammonoids (lithofacies 5; Fig. 6b). Lithofacies 5 is dated as lower Sakmarian by conodonts (Chuvashov et al. 1996). A first dyke episode took place shortly after the formation of the Tratau reef. This is evidenced by the fossil record of the dykes, which is different from the reef fabrics. The absence of taxa diagnostic of the reef growth and the relative abundance of gastropods and ostracods indicate reworking from an inner, protected reef setting or an autochthonous cryptic habitat in the fracture. Representatives of the dweller guild are the major constituents of the first fracture type. Gastropods and ostracods occur locally as dense clusters and are arranged together showing probably an in situ habitat with the organisms grazing on microbial matter. Members of the constructor and binder guilds are poorly represented because they were bound to the reef surface through encrustation and early cementation. Oolithic and lump layers (Fig. 6c) are not preserved in the sedimentary record of the reef and only occur in fractures. Therefore, these late Asselian relic sediments offer a good opportunity to complete the sedimentary succession, which is not preserved in the reef itself. The association of lithofacies 4c with emersive features (fenestrae, dissolution and dolomitization) at the top of the mud-dominated sedimentation permits the correlation of this episode with the late Asselian subaerial exposure evidenced throughout the East European platform by Aisenberg et al. (1983) and give information of the importance of the lowering of the relative sea level. Ages are based on the basis of the presence of ammonoids and conodonts (Mesogondolella lacerta and Mesogondolella parafoliosa) in the lithofacies 5 overlying the erosion surface observed in the greatest N800 dyke (Fig. 6b). (ii) A multiple phase of layers parallel or inclined at low angles to the fracture walls appears in the second assemblage of fissures (N130*-N140 ~ and N60~ Figs 6d & 7). This second type of tectonic-induced opening affects the previous fracturation and Asselian-Sakmarian host rocks. At least eight distinct neptunian lithofacies are recognized in some fractures. Some of them are recurring by repeated tectonically re-opening favouring development of vertical and steeply inclined walls. They present different ages from late Asselian to Artinskian (Fig.7). The arrangement of the lamination of the sediment-fill varies widely, from parallelism to obliquity and directly depends on the nature of the fill. Three phases of fracturing and infill may occur in a single fracture, indicating several stages in opening.
1.
2.
3.
The opening of fissures began on a smallscale with microfracturing of a lithified host rock. Walls were lined with thin ferruginous and/or phosphatic crusts, or radial fibrous calcite cements (lithofacies 1). Cracks were lined with alternating domeshaped to microbial laminae and fibrous calcite cements (lithofacies 2, Figs 6e & 7c; lithofacies 3, Figs 6h & 7d, e). Fissures opened into large, metre-scale dimensions, which were partially or completely filled with a late complex sediment succession (Fig. 7), composed of peloidal packstones (lithofacies 3), bioclastic packstones (lithofacies 5, Fig. 7h; lithofacies 6a and 6b, Fig. 7i), breccias (lithofacies 7, Fig. 7g) and, finally, argillaceous wackestones and claystones (lithofacies 8, Fig. 7f).
The multiple generations of sediments and the intercalated cement layers show that filling was episodic, with pulses of sediment injection and quiescent periods for cementation and microbial growth. Several fractures show an infill composed of polymictic breccias, included within bioclastic packstones-wackestones. Intraclasts, up to 50 cm in size, exhibit angular shapes and consist of internal sediments, microbialites and, sometimes, flanking bed fragments. The filling evidences episodes of tectonic instability by successive collapses of the fracture walls. The marine conditions of fissure infilling is indicated by the presence of: (i) foraminifers, gastropods, ostracods, conodonts and echinoderms in the finegrained matrix; (ii) radial fibrous calcite cements between the clasts; and (iii) stable isotope values (6180 from -1 to -2%o and 613C from 5 to 6%o) of radial fibrous calcite that show isotopic values similar to those recorded in subsurface and stromatactis. The second fracture assemblage (N130 ~ N140 ~ probably formed during the lower Sakmarian based on their transections through the first fracture assemblage and the presence of Mesogondolella lacerta and Mesogondolella parafoliosa (level A, lower Tastubian, Sakmarian) in lithofacies 5, Mesogondolella striata and Hindeodus sp. (level B, upper Tastubian, Sakmarian) in the same lithofacies, and Mesogondolella lata, Mesogondolella bisseli and Hindeodus sp. in lithofacies 6a (bioclastic packstones, level C, lower Sterlitamakian, Sakmarian). The final infilling lithofacies 8 corresponds to the Artinskian indicated by the presence of ammonoids, foraminifers and orthoceratids (Rauser-Chernoussova 1951; Chuvashov et al. 1996).
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX Both fracture assemblages are similar in that: (i) thin ferrugineous, phosphatic crusts, domeshaped microbialites and/or laminar layer of microbialites (lithofacies 2-3) directly line and encrust the wall of dykes. This indicates that the fractures remained for some time open before they were filled by external skeletal debris; (ii) a source from the surface is proposed for part of the filled sediment; and (iii) fissures disappear upwards and are overlain by upper Artinskian argillaceous wackestones. They differ in that: (i) fractures exhibit different orientations; (ii) in the first fracture assemblage, lithoclasts of wall rock and cryptobiontic organisms evidence a passive sediment-fill by gravitational-induced processes; this contrasts with the injection processes from the second fracture type. The evidence for fracture-fill cementation and rebrecciation indicates that the process of fracture develop during several phases; (iii) the age of fracturing and filling is upper Asselianlower Sakmarian for the first assemblage and upper Asselian-upper Artinskian for the second fracture assemblage. Lithofacies 5 (crinoidal packstones) is the last step of infilling for N80 ~ fractures, and argillaceous wackestones and claystones are the last sediments for the N130 ~ N140 ~ fractures. Microbial texture
Fractures were cryptic habitats similar to those in the skeletal framework on the subsurface of the Permian reef of New Mexico (Wood et al. 1996). As in recent reef niches, marine microbialites are common structures in cavities and fractures (Reitner et al. 2000) and contribute to fracture filling as fixing and binding organisms. Microbial structures in the fractures are represented by stromatoid laminations, muddy matrix, clottedpeloidal fabrics (thromboid structures) and micritic laminae. Two general types of microbial fabrics are commonly recognized. (i) Up to 10 cm-thick stromatoid laminae appear planardomical in shape, passing into clotted-like textures. Interdomal areas are filled by radial fibrous calcitic cements (Fig. 6e-g). They directly encrust fracture walls and synsedimentary re-opened fracture walls and act as pioneer organisms. They consist of dark micritic laminae. Organisms living in the fissures would be expected to be members of the Tratau framework communities as the problematic Archaeolithoporella and are observed in dykes and other cryptic sites of the Permian Capitan reef in Texas (Babcock 1977; Stanton & Pray 2004). Archaeolithoporella has been considered as alga (Mazzulo & Cys 1978; Wang et al. 1994; Kirkland et al.
223
1998), but Grotzinger & Knoll (1995) favour instead a microbialitic (Shen & Xu 2005) or organic-cement origin. It was apparently able to grow in very low light conditions as it is quite common in cryptic settings (Babcock 1979). In the Tratau reef Archaeolithoporella is frequently observed in association with stromatactislike cavities; microbialitic structures in some fractures seem to be in good agreement with Archaeolithoporella structures (Sarfati pers. comm). Archaeolithoporella encrusted radial fibrous cements or distinctly grew between cements layers, suggesting that the cement was of syndepositional marine origin. Microbialites are repeatedly interlayered with radial fibrous calcite (Fig. 6g). Following Wahlman (1988), the relation between stromatolitic laminations and cements suggests that microbes might have played a role in the precipitation of 'botryoidal' macrocements but might also simply represent periodic cessations of 'botryoidal' cement growth. Occurrence of several generations of stromatolites in a single fracture points to various episodes of opening (Fig. 7). A sharp contact separates the stromatolites from the overlying laminated peloidal packstones and the thrombolitic facies (Figs 6f & 7b). The peloids in contact with stromatolitic layers present a mesoscopic clotted fabric. (ii) Thromboids correspond to unlaminated peloidal grainstonespackstones (Fig. 6h). They are frequently associated with millimetre-thick layers of fibrous cements and microbial laminae, which appear as amalgamated thin-layered micritic laminae and form a consortium with thromboids. They form flat to slightly undulating laminae. The thrombolitic facies is made of peloidal, clotted and homogeneous micrite (Fig. 6h). Peloids are well sorted and rounded, generally up to 100 gm in diameter and are composed of dark coloured micrite embedded in a groundmass of calcite microspar. Peloidal matrix grades into clotted and/or homogenous micrite. Wahlman (1988) proposed microbial activity as an origin for peloidal fabrics. Kirkland et al. (1988) proposed an in situ formation for the peloids within a mucilageneous organic film. Such 'peloidal sediments' have been described by Chafetz (1986) in modern carbonates as a product of microbially induced precipitation.
Discussion on regional tectonostratigraphic evolution and origin of fracturation The north-south-trending area of transition between the eastern margin of the Russian platform and the preduralian basin is an excellent
224
E. VENNIN
area to study interacting structural and sedimentary processes. Orogenic collision was oblique, moving wave-like from south to north along the orogen (Puchkov 1997). Transverse NW-SE structural elements accommodated this oblique collision and subdivided the foreland into several semi-isolated basins. Tectonic processes in foreland basins have a clear 'first-order' control on the thickness of carbonate platform deposits and on its gross facies distribution (Jordan et al. 1988). The model of Palaeozoic collision in the Urals was complicated by the uneven outline of the passive margin, with a promontory in the middle Urals called the Ufimian amphitheatre (Puchkov 1997). In the Carboniferous, the southern Urals were characterized by an atypical collision and the formation of local tensional structures. During the Lower Permian a rigid collision followed, recording changes in intensity along the length of the orogen until the Late Permian (Puchkov 1997). The origin and growth of reefs and associated fracturation were partly governed by synsedimentary tectonism, which is evidenced by: (i) the spotty distribution of reefs, the presence of regional north-south-trending faults, and the sharp variations in facies and thicknesses on flanks (Rauser-Chernoussova 1951); (ii) the widespread development of fault-bound depressions on the top of several reefs; (iii) the presence of synsedimentary fractures and neptunian dykes within reefs (Vennin 1997; Vennin et al. 2002). Four geodynamic phases in the development of Tratau and Shaktau reefs are proposed (Vennin et al. 2002): (i) reef growth began on tectonically induced palaeohighs during the Gzhelian; (ii) large isolated pinnacles are affected by fractures in the Asselian, as indicated by the age of the infilled sediments. After late Asselian karstification and erosion, the reef growth continued through the Sakmarian and early Artinskian ending by uplift and erosion; (iii) a tilting of the platform favoured the uplift of the entire Sterlitamak area during the early Artinskian (Vennin et al. 2002) and, at a local scale, explains the differential subsidence recorded by the Tratau and Shaktau reefs; (iv) finally, the platform and reefs were drowned during the late Artinskian because of siliciclastic influx related to thrusting and closure of the palaeo-Ural Ocean during the Artinskian (Zempolich et al.
2002). Explanations for neptunian dyke formation are manifold. Some yield important information concerning the dyke formation in the Urals. Neptunian dykes crossing out in the Capitan reef carbonate margin are interpreted to have formed
in response to differential compaction induced by the reef-loading of foreslope sediments (Yurewicz 1977; Jagnow 1979; Melim & Scholle 1991). Stanton & Pray (2004) proposed gravity settling and downwarping of lithified platform carbonate to explain neptunian dykes aligned parallel to the platform margin of the Capitan reef. A high local depositional relief and sediment instability can result in passive gravitational movement and fracturing of the rock. However, Hunt et al. (2002) argued that these fractures and faults were preferentially localized within the damage zones of syndepositional faults and not exactly related to the syndepositional down-tothe-basin tilting of the platform as proposed by Saller (1996). This latter process is a response to compaction-induced differential subsidence. The origin of neptunian dykes can also be related to tectonic mechanisms that cause fracturing or initiate gravitational movement (Cozzi 2000). Although the Lower Jurassic Hierlatzkalk dykes from Hungary are similar to those of the Capitan reef, their origin is attributed to normal faulting of the platform rather than to sedimentationinduced subsidence (Galacz 1998). Neogene intramontane basins of SE Spain have a complex structural history. Kinematic rearrangements during the Late Tortonian resulted in a significant tilting of blocks and formation of neptunian dykes (Brachert et al. 2001). Tectonic and seismicity at the southern Minervois margin (France) resulted in fault growth and formation of neptunian dykes in Emsian mound facies. The dykes formed during or immediately after mound accretion and simultaneously with cementation of the stromatactoid cavity networks (Bourrouilh et al. 1998). Fractures with synsedimentary filling occur also in the Late Frasnian 'Les Wayons' buildups of Belgium. Vertical extension-related fractures affected the mound facies, truncating and circumventing the fossils (Playford 1984; Boulvain 2001). In the Tratau reef mound, small fissures connecting cavities and voids are attributed to passive gravitational movements but are only filled by fibrous radial cements. These fissures are systematically cut by the neptunian dykes described above. Explanations for tectonically induced origin of dykes invoke: (i) contemporaneous orthogonal dyke sets with an orientation corresponding to the regional tectonic fabric N90 ~ and N80 ~ (Rauser-Chernoussova 1951). Local tectonic events have been dated and integrated into a regional extensional tectonic episode, within a foredeep geodynamic model. The eastern margin of the Russian platform appears to have been governed from the Asselian
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX to the Artinskian by syndepositional movements on north-south-trending fault lines, parallel to the shelf edge. The formation of the fractures can be ascribed to differential movement between fracture-bounded blocks of host rocks resulting in dilation, or complete removal of blocks, presumably by gravitational mass movements. It is important to call attention to the paradox of contemporaneous normal faulting, subparallel to the strike of the reefbelt, and general compressional features in the eastern foreland. Distensional processes are recorded by the breakdown of the Lower Permian platform, and the common synsedimentary fractures and collapse structures recorded in the Tratau reef and in neighbouring reefal area (Shaktau). Winslow (1983) argued that dykes are restricted to the toe areas of thrust hanging walls. The same phenomena are also seen in the Alpine foldbelt (Laubscher 1978) and the Ouachita-Marathon forebelt (Hopkins 1968); (ii) the neptunian dykes cut the complete reef. They originated after lithification. This prevented a strictly passive gravitational-induced process affecting the reef flanks; and (iii) age of the neptunian sediment and their organization within the fractures are not consistent with a progressive opening of fractures towards the margin of the reef.
Conclusions 9 Neptunian dykes filled with skeletal material are well developed and an abundant feature of the Tratau reef. As a result of a detailed study of the fractures affecting the Tratau reef, the following conclusions are drawn. 9 Neptunian dykes provide insights concerning the reef biota and reef-growth episodes that are poorly preserved in subsurface. The dykes record fracture formation at the platform margin but represent orthogonal dyke sets with orientations corresponding to the regional tectonic fabric. 9 Synsedimentary fissures and fractures, seen as neptunian dykes, are related to synsedimentary tectonism rather than gravitational phenomena on the reef flank. The formation of the neptunian dykes is related to the lower Asselian break-up (distensional period) of the reef connected to the orogenic evolution of the eastern foreland basin. 9 Biostratigraphically diagnostic fossils in the dykes indicate that the fractures were formed from late Asselian to Artinskian time. The biostratigraphic analysis of the neptunian sediments indicates that there was a marked erosional gap (hiatus), spanning the
225
Sakmarian-Early Artinskian on the Tratau reef surface. Argillaceous wackestones and claystones are the latest neptunian sediments. They are correlated with the late Artinskian drowning of the platform and the closure of the palaeo-Ural Ocean during the Artinskian. 9 The fractures act as sediment traps for loose skeletal materials from the reef surface and act as niches for autochthonous biota (Archaeolithoporella, microbialites, rare ostracods and gastropods). Internal bioerosion is not observed on the walls, and bioclasts and endolitic biota seem to be absent. The framework reef communities are poorly preserved in the dykes because they were bound to the reef surface and thus were not a sediment source. Therefore, the dykes are filled by members of the vagile dweller and sessile binder (Archaeolithoporella) guilds. Fracture habitat was similar to the cryptic habitat within skeletal reef frameworks (as stromatactis-like cavities). 9 The submarine origin of the filled fractures is evidenced by the consistent occurrence of marine sediments and marine calcite cements. The author is grateful to H.-G. Herbig, B. Pratt and M. Aretz for helpful criticisms and suggestions, which have greatly improved this paper. I wish to express my thanks to all the scientists, Russian staff and Total Company who supported this research. This paper was improved by suggestions and comments made by B.I. Chuvashov (Institute of Geology and Geochemistry, Russia), B. Beauchamp (University of Calgary, Canada) and T. Boisseau (Total, France). D. Vachard (University of Lille, France) also contributed to improve the manuscript by his foraminifer determinations and M. Guiraud (University of Bourgogne, France) for his comments on tectonics.
References AISENBERG, D. Y., POLETAYEV,V. I., MAKHLINA,M. K. & SHCHERBAKOV,0. A. 1983. Carboniferous sea level changes on the East-European platform. X Congres International Stratigraphy Geology Carboniferous, 4, 65-71. BABCOCK, J. A. 1977. Calcareous algae, organic boundstones, and the genesis of the upper Capitan Limestones (Permian, Guadalupian), Guadalupe Mountains, West Texas and New Mexico. In: HILEMAN, M. E. & MAZULLO, S. J. (eds) Upper Guadalupian Facies, Permian Reef Complex, Crudalupe Mountains, New Mexico and West Texas. Permian Basin Section, SEPM Field Conference Guidebook, Midland, Texas, 77-16, 3-44. BABCOCK, J. A. 1979. Calcareous algae and algal problematica of the Capitan reef (Permian), Guadalupe Mountains, West Texas and New Mexico, USA. Bulletin des Centres de Recherches Exploration -Production Elf-Aquitaine, 3, 419-428.
226
E. VENNIN
BOULVAIN, F. 2001. Facies architecture and diagenesis of Belgian Late Frasnian carbonate mounds. Sedimentary Geology, 145, 269-294. BOURQUE, P. A. & BOULVAIN,F. 1993. A model for the origin and petrogenesis of the red stromatactis limestone of Paleozoic carbonate mounds. Journal of Sedimentary Petrology, 63, 607-619. BOURROUILH, R., BOURQUE, P. A., DANSEREAU, P., BOURROUILH-LE JAN, F. & WEYANT, P. 1998. Synsedimentary tectonics, mud-mounds and sealevel changes on a Paleozoic carbonate platform margin: a Devonian Montagne Noire example (France). Sedimentary Geology, 118, 95-118. BRACHERT, T. C., HULTZSCH, N., KNOERICH, A. C., KRAUTWURST, U. M. R. & STIOCKRAD, O. 2001. Climatic signatures in shallow-water carbonates: high-resolution stratigraphic markers in structurally controlled carbonate buildups (Late Miocene, southern Spain). Palaeogeography, Palaeoclimatology, Palaeoecology, 175, 211-237. CHAFETZ, H. S. 1986. Marine peloids: a product of bacterially induced precipitation of calcite. Journal of Sedimentary Petrology, 56, 812-817. CHUVASHOV, B. I. 1983. Permian reefs of the Urals. Facies, 8, 191-212. CHUVASHOV,B. I. 1984. On Joint Line Between Russian
Platform and Preduralyan Foredeep Metamorphism and Tectonic of Uralian Western Zones. Urals Scientific Centre of Academy of Sciences USSR Sverdlow'k, 111-120. CHUVASHOV, B., PROUST, J. N., BOISSEAU, Z. & VENNIN, E. 1996. On History of Development of
Sterlitamakian Shikhans (Early Permian Reefs of Southern Preduralye). Proceedings of the Institute of Geology and Geochemistry, 28, 25-32 [in Russian]. CozzI, A. 2000. Synsedimentary tensional features in Upper Triassic shallow-water platform carbonates of the Carnian Prealps (northern Italy) and their importance as paleostress indicators. Basin Research, 12, 133-146. GALACZ, A. 1998. Tectonically controlled sedimentation in the Jurassic of the Bakony Mountains (Transdanubian Central Range, Hungary). Acta Geologica Hungarica, 31, 313-328. GERASIMOV,N. P. 1940. Geology of the Eastern of Bear-
ing Region (West Slope of the Urals and Western Preduralyde). Izdatelstvo, Academy of Sciences USSR, 1-140. GROTZINGER, J. P. & KNOLL, A. H. 1995. Anomalous carbonate precipitates. Is the Precambrian the key to the Permian? Palaios, 10, 578-596. Guo, L. & RIDING, R., 1992. Aragonite laminae in hot water travertine crusts, Rapolano Terme, Italy. Sedimentology, 39, 1067-1079. HOPKINS, H. R. 1968. Structural interpretation of the Ouachita Mountains. In: CLINE, L. M. (ed.) A Guide
Book to the Geology of Western Arkoma Basin and Ouachita Mountains. Oklahoma City Geological Society, 104-108. HUNT, D. W., FITCHEN, W. M. & KOSA, E. 2002. Syndepositional formation of the Permian Capitan reef carbonate platform, Guadalupe Mountains, New Mexico, U.S.A. Sedimentary Geology, 154, 89-126.
JAGNOW, D. H. 1979. Cavern Development in the Guadalupe Mountains. Cave Research Foundation, Columbus, OH. JORDAN, J. E., FLEMINGS, P. B. & BEER, J. A. 1988. Dating thrust activity by use of foreland basin strata. In: PAOLA, C. & KLEINSPEHN, K. L. (eds) New Perspectives in Basin Analysis'. Springer, New York, 307-330. KIRKLAND, B. L., DICKSON, J. A. D., WOOD, R. A. & LAND, L. S. 1998. Microbialites and microstratigraphy: the origin of encrustations in the middle and upper Capitan Formation, Guadalupe Mountains, Texas and New Mexico, U.S.A. Journal of Sedimentary Research, 68, 956-969. KOROLYUK, I. K. 1985. Methods and Results of the
Study of the Permian 'Mt Shaktau' reef (Bashkirian section of the Western Urals). Moscow, Nauka. [in Russian]. LAUBSCHER, H. 1978. Foreland folding. Tectonophysics, 47, 325-337. LEES, A. 1964. The structure and origin of Waulsortian (Lower Carboniferous) reef of the West-central Eire. Philosophical Transactions of the Royal Society of London, 740, 485-531. MAZZULO, S. J. & CYS, J. M. 1978. Archaeolithoporellaboundstones and marine aragonite cements, Permian Capitan reef, New Mexico and Texas, USA. Neues Jahrbuch fiir Geologie und Paliiontologie, Monatshefte, 10, 600-611. MELIM, L. A. & SCHOLLE, P. A. 1991. The fore reef facies of the Permian Capitan Formation: the role of sediment supply versus sea-level changes. Journal of Sedimentary Research, 65, 107-118. NALIVKIN, V. D. 1950. Fazii i geologichekaia istoria ufimskogo Plato i Juresano-Sylvenskoi. Gostoptechizdat, 1-127 [In Russian]. PLAYFORD, P. E. 1984. Platform-margin and marginal slope relationships in Devonian reef complexes of the Canning Basin. Proceeding Geological Society
Australia and Petroleum Exploration Society Australia Symposium Perth, Australia, 190-214. PUCHKOV, V. N. 1997. Structure and geodynamics of the Uralian orogen. In: BURG, J. P. & FORD, M. (eds) Orogen through time. Geological Society. Special Publication, 121, 201-236. RAUSER-CHERNOUSSOVA,D. M. 1951. Facies of Upper Carboniferous and Artinskian deposits in the Sterlitamak-Ishimbaevo region of the pre-Urals, based on a study of fusulinids. Trudy Akademica Nauk SSSR, Series Geology, 43, 1-108 [in Russian]. REITNER, J., NEUWEILER,F. Er A/~. 1995. Mud-mounds a polygenetic spectrum of fine-grained carbonate buildups. Facies, 32, 1-69. REITNER, J., THIEL, V., ZANKL, H., MICHAELIS, W., WORHEIDE, G. & GAUTRET, P. 2000. Organic and biogeochemical patterns in cryptic microbialites. In: RIDING, R. & AWRAMIK, S. M. (eds) Microbial Sediments. Spinger, Berlin, 149-159. RIDING, R. 2000. Microbial carbonates: the geological record of calcified bacterial-algal mats and biofilms. Sedimentology, 47, 179-214. RIDING, R. 2002. Structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth-Science Reviews, 58, 163-231.
NEPTUNIAN DYKES IN PERMIAN REEF COMPLEX Ross, C. A. & Ross, J. R. P. 1990. Revised world maps. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Paleozoic Paleogeography and Biogeography. Geological Society, London, Memoirs, 12, 1-21. RUZHENCEV, V. E. 1956. Early Permian Ammonoids of the Southern Urals. Ammonoids of Artinskian Stage. Izdatelstvo, Academy of Sciences USSR, 1-275 [in Russian]. SALLER, A. H. 1996. Differential compaction and basinward tilting of the prograding Capitan reef complex, Permian, west Texas and southeast New Mexico, USA. Sedimentary Geology, 100, 1-10. SHAMOV, D. F. 1984. Faci6s des d6p6ts sakmariensartinskiens d'Ishimbaevskoe. Travaux de rInstitut des RecherchespOtrolikres d'Oufa, 2, 3-77. SHEN, J. W. & Xtr, H. L. 2005. Microbial carbonates as contributors to Upper Permian (GuadalupianLopingian) biostromes and reefs in carbonate platform margin setting, Ziyun County, South China. Palaeogeography, Palaeoclimatology, Palaeoecology, 218, 217-238. SLOSS, L. L. 1972. Synchrony of Phanerozoic sedimentary-tectonic events of the north American Craton and the Russian Platform. 24th International Geological Congress, 6, 24-32. STANTON,R. J. & PRAY, L. C. 2004. Skeletal-carbonate neptunian dykes of the Permian Capitan reef, Texas, U.S.A. Journal of Sedimentary Research, 74, 805-816. VENNIN, E. 1997. Architecture skdimentaire des bioconstructions Permo-CarboniJOres de l'Oural mkridional (Russie). Publications de la Soci6t6 g6ologique du Nord, 26. VENNIN, E., BO1SSEAU, T., PROUST, J. N. & CHUVASHOV, B. I. 2002. Influence of sea level change on reef architecture in Early Permian buildup complexes, southern Urals, Russia. In: ZEMPOLtCH, W. G. & COOK, H. E. (eds) Carbonate Reservoir and Carbonate Field Analogs. Society of Paleontologists and Mineralogists, Special Publications, 74, 205-218. VISSARIONOVA,A. Y. A. 1975. Field Excursion Guidebook for the Carboniferous Sections 03" the Urals (Bashkiria). Izdatelstvo Nauka, Moskva. WAHLMAN, G. P. 1988. Subsurface Wolfcampian (Lower Permian) shelf margin reefs in the Permian basin of west Texas and southeastern New Mexico. In: MORGAN,W. A. & BABCOCK,J. A. (eds) Permian Rocks of the Midcontinent. Society of Paleontologists and Mineralogists, Midcontinent Section, Special Publications, 1, 177-204. WAHLMAN, G. P. & KONOVALOVA,M. V. 2002. Upper Carboniferous-Lower Permian Kozhim carbonate bank, Subpolar pre-Ural Mountains, northern Russia. In: ZEMPOLICH,W. G. & COOK, H. E. (eds) Carbonate Reservoir and Carbonate Field Analogs. Society of Paleontologists and Mineralogists, Special Publications, 74, 219-241. WANG, S. H., FAN, J. S. & RIGBY, J. K. 1994. The characteristics and development of the Permian
227
reefs in Ziyun County, south Guizhou, China. Acta Sedimentologica Sinica, 14, 66-74. WEIDLICH, O. 2002. Middle-late Permian reef patterns. In: KIESSLING,W., FLI)GEL, E. & GOLONKA,J. (eds) Phanerozoic Reef Patterns. Society of Paleontologists and Mineralogists, Special publications, 72, 339-390. WEIDLICH, O. & FAGERSTROM,J. A. 1999. Influence of sea-level changes on development community structure, and quantitative composition of the upper Capitan massif (Permian) Guadalupe Mountains, Texas and New Mexico. In: SALLER,A. H., HARRIS, P. M., KIRKLAND, B. L. & MAZULLO, S. J. (eds) Geologic Framework of the Capitan Reef Society of Paleontologists and Mineralogists, Special Publications, 65, 139-160. WINSLOW, M. A. 1983. Clastic dikes swarms and the structural evolution of the foreland fold and thrust belt of the Southern Andes. Geological Society of America, Bulletin, 94, 1073-1080. WINTERER, E. L. & SARTI, M. 1994. Neptunian dykes and associated features in southern Spain: mechanics of formation and tectonic implications. Sedimentology, 41, 1109-1132. WOOD, R., DICKSON,J. A. D. & KIRKLAND,B. L. 1996. New observations on the ecology of the Permian Capitan reef, Texas and New Mexico. Palaeontology, 39, 733-762. YUREWICZ, D. A. 1977. The origin of the massive facies of the lower and upper Capitan Limestone (Permian), Guadalupe Mountains, New Mexico and West Texas. In: HILEMAN,M. E. & MAZZULLO, S. J. (eds) Upper Guadalupian Facies, Permian Reef Complex, Guadalupe Mountains, New Mexico and West Texas. Permian Basin Section. SEPM Field Conference Guidebook, Midland, Texas, 77-16, 45-92. ZEMPOLICH, W. G. 1993. The drowning succession in Jurassic carbonates of the Venetian Alps, Italy: A record of supercontinent breakup, gradual eustatic rise, and eutrophication of shallow-water environments. In: LoucKs, R. G. & SARG, J. F. (eds) Carbonate Sequence Stratigraphy - Recent Developments and Applications. AAPG, Memoirs, 57, 63-105. ZEMPOLlCH, W. G., COOK, H. E. Er AL. 2002. The role of biotic and abiotic constituents in the development and diagenesis of middle to late Paleozoic carbonate platform: outcrops and subsurface examples from the C.I.S. In: ZEMPOLICH, W. G. & COOK, H. E. (eds) Carbonate Reservoir and Carbonate Field Analogs. Society of Paleontologists and Mineralogists, Special Publications, 74, 123-180. ZONENSHAIN, L. P., KORINEVSKY,V. G., KAZMIN, V. G., PECHERSKY, D. M., KHAIN, V. V. & MATVEENKOV, V. V. 1984. Plate tectonic model of the South Urals development. Tectonophysics, 109, 95-135.
Permian reef and shelf carbonates of the Arabian platform and Neo-Tethys as recorders of climatic and oceanographic changes OLIVER WEIDLICH
Earth Science Department, College of Science, Sultan Qaboos University, PO Box 36, Al Khod, Sultanate of Oman (e-mail: weidlich@squ, edu. om) The present study provides an example of how Permian benthic carbonate production is controlled by oceanographic factors and climate. After the Permo-Carboniferous glaciation, shallow-marine benthic carbonate production exhibits strikingly different temporal and spatial patterns on the eastern Arabian plate and on isolated platforms of the SW Neo-Tethys. Close to the SE margin of the Arabian plate in the Haushi area, Cisuralian (late Sakmarian-early Artinskian) mixed carbonate-siliciclastic deposits of the Saiwan Formation and Guadalupian (Wordian) marl-carbonate alterations of the Khuff Formation are characterized by dominance of crinoids, brachiopods, bivalves and bryozoans. Scarcity of calcareous algae and zooxanthellate invertebrates indicate that members of the heterozoan association (bryonoderm facies) were responsible for carbonate production. Guadalupian carbonates of the Khuff Formation in the Haushi area show striking compositional differences to lateral equivalents in the subsurface where photozoan carbonates dominate. Contemporaneous carbonates of the Saiq Formation from the Saih Hatat (Oman Mountains) exhibit a twofold composition: siliciclastic sediments contain small tabulate corals of the heterozoan association. Most of the formation is carbonate of the photozoan association with fusulinaceans, smaller foraminifers, calcareous algae, rugose corals and aggregate grains. Lopingian carbonates of the Saiq Formation (Saih Hatat) are dominated by heterozoan carbonate production with crinoids, chaetetids, bryozoans and brachiopods, and resemble heterozoan bryonoderm carbonates. Shallow-water reef carbonates of the Neo-Tethys, preserved as blocks in the Ba'id area (Oman Mountains) and the Batain coast, are contrasting: in the Ba'id area Late Cisuralian-Early Lopingian blocks of photozoan carbonates have been found, containing calcareous sponges, rugose corals, richthofeniid brachiopods and calcareous algae. Cisuralian blocks from the Batain coast are very similar to the heterozoan bryonoderm facies, while Guadalupian reef blocks show striking similarities to photozoan chlorosponge associations. Combining climate modelling data with outcrop data, dominance of heterozoan carbonate production along the rim of Arabian plate and the Neo-Tethys can be explained by cool water during the Cisuralian and local upwelling of nutrient-rich water during the Guadalupian. Formation of Lopingian heterozoan carbonates of the Saiq Formation is least understood but could be the consequence of upwelling of saline and nutrient-rich oceanic bottom water. Comparing the ecological complexity of reefs, chlorosponge reefs built complex frameworks with a variety of ecological niches, if encrusters are present, whereas heterozoan reef communities are comparatively simple in their structure. Abstract:
Sedimentary dynamics of modern tropical rimmed carbonate platforms stimulated sedimentologists for a long time, although evidence increased that these depositional systems are inadequate models for many Late Palaeozoic limestone occurrences. Since the 1970s, coolwater carbonates have increasingly attracted the interest of sedimentologists (e.g. Lees & Buller 1972; Nelson 1988) and so did the reconnaissance of mud mounds approximately a decade later (e.g. Lees & Miller 1985; Bosence & Bridges 1995; Monty 1995; Pratt 1995). To overcome the shortcomings of tropical carbonates as an unequivocal depositional model for limestone genesis, Schlager (2000, 2003) synthesized existing descriptions
of the tropical, the mud-mound and the coolwater benthic carbonate factories. These three factories have been characterized by their variations with respect to depositional profile, depth range and sediment composition as follows.
Tropical factory This well-known type of benthic carbonate production prevails in shallow, oligotrophic, warm (monthly mean minimum temperature > 20 ~ and sunlit environments with minimal siliciclastic input. Production rates are highest in very shallow waters above a maximum depth of 25-30m. Biotically controlled carbonate
~kLVARO,J. J., ARi~TZ,M., BOULVAIN,F., MUNN~CK~,A., VACHARD,D. & V~NNIN, E. (eds) 2007. PalaeozoicReefsand Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special
From:
Publications, 275,229-253.0305-8719/07l$15.00 9 The Geological Society of London.
230
O. WEIDLICH
production is governed by photo-autotrophic organisms (calcareous algae and calcified invertebrates with symbiotic zooxanthellae) and heterotrophic metazoans. Abiotic or biotically induced carbonate production, for example the precipitation of synsedimentary cements, is locally significant as well. Ideal depositional profiles of the tropical factory are rimmed shelves, where either reefs or shoals significantly reduce the water energy of waves at the shelf edge and allow sedimentation production in protected lagoons. Following the nomenclature of James (1997), the benthic carbonate-producing communities belong to the photozoan association. Comparable sediments of Permian age are chloroforam and chlorosponge carbonates, containing calcisponges, Archaeolithoporella and large specimens of Shamovella (Beauchamp & Desrochers 1997).
Cool-water factory Biota of the heterozoan association (sensu James 1997) dominate carbonate production from sea level to a depth of several hundreds of metres mostly in temperate, cool and cold (mean monthly temperature <20 ~ environments. Carbonate production is maintained by calcified metazoans (biotically controlled carbonate production) with emphasis on heterotrophic biota, including mollusks, barnacles, brachiopods, crinoids and bryozoans. The importance of abiotic or biotically induced carbonate production is low-absent. These carbonates lack warmwater components of the photozoan assemblage, such as ooids, calcareous green algae and zooxanthellate organisms. Modern cool-water carbonates occur poleward from the tropical reef belt and within 30~ and 30~ of the equator. In low latitudes they frequently accumulate in upwelling areas along the eastern rims of modern oceans. Carbonates formed by heterozoan associations are therefore not necessarily restricted to cool-water environments. Other extrinsic factors such as increased nutrient or siliciclastic supply and rapidly rising sea level may cause a change from photozoan to heterozoan carbonate production. Characteristic depositional profiles are open shelves. Waves are thus important control factors of the shallow shelf areas owing to the absence of a protective shelf rim. For a correct interpretation of factors leading to the deposition of heterozoan carbonates, independent lines of evidence must be considered, including palaeolatitude and palaeoceanographic modelling data. Permian equivalents of the heterozoan association are bryonoderm (dominance of bryozoans, crinoids, brachiopods and siliceous sponges)
and hyalosponge carbonates (dominance of siliceous sponges) rich in cherts (Beauchamp & Desrochers 1997).
Mud-mound factory This carbonate factory is dominated by biotically induced and abiotic carbonate production. Microbialites and marine synsedimentary cements are of special importance while calcified metazoans contribute to the carbonate production to a varying extent. The mud-mound carbonate factory has no restrictions with respect to depths or latitude. Oxygen-poor and/or mesotrophic-eutrophic oceanic conditions may favour the dominance of this carbonate factory. The abundance of carbonates formed by the heterozoan association fluctuated throughout Earth history and culminated during the Palaeogene-Neogene, Permian-Carboniferous and Ordovician. Over the last years, Permian cool-water carbonates have received a lot of attention with emphasis on the Canadian Arctic Archipelago and the Barents Shelf (see Beauchamp & Desrochers 1997 and Stemmerik 1997 for an overview and further references). Published data analysing the composition of sediments clearly demonstrate the progressive climatic cooling of northern Pangea during the Permian. Few data have been compiled from the southern hemisphere (e.g. Rao 1981, 1988; Weidlich 2002a, b), despite the fact that the Carboniferous glaciation was asymmetrical with thick continental ice sheets over the South Pole. The purposes of this paper are: (1) to analyse the composition of Permian shallow-water carbonates from four localities of the Arabian plate and the Neo-Tethys; (2) to assign these sediments to the photozoan or heterozoan association; and (3) to interpret the control mechanisms of carbonate production.
Geological setting and stratigraphy of the study areas Permian deposits form the Arabian platform megasequences AP 5-6 (Sharland et al. 2001). Classic outcrops of tectonically undeformed Permian carbonates occur in central Saudi Arabia (Fig. l a). Slightly to moderately deformed carbonates of the Arabian plate can be studied in the Oman Mountains, notably in the Saih Hatat, Jebel Akhdar, Musandam Peninsula and Jebel Qamar. In addition, Permian carbonates crop out in the Haushi-Huqf area, interior Oman. Key localities for the study of Permian carbonates of the Neo-Tethys are the Ba'id area (Oman Mountains) and the Batain coast (Fig. lb).
PERMIAN REEFS OF THE ARABIAN PLATFORM
231
Fig. 1. Location and paleogeographic setting of the Haushi area, the Saih Hatat, the Ba'id area and the Batain coast. (a) Overview of the Arabian plate. (b) Position of the study areas in Oman. (c) Cisuralian palaeogeographic reconstruction of the study areas (after Golonka et al. 1994).
Outcrops representing the following palaeogeographic positions were chosen for this study: eastern margin of the Arabian plate: Haushi area (Fig. lc, locality 1); eastern margin of the Arabian plate: Saih Hatat (Fig. lc, locality 2); - SW Neo-Tethys: Batain coast (Fig. lc, locality 3); SW Neo-Tethys: Ba'id (Fig. lc, locality 4). -
-
-
Arabian plate Haushi area, central Oman. The Late Carbonifer-
ous-Early Permian Haushi Group begins with the 100-700 m-thick A1 Khlata Formation, a succession from glacial diamictites to glaciofluvial conglomerates and sandstones and, on top of the formation, to lacustrine deposits of the Rahab shale (Hughes-Clarke 1988; Dubreuilh et al. 1992), see Figures 2 & 3. Palynomorph data indicate an Asselian-earliest Sakmarian age for
the top of the AI Khlata Formation (Stephenson & Filatoff 2000). The onset of the first transgressive pulse after the Permo-Carboniferous glaciation commenced during the Sakmarian with the Saiwan Formation, which has an almost ubiquitous distribution in the subsurface of interior Oman (Fig. 2). The Sakmarian-early Artinskian Saiwan Formation conformably overlies the Rahab shale and lateral clastic equivalents or cuts into its base (Dubreuilh et al. 1992; biostratigraphic data are based on brachiopods, see Angiolini et al. 2003a): 50-65 m of cross-bedded or horizontally bedded, partly laminated mixed carbonate-siliciclastic sediment of varying composition and minor shale make up the Saiwan Formation (Angiolini et al. 2003a). The top of this formation is a regional unconformity. The Saiwan Formation thins out towards the local Haushi uplift with an increasing number of unconformities (Blendinger et al. 1990). Parts of the Saiwan Formation had been informally
232
O. WEIDLICH
Fig. 2. Overview, showing the stratigraphic range of the investigated carbonates. Note that the Aseelah Unit has a Lower Triassic age, while the investigated blocks have a Permian age. Timescale after Gradstein et al. (2004).
called Haushi Limestone, Bellerophon Limestone or Metalegoceras Limestone. Following an indistinct unconformity, the Akhdar Group begins with 70-100 m thick fluviatile, lacustrine and flood-plain deposits of the Gharif Formation (Hughes Clarke 1988; Angiolini et al. 2003a), see Figure 2. During the Guadalupian-Lopingian, marine conditions prevailed over the vast area of the Arabian plate. The sediments of the Khuff Formation form cyclic sequences from carbonates and marls to evaporites, with a minimum thickness of 670 m in the subsurface (A1-Jallal 1995; Alsharhan & Nairn 1997 and further references herein). Outcrops of the Haushi area show a reduced thickness of 30-40 m and indicate marine benthic carbonate production only during the Wordian (Angiolini et al. 2003b), see Figure 2. The Saiwan Formation was sampled at point 21~ 57~ and the Khuff Formation was sampled at point 21~ 57o33'36"E.
Saih Hatat, Oman Mountains. Permian deposits
of the Saih Hatat, with a thickness of 1000 m, are part of the Akhdar Group. Shallow-water tropical carbonate deposition started on the Arabian plate after the Permo-Carboniferous glaciation with the Saiq Formation, (see Figs 2 & 3). The oldest Permian limestones have a Wordian age. Previously, a 'Dzhulfian' age has been discussed (Glennie et al. 1974; Montenat et al. 1976; Le M6tour et al. 1994, 1995). The finding of Colaniella nana (determination by D. Vachard) provides significant evidence for a Changhsingian age of the carbonates in the study area. Close to the rim of the attached Arabian platform, carbonates of the Guadalupian-Changhsingian Saiq Formation consist of third-order sequences with fossiliferous transgressive systems tracts (TSTs) and monotonous highstand systems tracts (HSTs) with mud/wackestone and coral baffiestone (Weidlich & Bernecker 2003, 2006). The Changhsingian beds are bioturbated wacke- to floatstone with crinoids, brachiopods,
PERMIAN REEFS OF THE ARABIAN PLATFORM
233
Fig. 3. Simplifiedlithological logs of the study areas (see text for references).
bryozoans and, probably, sponges. The Saih Hatat, situated close to the rim of platform, was affected by repeated tectonic pulses that caused rapid facies changes and the deposition of mafic volcanism. Data of sections Aday 1 (23~ 58~ Aday 2A and 2B (23~ 58~ were used for interpretations (see Weidlich & Bernecker 2006 for details of the locations).
Neo-Tethys Batain coast, eastern Oman. The Batain nappes comprise Permian-Cretaceous volcanic rocks and sediments (Immenhauser et al. 2000), see Figure 2. Of interest for this paper are Permian carbonate blocks within the Lower Triassic Aseelah Unit (Fig. 3) (Saal Formation, see Hauser et al. 2002 for details). The onset of
234
O. WEIDLICH
shallow-water benthic carbonate production is under discussion. Probably, it could have started during Sakmarian-Artinskian times (Shackleton et al. 1990; Immenhauser et al. 2000; Vachard et al. 2001, 2002; Hauser et al. 2002) or in the late Artinskian-Kubergandian (D. Vachard pers. comm.). It culminated in the Guadalupian and no evidence of significant Lopingian shallowwater carbonate production has been found. Data of blocks east and west of a dust road about 9 km south of A1 Hadd (22~ 59~ were used for this paper. The area has been mapped as 'Batain Melange with limestone Blocks' by Shackleton et al. (1990). Ba'id area, Oman Mountains. The nappes of the Hawasina Complex document the evolution of the Neo-Tethys. The destruction of the shallowwater carbonate platform has complicated the palinspastic reconstruction. However, there is an increasing consensus that a Permian isolated platform existed close to the rim of the Arabian plate and that a Triassic isolated platform occurred in a more distal position with respect to the shelf margin of Gondwana (e.g. Pillevuit et al. 1997). Relics of Guadalupian and Middle-Late Triassic shallow-water carbonates have been preserved as breccias or enormous mega-blocks attaining a size of several kilometres. Guadalupian shallow-water blocks of the Ba'id and A1 Jil formations (Fig. 2) have been frequently found in the Ba'id area (e.g. Blendinger 1988), while Lopingian blocks are rare-absent. Data from a well-exposed section at Wadi Wasit (23~ 58~ were used for this paper (Fig. 3). Facies
variation
of the
Arabian
plate
S E m a r g i n o f the A r a b i a n plate: H a u s h i area
In the field, carbonate-rich sediment of the Saiwan Formation is characterized by horizontal bedding (Fig. 4a), planar cross-bedding (Fig. 4b), erosional features at the base of beds (Fig. 5a), normal grading (Fig. 5a) and convex-up orientation of shells (Fig. 4a). Recognizable biota include crinoids, bryozoans, brachiopods, mollusks and cephalopods. In addition, the occurrence of conodonts, ostracods and fish remains has been reported (Angiolini et al. 2003a). Grainstone or rudstone are the dominant carbonate textures. Biota in life position such as doublevalved brachiopods or bivalves are rare. Disarticulation of shells, abrasion and concentration of bioclastic debris indicate lateral transport. Micrite envelopes suggest that microborers were
present (Fig. 5c, d). Locally, dissolved peloids (circular-elliptical outlines, grain sizes between 0.2 and 1.0 mm) are common. The amounts of detrital quartz (mono- and polycrystalline quartz), feldspar, micas (muscovite and biotite) and lithoclasts vary from bed to bed and even within thin sections (Fig. 5b, e). The grain size is heterogeneous, ranging from silt to gravel. Quartz grains are predominantly subangular, rounded grains occur subordinately. Only two facies types can be distinguished: bioclastic rudstone with shell debris and detrital quartz (Fig. 5a~t); grainstone with detrital quartz, peloids and bioclasts (Figs 4b & 5e). The groundmass of grain-supported sediments commonly is cement. The first generation cement, filling a large percentage of the interparticle pores, is isopachous radiaxial fibrous calcite, indicative ofa synsedimentary cementation in the marine-phreatic realm (Fig. 5c, d). Bioclasts like bivalve shells with an aragonitic mineralogy were dissolved. Cementation of moldic porosity with radiaxial calcite (Fig. 5c, d) indicates dissolution and cementation in the marine realm. Crinoid remains are commonly surrounded by syntaxial overgrowth (Fig. 5f, g) and open intraparticle pores were finally cemented by calcite spar (Fig. 5c, d), probably of meteoric or burial origin. Peloids, crinoids and other bioclasts were partly replaced by hematite or phosphate (Fig. 5f, g). Interparticle pores of siliciclastic units are cemented by poikilotopic calcite. Field characteristics of carbonates of the Khuff Formation comprise horizontal bedding, bioturbation and erosive shell beds with bivalves or brachiopods. Sediment textures resemble the inventory of the Saiwan Formation; however, siliciclastic input is generally low, except for a few beds in the lowermost part. In thin sections, bivalves, brachiopods, crinoids and bryozoans are the most frequent skeletal grains. Noteworthy in these sediments is the rare occurrence of bioclasts of the calcareous algae Permocalculus sp., which is normally very abundant on the Arabian platform in Guadalupian strata. The following facies types were differentiated based on the biotic content: - p a r a u t o c h t h o n o u s bivalve or brachiopod floatstone (Figs 4c & 6a); transported bioclastic bivalve and brachiopod floatstone-rudstone (Figs 4d & 6b); - crinoid bryozoan brachiopod bivalve floatstone-rudstone (Fig. 6c, d); diverse packstone-grainstone with calcareous algae (Fig. 6e, f);
PERMIAN REEFS OF THE ARABIAN PLATFORM
235
Fig. 4. Field photographs and samples of the Saiwan and Khuff Formations, Haushi area. (a) Bedding plane of a tempestite with convex-up orientation of disarticulated bivalve shells; the sediment is a bioclastic rudstone with shell debris and detrital quartz, Saiwan Formation. Scale is 1 cm. (b) Planar cross-bedded submarine dune facies; the sediment is a grainstone with detrital quartz, peloids and bioclasts, Saiwan Formation. See hammer for scale. (c) Slab of parautochthonous bivalve floatstone, containing some crinoid fragments (whitish bioclasts), Khuff Formation. Note geopetal infill within many bivalves. Scale is 1 cm. (d) Photomicrograph, showing bioclastic bivalve floatstone (tempestite) and mudstone, Khuff Formation. Note bioturbation of the sediments. Scale is 2 cm.
236
O. WEIDLICH
Fig. 5. Photomicrographs of heterozoan carbonates, Saiwan Formation, Haushi area. (a) Slab, showing two amalgamated tempestites. The dashed line separates the storm deposits. Note normal grading indicated by the black arrow. Scale in mm. (b) Thin section overview of the bioclastic rudstone with shell debris and detrital quartz. Skeletal debris is from bivalves, crinoids and bryozoans. Note the intense cementation of interparticle pores. (c) & (d) Close-up view under plane polarized light (c) and crossed nicols (d), showing the densely packed debris of recrystallized bivalves with micrite envelopes. Porosity has been occluded by radiaxial fibrous calcite and calcite spar. Scale units in 0.1 mm steps. (e) Grainstone with detrital quartz, peloids and bioclasts. Scale units in 0.1 mm steps. (f) & (g) Close-up view of a hematized crinoid fragment with syntaxial overgrowth under plane polarized light (f) and crossed nicols (g). Relics of the original net-like microstructure have been preserved. Scale units in 0.1 mm steps.
PERMIAN REEFS OF THE ARABIAN PLATFORM
237
Fig. 6. Photomicrographs of heterozoan carbonates, Khuff Formation, Haushi area. All scale units in 0.1 mm steps. (a) Close-up view the parautochthonous bivalve floatstone. (b) Bioclastic bivalve and brachiopod floatstone (tempestite); note the productid pseudopunctate brachiopod fragment (upper right). (e) & (d) Close-up view under plane polarized light (c) and crossed nicols (d) of bryozoan, crinoid, brachiopod, bivalve rudstone. Note syntaxial overgrowth rimming the crinoid fragment. (e) Diverse packstone with dasycladacean algae, fenestellid bryozoans and bivalve fragments. The calcareous algae are densely packed fragments of Permocalculus sp. (f) Diverse grainstone with fragments of Permocalculus, bivalves and brachiopods. Note the dark microbial incrustations of some fragments.
238 -
-
O. WEIDLICH
partly bioturbated mudstone-wackestone (Fig. 4d); siliciclastic sediment dominated by detrital quartz.
Parautochthonous bivalve or brachiopod floatstone and transported bioclastic bivalve and brachiopod floatstone-rudstone are textural end members. The parautochthonous character is indicated by the presence of double-valved bivalves or brachiopods, whereas bioclastic floatstone predominantly contains shell debris. Erosion at the base of this type of sediment points to storm events. Shell beds with a high percentage of unfragmented and complete specimens tend to be taxonomically monotonous (Fig. 6a), whereas debris-rich shell beds contain a mixture of brachiopods and bivalves (Fig. 6b). Among all skeletal components of the diverse packstonegrainstone, calcareous algae show the highest degree of fragmentation. This biomodal taphonomy of allochems resulted either from transport of calcareous algae over longer distances across the shelf or from increased fragility of calcareous algae (Fig. 6e, f). All facies types mentioned above have in common that biogenic incrustations are rare. If present, encrustations are dominated by bryozoans and microbials (Fig. 6f). Bioturbation is a common phenomenon of the sediments regardless of texture or biotic composition of the skeletal grains. Within grain-supported sediment patches, interparticle pore spaces either filled with isopachous fibrous calcite, suggesting the presence of early marine-phreatic cementation, or bladed rims and/or blocky calcite occluded the remaining open interparticle pore space and moulds of aragonitic bivalves, suggesting meteoric and/or shallow burial cementation. Syntaxial overgrowth of crinoid fragments is a common phenomenon of the sediments of the Khuff Formation (Fig. 6c, d). Interpretation. Thinning and pinch-out of strata of the Saiwan and Khuff formations towards the Haushi uplift (see subsurface data presented by Blendinger et al. 1990) make a ramp the most probable depositional profile. A further facies interpretation of the subsurface data is impossible. Cross-bedded deposits of the Saiwan Formation from outcrops are interpreted as subtidal mixed carbonate-siliciclastic dunes. Except for some out-sized shells, the submarine dunes are typically grain- and packstone-dominated sediments with a fine sand fraction. Beds with horizontal bedding, high packing density of coarse reworked skeletal debris, normal grading and erosional features at the base of beds are tempestites. Amalgamation of tempestites indicates the frequent occurrence of storm events and
deposition of sediment above storm wave base. These features suggest the absence of a shelf rim and a shallow-water inner or middle ramp as depositional environment for the Saiwan Formation. Storm events repeatedly reworked the sediments base, a typical depositional profile of coolwater settings. The shallow-water interpretation is backed by the fact that the study area, is within the Haushi area which is regarded as a topographical high throughout the Permian (Fig. lc). The depositional environment of the Khuff Formation resembles that of the Saiwan Formation. However, the decreasing number of tempestites and the increase in low-energy mudstone indicate a deepening of the environment of the Khuff Formation from an inner and middle ramp to a middle and outer ramp position. The observed increase in water depth coincides with the thirdorder rise in sea level during the Wordian, which is well known from the Arabian plate (Sharland et al. 2001; Weidlich & Bernecker 2003). In the study area the biotic content of the sediments of the Saiwan and Khuff formations is typical of the heterozoan association and differs from coeval carbonates of the Arabian plate owing to the absence of sabkha evaporites and the scarcity of aggregate grains, ooids and phototrophic skeletal grains such as calcareous algae (see Vachard et al. 2005; Vaslet et al. 2005 and Insalaco et al. 2006 for descriptions of the photozoan association of the Khuff Formation). The sediments of the study area correspond to the bryonoderm association (Beauchamp & Desrochers 1997). During the Cisuralian a palaeolatitude of 35"-45 ~ (Golonka et al. 1994) and the rapid transition without a significant temporal hiatus from diamictites to mixed carbonate-siliciclastic sediments fuel the idea of a cool-water origin of the marine sediments of the Saiwan Formation. By contrast, the dominance of coeval tropical sediments across the Arabian plate complicates the interpretation of the Khuff Formation as a cool-water deposit in the Haushi area. The dominance of bryonoderm sediments and the local presence of highly fragmented algal thalli in a few beds suggest the mixing of bryonoderm (heterozoan) and chloroforam (photozoan) sediments. The mixed composition of these sediments supports the hypothesis of local upwelling of cool and/or nutrient-rich sea water, which has been predicted by climatic modelling using the programme PALEOCLIMATE (Golonka et al. 1994). Eastern margin o f the Arabian plate: Saih Hatat
Dolomitized Guadalupian-Lopingian limestones of the Saiq Formation exhibit a stacked
PERMIAN REEFS OF THE ARABIAN PLATFORM pattern of third-order cycles (Weidlich & Bernecker 2003). In the field, the carbonates have a dark colour and a varying biogenic content. Sediment structures comprise horizontal bedding, erosional features at the base of beds and alignment of fossils. Transgressive systems tracts (TSTs) yield a diverse biota, including calcareous algae, fusulinaceans, smaller foraminifers, rugose corals, bryozoans, brachiopods, molluscs, ostracods, microbes and rare sponges. Non-skeletal grains are peloids and aggregate grains. Firmgrounds and incipient hardgrounds are characterized by the presence of Thalassinoides burrows and laminated microbialites. Highstand systems tracts (HSTs) are either dominated by monotonous mudstones or contain tabulate and waagenophyllid rugose corals with a dendroid growth form (Weidlich & Flfigel 1995; Weidlich 1999). Subaerial unconformities are frequent within HSTs. The following facies types have been found: - m u d s t o n e - b i o c l a s t i c wackestone, massive, bioturbated or laminated, HST; coral baffiestone with dendroid rugose (e.g. Waagenophyllum sp.) and tabulate corals (Multithecopora sp.), HST (Fig. 7b, d, e); -siliciclastic sediments (medium- to finegrained sandstone, siltstone) early TST?; - coral bivalve floatstone, HST and TST (Fig. 7a, c); diverse floatstone-rudstone with calcareous algae, fusulinids, smaller foraminifers and aggregate grains, TST; - bivalve floatstsone, TST; - brachiopod floatstone, TST; - crinoid floatstone, TST; - bryozoan floatstone, TST; incipient hardgrounds with burrow structures (e.g. Thalassinoides) and/or laminated microbialites, TST. Locally, the Saiq Formation is enriched in terrigeneous siliciclastic detritus (clay and detrital quartz) and yields an impoverished fauna dominated by toppled colonies of the tabulate coral Multithecopora. No biogenic encrustations or boring traces have been observed. Dolomitized carbonates of the TST yield aggregate grains, calcareous algae (especially dasycladaceans including Mizzia and Goniolinopsis) and fusulinids (e.g. Neoschwagerina and Verbeekina sp.). The latter indicate a Guadalupian age of the sediments. Other tropical organisms are frequent, including cerioid wentzeMloid rugose corals and large alatoconchid bivalves (Fig. 7c), a group of thick-shelled bivalves that probably housed symbiotic zooxantellae. Larger bioclasts
239
of corals and bivalves have thick micrite envelopes, resulting from the destructive activity of microborers and the encrustation of microbes (Fig. 7c). In addition to microborers, macroborers of unknown taxonomic affinity penetrated reworked alatoconchid bivalve bioclasts. The absence of depositional relief suggests that rugose and tabulate corals formed biostromes. The matrix between calices of the dendroid corals is a mudstone-wackestone (Fig. 7b, d, e). A small percentage of the corals has been preserved in growth position. Toppled cerioid coral colonies were able to recover (Fig. 7a). A significant change in the biotic composition occurred at the end of the Guadalupian. Lopinging carbonates of the TST differ from Guadalupian TSTs owing to the lack of fusulinaceans, calcareous algae, cerioid corals, aggregate grains and ooids. Crinoids, brachiopods and bryozoans are much more common. Biotic carbonate production of this type lasted until the end of the Permian (Weidlich & Bernecker 2006).
Interpretation. The depositional environment of the marine carbonates of the Saiq Formation has been regarded as tropical. This view has to be modified to some extent as the biotic inventory of the Saiq Formation shows considerable changes in the Saih Hatat. Colonies of the tabulate coral Multithecopora from Guadalupian siliciclasticrich sediment correspond to the heterozoan association. It is very likely that photozoan carbonate production was inhibited by terrigeneous influx, high nutrient input and turbid water. Despite the presence of a warm and equatorial environment, no tropical carbonate production could be established in this environment. Conversely, most Guadalupian carbonates of the Saiq Formation lacking detrital quartz are typical of the photozoan association and correspond to the chloroforam association defined by Beauchamp & Desrochers (1997). These sediments are dominated by a high percentage of benthic carbonate production, which was directly or indirectly related to phototrophic processes. The presence of Colaniella nana in a bryozoan crinoid floatstone confirms a Lopingian age for the overlying carbonates. There, photozoan carbonate production switches to heterozoan carbonate production. Control mechanisms of this change remain somewhat speculative. The Lopingian low-amplitude rise in sea level cannot account for the change in biotic composition as tropical carbonate factories easily keep up with rapid sea-level rise during greenhouse times. Also, tectonic pulses during repeated rift phases of the Neo-Tethys cannot account alone for these compositional changes. The absence of clastic
240
O. WEIDLICH
"8
"~
~o o oo~z
S~oo~
0"~
"
=
o~
.
~
~
o .o~ ~
,,.~ ~
~ .~1 ~,~
~ ,~ ~ .~
~.~
~
~~
I~ ~0 o ,.~ ~ ~-~- ~~ "~
~
~
0
~--"~.
0
~
~ o ~
....
~.H
. . ~ ~. ~ ~.,
0
~~
~m r.~ ~
PERMIAN REEFS OF THE ARABIAN PLATFORM input eliminates terrestrial runoff as a control mechanism. A possible control mechanism for the compositional change could be upwelling of saline and eutrophic sea water, which disturbed the photozoan carbonate factory. The interpretation is backed by modelling data predicting Late Permian oceans to be increasingly stratified with saline bottom waters (Kidder & Worsely 2004). N e o - T e t h y s
S W Neo-Tethys: Batain coast In the field, the most obvious feature of the reef blocks is the locally intense silicification of sediment and biota. Of importance are crinoids, bryozoans, chaetetids, calcisponges, rugose corals, calcareous algae, smaller foraminifers, fusulinaceans, brachiopods, molluscs, Archeolithoporella hidensis, Shamovella and microbes. Many of the blocks are from bryozoan reefs, sponge reefs and rugose coral reefs. Distribution and preservation of the rich biota is heterogeneous, allowing the differentiation of the following facies types (Weidlich 2002a, b): - b r y o z o a n crinoid brachiopod baffiestone (Figs 8a-c & 9a-d); bryozoan crinoid brachiopod floatstonerudstone (Figs 8d & 9e-h); calcisponge baffiestone (Fig. 10a, b); - calcisponge microbe framestone (Fig. 10c, d); calcisponge floatstone-rudstone; - coral boundstone (Fig. 1lc, d). -
-
-
The exact spatial and temporal position of the reef blocks within a facies belt is still unsolved. Bryozoan crinoid brachiopod baffiestone and floatstone-rudstone represent textural end members of the same depositional environment. Indistinct bedding planes of a large block (5 • indicate that the metazoan communities represent level-bottom communities or biostromes without significant depositional relief. Baffiestone contains complete crinoids (see Fig. 9a) and fragmented debris. Differential taphonomy is also true for bryozoans, comprising large fenestellid bryozoan colonies as well as debris. Typical of the bafflestone are encrustations by the bryozoan Fistulipora, chaetetids and Shamovella. The latter is only represented by small specimens. The groundmass is either micrite or radiaxial calcite (Fig. 8a, b), microbial peloids with a clotted fabric occur in cavities. Careful petrography revealed no significant changes with respect to packing density or taphonomy of organisms, which could explain the presence of mud or cement. Stromatactis-like cavities with rims of radiaxial fibrous calcite and
241
geopetal infills of microbial peloids are an important feature of the bryozoan crinoid brachiopod baffiestone. The bryozoan crinoid brachiopod floatstone-rudstone is well sorted. Commonly, the bioclasts are rimmed by radiaxial calcite. The remaining pore space has been later filled with lime-mud or calcite spar. In floatstonerudstones, bioclasts are reworked and brachiopod shells are penetrated by microborers. The age of bryozoan crinoid brachiopod baffiestone and floatstone-rudstone is unclear owing to unequivocal biostratigraphic data. Considering the absence of diagnostic Guadalupian foraminifers, an almost complete crinoid corona (Fig. 8a) resembling Early Permian cladid crinoids from central Oman (Jell & Willink 1993) and the compositional similarities with the Saiwan Formation, a Cisuralian age is considered for these blocks despite the fact that a Guadalupian age cannot be ruled out (Vachard pers. comm.). This idea is supported by the presence of Protonodosaria sp., Pachyphloia sp. and Nodosinelloides sp. as these impoverished associations flourished in cool-water environments (DiazMatinez et al. 2000). End members of the calcisponge reefs are floatstone-rudstone, baffiestone (Fig. 10a, b) and framestone (Fig. 10c, d), yielding sphinctozoan sponges, inozoan sponges, chaetetids, various bryozoans, crinoids and brachiopods. Also, calcareous algae and smaller foraminifers occur.
Shamovella obscura, Archaeolihoporella hidensis and microbes incrusted larger skeletal metazoans and played a lead role as frame builders. The increase in encrusters, automicrite (in situ precipitate, see Reitner & Neuweiler 1995) and marine-phreatic radaxial calcite cement is responsible for the creation of a complex reef framework with cavities. Commonly, reef cavities have been filled with clotted microbial micrite or various types of bioclastic sediment. These cryptic environments represent growth framework porosity and result from the encrustation of larger metazoans (e.g. sponges by Archaeolithoporella and microbial crusts) (Fig. 10c, d). The presence of smaller foraminifers (Lasiotrochus tatoiensis) and fusulinaceans (Codonofusiella laxa) indicate a Guadalupian age for most of the blocks (Hauser et al. 2002; Vachard et al. 2002). Coral boundstone is represented by isolated boulders of cerioid coral colonies (Fig. 11c) while coral floatstone yields clasts of cerioid and dendroid colonies (Fig. 11d). Biogenic encrustations suggest that the rugose corals were part of a reef framework and did not grow as isolated colonies in a platform environment; some of the colonies show evidence of macroborer activity. Coral
242
O. WEIDLICH
Fig. 8. Photomicrographs of Permian heterozoan carbonates, Aseelah Unit, Batain coast. Arrows indicate the top of the sample. Scale bar is 1 cm. (a) & (b) Overview of crinoid bryozoan bafflestone under plane polarized light and crossed nicols. Patches of grain support fabric with radiaxial fibrous calcite pores alternate with areas dominated by micrite. Note stromatactis-like cavities. Parautochthonous facies. (c) Crinoid bryozoan floatstone with incipient boundstone patches of bryozoans and chaetetids. (d) Densely packed floatstone, containing crinoids, bryozoans and brachiopods. The bioclasts have thin rims of bladed calcite.
PERMIAN REEFS OF THE ARABIAN PLATFORM
243
Fig. 9. Photomicrographs of heterozoan carbonates, Aseelah Unit, Batain coast. All scale bars in 0.1 mm units. (a)-(d) Bryozoan crinoid brachiopod baffiestone; (e)-(h) bryozoan crinoid brachiopod floatstone-rudstone. (a) Close-up view of a crinoid corona, indicating the parautochthonous character of facies. (b) Bafflestone with fenestellid bryozoan fragments. (e) Bryozoan fragment incrusted by Shamovella (formerly Tubiphytes) obscura. (d) Close-up view of a brachiopod shell with isopachous radiaxial fibrous calcite. The remaining pore space has been filled by microbial peloids. (e) Crinoid fragment surrounded by fibrous cement. (f) Heavily bored brachiopod bioclast incrusted by Shamovella obscura. (g) Crinoid ossicle with bryozoan encrustation. (h) Brachiopod bioclast with Shamovella incrustation.
244
O. WEIDLICH
Fig. 10. Photomicrographs of sponge reef framework variation of photozoan carbonates, Aseelah Unit, Batain coast. (a) & (b) Calcisponge baffiestone; (c) & (tl) calcisponges microbe framestone. Note the increase in framework stability from (a) to (d). Arrows indicate the top of the sample. All scale bars are 1 cm. (a) Bafflestone with calcisponges, bryozoans and crinoids. Note the high percentage of mud. (b) Bafflestone with incipient framework patches, e.g. a brachiopod forms the substrate for other organisms (top of sample). The bioclastic wackestone-packstone groundmass is still dominant. (c) Framework with calcisponges (sphinctozoans and inozoans) and thick crusts Archaeolithoporella hidensis, Shamovella and microbes. The resulting large reef cavities were filled with various generations of bioclastic sediment. (d) Calcisponge microbe framework with thick biogenic crusts and radiaxial fibrous calcite, rimming growth framework pores. Note the cluster of dasycladacean algae (upper right).
PERMIAN REEFS OF THE ARABIAN PLATFORM
245
Fig. 11. Photomicrographs of photozoan rugose coral reef facies, Batain coast and Ba'id area. Arrows indicate the top of the sample. All scale bars are 1 cm. (a)-(c) Rugose coral calcimicrobe framestone. (a) Praewentzelella community. (b) Wentzelella community. Note the large reef cavity sheltered by the cerioid coral. (c) Cerioid coral colony with traces ofmacroborers (arrows). (d) Coral floatstone with clasts of cerioid and dendroid corals.
246
O. WEIDLICH
floatstone-rudstone contains a mixture of dendroid and cerioid coral bioclasts in a cement-rich matrix with abundant bryozoan, crinoid and smaller foraminifers fragments. Despite the scarcity of determinable fusulinaceans, a Guadalupian age of the blocks is suggested for both facies types by the presence of Neoendothyra reicheli and Rauserella sp.
Interpretation. The bryozoan crinoid brachiopod baffiestone and floatstone-rudstone is a typical element of the heterozoan association (bryonoderm facies sensu Beauchamp & Desrochers 1997) because of the lack of phototrophic organisms and Archaeolithoporella hidensis and the small size of Shamovella. The facies resembles Waulsortian mounds with respect to biotic composition, presence of stromatactis and cementation patterns. Wahlman (2002) described similar fenestellid bryozoan-Tubiphytes facies and concluded deeper- and cooler-water settings in the uppermost Carboniferous and Lower Permian of Arctic Canada and northern Russia. These depositional interpretations help to better understand the bryozoan crinoid brachiopod baffiestone and floatstone. Cool water is the most probable control mechanism of the heterozoan carbonates of the study area because of the similarity of some bryozoan taxa of the study area with cool-water faunas (Weidlich et al. 2004); no convincing additional evidence supporting a deep-water model exists. The dominance of sponge and coral reef facies during Guadalupian is well constrained with biostratigraphic data; vice versa missing biostratigraphic data suggest their scarcity during Cisuralian and Lopingian times in this region. The Guadalupian sponge and coral reef blocks clearly indicate the presence of the photozoan association and fulfil important criteria of the chlorosponge association (sensu Beauchamp & Desrocher 1997), including the presence of calcisponges, dasycladacean algae and thick crusts of Archaeolithoporella hidensis. A tropical, shallow-water environment can be concluded. Considering a Cisuralian age of the bryozoan crinoid brachiopod baffle and floatstonerudstone (based on preliminary data), the compositional changes from Cisuralian heterozoan-dominated carbonates to Guadalupian photozoan-dominated carbonates is interpreted to reflect increasing sea-water temperature. As a consequence, Cisuralian reef blocks are interpreted to reflect cool-water conditions during or shortly after deglaciation in a realm outside the tropical belt, while Guadalupian reef blocks are photozoan deposits,
indicating warm and shallow water. This interpretation of global warming is backed by stable isotope data that, on a global scale, point to an increase in water temperature (Korte et al. 2005).
S W Neo- Tethys: Ba'id area Reef blocks of the Ba'id area are massive and lack sediment structures. Comparable to blocks of the Batain coast, silicification is the most obvious feature. Many of the blocks are from algal-cement reefs, calcimicrobe reefs, sponge reefs and rugose coral reefs. Reef blocks contain a rich biota, including calcisponges, chaetetids, rugose and tabulate corals, richthofeniid brachiopods, crinoids, microbes, Shamovella, Archaeolithoporella, bryozoans, Lercaritubus, dasycladacean algae, solenopor-acean algae and foraminifera. In addition, brachiopods, molluscs, trilobites and ostracodes are present. About 100 taxa of reef builders form a diverse community, including 47 species of calcisponges, 25 species of rugose and tabulate corals, 15 species of calcimicrobes/calcareous algae/ problematica, nine species of bryozoans and four species of other reef-building biota (see Weidlich 2002a, b and further references herein). The following facies types occur: - chaetetid Archaeolithoporella framestone (Fig. 12a); - calcisponge Arehaeolithoporella framestone with bryozoans (Fig. 12b, c); sponge baffiestone (Fig. 12d); sponge floatstone-rudstone; - rugose coral calcimicrobe framestone (Fig. 1la, b); richthofeniid brachiopod baffiestone. -
-
-
The blocks have a Guadalupian age (Weidlich & Bernecker 2003), no evidence exists supporting the hypothesis of significant reef growth during Cisuralian or Lopingian times. The spatial distribution of the blocks remains problematical. Sponge reef blocks resemble those of the Batain coast and are not described in detail. Similar to the calcisponge Archaeolithoporella framestone, the chaetetid Archaeolithoporella framestone contains thick crusts of calcimicrobes. The rugose coral calcimicrobe framestone consists of a variety of subcommunities, including the Praewentzelella community, the cerioid coral community and the Waagenophyllum community (Weidlich & F1/igel 1995). The reef framework of the cerioid coral community is the most complex type, with a variety of ecological niches. For example, cryptic environment within or underneath the coral colonies were colonized by calcified metazoans and calcimicrobes (Fig. 1lb).
PERMIAN REEFS OF THE ARABIAN PLATFORM
247
Fig. 12. Photomicrographs of reef types of photozoan carbonates, Ba'id and A1 Jil Formations, Ba'id area. Arrows indicate the top of the sample. All scale bars are 1 cm. (a) Chaetetid Archaeolithoporella framestone with sphinctozoan sponge. The chaetetid is the recrystallized reef builder in the centre. (b) Calcisponge Archaeolithoporella framestone. Note the complex geometry of the cemented reef cavities. (e) Calcisponge Archaeolithoporella framestone with thick Archaeolithoporella-Shamovella crusts. Reef cavities have been filled with bioclastic packstone-grainstone, containing the dasycladacean alga Mizzia velebitana. (d) Calcisponge baffiestone in muddy matrix. Note that dark microbial crusts also occur within the sponge tissue.
248
O. WEIDLICH
Interpretation. Assuming an unbiased export of the blocks from the source area and an unbiased sampling procedure, algal cement reefs, sponge reefs and coral reefs are the most abundant reefs types, followed by phylloid algal reefs and richthofeniid brachiopod reefs. Reef carbonate production is typical of the chlorosponge association (Beauchamp & Desrochers 1997) indicating tropical and shallow water.
Climatic and oceanographic implications The tropical (photozoan) and the cool-water (heterozoan) factories contributed significantly to Permian carbonate production at the eastern margin of the Arabian plate and in the SW Tethys, while carbonate production was insignificant from the mud-mound factory. Unravelling in detail the control mechanisms of marine carbonate production is a core business of sedimentology and palaeontology. By combining microfacies, taxonomy, taphonomy and palaeogeographic reconstructions with computer modelling data, climatic and oceanographic changes become evident that favoured or hampered the precipitation of either heterozoan- or photozoan-dominated carbonates during the Permian: Cisuralian (Fig. 13a)
Sediment composition of the late Sakmarianearly Artinskian Saiwan Formation (Haushi area) and Cisuralian blocks of the Triassic Aseelah Unit (Batain coast) indicate that heterozoan bryonoderm carbonate production was dominant on the NW and SE flank of the Haushi uplift above storm wave base. Conversely, there is no evidence for significant Cisuralian photozoan carbonate production in the study area. A palaeolatitude of between 30~ and 40~ a phase of deglaciation and the prediction of cool-temperate oceanic water masses provide independent lines of evidence that the bryonoderm facies of the eastern Arabian plate and the SW Tethys reflects a cool-water environment. Reports of oolites (Blendinger et al. 1990) of the Saiwan Formation might indicate limited photozoan carbonate production above a thermocline WNW of the Haushi uplift (Artinskian oolites might have formed under similar conditions in Afghanistan; Vachard pers. comm.) or high-frequency climate cycles. Considering for the Saiwan Formation the inner ramp position close to the Haushi uplift above storm wave base, deep-water conditions can be ruled out as a control mechanism for the formation of heterozoan carbonates.
Guadalupian (Fig. 13b)
Photozoan carbonate production prevailed during the Guadalupian, with the exception of the sediments of the Khuff Formation in the Haushi area (Oman). Palaeogeographic interpretations agree that the eastern margin of the Arabian plate was between 20~ and 30~ palaeolatitude (Golonka et al. 1994) and that greenhouse climatic conditions prevailed. Climatic sensitive marine sediments, including sabkha evaporites of the Khuff Formation, document a hot and arid climate. Photozoan chloralgal carbonates of the Arabian plate (e.g. Khuff outcrops in Saudi Arabia, subsurface Khuff and Saiq Formation) yield a rich flora of calcareous algae, including many species of Gymnocodium, Permocalculus and Mizzia (Rezak 1959; Okla 1992, 1994; Vachard et al. 2005; Insalaco et al. 2006). With a more detailed perspective, however, the compositional variation of sediments from the eastern Arabian plate (Saih Hatat and Haushi area) and the SW Neo-Tethys provide evidence for a complex oceanographic pattern controlling sedimentation during the Guadalupian. Photozoan carbonate production dominated the SW Neo-Tethys (Ba'id area and Batain coast). The chlorosponge and chloroforam communities of the reef blocks indicate a healthy and productive tropical carbonate factory, with the highest taxonomic diversity of benthic communities in the SW Neo-Tethys. Chloralgal carbonates also dominate the Saiq Formation in the Saih Hatat, however, except for siliciclastic units, where small tabulate coral colonies flourished while terrigenous input was high. Most probably, clastic sediments were reworked during a rise in sea level and caused an increased turbidity of sea water, as well as a large supply of dissolved nutrients. These unfavourable oceanic conditions locally favoured heterozoan carbonate production (not plotted in Fig. 13b). The dominance of heterozoan carbonates of the Khuff Formation in the Haushi area is unexpected because of the overwhelming dominance of photozoan skeletal sediments elsewhere in the Khuff Formation on the Arabian platform. The presence of heterozoan carbonates in a tropical realm is explained best either by upwelling or by terrestrial input (palaeogeographic reconstructions rule out deep-water conditions, see Konert et al. 2001). Computer models predict a greater than 70% probability of upwelling at the eastern rim of the Arabian plate (Golonka et al. 1994). An increase of nutrients from upwelling is a mechanism to explain the local mixture of heterozoan and photozoan sediments, and explains the scarcity
PERMIAN REEFS OF THE ARABIAN PLATFORM
249
Fig. 13. Reconstruction, showing the controls of climatic and oceanographic changes on Permian carbonate production. Palaeogeography after Golonka et al. (1994) and Konert et al. (2001), water temperature (tropical v. temperate) after Ziegler et al. (1998) and upwelling areas after Golonka et al. (1994).
250
O. WEIDLICH
of calcareous algae and the dominance of bryonoderm in the Haushi area. Of interest are oceanographic modelling data (Ziegler et al. 1998), predicting the boundary between temperate and tropical sea water in this region. Lopingian (Fig. 13c) A Lopingian age for shallow-water platform carbonates from the Saih Hatat has been confirmed with the presence of Colaniella nana. As in the Guadalupian, palaeogeographic interpretations suggest a tropical realm (Golonka et al. 1994), and the coeval deposits of the Khuff Formation in the subsurface are exclusively dominated by carbonate-evaporite cycles. For the aforementioned reasons, the presence of Lopingian heterozoan carbonates at the eastern margin cannot be explained by cool water. Rising sea level can excluded as a potential control mechanism as well, because an amplitude of only a few metres would be expected during this greenhouse interval. Recently published data suggest upwelling of warm and saline oceanic bottom water that flooded the shelves during the Changhsingian transgression (Kidder & Worsely 2004). This is a potential mechanism that could have favoured carbonate production by the hetereozoan association. This hypothesis has to be tested using stable isotopes from pristine brachiopod shells or least-altered micrites.
Carbonate factories, reef frameworks and taphonomic processes Biotic and abiogenic reef-building processes and subsequent taphonomic alteration of photozoan and heterozoan reef factories were compared, as they are key control mechanisms determining the preservation of reef frameworks. Permian photozoan reefs are represented by mud-rich Tethyan reefs and cement-rich epeiric reefs (Weidlich 2002a, b). Applying the guild concept (see Fagerstrom & Weidlich 1999 and further references herein), reef building was dominated by the baffler and constructor guilds, including calcisponges, chaetetids, rugose and tabulate corals, bryozoans, as well as by dasycladacean, phylloid and problematic calcareous algae. Framework stabilization was maintained by encrusting sponges, chaetetids and bryozoans, sessile foraminifers and a variety of problematic organisms, including Shamovella, Archaeolithoporella and laminated microbialites, all of which belong to the binder guild. Local syndepositional cementation of radiaxial fibrous calcite and botryoidal cement was important for framework stabilization as well. Compared with
reefs rimming the landlocked Delaware and Zechstein basins, synsedimentary cementation processes played a minor role and biotic reefbuilding processes were more important. The resulting fabrics of reef blocks from the NeoTethys are complex when biogenic crusts (microbial crust, Archaeolithoporella and Shamovella) are present and the number of sheltered habitats increased for calcified metazoans (e.g. cryptic habitats). By contrast, Permian heterozoan reefs (bryonoderm association sensu Beauchamp & Desrochers 1997) are dominated by bafflers (crinoids and bryozoans) and commonly lack binders (especial microbialites). Textures are commonly baffiestone or floatstone-rudstone. Locally, incipient framestone patches occur. Although lime-mud is of overall importance as groundmass, radiaxial fibrous calcite is of local importance as well. This is probably a striking difference to modern and Cenozoic heterozoan carbonates, where synsedimentary cementation is less important. The comparison of photozoan chlorosponge and heterozoan bryonoderm reef fabrics exhibits significant differences with respect to ecological niches. Chlorosponge reefs consist of complex frameworks with a variety of ecolocical niches, if encrusters are present, while heterozoan reef communities are comparatively simple.
Conclusions This paper is aimed at the characterization of Permian shallow-water benthic carbonate production. Differentiation of photozoan and heterozoan carbonates of the Arabian plate led to a modified interpretation of depositional models, and is an attempt to interpret climatic and oceanographic controls on Permian carbonate production of the Arabian plate and Neo-Tethys. The existing model of tropical carbonates as carbonates of only shallow-water origin in this region has had to be revised as follows. ~ Close to the SE margin of the Arabian plate in the Haushi area, Cisuralian (late Sakmarianearly Artinskian) mixed carbonatesiliciclastic deposits of the Saiwan Formation and Guadalupian (Wordian) marl-carbonate alterations of the Khuff Formation are characterized by dominance of crinoid ossicles, brachiopods, bivalves and bryozoans. Scarcity of calcareous algae and invertebrates housing symbiotic algae indicate that members of the heterozoan association (bryonoderm facies) were responsible for carbonate production. Heterozoan-dominated
PERMIAN REEFS OF THE ARABIAN PLATFORM
9
9
9
9
carbonates of the Khuff Formation from the Haushi area significantly differ from the Khuff Formation elsewhere on the Arabian platform. Dolomitized Guadalupian carbonates of the Saiq Formation in the Saih Hatat exhibit a twofold composition: siliciclastic units that contain small tabulate corals are typical of the heterozoan association. Most of the Guadalupian Saiq Formation is dominated by carbonate of the photozoan association, containing fusulinaceans, smaller foraminifers, calcareous algae and a variety ofrugose corals (solitary, dendroid and compound taxa). Conversely, Lopingian carbonates of the Saiq Formation from the Saih Hatat are dominated by heterozoan carbonate production with crinoids, chaetetids, bryozoans and brachiopods in a lime-mud matrix and resemble heterozoan bryonoderm carbonates. Shallow-water carbonates of the Neo-Tethys, preserved as blocks in tectonic nappes in the Ba'id area and the Batain coast, are contrasting: in the Ba'id area Guadalupian reef blocks indicate photozoan carbonate production owing to the dominance of calcareous sponges, chaetetids, cerioid rugose corals, richthofeniid brachiopods and calcareous algae. Reef blocks from the Batain coast interpreted to be of Cisuralian age are dominated by the bryonoderm facies, while Guadalupian reef blocks from the same area show striking similarities to photozoan chlorosponge associations. The mentioned temporal and spatial variations of heterozoan carbonate production of the Arabian plate and the Neo-Tethys can be explained by cool-water conditions during the Cisuralian, local upwelling of nutrientrich water and siliciclastic input during the Guadalupian, and by upwelling of saline and warm bottom water during the Lopingian heterozoan. Chlorosponge reefs (photozoan association) consist of complex frameworks with a variety of ecological niches, if encrusters are present, while heterozoan reef communities are comparatively simple.
This study was financed by the German Science Foundation (project We180418-1,2). I thank the Ministry of Commerce and Industry in the Sultanate of Oman, especially the Director General of Minerals, Dr H. bin Mohamed A1-Azri for support. The enthusiasm of the organizing committee 'Climatic and Evolutionary Controls on Palaeozoic Reefs and Bioaccumulations', especially M. Aretz (University K61n), is greatly acknowledged. M. Bernecker (University Erlangen)
251
helped to improve the manuscript. The manuscripts benefited from important comments of the reviewers E. Vennin and D. Vachard. M. Bernecker improved an early version of the manuscript. D. Vachard helped to clarify biostratigraphic problems.
References AL-JALLAL, I. A. 1995. The Khuff Formation: Its regional reservoir potential in Saudi Arabia and other Gulf countries. In: AL-HUSSEINI,M. I. (ed.) The Middle East Petroleum Geosciences GEO'94, Volume 1. Gulf PetroLink, Bahrain, 103-119. ALSHARHAN,A. S. ~r NAIRN, A. E. M. 1997. Sedimentary Basins and Petroleum Geology of the Middle East. Elsevier, Amsterdam. ANGIOLINI, L., BALINI,M., GARZANTI,E., NICORA,A. & TINTORI,A. 2003a. Gondwanan deglaciation and opening of Neotethys: the Al Khlata and Saiwan Formations of Interior Oman. Palaeogeography, Palaeoclimatology, Palaeoecology, 196, 99-123. ANGIOLINI,L., BALINI,M., GARZANTI,E., NICORA,A., TINTORI, A., CRASQUIN,S. 8r MUTTONI, G. 2003b. Permian climatic and paleogeographic changes in Northern Gondwana: the Khuff Formation of Interior Oman. Palaeogeography, Palaeoclimatology, Palaeoecology, 191,269-300. BEAUCHAMP, B. ~r DESROCHERS, A. 1997. Permian warm- to very cold-water carbonates and cherts in northwest Pangea. In: JAMES, N. P. ~r CLARKE, J. A. D. (eds) Cool-water Carbonates. Society of Economic Paleontologists and Mineralogists, Special Publications, 56, 327-347. BLENDINGER,W. 1988. Permian to Jurassic deep water sediments of the eastern Oman Mountains: Their significance for the evolution of the Arabian margin of the south Tethys. Facies, 19, 1-32. BLENDINGER,W., VAN VLIET, A. & HUGHES-CLARKE, M. W. 1990. Updoming, rifting and continental margin development during the Late Palaeozoic in northern Oman. In: ROBERTSON,A. H. F., SEARLE, M.P. & RIES, A. C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 27-37. BOSENCE,D. W. J. & BRIDGES,P. H. 1995. A review of the origin and evolution of carbonate mounds. In: MONTY, C. L. V., BOSENCE,D. W. J., BRIDGES, P. H. & PRATT, B. (eds) Carbonate Mud Mounds. International Association of Sedimentologists, Special Publications, 23, 3-9. DiAZ-MART[NEZ, E., MAMET, B., ISAACSON, P. E. & GRADER, G. W. 2000. Permian marine sedimentation in northern Chile: new paleontological evidence from the Juan de Morales Formation, and regional paleogeographic implications. Journal of South American Earth Sciences, 131, 511-525. DUBREUILH,J., PLATEL,J. P., LE MI~TOUR,J., ROGER, J., WYNS, R., BI~CHENNEC,F. & BERTHIAUX, A. 1992. Geological Map of Khaluf, Sheet NF 40-15, Scale 1.250 000. Explanatory Notes. Ministry of Petroleum and Minerals, Muscat. FAGERSTROM, J. A. & WEIDLICH, O. 1999. Strengths and weaknesses of the reef guild concept and quantitative data: Application to the Upper
252
O. WEIDLICH
Capitan-massive community (Permian), Guadalupe Mountains, New Mexico--Texas. Facies, 40, 131156. GLENNIE, K. W., BOEUF, M. G. A., HUGHESCLARKE, M. W., MOODY-STUART, M., PILAAR, W. H. F. & REINHARDT, B. M. 1974. Geology of the Oman Mountains. Verhandlingen van het Koninklijk Nederlands geologisch mijnbouwkundig Genootschap, 31, 1-423. GOLONKA,J., ROSS, M. I. & SCOTESE,C. R. 1994. Phanerozoic paleogeographic and paleoclimatic maps. In: EMBRY, A. F., BEAUCHAMP,B. & GLASS, D. J. (eds) Pangea: Global Environments and Resources. Canadian Society of Petroleum Geologists, Memoir, 17, 1-47. GRADSTEIN, F. M., OGG, J., SMITH, A. G., BLEEKER, W. & LOURENS, L. J. 2004. A new Geological Time Scale, with special references to Precambrian and Neogene. Episodes, 27, 83-100. HAUSER, M., MARTINI, R., MATTER, A., KRYSTYN, L., PETERS, Z., STAMPFLI, G. & ZANINETTI, L. 2002. The break-up of East Gondwana along the northeast coast of Oman: evidence from the Batain basin. Geological Magazine, 139, 145-157. HUGHES-CLARKE, M. W. 1988. Stratigraphy and rock unit nomenclature in the oil-producing area of interior Oman. Journal of Petroleum Science, 11, 5-60. INSALAr E., VIRGONE, A. ~T AL. 2006. Upper Dalan Member and Kangan Formation between the Zagros Mountains and offshore Fars, Iran: depositional system, biostratigraphy and stratigraphic architecture. GeoArabia, 11, 75-176. IMMENHAUSER, A., SCHREUBS, G., GNOS, E., OTERDOOM, H. W. & HARTMANN, B. 2000. Late Palaeozoic to Neogene geodynamic evolution of the northeastern Oman margin. GeologicalJournal, 137, 1-18. JAMES, N. P. 1997. The cool-water carbonate depositional realm. In: JAMES, N. P. & CLARKE, J. A. D. (eds) Cool-water Carbonates. Society of Economic Paleontologists and Mineralogists, Special Publications, 56, 1-20. JELL, P. A. & WILLINK, R. J. 1993. Early Permian cladid crinoids from the Gharif Formation of Oman. Association of Australasian Palaeontologists, Memoir, 15, 305-312. KIDDER, D. L. & WORSLEY, D. 2004. Causes and consequences of extreme Permo-Triassic warming to globally equable climate and relation to the PermoTriassic extinction and recovery. Palaeogeography, Palaeoclimatology, Palaeoecology, 203, 207-237. KONERT, G., AFIFI, A. M., AL-HAJARI, S. A. & DROSTE, H. J. 2001. Paleozoic stratigraphy and hydrocarbon habitat of the Arabian plate. GeoArabia, 6, 407-442. KORTE, C., JASPER, T., KOZUR, H.W. & VEIZER, J. 2005. ~180 and ~13C of Permian brachiopods: A record of seawater evolution and continental glaciation. Palaeogeography, Palaeoclimatology, Palaeoecology, 224, 333-351. LEES, A. & BULLER,A. T. 1972. Modern temperate and warm-water shelf carbonate sediments contrasted. Marine Geology, 13, 1767-1773.
LEES, A. & MILLER, J. 1985. Facies variation in Waulsortian buildups. Part 2. Mid-Dinantian buildups from Europe and North America. Geological Journal, 20, 159-180. LE MI~TOUR, J., BI~CHENNEC, F., PLATEL, J.-P. & ROGER, J. 1994. Late Permian birth of the NeoTethys and development of its southern margin in Oman. In: AL-HUSSEINI,M. I. (ed.) The Middle East Petroleum Geosciences GEO'94, Volume 2. Gulf PetroLink, Bahrain, 643-654. LE MI~TOUR, J., MICHEL, J., B1~CHENNEC,F., PLATEL, J.-P. & ROGER, J. 1995. Geology andMineral Wealth of the Sultanate of Oman. Directorate General of Minerals, Ministry of Petroleum and Minerals, Muscat. MONTENAT, C., DE LAPPARENT, A. F., LYS, M., TERMIER, H., TERMIER, G. & VACHARD, D. 1976. La transgression permienne et son substratum dans le Jebel Akhdar (Montagnes d'Oman, P6ninsule Arabique). Annales de la Societk gkologique du Nord, 96, 239-258. MONTY, C. L. V. 1995. The rise and nature of carbonate mud-mounds: an introductory actualistic approach. In: MONTY, C. L. V., BOSENCE,D. W. J., BRIDGES, P. H. & PRATT, B. (eds) Carbonate Mud Mounds. International Association of Sedimentologists, Special Publications, 23, 11--48. NELSON, C. S. 1988. An introductory perspective on non-tropical shelf carbonates. Sedimentary Geology, 60, 3-12. OKLA, S. M. 1992. Permian algae and algal microfacies from Unazah, Quassim district, Saudi Arabia. Facies, 27, 217-224. OKLA, S. M. 1994. Fossil algae from Saudi Arabia revisited. Rivista Italiana Paleontologica Stratigraphica, 99, 441-460. PILLEVUIT,A., MARCOUX,J., STAMPFLI,G. & BAUD,A. 1997. The Oman Exotics: a key to the understanding of the Neotethyan geodynamic evolution. Geodinamica Acta, 10, 209-238. PRATT, B. R. 1995. The origin, biota and evolution of deep-water mud-mounds. In: MONTY, C. L. V., BOSENCE, D. W. J., BRIDGES, P. H. & PRATT, B. (eds) Carbonate Mud Mounds. International Association of Sedimentologists, Special Publications, 23, 49-123. RAO, C. P. 1981. Criteria for recognition of cold-water carbonate sedimentation, Berriedale Limestone (Lower Permian), Tasmania, Australia. Journal of Sedimentary Petrology, 51,491-506. RAO, C. P. 1988. Paleoclimate of some Permo-Triassic carbonates of Malaysia. Sedimentary Geology, 60, 163-171. REITNER, J. & NEUWEILER, F. 1995. Mud mounds: a polygenetic spectrum of fine-grained carbonate buildups. Facies, 32, 1-70. REZAK, R. 1959. Permian algae from Saudi Arabia. Journal of Paleontology, 33, 531-539. SCHLAGER, W. 2000. Sedimentation rates and growth potential of tropical, cool-water and mud-mound carbonate systems. In: INSALACO,E., SKELTON, P. W. & PALMER, T. J. (eds) Carbonate Platform Systems: Components and Interactions. Geological Society, London, Special Publications, 178, 217-227.
PERMIAN REEFS OF THE ARABIAN PLATFORM SCHLAGER, W. 2003. Benthic carbonate factories of the Phanerozoic. International Journal of Earth Sciences, 92, 445-464. SHACKLETON, R. M., RIES, A. C., BIRD, P. R., FILBRANDT, J. B., LEE, C. W. & CUNNINGHAM, G. C. 1990. The Batain Melange of NE Oman. In: ROBERTSON,A. H. F., SEARLE,M. P. & RIES, A. C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 673-696. SHARLAND, P. R., ARCHER, R. ET AL. 2001. Arabian Plate Sequence Stratigraphy. GeoArabia, Special Publication, 2. STEMMERIK, L. 1997. Permian (Artinskian-Kazanian) cool-water carbonates in north Greenland, Svalbard and the western Barents Sea. In: JAMES,N. P. 8r CLARKE, J. A. D. (eds) Cool-water Carbonates. Society of Economic Paleontologists and Mineralogists, Special Publications, 56, 349-364. STEPHENSON;M. H. & FILATOFF, J. 2000. Correlation of Carboniferous-Permian palynological assemblages from Oman and Saudi Arabia. In: AL HAJARI,S. & OWENS, B. (eds) Stratigraphic Palynology of the Palaeozoic of Saudi Arabia. GeoArabia, Special Publication, 1,168-191. VACHARD,D., GAILLOT,J., VASLET,D. & LE NINDRE, Y. 2005. Foraminifers and algae from the Khuff Formation (late Middle Permian-Early Triassic) of central Arabia. GeoArabia, 10, 137-186. VACHARD, D., HAUSER, M., MARTINI, R., ZANINETTI, L., MATTER, A. & PETERS, T. 2001. New algae and problematica of algal affinity from the Permian of the Aseelah Unit of the Batain plain (east Oman). Geobios, 34, 375-404. VACHARD, D., HAUSER,M., MARTINI, R., ZANINETTI, L., MATTER,A. & PETERS,T. 2002. Middle Permian (Midian) foraminifera assemblages from the Batain Plain (eastern Oman): Their significance to NeoTethyan paleogeography. Journal of Foraminiferal Research, 32, 155-172. VASLET, D., LE NINDRE, Y.-M. Er ,4L 2005. The Permian-Triassic Khuff Formation of central Saudi Arabia. GeoArabia, 10, 77-134.
253
WAHLMAN, G. P. 2002. Upper Carboniferous-Lower Permian (Bashkirian-Kungurian) mounds and reefs. In: KIESSLING,W., FLOGEL,E. & GOLONKA,J. (eds) Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 271-338. WEIDLICH, O. 1999. Taxonomy and reefbuilding potential of Middle to Upper Permian Rugosa and Tabulata in platform and reef environments of the Oman Mountains. Neues Jahrbuch fiir Geologie und Paliiontologie A bhandlungen, 211, 113-131. WEIDLICH, O. 2002a. Middle and Late Permian reefs distributional patterns and reservoir potential. In: KIESSLING, W., FLUGEL, E. & GOLONKA,J. (eds) Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 339-390. WEIDLICH, O. 2002b. Permian reefs re-examined: extrinsic control mechanisms of gradual and abrupt changes during 40 my of reef evolution. Geobios, 24, M6moire Special, 287-294. WEIDLICH, O. & BERNECKER,M. 2003. Supersequence and composite sequence carbonate platform growth: Permian and Triassic outcrop data of the Arabian platform and Neo-Tethys. Sedimentary Geology, 158, 87-116. WEIDHCH, O. & BERNECKER, M. 2006. Differential severity of Permian-Triassic environmentalchanges on Tethyan shallow-water carbonate platforms. Global and Planetary Change, 54. WEIDLICH, O. & FLUGEL, H. W. 1995. Upper Permian (Murgabian) rugose corals from Oman (Ba'id area, Saih Hatat): Community structure and contributions to reef-building processes. Facies, 33, 229-264. WEIDLICH, O., ERNST, A. & SCHAFER, P. 2004. Early Permian reefs form the southwestern Tethys (Batain region, Sultanate of Oman) - Cool-water evidence confirmed by bryozoan paleogeography, biotic composition, and reef taphonomy. In: 13th International Bryozoology Association, Concepci6n, Chile. ZIEGLER, A. M., GrosS, M. T. & HULVER,M. L. 1998. A mini-atlas of oceanic water masses in the Permian Period. Transactions of the Royal Society of Victoria, 110, 323-343.
Late Permian limestones and the Permain-Triassic boundary: new biostratigraphic, palaeobiogeographical and geochemical data in Caucasus and eastern Europe J. M. T H t ~ R Y 1, D. V A C H A R D 2 & E. D R A N S A R T 3
19 Jardin Gabrieli, 37000 Tours, France (e-mail.'jm.
[email protected]) 2UMR 8014, Universitb de Lille 1, 59655 Villeneuve d'Ascq Cedex, France 3 E M T T Etudes MOtallurgiques et Traitement Thermique, 1 avenue du Chater, 69340 Francheville, France Abstract: This paper presents biostratigraphical and palaeogeographical correlations, in a post-Hercynian and pre-Cimmerian tectonic framework, between stratigraphic sections of Late Permian age in two important and remote areas, with some considerations concerning neighbouring countries. It concerns mainly the Nikitin sequence (Kuban, Russia) and the Bfikk Mountains (Hungary), and describes the carbonate environments, microfauna diversity, foraminiferal assemblages, calcisponge and brachiopod constructions, and gymnocodiacean accumulations. This study is extended to former Transcaucasia (southern Armenia, Georgia and Adzerbadjan), the Alborz Belt in northern Iran, Italy and the Carnic Alps. It emphasizes a relatively simple biosedimentary evolution, which permits a confident palaeogeographic reconstruction. New geochemical results provided some additional markers, in particular small microspherules consisting of Cr/Ni spinels of cosmic origin. They occur in several sections in a defined position at the Permian-Triassic boundary based on biostratigraphic correlations (conodonts, foraminifers, last Permian reefal phenomena). These geochemical data might be related to a possible meteorite impact, Nevertheless, another global alternative phenomenon occurs. From the Latest Permian to the Earliest Triassic, the carbonate production or biomineralization is preferentially concentrated in the confined, intertidal or lagunal zones, and its evolution is relatively progressive from gymnocodiacean accumulations of the Latest Permian to the microbialites of the Earliest Triassic.
The Late Permian, prelude to the great biological crisis of the Permian-Triassic boundary (PTB), is located between two huge losses of the global carbonate production, respectively, from 89 and 99.99% (Weidlich 2002). Nevertheless, some sponge (sphinctozoans, inozoans, sclerosponges/ chaetetids) or brachiopod (richthofenids) reefmounds (see definition in Riding 2002) are known during this period, especially in Greece (Skyros and Chios Islands) and South China (Sichuan, Hubei, and Guangxi) (e.g. Flfigel & Reinhardt 1989; Weidlich 2002). Bioaccumulations, dominated by gymnocodiaceans red algae (Permocalculus with subordinate Gymnocodium) and dasycladales green algae Mizzia, are locally well known in the Middle East (southern Iran, UAE and Saudi Arabia: e.g. Okla 1992; Vachard et al. 2005; Vaslet et al. 2005) and western Europe (Bellerophon Limestones: e.g. No6 1987). The stromatolitic limestones associated with the Zechstein facies in the UK, Germany and Poland are also well known (Kiersnowski et al. 1995; Smith 1995). Other interesting areas with Late Permian carbonates are located between the
Btikk Mountains (Hungary) and the Transcaucasia in Armenia. These two key points for the palaeogeography of this sector of PalaeoTethys and/or Neo-Tethys were studied during the French programme of reconstruction of the northern Peri-Tethys at the PTB. More recently, we have also investigated other Late Permian segments exhibiting the PTB, in Greece (Eubea, Attica, Hydra), Turkey (eastern Taurus: Hazro section), Iran (Zagros: e.g. Koh-e Surmeh and Koh-e Dena sections) and South China (Guangxi: Laren section). The preliminary results (see Gaillot 2006) are congruent with the conclusions exposed in this work. The Late Permian is also particularly interesting because of the diversity of its physical and geological events, notably the assemblage of the Pangea (e.g. Erwin 1995; Weidlich 2002). Hence, the biotools permitting the control of the migration of the tectonic plates are very important to characterize the orogenic timetable of the successive accretions of terranes or plates. The benthic foraminifers are the most important of these biotools because they have no pelagic stages or
From:/~LVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations: Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 255-274. 0305-8719/07/$15.00 9 The Geological Society of London.
256
J.M. THI~RY ET AL.
pelagic larvae (e.g. Debenay et al. 1996), and need a continuous expansion of shelves, platforms or ramps to migrate from plate to plate. Moreover, foraminifers provide the highest resolution biostratigraphy in the CarboniferousPermian carbonate series because they are more abundant and evolve more rapidly than the conodonts. The problems of chronology concern principally the southern boundary of Armenia (Transcaucasia) and the Turkish Pontides, whose geohistory is still very controversial (Robinson et al. 1995). It is still questionable if Iran, as an archipelago of microplates or an independent plate, was accreted on Transcaucasia at the end of the Norian in a pre-Cimmerian orogenesis or only in the Middle Jurassic; the latter based on the age dating of the granites known from sutures. The purpose of this paper is to show some palaeobiogeographical correlations between western Palaeo-Tethys and western Neo-Tethys in eastern Europe thanks to some bioconstructions of richthofenids and calcisponges, some microflora of gymnocodiaceans and brief considerations about the foraminiferal content. Principal investigated areas are NW Caucasus, Transcaucasia and the Biikk Mountains, occasionally compared with neighbouring plates, microplates or tectostratigraphic terranes. A second topic is to mention the discoveries of microspherules, as possible evidence for a meteoric event at the PTB. This phenomenon could be related with the meteoritic impacts known at Araguainha, Brazil (Crosta et al. 1981; Th6ry et al. 2003), or at Bedout, NW Australia (Becket et al. 2004). The last point of discussion is to consider the Late Permian-Early Triassic interval, as a replacement period of gymnocodiacean bioaccumulations by stromatolitic constructions, i.e. as a relatively continuous phenomenon, despite the loss of the carbonate production and the recorded geochemical shifts.
Nikitin section, NW Caucasus (Russia) Previous w o r k
The Permian-Triassic succession at Nikitin is relatively well known, taking into account the numerous Russian palaeontological publications in the last 50 years (e.g. Miklukho-Maklay 1954; Kotlyar et al. 1999b; Pronina-Nestell & Nestell 2001). A good summary was given by Gaetani et al. (2005). Miklukho-Maklay (1954) divided the Late Permian series, from base to top, into: (1) the Kutan Formation; (2) the Nikitin Formation,
with three biozones; (3) the Urushten Formation; and (4) the Abag Formation, now considered to be Triassic in age. In fact, the three first formations are coeval and updated as late Changhsingian (Kotlyar et al. 1999b; ProninaNestell & Nestell 2001) and correspond, respectively, to terrigenous, calcareous and reefal facies, laterally equivalent. The late Changhsingian foraminifers were described by Miklukho-Maklay (1954) and recently redefined and/or re-illustrated by Pronina-Nestell & Nestell (2001). The fauna was revised by Kotlyar et al. (2004). The Nikitin section
The field section at Nikitin is upstream from the Malaya Laba River, tributary of the Kuban River (Figs l a, b & 2a). It is located in the Perodovoy Chain in the mountains of NW Caucasus (Kuban province, Russia) (Kotlyar et al. 1994, 1997, 1999b). At the base of the ravine at Nikitin, upstream from the Malaya Laba River (Fig. 2a), some leaf imprints Sphenophylllum oblongifolium Gemar & Kaulfuss, and Pecopteris polymorpha Brongniart (determination Broutin) were collected in the black shales by Mikerina and Pinchuk during our 1999 field trip. In western Europe they are generally dated as Westphalian D, but here they correspond to the Gzhelian stage, which is considered as equivalent to the late Stephanian. These black shales are overlain by few metres of Asselian red shales, followed by Asselian-Autunian conglomerates (Aksuatskaya Formation) with some carbonaceous remains (Fig. 2a). According to Kotlyar et al. (1997), this series corresponds to the German Lower Rotliegend. These conglomerates, approximately 10 m thick, are topped by alternations of limestone with shales of Murgabian age. This last zone of conglomerates corresponds to a nick point and cascade in a water torrent (Fig. 2a). The overlying Dorashamian-Changhsingian (i.e. the 'Kutanian', 'Nikitian' and 'Urushtenian' facies) is approximately 105 m thick (Kotlyar et al. 1997). The 'Kutanian' sandstone overlies the Early Permian Red Beds or older formations unconformably with an important stratigraphic gap. This unconformity corresponds to a Saalian swell, stretching along the former Hercynian Cherskyi suture (Fig. 1, and see also Fig. 4 later). The Nikitian facies is composed of black bioclastic limestones, rich in microfauna, alternating with shales and algal limestones. The smaller foraminifers are represented by the genera Neoendothyra, Dagmarita, Paraglobivalvulinoides, Hemigordius sensu lato, Agathammina, Nodosinelloides, Geinitzina, Pachyphloia,
LATE P E R M I A N EASTERN EUROPE AND CAUCASUS
257
.9
Z
r~
~ 0
, ~
~:.,~ _oZ Or..
,~
0
.,...
~e
~ . ~m
~N o,.~
X=
J.M. THI~RY ETAL.
258
-.I
t~
G
o
o,.-
~o
r~O ~
~
.~
0
--
0
'~
t9[ -
O0
0
oo
___
O0
~
~)
o~
~~176 \ ~
)'T
,T. , ......:, . . . . . .
x ~'
~,,~1t~ ::: iii o;o.o; oloi4olo~olo~ H-H~,~t>FI~FI ~;'~I~,l .... I k4 I..,~l-4~l.q>l-J>l
I
I
o
00~
~--~
N~ ~= "g c ; g E = o '-~o2
E
Z
~
"--
---
~
~
S~
o
= o
:
s ~[-
~
.=.-~~._~ :3
ILl
~
x
BBBBBB13 I~
~
~
o _~
r~
~.~ ~.o
ru
=
....~ " ~
.N~
8 -6
I~
.,..~
._.
-= e'~
Era
~~ -
"~
O N
~i ~
~ ~
. .. .-~ ,.'.'.~
---
~
._~
"~..~
~-
.__
.~
o
--
~
con" ILl l
~r~
\
'.'. f'-, ~ 2 _~
_~ 9r -
~Z ..~.=
~
v =
g. ~
.-
~
far)
E
~'~
me_
~
"~N~
~
.~'a 0~..~
o
.~N~__
BBBBB v ,:~
O..,O
~e
. ,..,~
e4 ~
LATE PERMIAN EASTERN EUROPE AND CAUCASUS
Robuloides, Lingulina, Pseudolangella, 'Frondina" and advanced Colaniella; and the fusulinids Palaeofusulina nana Likharew, P. wangi Sheng, and P. sinensis Sheng with also species of Reiehelina, Codonofusiella and Parananlingella (Kotlyar et al. 1999b; Pronina-Nestell & Nestell 2001). Our samples contained bituminous bioclastic packstones with Diplosphaerina ex gr. inaequalis (Derville), Paraglobivalvulina sp., Pseudovermiporella nipponica (Endo), Midiella zaninettiae (Altiner), Reichelina media Miklukho-Maklay, Palaeofusulina wangi and Nodosinelloides sp., with more than 60% of the gymnocodiacean red algae (predominant Permocalculus sp., at the stage Dzhulfanella, and subordinate Gymnocodium exile Mu). The Nikitian facies contains also some sponge mounds (Gaetani et al. 2005). The Urushtenian facies consists of bioclastic limestones with abundant algae Permocaleulus and brachiopods. The microfauna is rich and diversified, and, compared with the Nikitian limestone, is characterized by Lasiodiscus,
Pseudomidiella, Graecodiscus, Neodiscus, Multidiscus, Hubeirobuloides, Nankinella and Parareiehelina (Pronina-Nestell & Nestell 2001). One of our samples at the top of the Urushtenian facies ('sample n ~ 2' in Fig. la) contains
Gymnocodium
exile, Shamovella obscura Permocaleulus fragilis (Pia), Palaeofusulina wangi, Pseudovermiporella nipponica, Hemigordiellina sp., Midiella zanitettiae and Nodosinelloides sp. This sample demon(Maslov)
and
strates definitively the contemporeanous character of the Nikitian and Urushtenian facies (Palaeofusulina wangi has not been discovered in the Urushtenian facies by Pronina-Nestell & Nestel! 2001, table 1 p. 211), and indicates, at least locally, stable palaeoecological and, consequently, geochemical conditions up to the PTB. The brachiopod assemblage of this facies is composed of bioherm-building genera, especially Leptodus, Richthofenia and Permianella (Kotlyar et al. 1999b). The colonial rugose coral Waagenophyllum is also mentioned (Kotlyar et al. 1999b). The top of Urushtenian facies consists of burrowed mudstones. Our samples show local pyritizations and only one undetermined
Nodosinelloides. The Changhsingian Lagoon The facies are generally rich in organic matter and were preserved in a restricted environment. They could correspond with a bay or lagoon, on a rimmed carbonate ramp (Burchette & Wright 1992). In particular, the boreholes associated
259
with the field sections revealed a lagoon protected in the south (Fig. la, b). This lagoon stretches for at least 50 km towards the NW from the Belaya to Raskol-Cliff (Fig. lb). It generally exhibits an assemblage of algae, bryozoans, brachiopods, sponges, Shamovella obscura, Gymnocodium exile, Reichelina media and Geinitzina spp. Pronina-Nestell & Nestell (2001) described similar assemblages of foraminifers dominated by Palaeofusulina nana at Gefo, west of Nikitin (Fig. lb). The lagoon is limited towards the north by an emerged zone that was made clearly visible by the drilling at Bagovskaya (Figs lb & 2b). According to Pinchuk (pers. comm.), in this borehole the Jurassic deposits (1044m thick) rest on Permian deposits biostratigraphically characterized up to the Murgabian (middle Middle Permian) crinoidal sandstone by Platycrinites ex gr. laevis Millot. They were deposited on a sandy shoal where phyllitic components from the Precambrian metamorphic substratum and conglomeratic sandstones are also present. These metamorphic Proterozoic green slates (Fig. 1) were penetrated at the base of the drill hole up to -3500m (Fig. 2b). Nevertheless, the presence of the palynomorph Protohaploxypinus sp. and the smaller foraminifer Geinitzina cf. ovata Lange (Pinchuk & Mikerina pets. comm.), just below the Jurassic, would indicate some Late Permian deposits before the emergence and erosion (Fig. 2b). We agree with this hypothesis as we consider that 'Geinitzina' ovata is not a true Geinitzina but most probably a Pachyphloides or a very primitive Colaniella, both taxa whose FAD (first appearance datum) is Late Permian in age (Gaillot & Vachard unpublished data). Consequently, although limited by a fault (Fig. lb), the northern extension of the lagoon is probably entirely preserved. As another piece of evidence of the coastal line, we observed a reddish sandy molasse with granite and Palaeozoic components in the Rufabgo ravine, tributary of the Belaya River (Fig. l b). The horizon is dated as Early Triassic in the transgressive Yatyrgvart Formation. Pinchuk & Kotlyar (pets. comm.) have observed Ammodiseus minutus Dunn here (Fig. l a) and a perhaps locally characteristic Nodosaria, as in the neighbouring section at Balka Svinachay near the Sakhray River (Fig. la, b). We suppose that these two foraminifers might be, respectively: (a) some primitive, planispirally, undivided, tubular cornuspirid miliolids, which are the first to reappear after the PTB, probably Rectocornuspira kahlori Br6nnimann, Zaninetti & Bozorgnia, mentioned as occurring immediately on the PTB in several localities of southern Turkey and
260
J.M. THI~RY ETAL.
southern Iran; and (b) a nodosarioid, relatively common in the same levels, the genus Polarisella (see Vachard et al. 2005; Gaillot 2006).
The reefal phenomenon An important reefal barrier stretched for a length of at least 50-70 km south of the lagoonal area. It settled there with up to 70-100 m of buildup, between the Belaya, Bolshaya and Malaya Laba river basins. The reefal sequence spread out into the Urushtenian facies with bioclastic, biogenic limestones with algae, foraminifers, sponges, bryozoa and reef-building brachiopods. Two field sections of late Changhsingian correspond to the back-reef area at Nikitin ravine and at Raskol-Cliff, west of the Belaya River at the presumed end of the lagoon. Some field sections have been studied at Khuko in the Main Range, 36 km N N W of Soci; we found there in an offreef area olistolites of late Changhsingian sediments. Near the Nikitin section, Severnaya ravine already corresponds to an off-reef field section only 2 km SSE from Nikitin (Fig. lb). The Urushtenian facies provides the best classification concerning the reefal phenomenon as a result of our field sampling at Nikitin ravine. There, beds of bioclastic Urushtenian limestones are massive, layered and about 5 m thick. The limestones have inclusions of marl limestone that contain accumulations of small foraminifers. Associated brachiopods bioherms of Richthofenia, often with Crurithyris or Leptodus, are repeated in association with sponges, and bryozoa in most of the beds. Some accumulations of algae Gymnocodiaceae (Permocalculus, Gymnocodium) and Dasycladales Mizzia form another characteristic biota.
Palaeobiogeographic compar&on with Transcaucasia and southern Crimea The best known Middle-Late Permian Transcaucasian sections are those at Arpa (southern Armenia), Sovetashen (Armenia) and Djulfa (Adzerbadjan) (Figs 3 & 4). The top of the Dorashamian-Changhsingian contains cherts and silicifications at Arpa and Khashik, and red marls at Sovetashen (Rostovtsev & Azaryan 1973; Alekseev et al. 1983; Zakharov et al. 1996). The different taxa of foraminifers clearly suggest the different palaeobiogeographical assignments of NW Caucasus and Transcaucasia during the whole of the Permian. In particular, the fusulinids Orientoschwagerina abiehi MiklukhoMaklay and Eopolydiexodina persica (Kahler) mentioned by Akopian (1974) and Kotlyar et al.
(1989) in Armenia also exist in Iran (Abadeh: see Kobayashi & Ishii 2003a, b), many species of Codonofusiella are common between Armenia (Kotlyar et al. 1984) and Abadeh (IranianJapanese Research Group 1981), and the genus Pseudodunbarula is known both in Transcaucasia (Kotlyar et al. 1984) and in the Abadeh area (Mohtat-Aghai & Vachard 2005). The smaller foraminifers identified by Pronina (1988) and Mohtat-Aghai & Vachard (2005) are also similar. Contrary to NW Caucasus, Changhsingian reef mounds are apparently lacking in Transcaucasia. Stromatolites are known in Armenia at the base of the Triassic, in all probability correlatable with those of Abadeh recently re-described by Heydari et al. (2000, 2003). During the Triassic a phase of rifting is particularly well known from the Carnian (Trachyceras Zone) during which, according to Vuks (2000), the subsidence is at a maximum, as previously supposed by Nikishin et al. (1998a, b). This pre-Norian phase precedes the collision of the Transcaucasian plate with the northern boundary of the Caucasian chain of the Cimmerian orogenesis. Subsequent to an emergence, creating an unconformity, a Jurassic suite of volcaniclastic sediments was deposited. The emergence is even more pronounced south of the Caspian, in the Alborz Mountains, by a layer of laterite (Stampfli 1978). Another important field section was studied by the first author at Dizi (Fig. 4, see later), in the region of Svenetis Kedi in Georgia, NE Transcaucasia. It is located at an altitude of about 3000 m on the banks of the high Inguri River, 80 km east of Suchimi. Located south of the Main Range of the Greater Caucasus, it appears to belong to the northern boundary of the Trancaucasian Terrane (Georgian Block: Khain 1982). The section shows a continuous sequence from the Eifelian to the Triassic epochs and is separated from the Jurassic by an unconformity. According to the work of the Oceanological Institute of Moscow (Kazmin pers. comm.), these series are folded and were overturned to the north, as at Nikitin, during the pre-Jurassic Cimmerian orogeneses, and are here metamorphosed to the greenschist facies. In the lower series, the jaspers and marbles are dated by conodonts as Devonian and Early Carboniferous. These horizons correlate with the levels known in southern Armenia at Danzik near Arpa (Fig. 3) and dated by Bonnet (1947). This upper part of the sequence also contains sandstones, often coarse grained, undoubtedly belonging to the Triassic. They cover a lower complex of Palaeozoic, which presents a chaotic facies of
LATE P E R M I A N EASTERN EUROPE AND CAUCASUS
261
Fig. 3. (a) Recent geological structures with an attempt at a palaeogeographical extension of Tethys Ocean during Dorashamian time in South Armenia area. (b) Sovetashen Permo-Triassic section according to Aleskeev et al. (1983).
262
J.M. THI~RY ET AL.
slides and relics of mudflows formed at the continental slope (Kazmin & Sborshchikov 1989). Their Permian olistolites are similar to those of the Taurida Formation of the Crimea, recently re-examined by Kotlyar et al. (1999a). Moreover, Taurida of Crimea and Transcaucasia have many palaeobiological points in common during the Middle-Late Permian owing to: (a) the diversity of the Parafusulina sensu lato (i.e. Skinnerella, Paraskinnerella and rare true Parafusulina) (compare Kotlyar et al. 1999a and Akopian 1974); (b) the presence of the neoschwagerinids genera: Cancellina, Armenina, Neoschwagerina ex gr. simplex and Praesumatrina (compare Kotlyar et al. 1999a and Leven 1998); (c) the diversity of Eopolydiexodina (compilation in Vachard & Bouyx 2002); and (d) the absence of Changhsingian reef mounds, although richthofenids and calcisponges have been present in Crimea since the Midian (Kotlyar et al. 1999a). The Dizi zone may mark the margin of the Georgian block and the proximity of the Hercynian suture of the Caucasus with the Russian platform (Fig. 4). Consequently, the Hercynian sector of NW Caucasus appears biogeographically separated from the microplate Crimea Taurida-Georgian Block-Lesser Caucasus-Armenia-Djulfa-Alborz. This microplate was initially named 'Extragondwan Realm' by Vachard (1980), but is most generally described in the literature under the name of 'Cimmerian Terranes'. The microplate Crimea-TranscaucasiaAlborz is also characterized by PalaeozoicTriassic olistolites present in Armenia at Sevan Lake, where a Permo-Triassic rift appears to exist (Bonnet 1947; Stampfli & Pillevuit 1993). Here a thin crust (Ershov et al. 1998) was formed, which was uplifted in front of a Cimmerian magmatic belt. A back-arc Permian-Triassic basin extended south of Sevan Lake, at the northern margin of the Neo-Tethys Ocean. This area showed a great seismic and volcanic instability, which even expressed itself in Permian-Triassic sedimentation. This instability was again perceptible in Holocene times by intense volcanism (Karakhanian et al. 2002, fig. 3). The Armenian and Georgian blocks have aggraded against one another since the Bajocian (Khain 1982) and, in turn, against the Russian platform, north of the Dizi area. The Cimmerian Orogeny is thus recognized more or less at the base of the Norian in the three regions studied above: Kuban, Georgia and Armenia. It corresponds to an extension of the Neo-Tethys before its major expansion during the Bajocian-Callovian (second phase of the Cimmerian Orogeny). We suggest that the
Palaeotethys probably closed partially at the base of the Late Permian (Saalian Orogeny), as indicated by the pre-Murgabian conglomerates observed in the Nikitin section, in Bulgaria (Yanev 2000), south of the Caucasus, and in the Alborz Mountains (Stampfli 1978; Stampfli et al. 1991). In the Nessen Formation (Latest Permian) in this region (Lys et al. 1978), the microfauna and algae appear to be very similar to those in certain southern Armenian sections (Khashik: Pronina 1988).
Biikk Mountains (Hungary) and neighbouring countries In Hungary, the PTB is known from the Biikk Mountains and the Transdanubian Range (e.g. Haas et al. 2006 with bibliography) (Figs 5 & 6a). The Bfikk Mountains are located in NE Hungary, near the border with Slovakia and Ukraine (Fig. 6). There, Balogh (1980) divided the Late Permian-Early Triassic series, respectively, into Nagyvisny6 and Gerennav~r limestones. The Nagyvisny6 Limestone (250-280 m thick) consists of black and dark grey, thinbedded limestones with shaly intercalations. The Gennevfir Limestone (110-140 m thick) contains, at its base, successively 50 cm of mudstones or Earlandia wackestones (with the conodont Hindeodus parvus recognized as characteristic of the PTB) and 8.5 m-thick stromatolites (Hips & Haas 2006). We have studied several field sections in the Btikk Mountains (Fig. 6a). Other good exposures of the PTB are located in Serbia, west of Belgrade, as well in a bore core at Gartnerkofel 1 (Austria) (Fig. 6a). The latter corresponds to the Bellerophon Formation, which appears in the Latest Permian as a significant bioclastic limestone with gymnocodiacean algae. In the Btikk Mountains (Nagysvisny6), the microfauna were accurately described from black bituminous limestone by B6rczi-Makk (1992) and B~rczi-Makk et al. (1995). We found in our samples accumulations of Permocalculus with rarer Gymnocodium, and foraminifers Earlandia, Paraglobivalvulina, Septoglobivalvulina? ex gr. decrouezae K6yluoglu & Altiner, Dagmarita, Paradagmarita, Shindella?, Pseudovermiporella, Hemigordius, Multidiscus, Geinitzina, Pachyphloia and Robuloides. Finally, we found only Changhsingian assemblages in our samples from the Btikk Mountains, contrary to B~rcziMakk (1992), which emphasized a Midian age for the series, or a very reduced DorashamianChanghsingian part (B~rczi-Makk et al. 1995). This Changhsingian age is indicated by: (a) the
LATE PERMIAN EASTERN EUROPE A N D CAUCASUS
263
+.;
. .,..~
o
9
b..,
O
et0
+,.,
~ ,,,,~
~ ,,,,~
,4
264
J.M. THl~RY E T
_<
c
03~
.m
0~i
~
~5
v
c
._= =~ =~ .=~
._v
v
c
E c~ ~=._
z
< []
AL.
~
=.~ ~
~rr
m
~" CO o.E
..4 -~,-~ .,.,0 .,~ . - . ~ ..-. ~ o
~_=~
~_~~ ~ z v
.--
I~-
=
o~
o 0
w
"--n
COO
9 CY
"~
"E
3
s 0_
~9
ccC _c 0~o c-
>n,' <
m
.._..
[-
x: 03
0
(.9
b
Z
c
7-
_o ~ _ o . ~ -o ~- o o 0)-(3
b~ C
q)~o
=s .~
00
~ob v 7-
9
Z
O {b
n~ I---
(-
~m >..
~
z <
~ ~ ~ ~.~.~
c
CO:
Z
.c
7-:o.mE
0,
"5 "~
,L, o > $
o
.c_ r
9-~
c O Ocxl
cOo
~
~-
N
:=l
"7
rr~ o.
mOc~
m
&
o
c
~'o O
!
c~
c
9
~ o
~o 0
t-~ I--
rn .~_ 0 LL
t,_
>-
4~
o
O ~.~ o
<
"~
}-
O
c o ~_
~.~ ~ ~
o') ~ -
=.q" o
8
oo-~ v <1-
~.-
e CO
o
rn
c o
rfl 1-13_
(.-
r,D " ~
,--
0 E3
0!sse!J1/~lJe:l
ue!wJecl a),el
~0
LATE PERMIAN EASTERN EUROPE A N D CAUCASUS
265
"7
O
O
o
0
~
.
m O ~'~
' ~
2~
266
J.M. THI~RY ET A L.
presence of a primitive Paradagmarita (denominated Globivalvulina cyprica by B6rczi-Makk et al. (1995, plate 6, fig. 4); (b) Shindella? sp. (juvenile or dwarf specimens identified as Codonofusiella nana Erk by B6rczi-Makk et al. 1995, plate 10, figs 1-6, and Codonofusiella spp., plate 11, figs 1, 2?, 5 & 6?) since the base of the series; (c) advanced representatives of Globivalvulina decrouezae that look like Paraglobivalvulina nitida Lin, Li & Sun from South China (B6rczi-Makk et al. 1995, plate 6, figs 1, 2, 3?, 5, plate 7, figs 1-5); and (d) Multidiscus with striated walls (B6rczi-Makk et al. 1995, plate 16, fig. 3 & 4, plate 20, figs 1, 2), plate 21, fig. 3) previously only observed by us in the late Changhsingian of Greece (Vachard et al. 1993, plate 8, figs 16-18). Nevertheless, two other Changhsingian markers: Colaniella (B6rcziMakk et al. 1995, plate 1, fig. 12) and the Kamurana? sp. (B6rczi-Makk et al. 1995, plate 14, figs 1-3) seem to be misinterpreted. According to Balogh (1964), the brachiopod reefs with Tchernyschevia are shared by both areas, Hungary and N W Caucasus. Although some authors have previously correlated the Biikk Montains with the NW Caucasus in the Changhsingian (Balogh 1964), more similarities appear between the Hungarian foraminiferal and algal assemblages and those of Taurus, Zagros, Saudi Arabia (Vachard et al., 2005; Gaillot 2006) than those of NW Caucasus. These similarities are as follows: (a) Hemigordiopsis renzi Reichel should be associated with Rectostipulina quadrata Jenny-Deshusses (B6rczi-Makk et al. 1995, text-fig. 10, p. 212); as this later species is considered by all the authors as characteristic of the Late Permian, 'Hemigordiopsis renzi' is most probably a new genus discovered by Gaillot (2006) in the Taurus and Zagros; (b) 'Globivalvulina cyprica' is, in fact, a very primitive Paradagmarita, early Wuchiapingian in age (Gaillot 2006); (c) Paraglobivalvulina mira is part of the group Septoglobivalvulina? decrouezae, well known in the Taurus and Saudi Arabia (e.g. Vachard et al. 2005); and (d) the return of the old genus Climacammina at the end of the Changhsingian (Gaillot 2006). Nevertheless, the bioconstructions with richthofenids and calcisponges similar to NW Caucasus are present in Biikk Mountains, as well as in Skyros Island (Greece). Hence, we suggest that Greece and B/ikk constitute microplates situated in the centre of the Palaeo-Tethys, but submitted to a northern system of currents, allowing communication with South China because of the very limited, on the world scale, late Changhsingian reef development. Another
marker of the links between South China, NW Caucasus and Greece is the fusulinid genus Parareichelina Miklukho-Maklay emend (Rauzer-Chernousova et al. 1996). Indeed, it is known only from South China (Wang 1974), NW Caucasus (Rauzer-Chernoussova et al. 1996) and in the Salamis Island in Greece (Altiner & Ozkan Altiner 1998, under the synonym name Baudiella) (Fig. 7). West of the Biikk Mountains, the biodiversity of the Late Permian microfauna is poor (or at least poorly known). But some algae can be palaeogeographical markers of the PeriHercynian Domain, e.g. Macroporella preromangica Praturlon (see No6 1987, plate 29, fig. 4). Data are given from the sections of NE Serbia (Pantic-Prodanovic 1996) at Krupanj, as well as at Jadar (Pesic et al. 1986; PanticProsanovic & Radosevic 1981); in the Vlasic zone, in the Panica Cave, and at Gaj, where the reefal coquinas with the brachiopod Richthofenia are overlain by the same abundant assemblage with Geinitzina and Permocalculus spp. at the end of the Permian. West of Villach (Carnic Alps, Austria), the drill core of Gartnerkofel 1 (Figs 5 & 6) clearly shows the PTB (Holser et al. 1991). At the top of the Tesero horizon, a bioturbated biomicrite and wackestone-mudstone are present in an internal platform with Hemigordius and Globivalvulina. As in Serbia, this level changes at the base of the Triassic into oolitic beds, as found in NE Hungary (Haas et al. 1986). The same facies is found at Gerennav/tr in the Biikk Mountains, but there the microspherules are of cosmic origin.
New geochemical data at PTB Cosmic microspherules
If compared to the K-T (Cretaceous-Tertiary) crisis, a catastrophe on a global scale that has led to the well-known mass extinction in a relative short time, the PTB would contain, in its deposits of many localities, microspherules of a cosmic origin. In our samples of Nikitin, the first ones were found by EMTT in the Urushtenian limestones at the top of the Late Permian. They are small agglomerates of microspherules consisting of Ni or Cr spinels (Fig. 8). They could be the result of the fusion at the surface of a giant meteorite, such as the one known from Araguainha in Brazil or two candidates of the same period at Bedout in Australia (Becker et al. 2004) or at Kursk in Russia (Masaitis 1999). The latter one is at a distance of 940 km from the Kursk impact and provides a remarkable layer of shocked minerals, traces of which we ought to find at
LATE PERMIAN EASTERN EUROPE AND CAUCASUS England
267
A ' Zeo,;,e,.,Sy
( t . ~k, ~,
'
'
_.)
Poaa.d
i Transdanubian
Range
I aria
~
~ | |
/
Kazakhstan
~!
Italy YTA
(Tunisia)
man (Turkey)
(Iran)
GONDWANA Fig. 7. Palaeobiogeographical map of Late Permian plates and microplates: 1, mountains; 2a, mainlands; 2b, seas; 3, late Changhsingian reefs; 4, Parareichelina localities; 5, possible migration ways.
Nikitin. The hypothesis of a meteoritic fall at the PTB has generally been abandoned (see Weidlich 2002), but is it still advocated by some authors (Retallack et al. 1998; Farley & Mukhopadhyay 2001; Kaiho et al. 2001; Becker et al. 2001, 2004; Basu et al. 2003; Erwin 2003; Koeberl et al. 2004). These microspherules can also be observed in Transcaucasia and the Raskol-Cliff in the NW of the Caucasus (sample 1, Kotlyar et al. 1997; see also Zakharov et al. 1996). Similarly, Bachmann & Kozur (2003) discovered microspherules of cosmic origin at Nelben (Germany). Some consist of spinels with 2% Ni and 2% Cr, and up to 0.17% Ir. They are coeval with the FAD (first appearance datum) of the conodont Hindeodus parvus, which is the main marker of the base of the Triassic (Kozur 1981, 1999). In Hungary we have also collected a sample in the reefal limestone located at the top of the Changhsingian stage. Fluorescence analyses reveal a content of 17 ppm Ni and 65 ppm Cr. Magnetic particles were separated by EMTT from the spinel microspherules. The oxide concentrations are: FeO = 86%, A1203= 6%, SiOz = 1%, CrzO~= 3% and NiO = 4%. These spinels were formed during the fusion of a meteorite while crossing the
atmospheric layers (Robin & Rocchia 1998), and whose impact must have taken place exactly at the PTB. Isotopic anomalies
Important geochemical and isotopic anomalies, revealed by our samples, concern particularly the organic 6~3C and the Ca/Mg ratio. At Raskol-Cliff (west of the Belaya Rivet'), Ca/Mg reaches a maximum of 193 200 at the top of the Urushtenian facies. This maximum is comparable with that in Transcaucasia. The highest value of 6~3C is reached near Djulfa, in the Araxoceras zone, with 2.5% against -0.1%, with a maximum of 0.1%, in the lnduan. This would translate, at the PTB, in an abrupt decline in the 613Cand with this a decline in the photosynthesis and bioproductivity of the sea with a deficiency in heavy oxygen (6'80). After a brief period of a cold climate, without doubt owing to a gigantic basaltic eruption in Siberia (Norilsk), with a maximum ~13C, one observes in the late Induan substage an abrupt increase in temperature. According to recent data (Schultz 2005), the Middle Triassic would have been one of the hottest periods governing the Earth.
268
J.M. THI~RY ET AL.
Fig. 8. Microspherules consisting of Cr or Ni spinels at Nikitin (Kuban, Russia) and in the Biikk Mountains (Hungary) sampled in the PTB deposits. (a) & (b) External view of spherules found at Nikitin. (c) & (d) Polished cut-out of the internal structure of spherule from the Biikk Mountains. Several investigations of gl3C and g~80 have led to the same conclusions at the PTB (especially at the base of the Mazzin Formation) (Fig. 5), as in Transcaucasia, Kashmir and Nepal (e.g. Holser et al. 1991; Baud et al. 1996). The variation of their content falls to a minimum during a period close to 3 Ma. This would be the result of an increase in aridity, accompanying a lowering of ocean levels, followed by a flow of carbon gas in the atmosphere coming from Siberia. In the upper Mazzin Formation, an abrupt increase in temperature of 5 ~ and a rise in sea level took place in a transgression, which occurred worldwide. These different phenomena would have been the only ones recognized as responsible for the greatest extinction of the Phanerozoic. Carbonate productivity
Correlatively, a palaeoclimatic proxy can be provided by the accumulations of Permocalculus
remains. They are already present in the Midian microfacies, e.g. in Djebel Tebaga (Tunisia) (Glintzboeckel & Rabat6 1964; Termier et al. 1977; Vachard & Razgallah 1993). In fact, they are present from the Kubergandian-Roadian in Aghanistan (Vachard 1980; Vachard & Montenat 1981: with the so-called Dzhulfanella). The presence of these Permocalculus accumulations in the Lopingian, after the important loss of carbonate production mentioned by Weidlich (2002), is puzzling. Permocalculus are very littoral algae (Vachard et al. 2003). Our hypothesis is that, whereas the carbonate rate of the sea waters greatly decreased in the ocean, far off the coasts, the innermost platform remained relatively favourable to the carbonate production. Hence, after the disappearance of carbonate microfauna and microflora at the PTB, the carbonate environments were confined to the intertidal zone, dominated by stromatolites.
LATE PERMIAN EASTERN EUROPE AND CAUCASUS The conditions were similar to those of the Early Palaeozoic rather than an environment for crisis survivors of the microbial reefs or 'anachronistic facies' (contrary to Sepkoski et al. 1991; Schubert & Bottjer 1992; Baud et al. 1997), although these are present in many bases of Triassic sequences from SW Turkey to South China (Heydari et al. 2000, 2003; Baud et al. 2005; Mohtat-Aghai & Vachard 2005). We prefer the hypothesis of Riding & Liang (2005). These authors explain that the abundance of microbial carbonate is inversely proportional to that of the metazoans or, in this case, of the advanced red algae. These conditions existed first in the Zechstein carbonate equivalents with their stromatolites (Kiersnowski et al. 1995; Smith 1995). Riding (2005) indicated that the reefal microbial carbonates were already abundant prior to the end-Permian period owing to an elevated seawater saturation state. In the stromatolitic zone, i.e. upper intertidal-mediolittoral zone, stromatolites, thrombolites and other microbialites are associated with particular cements (Heydari et al. 2000, 2003; Heydari & Hassandzadeh 2003; Baud et al. 2005). The Triassic system appears in this case as a reconquest of the oceans by the carbonate production. This progressive change of the carbonate geochemistry can be accelerated by other shifts (Magaritz et al. 1988; Heydari et al. 2000; Heydari & Hassandzadeh 2003), and the influence of diverse falls: meteorite, sea level, temperature or oxygenation (see the summary of Heydari et al. 2003); or gaps: reef, chert or coal (see the summary in Pruss et al. 2005).
Palaeobiogeographical correlations According to the geodynamic syntheses about Bulgaria, Slovakia, Hungary and Serbia (Balogh 1964; Pesic et al. 1986; Vozar & Vozarova 1988; Pantic-Prodanovic 1996; Yanev 2000; Bachmann & Kozur 2003), correlations exist between these areas and the zone NE of the Caucasus. The work of Lemaire et al. (1998) on Poland and the Tornquist rift, as well as that of Nikishin et al. (1998a, b) on the Caucasus, show the tectonic relations with the neighbouring countries of the Black Sea of Seng6r et al. (1984), assurring good correlations on the northern periphery of the European Peri-Tethys. Balogh (1964) considered the Lvov (L'viv, Ukraine) Gulf the most likely connection; moreover, the Teisseyre-Torquist Gulf could fulfil the same role. The studies by Vozar & Vozarova (1988), Karamata et al. (1994), Lamarche et al. (1998) and Yanev (2000) show the impossibility of this interpretation for tectonic and magrnatic reasons, and because of
269
the presence of evaporites at the two ends of these gulfs. Near Varna in Bulgaria and at Zemplicum in the NE of Slovakia, close to the Bfikk Mountains in Hungary, the same types of volcano-sedimentary strata (rhyolites, latites) were deposited during the Latest Permian. According to Yanev (2000), the Rhodope, Moesia and Tisza terrains or plates were converged together north of Gondwana at the end of the Devonian and entered into collision much later during the Late Permian subsystem. The final tectonic phase was the pre-Cimmerian equivalent to the one south of the Russian platform. According to Kotlyar et al. (pers. comm.), the closest relationships start from the Late Permian and become more refined during, or at the base of, the Norian stage. The species Miliolipora cuvillieri Br6nnimann & Zaninetti, found by Vuks (2000) in the Balka Svinachay (Sakhray River) with Involutina of the Norian, can be correlated with the Balkans (Budurov & Trifonova 1994). As already indicated by Banks & Robinson (1997), based on strike-slip faults that traverse the Black Sea from NW to SE, the pre-Norian and Cimmerian tectonic phases could correspond to those in the Balkans at Alg61 (Pontides) followed by the Taurides and those in turn with the NW Caucasus (Kotlyar et al. 1999a). Nevertheless, these correlations are not demonstrated during the Late Permian. Hence, our work attempts to provide some palaeogeographical data (Fig. 7). The NW Caucasus, which belongs to the Peri-Hercynian Domain (Vachard 1980), is closely related with South China in the late Changhsingian, but its foraminiferal algal and foraminiferal population differ considerably from those of Transcaucasia, which are correlable with southern Crimea and Alborz Mountains. This unit belongs to the Extragondwanian Domain (Vachard 1980; equivalent to the Cimmerian blocks of the authors). The Bfikk Mountains and Greece are poorly constrained, but seem to belong to one or two independent microplates. The Perigondwanian Domain is considered here following the interpretation of Vachard et al. (2005). It extends to the Tebaga in Tunisia. The eastern border of the Palaeo-Tethys-Neo-Tethys confluence is constituted by Italy and the former Yugoslavia. Within the Peri-Hercynian Domain two confined seas existed: the Zechstein Sea and the Bellerophon Sea. The proposed palaeogeography is consistent with the distribution of the Permian markers firstly emphasized by Seng6r et al. (1988), namely Eopolydiexodina, Shanita, Palaeofusulina and Colaniella.
270
J.M. THI~RY ET AL.
Conclusions 9 The discovery of new representatives of Palaeofusulina wangi confirms the coeval character of the Nikitian and Urushstenian facies of the Malaya Laba River (NW Caucasus, Russia), Nikitin (Russia) and their assignment to the late Changhsingian. The 'Nikitian' facies corresponds to a large lagoon, behind the 'Urushtenian' small reefmounds with sponges and/or brachiopods. Similar bioconstructions are only known in the Greek Islands and South China. ~ Owing to the presence of very particular late Changhsingian reefs and some foraminifers, like Parareichelina (= Baudiella), the correlations of N W Caucasus assemblages extend further towards the east in a large part of the Palaeo-Tethys, up to South China (Meishan section) and to the south up to Greece (Salamis Island). ~ Late Changhsingian reefs are lacking in Transcaucasia, and the foraminiferal microfauna of N W Caucasus and Transcaucasia differ palpably. Those of Transcaucasia have more in common with the assemblages from Alborz and Abadeh areas (Iran), and prove the discontinuity of the carbonate platform southwards, and consequently the independence of the blocks. 9 The identity of the Permian biological assemblages can be extended from Crimea to Alborz. This unit is the Permian Extragondwan Domain, characterized by the identity of the populations and the absence of Hercynian orogenesis. ~ The series of the Biikk Mountains is probably only Changhsingian in age, and no evidence of the Midian age can be provided. The Biikk Mountains appear as a transitional territory between N W Caucasus and Transcaucasia because they possess Changhsingian reefs, but their foraminifers resemble more closely the Perigondwan assemblages. 9 In the studied areas, microspherules are often discovered at the PTB and can provide arguments against the meteorite hypothesis. ~ In our Nikitin samples, agglomerates of microspherules consisting of Ni or Cr spinels were found at the top of the Urushtenian limestones. They can be caused by fusion at the surface of a giant meteorite, such as the ones known from Araguainha in Brazil, at Koursk in Russia or Bedout in Australia, and in this case the PTB is very abrupt. 9 Another common characteristic is the location of carbonate bioaccumulations. The
gymnocodiacean Permocalculus accumulations are common everywhere from N W Caucasus to Hungary, and correspond to prolific carbonate production, quite puzzling during a period of crisis for limestone deposits elsewhere. Comparison between the very littoral Permocalculus bioaccumulations of the Late Permian and the microbialites of the Early Triassic, allows one to enunciate the hypothesis that, during this very critical period, only the innermost platforms remained favourable to carbonate production. Consequently, the PTB should be only conventional along a sedimentological and geochemical continuum affecting the carbonate production and/or carbonate biomineralization. The palaeogeographical reconstruction indicates four groups of terranes in the studied area, from north to south: Peri-Hercynian Domain; Group Biikk-Greece; Extragondwanian Domain; and Perigondwanian Domain. The individuality of these areas is complete up to the PTB, and no element allows clear discrimination between the Palaeo-Tethys and Neo-Tethys in this area. We have benefited by the projects Unesco IGCP 329 and 384. In Russia, we thanked the Director General of Kubangasprom, Dr U.M. Basarigin, and the members of our team at Krasnodar, Dr S.L. Proshliakov, Dra. T.V. Pinchuk and Dra. T.B. Mikerina, as well as S. Proshliakov, Y. Vuks and G. Kotlyar. We appreciated the help of all the colleagues of the laboratories of Krasnodar (Russia), Sankt-Petersburg (Russia) and Budapest (MAFI and ELTE, Hungary). We are grateful to Mrs Siegl-Farkas, L. Pille and T. Vachard, Mr J. Gaillot, J. Dercourt, C. Detre, I. Veto, P. Solt and A. Baud for their advice; and especially to J. Broutin, who determined Gzhelian leaf imprints from Nikitin. The careful reviews of O. Weidlich and M. Aretz have contributed profoundly to modifying this manuscript.
References AKOPIAN,V. T. 1974. Atlas of Fossil Faunafrom Armenia SSR. Akademiya Nauk Armenia SSR, Institut GeologicheskykhNauk, 1-836 [in Russian]. ALEKSEEV,A. S., BARSUKOVA,L. D., KOLESOV,G. M., NAZAROV, M. A. & GRIGORYAN,A. G. 1983. The Permian-Triassic boundary event: geochemical investigation of the Transcaucasia section. Lunar and Planetary Science, 14, 7-8. ALTINER, D. & OZCAN ALTINER, S. 1998. Baudiella stampflii n. gen. n. sp., and its position in the evolution of Late Permian ozawainellid fusulines. Revue de Palkobiologie, 17, 163-175. BACHMANN, G. H. & KOZUR, H. W. 2003. First evidence of a microspherule interval around the
LATE PERMIAN EASTERN EUROPE AND CAUCASUS continental Permian-Triassic boundary, Germany, and its correlation with the marine realm. Acta Scientiarum Naturalium Musei Moraviae Occidentalis Trebic, 41, 143-146. BALOGH, K. 1964. Die geologischen Bildungen des Biikk Gebirges. Annales Instituti Geologici Publici Hungarici, 48, 2 [in Hungarian and German]. BALOGtt, K. 1980. Correlation of the Hungarian Triassic. Acta Geologica Hungarica, 26, 83-101. BANKS, C. J. & ROBINSON,A. G. 1997. Mesozoic strikeslip back-arc basins of the Western Black Sea Region. In." ROBINSON, A. G. (ed.) Regional and Petroleum Geology of the Black Sea and surrounding regions. AAPG Memoir, 68, 53-62. BASU, A. R., PETAEV, M. I., POREDA, R. J., JACOBSEN, S. B. & BUCKER, L. 2003. Chondritic meteorite fragments associated with the Permian-Triassic Boundary in Antarctica. Science, 302, 1388-1392. BAUD, A., ATUDOREI, V. & SHARP, Z. 1996. Late Permian and Early Triassic evolution of the Northern Indian margin: carbon isotope and sequence stratigraphy. Geodinamica Acta, 9, 57-77. BAUD, A., CIRILLI, S. & MARCOUX, J. 1997. Biotic response to mass extinction. Facies, 36, 238-242. BAUD, A., RICHOZ, S. & MARCOUX, J. 2005. Calcimicrobial cap rocks from the basal Triassic units: western Turkey occurrences (SW Turkey). Comptes Rendus Palevol, 4, 569-582. BECKER, L., POREDA, R. J., BASU, A. R., POPE, K. O., HARRISON,T. M., NICHOLSON,C. & IASKY,R. 2004. Bedout: A possible end-Permian impact crater offshore of northwestern Australia. Science, 109, 925 (see http://www.sciencemag.org/cgi/content/ abstract/1093925vl). BUCKER,L., POREDA,R. J., HUNT, A. G., BUNCH,T. E. B. & RAMPINO, M. R. 2001. Impact event at the Permian-Triasssic Boundary: Evidence from extraterrestrial noble gases in fullerenes. Science, 291, 1530-1533. BERCZI-MAKK, A. 1992. Midian (Upper Permian) Foraminifera from the large Mihalovits quarry at Nagyvisnyd (North-Hungary). Acta Geologica Hungarica, 35, 27-38. BI~RCZI-MAKK, A., PELIKAN, P. & CSONTOS, L. 1995. Data on the Upper Permian foraminifer fauna of the Nagyvisny6 Limestone Formation from borehole M~lyinka-8 (northern Hungary). Acta Geologica Hungarica, 38, 185-250. BONNET, P. 1947. Description g6ologique de la Transcaucasie m6ridionale. Mkmoires de la Socidtk GOologique de France, 53, 1-263. BUDUROV, K. & TRIFONOVA, E. 1994. Progress in concepts about conodont of the Triassic in Bulgaria. In: GUEX, J. & BAUD, A. (eds) Recent Development on Triassic Stratigraphy. Mdmoires de Gkologie (Lausanne), 22, 9-14. BURCHErrE, T. P. & WRIGHT, V. P. 1992. Carbonate ramp depositional systems. Sedimentary Geology, 79, 3-57. CROSTA, A. P., GASPAR, J. C. & FORNONI, C. 1981. Feigoes do metamorfismo de impacto no Domo de Araguainha. Revista Brasileira de Geociencias, 11, 139-146.
271
DEBENAY, J. P., PAWLOWSKI,J. & DECROUEZ,D. 1996. Les Foraminifkres actuels. Masson, Paris. ERSHOV, A. V., BRUNET, M. F., NIKISHIN, A. M., BOLOTOV, S. N., KOROTAEV,M. V. & KOSOVA,S. S. 1998. Evolution of the eastern Fore-Caucasus basin during the Cenozoic collision: burial history and flexural modelling. In: CRASQUIN-SOLEAU, S. BARRIER, E. (eds) Peri-Tethys Memoir 4, Epicratonic Basins of Peri-Tethyan Platforms. Mkmoires du MusOum national d'Histoire naturelle, 179, 111-130.
ERWlN, D. H. 1995. The end-Permian mass extinction. In: SCHOLLE, P. A., PERYT, T. M. & ULMERSCHOLLE, D. S. (eds) The Permian of Northern Pangea, Paleogeography, Paleoclimates, Stratigraphy, Volume 1. Springer, New York, 20-34. ERWIN, D. H. 2003. Impact at the Permo-Triassic boundary: a critical evaluation. Astrobiology, 3, 67-74. FARLEY, K. A. & MUKHOPADHYAY, S. 2001. An extraterrestrial impact at the Permian-Triassic boundary? Science, 293, 2343a-2344a. FLUGEL, E. & REINHARDT, J. 1989. Uppermost Permian reefs in Skyros (Greece) and Sichuan (China): implications for the Late Permian extinction event. Palaios, 4, 502-518. GAILLOT, J. 2006. The Khuff Formation (Late Permian, Middle East): high-resolution biostratigraphy by means of foraminifers and algae, sequence stratigraphy, depositional environments, and new approach of the Permian~Triassic crisis. PhD thesis, University of Lille. GAETAN1, M., GARZANTI,E., ET AL. 2005. Stratigraphic evidence for Cimmerian events in NW Caucasus (Russia). Bulletin de la SociOtk gkologique de France, 176, 283-299. GLINTZBOECKEL,C. & RABATI~,J. 1964. Microfaunes et microfaciks du Permo-Carbonifkre du Sud-Tunisien. International Sedimentary Petrology Series, 7, 1-45. HAAS, J., DEMI~NY,A., HIPS, K. & VENNEMANN,T. W. 2006. Carbon isotope excursions changes in marine Permian-Triassic boundary sections in Hungary. Palaeogeography, Palaeoclimatology, Palaeoecology, 237, 160-181. HAAS, J., Gt)CZAN, F., ORAVECZ-SCHEFFER, A., BARABAS-STUHL, A., MAJOROS, GY. & BERCZIMAKK, A. 1986. The Permian-Triassic boundary in Hungary. Memorie della Societa Geologica Italiana, 34, 21-241. HEYDARI, E. & HASSANDZADEH, J. 2003. Deev Jahi Model of the Permian-Triassic boundary mass extinction: a case for gas hydrates as the main cause of biological crisis on earth. Sedimentary Geology, 163, 147-163. HEYDARI, E., HASSANDZADEH,J. & WADE, W. J. 2000. Geochemistry of central Tethyan Upper Permian and Lower Triassic strata, Abadeh region, Iran. Sedimentary Geology, 137, 85-99. HEYDARI, E., HASSANDZADEH, J., WADE, W. J. & GHAZI, A. M. 2003. Permian-Triassic boundary interval in the Abadeh section of Iran with implication for mass extinction: Part 1 - Sedimentology. Palaeogeography, Palaeoclimatology, Palaeoecology, 193, 405-423.
272
J.M. THI~RY ET AL.
HIPS, K. & HAAS, J. 2006. Calcimicrobial stromatolites at the Permian-Triassic boundary in a western Tethyan section, Biikk Mountains, Hungary. Sedimentary Geology, 185, 239-253. HOLSER, W. T., SCHONLAUB,n . P., BOEKELMANN,K. & MAGARITZ, M. 1991. The Permian-Triassic boundary in the Carnic Alps of Austria (Gartnerkofel Region). Abhandlungen der geologische Bundesanstalt, 45, 213-232. IRANIAN-JAPANESE RESEARCH GROUP. 1981. The Permian and Lower Triassic systems in Abadeh region, Central Iran. Memoirs of the Faculty of Sciences, Kyoto University, Series of Geology & Mineralogy, 47, 61-133. KAIHO, K., KAJIWARA, Y. ET AL. 2001. End-Permian catastrophe by a bolide impact: evidence of a gigantic release of sulfur from the mantle. Geology, 29, 815-818. KOEBERL, C., FARLEY, K. A., PEUCKER-EHRENBRINK, B. & SEPHTON, M. A. 2004. Geochemistry of the end-Permian extinction event in Austria and Italy: No evidence for an extraterrestrial component. Geology, 32, 1053-1056. KARAKHANIAN, A., DJRBASHIAN, R., TRIFONOV, V., PHILIP, H., ARAKELIAN, S. & AVAGIAN, A. 2002. Holocene-historical volcanism and active faults as natural risk factors for Armenia and adjacent countries. Journal of Volcanology and Geothermal Research, 113, 319-344. KARAMATA, S., KRISTIC, B., DIMITRIJEVIC, M. D., KNEZEVIC,V., DIMITRIJEVIC,M. N. & FILIPOVIC,I. 1994. Terranes between the Adriatic and the Carpatho-Balkan Arc. Bulletin de l'AcadOmie Serbe des Sciences et des Arts, Classe des Sciences mathdmatiques et naturelles, Sciences naturelles, 108, 35, 47-68. KAZMIN, V. G. & SBORSHCHIKOV,I. M. 1989. Paleozoic and Early Mesozoic deformations on the Caucasus, and their place in the Tethys history. In: BELOV, A. A. & SATIAN, M. A. (eds) Geodynamics of the Caucasus. Nauka, Moscow, 46-54 [in Russian]. KHAIN, V. YE. 1982. Comparaison of Fixist and Mobilist models of the tectonic evolution of the greater Caucasus. Geotectonics, 16, 249-257. KIERSNOWSKI,H., PAUL, J., PERYT, T. M. & SMITH, D. B. 1995. Facies, paleogeography, and sedimentary history of the southern Permian Basin in Europe. In: SCHOLLE, P. A., PERYT, D. & ULMER-SCHOLLE,D. S. (eds) The Permian of Northern Pangea, Volume 2. Springer, New York, 119-136. KOBAYASHI, F. & ISHII, K. I. 2003a. Paleobiogeographic analysis of Yahtashian to Midian fusulinacean faunas of the Surmaq Formation in the Abadeh region, Central Iran. Journal of Foraminiferal Research, 33, 155-165. KOBAYASHI, F. & ISHII, K. I. 2003b. Permian fusulinaceans of the Surmaq Formation in the Abadeh region, Central Iran. Rivista Italiana di Paleontologia e Stratigrafia, 109, 307-337. KOTLYAR, G. V. 1997. Correlative reference levels of the Permian. Stratigraphy and Geological Correlation, 5, 131-146 (translated from Stratigrafiya Geologicheskaya Korrelyatsiya, 5, 35-50).
KOTLYAR, G. V., BAUD, A. ErAL. 1999a. Permian and Triassic exotic limestone blocks of the Crimea. Geodiversitas, 1,299-322. KOTLYAR,G. V., NESTELL,G. P., ZAKHAROV,YU. D. & NESTELL, M. K. 1999b. Changhsingian of the northwestern Caucasus, southern Primorye and southeastern Pamirs. Permophiles, 35, 18-22. KOTLYAR, G. V., PRONINA, G. P., ROSTOVCEV, K. O., ZAKHAROV, YU. D. & VUKS, V. YA. 1997. A Field Guide to the Permian and Triassic in Western Caucasus. International Field Trip on the Permian and Triassic in Western Caucasus, 9-18 July 1997. VSEGEI, St Petersburg and Krasnodar. KOTLYAR, G. V., SADOVNIKOV,G. N. & SEGEI,V. 1994. Events related to the Permian/Triassic boundary in Tethys and Biarma. In: GUEX, J. & BAUD, A. (eds) Recent Developments on Triassic Stratigraphy. Memoires de GOologie, Lausanne, 22, 63-68. KOTLYAR,G. V., ZAKHAROV,YU. D., KROPACHEVA,G. S., PRONINA,G. P., CHEDIYA,I. O. & BURAGO,V. 1. 1989. Pozdnepermskii etap evolyutsii organicheskogo mira, Midinskii yarus SSSR. [Evolution of the Latest Permian Biota, Midian Regional Stage of the USSR.] Leningrad 'Nauka', Leningradskoe Otdelennie [in Russian]. KOTLYAR, G. V., ZAKHAROV, YU. D. ET AL. 1984. Pozdnepermskii etap evolyutsii organicheskogo mira, Dzhulfinskii i Dorashamskii yarusy SSSR. [Evolution of the Latest Permian Biota, Dzhulfian and Dorashamian Stages of the USSR.] Akademiya Nauk SSSR, Dalnevostochnyi Nauchnyi Tsentr, Biologo-Pochvennyi Institut [in Russian]. KOTLYAR, G. V., ZAKHAROV,YU. D. & POLUBOTKO,I. V. 2004. Late Changhsingian fauna of the northwestern Caucasus, mountains, Russia. Journal of Paleontology, 78, 513-527. KOZUR, H. W. 1981. The main events in the Upper Permian and Triassic conodont evolution and its bearing to Upper Permian and Triassic stratigraphy. Rivista Italiana di Paleontologia, 85, 741-766. KOZUR, H. W. 1999. The correlation of the Bundsandstein and Muschelkalk with the Tethys scale. In: BACHMAN, G. H. & LERCHE, L. (eds) Epicontinental Triassic. Zentralblatt fiir Geologie und Paliiontologie, Teil 1, 1998, 701-725. LAMARCHE, J., SWIDROWSKA,J. ET AL. 1998. Development and deformation of a Mesozoic basin adjacent to the Teisseyre-Tornquist Zone: The Holy Cross Mountains (Poland). In: CRASQUIN-SOLEAU, S. & BARRIER, E. (eds) Peri-Tethys Memoir 4: Epicratonic Basins of Peri-Tethyan Platforms. Mdmoires du Musdum National d'Histoire Naturelle, 179, 75-92. LEMAIRE, M. M., GUREVITCH, E. L., NAZAROV, K., WESTPHAL, M., FEINBERG, H. & PozzI, J. P. 1998. Paleomagnetism of Permian to Jurassic formation from Turan Plate. In: CRASQUIN-SOLEAU, S. & BARRIER, E. (eds) Peri-Tethys Memoir 3." Stratigraphy and Evolution of Peri-Tethyan Platforms. MOmoires du Museum National d'Histoire Naturelle, 177, 71-87. LEVEN, E. YA. 1998. Permian fusulinid assemblages and stratigraphy of the Transcaucasus. Rivista Italiana di Paleontologia e Stratigrafia, 104, 299-328.
LATE PERMIAN EASTERN EUROPE AND CAUCASUS LYS, M., STAMPFLI,G. & JENNY,J. 1978. Foraminif6res du Carbonif6re et du Permien de l'Elbourz oriental (Iran Nord-Est). Note Laboratoire Pal~ontologie Genkve, 2, 63-99. MAGARITZ, M., BAR, R., BAUD, A. & HOLSER, W. T. 1988. The carbone-isotope shift at the Permian/ Triassic boundary in the southern Alps is gradual. Nature, 331,337-339. MASAITIS,V. L. 1999. Impact structures of northeastern Eurasia; The territories of Russia and adjacent countries. Meteoritics & Planetary Science, 34, 691-711. MEBE. 2005. In: BARRIER, E. GAETANI, M. (eds) MEBE News No. 1 Mars 2005 Middle East Basins Evolution, 1-15. MIKLUKHO-MAKLAY, K. V. 1954. Foraminifery verkhnepermskikh otlozhenii severnogo Kavkaza (Foraminifers of the Late Permian deposits of the northern Caucasus). Trudy VSEGEI, Gosgeoltekhizdat, 1, 1-163 (in Russian; French translation Bureau des Recherches G6ologiques et Mini6res No. 2683, 1-216). MOHTAT-AGHAI, P. & VACHARD, D. 2005. Late Permian foraminiferal assemblages from the Hambast Region (Central Iran) and their extinctions. Revista Espa~ola de Micropaleontologia, 37, 205-227. NIKISHIN,A. M., CLOETINGH,S. ET AL. 1998a. Scythian Platform: Chronostratigraphy and polyphase stages of tectonic history. In: CRASQUIN-SOLEAU, S. & BARRIER, E. (eds) Peri-Tethys Memoirs No. 3, Stratigraphy and Evolution of Peri-Tethyan Platforms. Mkmoires du Musdum National d'Histoire Naturelle, 177, 151-162. NIKISHIN, A. M., CLOETHING, S., BRUNET, M. F., STEPHENSON, R. A., BOLOTOV, S. N. & ERSHOV, A. V. 1998b. Scythian Platform, Caucasus and Black Sea region: Mesozoic-Cenozoic tectonic history and dynamics. In: CRASQUIN-SOLEAU, S. & BARRIER, E. (eds) Peri-Tethys Memoirs No. 3 Stratigraphy and Evolution of Peri-Tethyan Platforms. MOmoires du MusOum d'Histoire Naturelle, 177, 163-176. No~, S. 1987. Facies and paleogeography of the marine Upper Permian and of the Permian-Triassic Boundary in the Southern Alps (Bellerophon Formation, Tesero Horizon). Facies, 16, 89-142. OKLA, S. M. 1992. Permian algae and algal microfacies from Unayzah, Quassim district, Saudi Arabia. Facies, 27, 217-224. PANTIC-PRODANOVIC,S. 1996. The micropaleontologic and biostratigraphic characters of Upper Permian and Lower Triassic sediments in Northwestern Serbia-Cer Mountain, Petkovica and Necaje localities. Annales Gdologiquesde la P~ninsule Balkanique, 58, 129-168. PANTIC-PRODANOVIC, S. & RADOSEVIC,B. 1981. Biostratigraphic and sedimentologic characteristics of Upper Permian and Lower Triassic sediments in the Gucevo Mt area, Yougoslavia. Acta Geologica Academiae Scientiarum Hungaricae, 24, 99-11. PESIC, L., RAMOV,A., SREMA,J., PANTIC-PRODANOVIC, S., FILIPOVIC, C., KOVACS, S. & PELIKAN, P. 1986. Upper Permian deposits of the Jadar region and their position within the Western Paleotethys.
273
Memorie della Societa Geologica Italiana, 36, 211-219. PRONINA, G. P. 1988. The Late Permian smaller foraminifers of Transcaucasus. Revue de PalOobiologie, 1, Special Volume 2, 89-96. PRONINA-NESTELL,G. P. & NESTELL, M. K. 2001. Late Changhsinghian foraminifers of the northwestern Caucasus. Micropaleontology, 47, 205-234. PRUSS, S. B., CORSETTI, F. A. & BOTTJER, D. J. 2005. The unusual sedimentary rock record of the Early Triassic: A case study from the southwestern United States. Palaeogeography, Palaeoclimatology, Palaeoecology, 222, 33-52. RAUZER-CHERNOUSOVA, D. M., BENSH, F. P. ET AL. 1996. Spravochnik po sistematike foraminifer Paleozoya; Endothyroidy, Fuzulinoidy. [Referencebook on the Systematics of Paleozoic Foraminifers; Endothyroida and Fusulinoida.] Rossiiskaya Akademiya Nauk, Geologicheskii Institut, Moskva 'Nauka' [in Russian]. RETALLACK, G. J., SEYEDOLALI, A., KRULL, E. S., HOLSER, W. T., AMBERS, C. P. & KYTE, F. T. 1998. Search for evidence of impact at the PermianTriassic boundary in Antarctica and Australia. Geology, 26, 979-982. RIDING, R. 2002. The structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth Science Reviews, 58, 163-231. RIDING, R. 2005. Phanerozoic reefal microbial carbonate abundance: comparisons with metazoan diversity, mass extinction events, and seawater saturation state. Revista Espagola de Micropaleontologia, 37, 23-39. ROBIN, E. & ROCCHIA, R. 1998. Ni-rich spinel at the Cretaceous-Tertiary boundary of E1 Kef, Tunisia. Bulletin de la Societk Gkologique de France, 169, 365-372. ROBINSON,A. G., BANKS,C. J., RUTHERFORD,M. M. & HIRST, P. P. 1995. Stratigraphy and structural development of the eastern Pontides, Turkey. Journal of the Geological Society, London, 152, 861-872. ROSTOVTSEV, K. O. & AZARYAN, N. R. 1973. The Permian and Triassic boundary in Transcaucasia. In: LOGAN, B. & HILLS, L. V. The Permian and Triassic Systems and Their Mutual Boundary. Canadian Society of Petroleum Geologists, Memoir 2, 89-99. SCHUBERT, J. K. & BOTTJER, D. J. 1992. Early Triassic stromatolites as post-mass extinction disaster forms. Geology, 20, 883-886. SCHULTZ, C. L. 2005. Biostratigraphy of the non marine Triassic: Is a global correlation based on a Tetrapod Fauna possible? In: KOUTSOUKOS, E. A. M. (ed.) Applied Stratigraphy, Dordrecht, Springer, 23, 123-145. SENGOR, A. M. C., ALTINER,D., CLIN, A., USTA0SMER, T. & Hst3, 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) Gondwanaand Tethys. Geological Society, London, Special Publications, 37, 119-181. SENGOR, A. M. C., YILMAZ, Y. & SUNGURLU,O. 1984. Tectonics of Mediterranean Cimmerides: nature and evolution of the western termination of Paleotethys. In: DIXON, J. E. & ROBERTSON, A. H.
274
J.M. THl~RY ET AL.
F. (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 17-112. SEPKOSKI, J. J., BAMBACH,R. K. & DROSER, M. L. 1991. Secular changes in Phanerozoic event bedding and the biological imprint. In: EINSELE,G., RICKEN, W. & SEILACHER, A. (eds) Cycles and Events in Stratigraphy. Springer, Berlin, 371-385. SMITH, D. B. 1995. Marine Permian of England. Chapman & Hall, London. STAMPFLI,G. 1978. Gkologie gknkrale de l'Elbourz oriental au Sud de Gonbad-e-Qabus (Iran NE). PhD thesis, Geneva. STAMPFLI, G. & PILLEVUIT, A. 1993. An alternative Permo-Triassic reconstruction in the Tethyan realm. In: DERCOURT, J., RICOU, L. E. & VRIELNYCK, D. (eds) Atlas Tethys Paleoenvironmental Maps. Explanatory Notes. CCGM-CGMW, Gauthier-Villars, Paris, 55-62. STAMPFLI, G., MARCOUX,J. & BAUD,A. 1991. Tethyan margins in space and time. Palaeogeography, Palaeoclimatology, Palaeoecology, 87, 373409. TERMIER, H., TERMIER, G. & VACHARD, D. 1977. Monographie pal6ontologique des affieurements permiens du Djebel Tebaga (Sud Tunisien). Palaeontographica, A, 156, 1-109. THI~RY, J. M., BILAL,E., CROSTA,A. & SOLT, P. 2003. Discovery of microspherules ejected from the Araguainha impact crater - Attempt to stratigraphical correlations. Acta Scientiarum Naturalium Musei Moraviae Occidentalis, Trebic, 41, 133-135. VACHARD, D. 1980. T6thys et Gondwana au Pal6ozoique sup6rieur; les donn6es afghanes: biostratigraphie, micropal6ontologie, pal6og6ographie. Documents et Travaux de rlnstitut Gkologique Albert de Lapparent, 2, 1463. VACHARD,D. & BOUYX,E. 2002. Les Eopolydiexodina g6antes (Foraminiferida, Fusulinina) du Permien moyen d'Afghanistan, remarques pr61iminaires. Annales de la Sociktk Gkologique du Nord, 2e skrie, 9, 163-189. VACHARD,D. ~; MONTENAT,C. 1981. Biostratigraphie, micropal6ontologie et pal6og6ographie du Permien de la r6gion de Tezak (Montagnes Centrales d'Afghanistan). Paleontographica, B, 178, 1-88. VACHARD, D. & RAZGALLAH,S. 1993. Discussion sur l'~tge murgabien ou midien des s6ries permiennes du
Jebel Tebaga (Sud de la Tunisie). Rivista Italh~na di Stratigrafia e Paleontologia, 99, 327-356. VACHARD,D., GAILLOT,J., VASLET,D. & LE NINDRE, Y. M. 2005. Foraminifers and algae from the Khuff Formation (late Middle Permian-Early Triassic) of central Saudi Arabia. GeoArabia, 10, 137-186. VACHARD, D., MARTINI, R., ZANINETTI, L. & ZAMBETAKIS-LEKKAS, A. 1993. Rdvision micropal6ontologique (foraminif6res, algues) du Permien inf6rieur (Sakmarien) et sup6rieur (Dorashamien) du Mont Beletsi (Attique, Grace). Bollettino Societd Paleontologica Italiana, 32, 89-112. VACHARD, O., ZAMBETAKIS-LEKKAS,A., SKOURTSOS, E., MARTINI, R. & ZANINETTI, L. 2003. Foraminifera, algae and carbonate microproblematica from the late Wuchiapingian/Dzhulfian (Late Permian) of Peloponnesus (Greece). Rivista Italiana di Paleontologia i Stratigrafia, 109, 339-358. VASLET, D., LE NINDRE, Y. M. ET AL. 2005. The Permian-Triassic Khuff Formation of central Saudi Arabia. GeoArabia, 10, 77-134. VOZAR, J. • VOZAROVA,A. 1988. Late Paleozoic in West Carpathians. Geologicky ustav Dionyza sutra [in English and Slovak]. VUKS,V. YA. 2000. Triassic foraminifers of the Crimea, Caucasus, Mangyshlak and Pamirs (biostratigraphy and correlations). Zentralblatt fiir Geologie und Paliiontologie, Teil I, 11-12, 1353-1365. WANG, L. W. 1974. A Handbook of the Stratigraphy and Paleontology in Southwest China. Nanking Institute of Geology and Paleontology, Science Press, Academica Sinica, 1454 [in Chinese]. WEIDLICH, O. 2002. Middle and Late Permian reefsdistributional patterns and reservoir potential. In: Phanerozoic Reef Patterns. Society of Economic Paleontologists and Mineralogists, Special Publications, 72, 339-390. YANEV, S. 2000. Paleozoic terranes of the Balkan Peninsula in the framework of Pangea assembly. Palaeogeography, Palaeoclimatology, Palaeoecology, 161,151-177. ZAKHAROV,Y. D., IGNATIEV,A. V., KOTLYAR,G. V., UKHANEVA, N. G. 8s CHEBARDZHI, A. K. 1996. Stable carbon isotopes and Ca/Mg ratio in the Permo-Triassic carbonates and Mass-Extinction of organisms. Geology of the Pacific Ocean, 13, 1-20.
Estimation of palaeoenvironmental changes: can analysis of distribution of tabulae in tabulates be a tool? M I K O L A J K. Z A P A L S K I 1'2, B E N O i T H U B E R T 2'3 & B R U N O M I S T I A E N 2
1Warsaw University, Faculty o f Geology, Zwirki i Wigury 93, 02~989 Warszawa, Poland 2Laboratoire de PalOontologie stratigraphique F L S T & ISA, U M R 8014 du CNRS, 41 rue du Port, 59046 Lille COdex, France (e-mail."
[email protected]) 3UnitO de Recherche de POtrologie sOdimentaire, B20, UniversitO de Likge, Sart-Tilman B-4000 Likge Abstract: Growth periodicity (cyclomorphic variation) in corals is expressed by various features, among them changes in the distribution of tabulae. A method potentially useful in analysis of periodical environmental changes is proposed herein. Measurement of spaces between tabulae in tabulate corals and preparation of a histogram converted into a trend curve may show relative periodical fluctuations of the environment. Such an analysis, exemplified here on Givetian Pachyfavosites sp. from the Avesnois (northern France), shows that this method may be used as a tool for estimation of environmental changes.
Growth periodicity (cyclomorphic variation) is a phenomenon occurring in both fossil and modern corals, and has been known since the classic papers of Ma (1933, 1934a, b). It is expressed in tabulate corals in rhythmical changes of distribution of skeletal elements, such as septal spines and tabulae, and also changes in thickness of wall and other elements. In tabulate corals it has been recognized by various authors, for example Powell & Scrutton (1980), Nowifiski (1991) and, more recently, by Young & Elias (1999) and Young & Kershaw (2005). The development of research on growth periodicity brought numerous interpretations of this phenomenon. Shortly after the discovery of cyclomorphism (Ma 1933), Krempf (1935) interpreted it (on scleractinians) as being caused by periods of sexual activity. Two years later, Wells (1937) following Krempf's suggestion, stated that banding of fossil rugose corals may correspond to periods of planulation. Scrutton (1964) proposed a correlation of growth periodicity of rugose corals with lunar cycles. Powell & Scrutton (1980) and Nowifiski (1991) on tabulate corals, and Young & Kershaw (2005) on tabulate and rugose corals (and stromatoporoids), interpreted this as the effect of periodic (seasonal) changes of environment. Such an interpretation is broadly accepted by scleractinian researchers (Lough 2004; Peirano et al. 2004; Shen et al. 2005; Yu et al. 2005). The analysis of growth rate as an environmental indicator was presented for the first time by Shinn (1966) based on scleractinians (Acropora
sp.). The aim of this paper is to present the possibility of the application of tabulae distribution analysis for evaluation of palaeoenvironmental changes. The presentation of this method is supported by an example, namely the analysis of tabulate coral (Pachyfavosites sp.) from the Givetian of Avesnois (cf. Delattre & Waterlot 1970), northern France (Fig. 1). Material
The material comes from the 'Croix de Bourges' outcrop, located in Avesnois, northern France (Fig. 1), in the western part of the southern wing of the Dinant Synclinorium. Samples were collected in 2004 and 2005, and the whole material consists of massively occurring colonies, with variable forms and sizes, showing different preservation. The outcrop exposes the sediments around the Eifelian-Givetian boundary. Owing to the absence of conodonts, the only stratigraphical marker found associated with the analysed corals was abundant Stringocephalus sp., indicating a Givetian age. The material was collected from the rubble, thus the time relations between colonies are unknown. The associated fauna is numerous and contains mainly bioconstructors: stromatoporoids, other tabulate (pachyporids, alveolitids and auloporids), and rugose corals (mainly acanthophyUids), brachiopods and crinoids. Sixty-seven thin sections were prepared out of 99 colonies, most of them large (up to 100 cm2). Finally, seven best-preserved coralla of Pachyfavosites sp. (Fig. 2) were chosen for the analysis.
From: ikLVARO,J. J., ARETZ,M., BOULVAIN,F., MUNNECKE,A., VACHARD,D. & VENNIN,E. (eds) 2007. Palaeozoic Reefs and Bioaccumulations." Climaticand Evolutionary Controls. Geological Society, London, Special Publications, 275, 275-281. 0305-8719/07l$15.009 The Geological Society of London.
276
M.K. ZAPALSKI ETAL.
Fig. 1. Sampling locality. The analysed coralla have diverse morphologies - low domical, mushroom shaped and bulbocolumnar (terminology after Young & Elias 1999). The material is housed in the collections of the Facult6 Libre des Sciences et Technologies, Lille (abbreviated as GFCL). Methods
As measuring the spaces between tabulae is the basis of this method, favositid tabulates were chosen. The main advantages of this group are: -
-
relatively large corallite diameter; corallites are usually straight, not curved or meandering; usually flat tabulae.
This group of corals allows the measurement process to be as simple as possible. The methodology of measurements of spaces between tabulae (with a regard to the form of tabulae) is absent in the literature, thus a unification is proposed here and the eight most often occurring common situations with their measurements are presented in Figure 3. In each specimen a few (usually five) of the longest neighbouring corallites (each of these corallites contained at least 30 tabulae, and two couplets of dark-light band), preferably from the centre of the colony, were chosen. To avoid mistakes in measurements we have chosen only well-developed corallites without any teratological changes. Distances (spaces) between tabulae were measured space-by-space (as shown in Fig. 4). When possible, distances between the successive tabulae inside a continuous corallite were measured. In numerous cases it was impossible to measure all spaces along a single corallite, so sometimes the measurements were compiled from neighbouring corallites (see Young & Kershaw 2005; thus one transect corresponds in some cases to a few adjacent corallites). The number of measured corallites in a corallum depended on its shape - the highest number of corallites was measured in low domical coralla because the corallites are straight and easy to measure. From each transect 33-104 spaces (in a total of 35 corallites) were measured. Two sets of histograms showing the size of the sequential spaces (Fig. 4) were prepared. Polynomial trend curves (coefficient 12 in one set, coefficient 19 in the other one) were added to the histograms in order to observe larger and smaller scale cycles. All graphs were prepared using IgorPro 4.0 software.
2. Pachyfavosites sp., specimens GFCL 4714 and GFCL 4715, Givetian, Croix de Bourges, Avesnois, France. (a) Longitudinal section, (b) transverse section.
Fig.
PALAEOENVIRONMENTAL ANALYSIS: CORAL TOOL
/
277
s d
/
!
Fig. 3. Scheme of eight most typical tabulae and measurements of spaces between them, marked by arrows.
~k .N
LD I
HD I
I
LD I
t
t
=o LD
e.J cO
D
HD
LD
9
10
11
12 v
Space number Fig. 4. Scheme explaining construction of diagrams (Figs 5 & 6). The size of spaces (in mm) between the successive tabulae is presented as bars. One bar corresponds to one space.
Results
The m a i n types of trend curves obtained will be presented below.
Regular curve C1 (Fig. 5a). It expresses regular patterns, showing clear cyclicity. Two C1 curves have been obtained (out of 35).
278
M.K. ZAPALSKI E T AL.
Discussion
Fig. 5. Histograms showing growth periodicity in Pachyfavosites sp. (a) Curve Cl, (b) curve C2 and (c) curve C3. All curves: coefficient 19. Specimens GFCL 4716 (bulbo-columnar corallum), GFCL 4717 (mushroom-shaped corallum) and GFCL 4718 (irregular near bulbo-columnar corallum).
2.
3.
Regular curve with disturbances C2 (Fig. 5b). It is less regular than the previous one; however, periodic repeats of cycles can still be clearly seen. Eleven curves of this type have been obtained (out of 35). Irregular curve C3 (Fig. 5c). This curve does not show a recognizable periodic pattern. Twenty-two curves of this type have been obtained (out of 35).
The thickness of high density (HD) bands varies from 3 to 8 mm, while the thickness of the low density (LD) bands ranges from 9 to 16 mm, but it must be stated that in numerous cases the border between LD and HD cannot be clearly defined.
The alternation of high/low density (HD/LD) bands in some tabulate corals induced numerous discussions and environmental interprctations of growth periodicity. If we assume that corals built their tabulae regularly, the distances between the tabulae depend only on the corallite growth rate; when conditions are favourable the individual grows faster than during unfavourable periods. The tabular spacing increases during favourable conditions and is in relation with the growth rate. The dense skeletal growth - high density (HD) bands - was correlated as summer (Weber et al. 1975), later as autumn (Steam et al. 1977) and as winter (Knutson et al. 1972). Powell & Scrutton (1980) proposed the following interpretation of growth periodicity: high density bands were secreted during cold periods, low density (LD) bands during warm ones. In addition, Shinn (1966) states that growth rate is strictly correlated with temperature. The temperature is the most often considered as the main cause of this phenomenon (Carricat-Ganivet et al. 2000; Abram et al. 2001; Peirano et al. 2004); however, other factors are also mentioned, such as insolation (Reynaud et al. 2004) or CO2 content in the atmosphere (Langdon 2002) and pH depending on it (Kleypas & Langdon 2002). Other authors also mention light extinction, sedimentation rate, dissolved nutrients and wave energy as environmental factors causing changes in coral growth rate (Cruz-Pinon et al. 2003). However, in our opinion, the temperature changes seem to be the primary factor causing the changes in salinity and nutrient contents. Only the sedimentation rate and wave energy seem to be independent of the temperature changes; however, we did not include these factors in order to get a more simple view on growth rates. Thus, the discussion below follows the assumption that the temperature is the principal cause of changes in growth pattern. 1. The C1 curve (Fig. 5a - curve with coefficient 19, see also Fig. 6a, b) shows regular and symmetric patterns. Using the histogram allowed us to identify four areas: two corresponding to high density (HD) growth (12-17th and 41st~9th spaces) and two for low density (LD) (25th-31 st and 52nd-57th spaces). These zones are located around the maximum and minimum values of spaces between each tabula and correspond to the inflection changes of the trend curve. The regular pattern of the trend curve is shown by the number of spaces between the two maxima (27 spaces) and the two minima (25-30 spaces). At the 15th space (see Fig. 5a) the curve reaches its local minimum
PALAEOENVIRONMENTAL ANALYSIS: CORAL TOOL
279
Fig. 6. Histograms showing growth periodicity in Pachyfavosites sp., comparison between curves with coefficient (a, e, e) 19 and (b, d, f) 12. (a) & (b) Curve C1, (e) & (d) curve C2 and (e) & (f) curve C3. Specimens GFCL 4716 and GFCL 4719 and GFCL 4720 (all bulbo-columnar coraUa).
and starts to ascend. The growth was fast, which means that environmental conditions became more favourable for the animal (at the beginning of a warm period). This period was relatively short, compared with the following one, and the change of environment was relatively rapid. Around the 27th space the curve reaches its local maximum, at which point the conditions were probably the best and the increase of the corallite was the fastest. After this peak the growth rate began to decrease. The decrease of growth rate was divided into two phases: the first one, between spaces 27 and 38, was slower, but between spaces 39 and 45 it was faster. It may show that the life conditions (e. g. the temperature) were changing in two steps. Around the 46th space the curve again
reaches its local minimum. In comparison to the previous one it may be stated that the growth rate was slightly slower than during the previous unfavourable period (cold period). The next maximum, around the 54th space, is lower than that around the 27th space, which may indicate less favourable conditions during the warm period. In the following descending wing two parts are clearly visible. They probably indicate a two-step change of environment, similar to that in the preceding cycle. The C2 curve (Fig. 5b) also shows a regular pattern; however, it is not as regular as the previous one. It is more difficult to clearly differentiate HD and LD zones within this curve. In this diagram two interpretations are proposed. The first one is to describe
280
3.
M.K. ZAPALSKI E T AL. this diagram as irregular. Three HD zones (around the 5th, 23rd and 45th spaces) and three LD zones (around the 15th, 34th and 54th spaces) and strong differences between local minima indicate a different growth rate caused by unfavourable environmental conditions, but changing in time. The regularity between each minimum and maximum cycle, seen in C 1 curve, is absent in this case. The second interpretation is to consider the 15th and 50th spaces not as minima but as microenvironmental disturbances during a warm period. Thus, the two maxima are the 15th and the 50th spaces. The only minimum is the 34th space and characterizes the cold period. In this case, the regular space between two maxima is similar to that in the C1 curve (around 30 spaces). The amplitude of the curve is thus the most important feature. If we assume the temperature as the main factor (see above) of these changes (causing also, for example, changes in salinity) it can be stated that the temperature during the second local maximum (50th space) was lower than during the first one (! 5th space). The curve C3 (Fig. 5c) seems to be generally irregular, and no clear cycles can be observed. This may indicate an unchanging environment, but also other factors, such as sedimentation rate or waving mentioned above, may have an influence on this growth pattern.
It can be clearly seen that the polynomial curves (Fig. 6a, c, e) of the factor 19 are more detailed and they may show small environmental fluctuations, while the '12' curves (Fig. 6b, d, f) show large-scale environmental cycles. Curves obtained from the same corallum may differ and can be assigned to two types of curves. This may pose some problems with their interpretation (as in curve C2); however, the differences between them may be explained by possible different microconditions around the colony, such as local currents caused by the morphology of its surface or different insolation. Such big differences in growth rate among closely related corals from one site may be explained by different ages of the coralla and their different shapes. Also, they were not collected in situ and their stratigraphic age is probably not the same. Last, but not least, these differences may be caused by non-periodic environmental changes and disturbances. It can be also noted that in one corallum of Pachyfavosites sp. (specimen GFCL 4720) we found a temporary fusion of corallites (Fig. 7). It
Fig. 7. Temporary fusion of corallites in Pachyfavosites sp., specimen GFCL 4720, Givetian, Croix de Bourges, Avesnois, France. Longitudinal section. Scale bar is 50 ~m.
occurs at the end of a HD zone and such a break in secretion of walls might have been caused by an adaptative strategy between two neighbourhooding individuals. This period constitutes the beginning of a warm period and consequent increase in growth. The shape of the corallum is probably in a relationship with environmental changes, such as wave energy and sedimentation rate. The relation between the corallum shape and growth rate (thus also environmental change) requires elaboration on a statistically larger sample.
Conclusions As stated by Powell & Scrutton (1980), growth periodicity in corals seems to be caused by seasonal changes, thus correlated mainly with the temperature (Shinn 1966; Carricat-Ganivet et al. 2000; Abram et al. 2001; Peirano et al. 2004). The above-presented analysis of the tabulae distribution in favostid tabulates may allow us to recognize relative changes in temperature. The C1 curve patterns expose two major cases: a strong slope during winter/spring, showing a quick increase of temperature just after a cold period; and a two-step slope during summer/ autumn, showing a progressive passage to a cold period. The deviation of the trend curve between two seasons is an important feature allowing evaluation of changes in temperature. When the curves have a regular pattern, seasons are similar year after year; when the curve deviations are strong (like in curves C2), the differences between the cold and warm periods are remarkable. Growth periodicity is not always easy to show. The absence of clear cyclicities seems to
PALAEOENVIRONMENTAL ANALYSIS: CORAL TOOL show a rapid evolution o f disturbances of climate during one seasonal period (C3 curves). O t h e r factors, not related with the t e m p e r a t u r e (e.g. sedimentation rate), m a y also have an i m p o r t a n t role in forming the g r o w t h banding, but only the t e m p e r a t u r e seems to be fluctuating periodically. We are deeply grateful to J. Trammer (Warsaw) for inspiring discussions, A. T. Halamski (Warsaw) and D. Elliot (Flagstaff) for remarks on the manuscript. We want also to express our thanks to G. Young and R. Elias (both Winnipeg) for giving valuable comments, P. Deville (Lille) for preparing thin sections, G. Boutry for discussions on statistical methods and M. Niechwedowicz (Warsaw) for providing some bibliographical references.
References ABRAM, N. J., WEBSTER,J. M., DAVIES,P. J. • DULLO, W. C. 200 I. Biological response of coral reefs to sea surface temperature variation: evidence from the raised Holocene reefs of Kikai-jima (Ryuku Islands, Japan). Coral Reefs, Berlin, 20, 221-234. CARRICART-GANIVET,J. P., BELTRAIN-TOORRES,A. U., MERINO, M. 8~;RUIZ-ZARATE, M. A. 2000. Skeletal extension, density and calcification rate of the reef building coral Montastrea annularis (Ellis and Solander) in the Mexican Caribbean. Bulletin of Marine Science, Miami, 66, 215-224. CRUZ-PINON, G., CARRICART-GANIVET, J. P. & ESPINOZA-AVALOS,J. 2003. Monthly skeletal extension rates of the hermatypic corals Montastraea annularis and Montastraea faveolata: biological and environmental controls. Marine Biology, Berlin, 143, 491-500. DELATTRE,C. 8~WATERLOT,G. 1970. Carte Gkologique 1150000. Trdlon, Carte XXVIII-7. Service g6ologique national, Orl6ans. KLEYPAS, J. & LANGDON, C. 2002. Overview of CO2induced changes in seawater chemistry. Proceedings
of the 9th International Coral Reef Symposium, Bali. Indonesian Institute of Sciences, International Society for Reef Studies, 2002, Volume 2, 1085-1089. KNUTSON, D. W., BUDDEMEIER,R. W. & SMITH, S. V. 1972. Coral chronometers: seasonal growth bands in reef corals. Science, Washington, 177, 270--272. KREMPF, A. 1935. Inscription mar6graphique des cycles de r&rogradation des cycles de la lune par certains coraux constructeurs de rr Comptes Rendus de l'Acad~mie des Sciences, Paris, 198, 1708-1710. LANGDON, C. 2002. Review of experimental evidence for effects of CO2 on calcification of reef builders.
Proceedings of the 9th International Coral Reef Symposium, Bali. Indonesian Institute of Sciences, International Society for Reef Studies, 2002, Volume 2, 1091-1098. LOUGH, J. M. 2004. A strategy to improve the contribution of coral data to high-resolution paleoclimatology. Palaeogeography, Palaeoclimatology, Palaeoecology, 204, 115-143.
281
MA, T. Y. H. 1933. On the seasonal change of growth in some Palaeozoic corals. Proceedings of the Imperial Academy, Tokyo, 9, 407-409. MA, T. Y. H. 1934a. On the seasonal change of growth in a reef coral, Favia speciosa (Dana), and the water-temperature of the Japanese Seas during the latest geological times. Proceedings of the Imperial Academy, Tokyo, 10, 353-356. MA, T. Y. H. 1934b. On the growth rate of reef corals and the sea water temperature in the Japanese Islands during the latest geological times. Science
Reports of the Tohoku Imperial University, Second Series, Sendai, 16, 166-169. NOWIlqSKI, A. 1991. Late Carboniferous to Early Permian Tabulata from Spitsbergen, Palaeontologica Polonica, Warsaw, 51, 1-74. PEIRANO, A., MORRI, C. ET AL. 2004. The Mediterranean coral Cladocora caespitosa: a proxy for past climate fluctuations? Global and Planetary Change, 40, 195-200. POWELL, J. H. & SCRUTTON, C. T. 1980. Periodic development of dimetrism in some favositid corals. Acta Palaeontologica Polonica, Warsaw, 25, 477-491. REYNAUD, S., EERRIER-PAGES, C., BOISSON, F., ALLEMAND, U. & FAIRBANKS,R. G. 2004. Effect of light and temperature on calcification and strontium uptake in the scleractinian coral Acropora
verweyi. Marine Ecology Progress Series, Oldendorfl Luhe, 279, 105-112. ScRvrroN, C. T. 1964. Periodicity in Devonian coral growth. Palaeontology, Oxford, 7, 552-558. SHEN, C.-C., LEE, T. ET AL. 2005. An evaluation of quantitative reconstruction of past precipitation records using coral skeletal Sr/Ca and 5180 data. Earth and Planetary Science Letters, 237, 370-386. SI-IIYN, E. A. 1966. Coral growth-rate, an environmental indicator. Journal of Paleontology, Lawrence, 40, 233-241. STEARN, C. W., SCOFFIN, T. P. & MARTINDALE, W. 1977. Calcium carbonate budget of a fringing reef on the west coast of Barbados, Part I. Zonation and productivity. Bulletin of Marine Sciences, Miami, 27, 479-510. WEBER, J. N., DEINES, P., WHITE, E. W. & WEBER, P. H. 1975. Seasonal high and low density bands in reef coral skeletons. Nature, London, 255, 697-8. WELLS, J. W. 1937. Individual variation in the rugose coral species Heliophyllum halli Edwards and Haime, Palaeontographica Americana, Ithaca, 2, 1-22. YOUNG, G. A. 8z ELIAS, R. J. 1999. Relationships between internal and external morphology in Paleofavosites (Tabulata): the unity of growth and growth form. Journal of Paleontology, Lawrence, 73, 580-597. YOUNG, G. A. & KERSHAW, S. 2005. Classification and controls of internal banding in Palaeozoic stromatoporoids and colonial corals. Palaeontology, Oxford, 48, 623-651. Yu, K.-F., ZHAO J.-X., COLLERSON, K. D., SHI, Q., CHEN, T.-G., WANG, P.-X. & LIU, T.-S. 2005. Storm cycles in the last millennium recorded in Yongshu Reef, southern South China Sea. Palaeogeography, Palaeoclimatology, Palaeoecology, 210, 89-100.
Index Page numbers in italic denote figures; page numbers in bold denote tables
Acanthina 44, 45 Acropora 182 Aegiromena 7, 78 Afiacyathus alloiteaui 32, 35, 41, 42 Agathammina 256 Aisemont Formation 144, 145, 146-149, 148, 149 ajacicyathids 35, 35, 37, 40 A1 Jil Formation 232, 234, 247 A1 Khlata Formation 231,232 Alborz Mountains 260 algae calcareous 104, 105 Atokan Sierra Agua Verde 203, 204 green, Frasnian carbonate mounds 130, 135 Alpine Nappe System 95 Alps Silurian-Upper Devonian reefs 95-105 Alticola Limestone 97 Alveolites 9, 129, 130, 146, 148, 150, 151-153, 157 Frechastraea association 154, 155-160 Alveolites-stromatoporoid biostrome 153-155, 158-160, 159 Amphipora mounds 103, 104 Frasnian 130, 135 Amplexocarinia 179, 180 Ancyrognathus triangularis 145, 146 Anger Crystalline Complex 96 Anomalocystites incipiens, systematic palaeontology 81-82 Anthomorpha 35, 37, 40, 43 Anti Atlas geology 52-54 microbial reefs 66-67, 68 shell beds, Lower-Ktaoua Formation 6-7 apatite 17, 23, 24, 25 see also collophane Aphralysia 173, 179, 180 Arabian plate Permian carbonates 230-251 aragonite, dissolution, Gotland carbonate 7-9 archaeocyathan-microbial reef complexes, Lemdad syncline 61 64, 62, 66, 67 archaeocyaths La Sentinella reef complex 31~49 Lemdad syncline 52 Archaeolithoporella 214, 215,223, 230, 247, 250 Archaeopharetra 35, 37, 40, 47 Arche member Frasnian carbonate mounds 125 139 Aristocystites cf. bohernicus 73 Armenian Block 262 Armorican Massif geology 73-74 Middle Ordovician echinoderms 72-73 Aseelah Unit 232, 233 Permian carbonate facies 242-244, 248 Atlas Mountains Lower Cambrian 51-54 Atokan, late chaetetids, Sierra Agua Verde 189-198 encrinite, Sierra Agua Verde 201-207
Aulopora 153, 154, 155, 157 Austria, Silurian-Upper Devonian reefs 95-105, 96 Av6ne-Mendic autochthonous unit 17, 18, 19 phosphorites 21-22, 22 bacteria, iron-oxidizing 132, 134, 135 Baelen facies 166-167, 168, 169, 184 Ba'id area Permian carbonate facies 245, 246, 247 Permian palaeoclimatology 248 stratigraphy 232,233,234 Ba'id Formation 232,234,247 Baryshnikovia 191 Basal Nappe System 95-96 Batain coast Permian carbonate facies 241,242-245, 246 Permian palaeoclimatology 246, 248 Permian stratigraphy 232, 233 234 bathymetry, Chaetetes 194 Baudiella 266, 270 Baugnde section, Dinant Synclinorium 147 bedforms, sand, Lemdad syncline 54, 56 Belgium Eifelian-Givetian carbonate platform 109 122 Famennian and Dinantian reefs 163-186 Frasnian carbonate mounds 125-139 Frasnian phillipsastraeidae biostromes 143-160 Bellerophon Limestone 232, 255, 262 Berabichia cf. vertummnia 52 bioaccumulation 1 bioconstruction controlling factors 181-184 definition 166 Strunian 169-170 biogenic concentration 2 bioherrns definition 164, 166 Famennian reefs 167 La Sentinella reef complex 30-37, 46~17, 49 Ounein reef complexes, Lemdad syncline 61, 63 Warnantian 180 Wenlock patch reef 88-89, 93 biostromes archaeocyathan-microbial, Cambrian 64, 67, 68 coral Tournaisian 172 Vis6an 180-181 definition 2, 98, 164, 166, 170 phiUipsastreid, Frasnian 146-160 Alveolites-stromatoporoid 153-155, 158-160 Frechastraea-A lveolites 155-160 stromatoporid, Famennian reefs 169-170 Bischofalm Facies 97 bivalves Middle Ordovician, Brittany 73, 74, 78 Permian 234, 238,239 Siluro-Devonian, Carnic/kips 97
284
INDEX
blavi6rite 18, 19, 21 Blumau Formation 98, 102 Bohemograptus bohemicus tenuis 102 Bomel reef 174, 175 Botomaella 29, 32, 34, 35, 36, 37, 38, 41, 45, 46, 47, 48 Botoman stage, carbonate factories, High Atlas 51 68 Bouffioulx Member, Vis6an reef177, 178 Bouvigne, Vis6an reef 177 Brabant Massif 163-164, 164, 169 brachiopods Atokan, Sierra Agua Verde 204 Eifelian-Givetian 112, 114, 116 Famennian reefs 169 Frasnian carbonate mounds 129-130, 131,132, 135 Middle Ordovician, Brittany 73, 74, 78 Ordovician shell beds 6-7 Permian 12, 13, 214, 234, 238, 239, 241,242, 243, 246 Siluro-Devonian 97, 102, 104 Vis6an, Belgium 11,173, 174, 175, 178 breccia, dissolution, Vis6an, Belgium 176-178 Brittany, Middle Ordovician ophiuroid-stylophoran assemblage 72-83 brittlestar beds modern 79-80 see also ophiuroids Brusque nappe 18, 19 bryozoans Famennian reefs 167, 168, 169, 170 fenestellid, Vis6an, Belgium 173-174, 175, 176-178, 179 Frasnian carbonate mounds 130, 132 Ordovician, and shell bed stabilization 6-7 Permian 214, 234, 239, 241,242,243,246 Siluro-Devonian, Carnic Alps 97 buildups, definition 3 Biikk Mountains, Permian carbonates 255,262, 264, 265, 266 calcimicrobes Ba'id area 246 Carnic alps 97 Fammenian 169, 179 La Sentinella reef complex 41, 47 Lemdad syncline 59-60, 63 Vis6an reefs 173 Wenlock patch reef 89-90 calcisponges, Permian Neo-Tethys 241,244, 246,247 Olinal~i Formation 12, 13 calcite, diagenetic cement 23, 25, 26, 46, 215-216 Calcivertella 191 Calix 74, 78 Calymene 73 Cambrian Lower High Atlas reefs 51-68 La Sentinella reef complex 2949 shelled phosphorites 17-27 shell beds 3 Lemdad sycline 58-59 Micmacca Breccia 5 6, 5 'Cambrian explosion' 1, 17 Campine Basin, microbial reefs 163,164, 179-180 Cancellina 262 carbonate facies associations, Lemdad syncline 57-59
carbonate factories cool-water 230, 246, 248,250 Eifelian-Givetian 109 mud-mound 230 tropical 229-230, 246, 248,250 carbonate mounds, Frasnian, palaeoceanography 138-139 carbonate platforms Carboniferous 10-12, 173 Eifelian-Givetian, SW Belgium 109-122, 110 Silurian, Gotland, cementation 7-9 Vis6an, Belgium 173 carbonate productivity, Permian-Triassic boundary 268-269 Carboniferous carbonate platforms Lives Formation 10-11,173 Montagne Noire 11-12 Lower see Mississippian Upper see Pennsylvanian Cardiola Beds 97 Cardiola docens 104 Cardiola Formation 104 Carnic Alps 96-98, 101,102,103, 104-105 Caucasus Permian Triassic biostratigraphy 256-262 cement diagenetic Lower Cambrian phosphorites 22, 23, 24-25, 24, 26 Permian, Tratau reef215 216, 217, 222, 223 fibrous 9 10 Permian Arabian Plate 238, 250 Tratau reef 215-216, 217, 222, 223 La Sentinella reef complex 46 cementation Permian, Arabian Plate 238, 250 Silurian carbonates 7-9 Vis6an microbial reefs 177-178, 179-180 cephalopods Middle Ordovician, Brittany 74, 78 Permian, Arabian Plate 234 Siluro-Devonian, Carnic Alps 97 Chaetetes 192, 192, 194, 195, 196, 197 chaetetids Atokan Sierra Agua Verde 189-198, 192, 193, 204, 206 environment 193-194, 195, 196, 201 Permian, Neo-Tethys 241,242,246, 247 Chancelloria 32, 33, 41 Changhsingian Lagoon 257, 259 channels, pyroclastic 56 chlorosponge association 246, 248 Cimmerian Orogeny 260, 262 Cladogirvanella? 41 Climacammina 191,203, 266 Coates Quarry, Wenlock patch reefs 88, 88, 89-90, 89 Codonofusiella 12, 241,259, 260, 266 Colaniella 232, 239, 250, 259, 266, 270 collophane 21, 24-25 Colpocoryphe 78 community guilds 4, 250 Composita bed 11, 174 conodonts Atokan, Sierra Agua Verde 204 Siluro-Devonian 97, 102
INDEX Coplacoparia 78 coquina 1 coquina-reef transitions 2, 4-13 corals Atokan, Sierra Agua Verde 204 controlling factors 182 Eifelian-Givetian 112, 114, 116 Famennian reefs 169, 170 Frasnian carbonate mounds 9, 129, 130, 131,132, 134, 135 extinction 163, 191,194, 198 Phillipsastraeidae 146-160 growth periodicity 275, 278 Montagne Noire carbonate platform 11-12 Permian 214, 239, 240, 241,245, 246 rugose 143, 149, 150, 151,155 Siluro-Devonian 104 tabulate 104, 143, 149, 155 as indicator of palaeoenvironmental change 275-281 Tournaisian 172 Visran 179-182 Coscinocyathus 32, 33, 35, 36, 37, 39, 41, 42 Cretaceous-Quaternary shell concentration 3 cricoconarids, Frasnian carbonate mounds 130, 132 Crimea, Permian palaeobiogeography 262 crinoids Atokan Sierra Agua Verde 196, 201-207, 205 Eifelian-Givetian La Couvinoise quarry 112, 114 Les Monts de Baileux quarry 112, 114, 116 Famennian reefs 167, 168, 169 Frasnian carbonate mounds 9, 129-130, 130, 131, 132, 134, 135 Ordovician, Brittany 78 Permian Arabian Plate 234 Neo-Tethys 239, 241,242, 243, 246 Tratau reef 214 Siluro-Devonian, Graz Palaeozoic 102 Wenlock patch reefs 88 Croix de Bourges outcrop, Pachyfavosites 275 crossbedding, Lemdad syncline 54, 56 currents, tidal, Lemdad Formation 54 cyanobacteria, Frasnian carbonate mounds 9, 130, 135 Cyclacantharia 13 Cyclocaudex insaturatus 204,205 Cyclocrista martini 204, 205 cyclomorphism 275 see also growth periodicity Cycloscapus laevis 204, 205 Cyrtospirifer condrusorum 145, 146 Dagmarita 256, 262 Dalmanites macrophtalma 73 Deckerella 191 Devonian carbonate mounds 9-10 Middle, Dinant Synclinorium 111 reefs, Austria 95-105 diagenesis, Silurian carbonates 2 9 diagenetic concentration 2 Diamphidiocystis 78 Dictyocyathus 35, 37, 40, 41
285
Dinant Synclinorium 109, 110, 126, 127, 147, 163, 164, 169 Frasnian carbonate mounds 9-10, 125-139, 143, 144 lithostratigraphy 111,145 Waulsortian mounds 163, 164, 170-173, 172 Dinantian, reefs, Belgium 163-186 palaeoenvironment 182-184 stratigraphy 165 Diplosphaerina 214, 259 disphyllids, Frasnian extinction 143, 146 Disphyllum 131,146 dissolution, Tratau reef 215 Dizi, Transcaucasia, Permian palaeobiogeography 260, 262 Dobrotiva Formation, Llandeilian mitrates 78 dolomite argillaceous, Frasnian 151, 153 Eastern Alps 98, 99, 104 'Heraultia beds' 21-22 La Sentinella reef complex 32, 33, 36 Donaldina robusta 204 dykes, neptunian 211,213 formation 224-225 Tratau reef 216-217,218, 219, 220,221,222-225 Dzhulfanella 259, 268 Eastern Alps 95-96, 98, 99,100, 102, 104, 105 echinoderms Middle Ordovician Brittany 71-83, 76, 77, 78 ecology 78-80 taphonomy 74-75, 76, 77-78 Ordovician shell beds 6-7, 6 see also crinoids; ophiuroids; stylophorans Eifelian, palaeogeography 110 Eifelian-Givetian, carbonate platform, SW Belgium 109-122, 111 encrinites Olinal~ Formation 12 Sierra Agua Verde 202-207 Endothyra ex gr. bowmani 191,203 Engihoul Quarry 173 174,175 Engis section 148-151,152, 153,154,155,157,158 Eopolydiexodina 260, 262, 270 Eorhipidomella 78 Eoschubertella texana 191,204 Eostaffella grozdilovae 191,203 epiclasts, Lemdad syncline 54 Epiphyton La Sentinella reef complex 29, 32, 34, 35, 37, 41, 43, 44, 4547, 48 Lemdad syncline 56, 59-60, 60, 61, 65, 66, 68 Erismacoscinus 32, 33, 35, 37, 40, 41, 42, 43 Eugonophyllum 191,193, 194, 197, 203 Evieux Formation 169, 184 extinction 1, 184 Frasnian 143, 146, 150 see also Lower Kellwasser Event; Permian-Triassic Boundary Hangenberg Event 170, 184 'Extragondwan Realm' 262 Famennian reefs, Belgium 163-186 microbial (Baelen type) 166-167, 168, 169, 184 microconchid 166, 169
286
INDEX
palaeoenvironment 182-184 Strnnian bioconstructions 166, 169-170, 170, 184 fenestrae Frasnian carbonate mounds 9-10, 129-130, 130, 131 Tratau reef 217 Findenig Facies 97 Fistulipora 241 fluorapatite 24 Fond-des-Cris section 147-148, 153, 154 foraminifera Atokan, Sierra Agua Verde 203, 204, 206 benthic, late Permian 255-256 Famennian reefs 167, 169 Permian Neo-Tethys 239, 241,246 Tratau reef 214 fractures Tratau reef 215-225 microbial texture 223 see also dykes, neptunian frame reefs 59 framework, definition 166 Frasnian carbonate mounds Belgium 9-10, 125-139,126 Kellwasser events 138, 144, 146, 150, 153, 160, 163 phillipsastreid biostromes, Belgium 143-160 Frechastraea 143, 146, 147, 148,149, 150, 151,152, 153, 154, 155 160 Frechastraea-Alveolites biostrome 154, 155-160, 159 "Frondina' 259 Fusulinella 191,193, 194, 196, 197, 204, 206 fusulinids Atokan, Sierra Agua Verde 190-191,193,203,204 Permian, Olinal~iFormation 12 Gandinocyathus 32, 35, 41, 42 Garwoodia 169 gastropods Atokan, Sierra Agua Verde 204 Eifelian-Givetian 112, 114, 116 Frasnian carbonate mounds 132 Siluro-Devonian 97, 102 Geinitzina 256, 259, 262, 266 Gennev~ir Limestone 262 Georgian Block 262 Gharif Formation 232 Girvanella crusts, Lemdad sycline 59, 60, 61, 66, 68 Frasnian carbonate mounds 130 La Sentinella reef complex 29, 32-35, 36, 38, 41, 44, 45, 46, 47, 48 Wenlock patch reefs 89 Givetian, carbonate platform, SW Belgium 109-122 glauconite, Lemdad syncline 63~o4 Globivalvulina 191,203, 266 Glyptarca 78 Gondwana, western margin, Great Atlasian Reef Complex 67 Goniolinopsis 239 Gordonophyton 47 Gotland, Silurian carbonate platforms 7-9 H6gklint Formation, Wenlock patch reefs 88-93 Gourin, Brittany, Middle Ordovician ophiuroid-stylophoran assemblage 72 83
Graecodiscus 259 Grande Br6che deposits 176 Grands-Malades Formation 176, 178 graptolites Middle Ordovician, Brittany 73, 74 Siluro-Devonian 97, 102 Graz Palaeozoic, Eastern Alps 95-96, 100, 102, 104, 105 Great Atlasian Reef Complex 67 Greece, Permian carbonates 266 growth periodicity 275, 278 growth rates, corals 182, 204-5 guilds see community guilds Gurktal Nappe 99 Gymnocodium 248, 255, 259, 260, 262 haematite, Frasnian carbonate mounds 129, 132 Hangenberg Event, extinctions 170, 184 Hankaxis 130, 146, 151,155, 157 Hanonet Formation 109, 111-122 hardgrounds 8 9 phosphatic, Marcory Formation 20-21, 20 Haushi, Arabian plate Permian carbonate facies 234-238 Permian carbonates, palaeoclimatology 238,248 stratigraphy 231 232, 233 Hawke Bay Regression 66 Hemigordiellina 259 Hemigordiopsis renzi 266 Hemigordius 256, 262, 266 'Heraultia beds' 17, 18, 19 diagenetic cements 22, 23, 24-25 phosphorites 21 22, 22 Heterosteleschus keithi 204, 205 heterozoan carbonates 230, 238,239,242, 243, 246, 248,249, 250, 251 Hexagonaria 145, 146 High Atlas platform geology 52-54 microbial reefs 65-67, 68 highstand systems tracts, Permian 232, 239 Hindeodus 222, 262, 267 Hobbs Quarry, Wenlock patch reefs 88, 88, 89, 90, 93 Hochlantsch Nappe 96 H6gklint Formation, Wenlock patch reefs 88, 89, 90, 91, 92 93 Hohe Warte Limestone 104 Hubeirobuloides 259 Hungary, Permian carbonates 255, 262, 264, 265, 266 Ichnusocyathus 32, 36, 37 Illaenus 73 Inessocyathus spatiosus 41 infill, Tratau reef 215, 216, 217-223 Insolentitheca horrida 191,203 Intermediate Nappe System 96 Ireviken, H6gklint patch reef 90, 91 iron, oxidizing bacteria 132, 134, 135 Isabelinia 78 Issafen Formation 52, 53-54, 53, 65 Issinella 130 Jbel Wawrmast Formation 5, 53 Jemelle Formation 109, 112
INDEX Kalkschiefer Nappe 96 kalyptrae 29, 47, 67 Kamaena? 191,203 Kamurana? 266 Kellwasser see Lower Kellwasser Event Khuff Formation 232, 234, 235, 237, 238, 248, 250 Kok Formation 97 Kollerkogel Formation 103, 104 Komia 191,193, 196, 197, 203 Koninckopora 180 K6tschberg Formation 102 La Couvinoise quarry 109, 111, 112 magnetic susceptibility 119, 120, 121-122 sedimentology 113,114, 115, 116 palaeoenvironment 116-117, 118 119 La Joya Formation 189 Atokan encrinite 201-207 taphonomy 204, 206-207 biostratigraphy 190-191,191,203 La Mallieue section 148-151,152, 153,154, 156, 157 La Sentinella reef complex 29-49, 33 bioherms 30-32 Lower 31, 32-34, 33, 36, 47, 49 Upper 31, 33, 34-35, 34, 37, 47, 49 cements 43, 46 comparison with other bioconstructions 47, 49 evolution 45, 4647, 48 facies 32-35, 36, 37, 38, 39, 46-47 fauna 37, 38, 39, 40-41 geology 29-30 matrix 45-46 microbial assemblages 41 Lady Burn starfish beds, taphonomy 75 Lambermont Formation 144, 146, 147 Lamprosterigma 204, 205 Lasiodiscus 241,259 Lastours Formation 17, 18, 19, 21 Laufnitzdorf Nappe 96 Lea Quarry, Wenlock patch reefs 88, 88, 8%90 leiolite 89, 90, 92 Lemdad Formation 52, 53, 65 Lemdad syncline carbonate facies associations 57-59 geology 52-54 palaeogeography 65-66 parasequences 55, 64-65 pyroclastic deposits 56-57 shell beds 5, 58-59 stratigraphy 53 volcanoclastic facies associations 52-57 Leptodus 259, 260 Lercaritubus 246 Les Monts de Baileux quarry 109, 112, 114 magnetic susceptibility 119, 120, 121-122 microfacies 115, 116 palaeoenvironment 117-118, 119 sedimentology 113 level-bottom communities 169 limestone argillaceous Eifelian-Givetian 112 Frasnian 9, 147-151, 153 Vis6an 11
287
Cambrian, La Sentinella reef complex 30-37 Carnic Alps 97, 98, 104 dolomitic, 'Heraultia beds' 21~2 Frasnian 129, 130 late Permian 255 Micmaeca Breccia 5-6 Much Wenlock Limestone Formation 88 90, 93 phosphatic, Marcory Formation 20-21 see also dolomite Lingulina 259 Lion member Frasnian carbonate mounds 125, 127, 131,133 buildup architecture 136, 137-138 buildup facies 130, 131, 132 facies interpretation 132, 135-136, 137 palaeoceanography 138-139 Lithostrotion 182 Lives Formation, Vis6an carbonate platform 10-11,173, 174, 178 Llandeilian, Brittany, ophiuroid assemblage 72-83 Lower Kellwasser Event 138, 144, 146, 150, 153, 160, 163 Lower-Ktaoua Formation, Anti Atlas, Morocco 6-7 lumachelle 1 Lustin Formation 138, 144, 145, 146, 147, 148, 149, 150 Macgeea 146 Macroporella preromangica 266 magnetic susceptibility 119, 120, 121 Maizeret Member 176 Marcory Formation 17-18, 18, 19 diagenetic cements 22, 23, 24-25 phosphorites 20-21 marl, Gotland carbonate 7, 9 Matoppa Formation 29, 32 geology 29-30, 31 Mediocris breviscula 191 Megaerella Limestone 97 M61agues nappe 18, 19 phosphorites 20-21 Mesogondolella 222 Mesophyllum 104 Metalegoceras Limestone 232 metazoans 1-2 'Cambrian explosion' 17 Mexico Olinalfi Formation, Permian shell beds 12-13 see Sierra Agua Verde Michelina 172, 193, 204 Miemacca Breccia, microbial mats 5-6, 5 Micmacca? albesensis 22 micrite Alticola Limestone 97 La Sentinella 45 Permian Arabian Plate 234 Neo-Tethys 239 Tratau reef 215-216, 223 Wenlock patch reefs 89, 90-92 microbial crusts 11-12 La Sentinella 34, 35, 37, 41, 44, 45, 46, 47, 48 microbial mats Devonian gel 10 Frasnian carbonate mounds 10, 129, 130, 135 Micmacca Breccia 5-6, 5
288
INDEX
nomenclature 4 see also stromatolites; thrombolites microbial mounds 104 microbial reefs Famennian 167 Lemdad syncline 59-64, 66-67, 66 Vis6an 173-176, 176-180 microbiolites 230, 250 Tratau reef 222, 223 microconchid reefs Famennian 169-170 Vis6an 173, 178-179 microspherules, cosmic, Permian-Triassic boundary 256, 266-267 Midiella zaninettiae 259 Miliolipora cuvillieri 269 Millerella 191,203 Millingsklint, H6gklint patch reef 90 Minervois nappe 18, 19 Mississippian reefs, Belgium 163-166, 170-186 shell beds, microbial communities 10-12 Mitcheldeania 169 Mitrocystella 72, 74, 77 78, 78, 81-83 Mizzia 239, 247, 248, 255,260 molluscs see bivalves; cephalopods Montagne Noire Carboniferous carbonate platform 11-12 geology 17-19, 18 Lower Cambrian phosphorites 19-27 palaeogeography 19 Mooreanteris waylandensis 204, 205 Morocco Anti Atlas, geology 52-54 Atlas Mountains, 'Sala'irian' tectonic phase 51 Botoman reefs 51-68 High Atlas, geology 52 54 mounds carbonate, Frasnian, Belgium 9 10, 125-139 cryptalgal 3 definition 34, 4, 166 microbial 3, 104 mud 3 carbonate factories 230 Famennian reefs 166-167,168, 169 Olinahi Formation 12 Waulsortian 163, 164, 170-173, 172, 184, 246 skeletal 3 Much Wenlock Limestone Formation, patch reefs 88-90, 93 mud-mounds see mounds, mud Multidiscus 259, 262, 266 Multithecopora 194, 239, 240
obrution deposits, Middle Ordovician echinoderms 74-75, 78 Olinal~ Formation, Permian shell beds 12-13 olistolites, Permian 262 oncoids 57, 58 oncolite 11, 30, 31, 32, 36, 47, 48 Ophiura 71, 74-75 ophiuroids Middle Ordovician, Brittany 72-83 ecology 78-80 taphonomy 74-78 modern 71, 74-75, 79-80 Ordovician biodiversification 3 Middle, Brittany, ophiuroid assemblage 72-83 shell beds 3 Lower-Ktaoua Formation 6-7 Orientoschwagerina abichi 260 Orionastrea 10 Orthis berthoisi 73 Orthoceras 112 ostracods Famennian reefs 167 Frasnian carbonate mounds 132 Permian Neo-Tethys 239, 246 Tratau reef 214 Siluro-Devonian, Southern Burgenland 102 Ounein, archaeocyathan-microbial reef complexes, Lemdad syncline 52, 53, 55, 61-64, 62, 65
Nagyvisny6 Limestone 262 Namur Synclinorium 125, 127, 143, 144, 145, 146, 163,164, 173, 178 Engis and La Mallieue sections 148-153,155, 156, 157, 158 Namur-Dinant Basin geology 143-144, 146 Nankinella 259 nautiloids Ordovician 78 Siluro-Devonian Austria 97, 102
Pachyfavosites, palaeoenvironmenal change 275-281 Pachyphloia 241,256, 262 Pachyphloides 259 Pachyporidae 146 Palaecoma spinosa systematic palaeontology 80-81 Palaeo-Tethys, eastern Europe, palaeobiogeography 256, 262, 268 Palaeoaplysina 213, 214 palaeobathymetry, Frasnian carbonate mounds 137, 138 palaeoberesellids, Famennian reefs 167 palaeobiogeography, late Permian correlations 256, 260, 262, 268, 269-270
Navalacria compacta 145 nebuloids, Frasnian carbonate mounds 9-10, 130 Neo-Tethys eastern Europe, palaeobiogeography 256, 262,268 Permian carbonate facies 241,242-245, 246, 247, 248,250 Permian palaeoclimatology 246, 248-250 stratigraphy 232,233-234 Neoendothyra 246, 256 Neoproterozoi~Cambrian transition 17 Neoschwagerina 239, 262 neptunian dykes see dykes, neptunian Neseuretus 74, 78 Neuville Formation 144, 145, 146 Nikitin section, 256-262,258 Nipperella 191,204 Nobiliasaphus 78 Nodosaria 259 Nodosinelloides 241,256, 259 nodules, Frasnian carbonate mounds 132 nomenclature 1-4
INDEX palaeobiology, Dinantian reefs, Belgium 181-182 palaeoceanography, Frasnian carbonate mounds 138-139 palaeoclimatology, Permian Arabian plate 238, 239, 241,248-250 Neo-Tethys 246, 248-250 palaeocurrents, Lemdad sycline 54, 56, 58 palaeoenvironment, Dinantian reefs, Belgium 182-184 palaeoenvironmental change, Pachyfavosites 275-281 Palaeofusulina 259, 270 palaeogeography Atokan 197 Lemdad syncline 65--66 Palaeomicrocodium 130 palaeosiphonocladeles, 130, 132, 135 Panderia 78 Parabarrandia 78 Paradagmarita 262, 266 Parafusulina 262 Paraglobivalvulina 259, 262, 266 Paraglobivalvulinoides 256 Parananlingella 259 Parareichelina 259, 266, 268, 270 parasequences, Lemdad syncline 55, 64-65 Paraskinnerella 262 Pardailhan Formation 18, 19 patch reefs 11 Vis6an 173, 176, 178 Wenlock 87-93 pavements abrachiopod 102, 104 Cardiola shell 104 colonized, Siluro-Devonian 98, 99 Pecopteris polymorpha 256 peloids Cambrian, Lemdad syncline 57, 59, 61 Eifelian, Belgium 117 Famennian reefs 167 Frasnian carbonate mounds 129, 130, 132, 135 Permian Arabian Plate 234 Neo-Tethys 239 Tratau reef 215,223 Vis6an, Belgium 177, 178, 179 Peneckiella 146, 155 Pennsylvanian, Sonora biostratigraphy 190-191 geology 189-190 late Atokan chaetetids 189-198 late Atokan crinoids 201-207 Pentagonopteris insculptus 204, 205 Pentaridica simplicis 204, 205 Permian carbonate factories 230, 248,249, 250 carbonates, Arabian plate 230-251 late, palaeobiogeography, eastern Europe 256-270 shell beds, Olinalfi Formation 12-13 Tratau reef 213-225 Permian-Triassic boundary 255, 262, 264 C isotopic anomaly 267-268 Ca/Mg ratio 267-268 carbonate productivity 268-269 cosmic microspherules 256, 266-267 Permocalculus 234, 237, 248, 255, 259, 260, 262, 266, 268-269, 270
289
Petit-Mont Member Frasnian carbonate mounds 9-10, 125, 128, 133 buildup architecture 136-137, 138 buildup facies 129-130, 134 facies interpretation 132, 134, 134-135, 136-137 palaeoceanography 138 Phacopidina 74, 78 Philippeville Anticlinorium Frasnian carbonate mounds 9-10, 125, 126, 127,128, 144, 144 Philippeville Formation 144, 145 Phillipsastraea 9, 129, 130, 144, 146, 148, 151,154, 155, 156 phillipsastraeidae biostromes, Frasnian, Belgium 143-160 phosphate shells, Lower Cambrian 17-27 phosphogenesis 17, 24-25 phosphorites, Lower Cambrian, Montagne Noire 17-27 photozoan carbonates 230, 238, 239, 244, 246, 247, 248, 249, 250, 251 Plabutsch Formation 104 Placoparia 74 Plaesiacomia 74 Platycrinites ex gr. laevis 259 Pl6ken facies 97 Polydiexodina capitanensis 12 Polygnathus 111 Porocoscinusflexibilis 37, 41, 43 Praenucula 78 Praesumatrina 262 Praewentzelella 240, 245, 246 Prayon section 147-148, 153 Preptopremnum 204, 205 Prionocheilus 74, 78 Proaulopora, Lemdad syncline 61, 62, 65 Proninella 130 Protohaploxypinus 259 Protonodosaria 241 Protopharetra 32, 35, 37, 39, 40, 41, 43 Psammichnites 19 Pseudodunbarula 260 Pseudostaffella 19l, 203 Pseudovermiporella 259, 262 pyrite 21, 23, 24 pyroclastic deposits, Lemdad syncline 54, 56-57, 65, 67 Radiosphaeroporella 130 radiospheres, Frasnian carbonate mounds 130, 132 Rannach Nappe 96 Rauchkofel Limestone 97, 98 microbial mounds 102, 104 Rauserella 246 Razumovskia 29, 32, 33, 34, 35, 38, 39, 41, 44, 45, 46, 47, 48 Receptaculites 9, 112, 129 Rectocornuspira kahlori 259, 266 Redonia 73, 78 reefs definition 1-2, 3-4, 4, 98, 164, 166 development controlling factors 181 Famennian and Vis6an 181-184 Reichelina 259 Renalcis bushes, Lemdad syncline 59, 60, 61 Carnic Alps 103, 105
290 Frasnian carbonate mounds 130, 131 La Sentinella reef complex 29, 34, 35, 37, 39, 4041, 43, 44, 45-47, 49 Ribeiria 78 Richthofenia 259, 260, 266 richthofeniids 12, 13 rifting Carnian 260 Moroccan Atlas 51 Robuloides 259, 262 Rothpletzella 89, 130 Rugosa 150-151,155 Frasnian extinction 143 Russia, neptunian dykes, Ural Mountains 211-225 Saih Hatat, Arabian plate Permian carbonate facies 238-239,240, 248, 250 Permian stratigraphy 232-233 Saiq Formation 232,232 Permian carbonate facies 238-239,240 Saiwan Formation 231,232, 234,235 Permian carbonates 234, 235,236, 238, 248 'Salairian' tectonic phase, Moroccan Atlas 51 salinity, effect on bioconstruction 183 Salterocoryphe 74, 78 sandforms, Lemdad syncline 54, 56 Sardinia, La Sentinella reef complex 29-49 Schistes de Postolonnec Formation 72 74 Middle Ordovician echinoderms 74, 76, 77, 78, 78 schistosity, crenulation 20 Sch6ckel Nappe 96 Scoliopora 153, 154, 155 Scyphocrinites 97 sea level Cambrian, Lemdad syncline 68 effect on bioconstruction 183, 184 Frasnian Aisemont Formation 144 Carbonate mounds 137, 138-9 Permian Arabian plate 238, 239, 241,248,250 Neo-Tethys 246, 248 sedimentary concentration 2 Selenopeltis 78 Senceliaepora 9, 129 Septatrypa subsecreta 102, 103 Septoglobivalvulina? ex gr. decrouezae 262, 266 serpulids, Upper Silurian 98, 102, 103 shale, Lemdad syncline 57 Shamovella 230, 241,243, 244, 246, 247, 250 Shanita 270 shell beds 3 Lemdad sycline 58-59 Lower-Ktaoua Formation 6-7 Micmacca Breccia 5-6 modes 3 shell concentration 3 definition 1, 2 Lives Formation 10-11 stabilization 5-12 Shindella? 262, 266 shoals epiclastic 54 grainstone 57 pyroclastic 56-57
INDEX Sierra Agua Verde 202 biostratigraphy 190-191,203-204, 203 late Atokan chaetetids 189 198 late Atokan encrinite 201-207 Sonora, geology 189-190 Silurian Middle, patch reef 87-93 shell beds, early diagenesis 7-9 Siluria~Upper Devonian, reefs, Austria 95-105 Siphonodendron 180-181, 182 skeletal concentrations 2 modes 3 Skinnerella 262 Southern Alps 96-98, I01,104~105 Southern Burgenland, Eastern Alps 98,100, 102, 102, 103 Souvrain-Pr6 Formation 166, 167 Sphaerocodium 151, 154, 155 Sphenophyllum oblongifolium 256 sponges Famennian reefs 167 Frasnian carbonate mounds 129, 132, 134 Stachyodes 103, 104 Staffella 191,194, 203 storms effect on bioconstruction 117, 183 effect on shell accumulations 3, 4 nebuloids 9-10 Striatopora suessi 103 Stringophyllum 104 stromatactis Famennian reefs 167, 171 Frasnian carbonate mounds 9-10, 129, 130, 131,132, 134 La Sentinella 45 Permian, Tratau reef 215,216, 217 stromatactoids Famennian reefs 167, 168 Vis6an 177, 179 stromatoids crusts, Lemdad syncline 59 intergrowth, Lemdad syncline 6(~61, 60, 66 nomenclature 4 stromatolites 3 agglutinated 59, 60 Frasnian Arche and Lion members 130 Frasnian Petit-Mont member 130 Micmacca Breccia 5 6 nomenclature 4 Permian Olinal~ Formation 12, 13 Tratau reef 223 Vis6an, Belgium 173, 174, 176, 178-179 stromatoporoids Carnic Alps 103, 104 Eifelian Givetian La Couvinoise quarry 112, 114, 117 Les Monts de Baileux quarry 112, 114, 116 Famennian reefs 169-170 Frasnian carbonate mounds 9, 129, 130, 131,132, 135 Alveolites association 148, 149, 153 155 Vis6an 182 Wenlock patch reefs 89 'structure grumeleuse' 135, 215 Strunian bioconstructions, Famennian reefs 163,166, 169-170,170, 182, 184
INDEX stylolites 7, 20, 97, 130 stylophorans, Middle Ordovician, Brittany 72, 74-80 Syringopora 193 Syzrania 191
Tabulaeyathus insperatus 35, 41, 42 Tabulata Frasnian extinction 143 as indicator of palaeoenvironmental change 275-281 Tabulophyllum 146, 155 Taeniaster 72, 74, 78, 78, 80-81 taphonomic feedback 2, 206 taphonomy Atokan crinoids, Sierra Agua Verde 204, 206-207 Middle Ordovician echinoderms, Brittany 74-78 Taphrelminthopsis 19 Tarthinia 47 Tauride Formation 262 Taylorcyathus vologdini 35, 40, 41 Tchernyschevia 266 tectonism, synsedimentary Lemdad syncline 65-6, 67 Tratau reef 211,216, 219, 223-225 temperature effect on bioconstruction 183 and growth periodicity, tabulate coral 278-281 tempestite 204, 206, 235, 238 Tethys Ocean, Dorashamian 261,263 Thalassinoides 239 Thamnophyllum 9, 104, 129 Thamnopora 9, 104, 129, 130, 146, 155 Thecostegites 9, 129 thromboids intergrowth, Lemdad syncline 60-61, 60, 66 nomenclature 4 Tratau reef 223 thrombolites 3, 4, 32, 36, 47 Frasnian carbonate mounds 130, 131, 135 La Sentinella reef complex 32, 36, 46, 47 nomenclature 4 Vis6an 176, 177, 178 tides, influence on epiclastic strata 54 Tiocyrspis bironensis 145, 146 Tommotian Atdabanian transition 18, 19 Tournaisian reefs 170-173, 184 Transcar Quarry 176 Transcaucasia, Permian palaeobiogeography 260, 262 transgressive systems tracts Atokan 194, 201 Permian, Arabian plate 232, 239 transgressive-regressive cycle Frasnian Aisemont Formation 144, 145, 146-151,148, 153, 160 carbonate mounds 138 9 Permian, Tratau reef212 transition 1 coquina-reef 2, 4-13 mound-reef 4 Neoproterozoic-Cambrian 17
291
Tratau reef212, 213,213-225 diagenesis 21 4-216 fossil record 222 microbial texture 223 synsedimentary tectonism 211, 216, 219, 223-225 Trelonella 130 trilobites Ashgillian, Lower Ktaoua Formation 7 Cambrian Av6ne-Mendic Unit 22 Lemdad syncline 59, 61 Eifelian-Givetian La Couvinoise quarry 114 Les Monts de Baileux quarry 112, 116 Frasnian carbonate mounds 132 Middle Ordovician, Brittany 73, 74, 78 Siluro-Devonian, Carnic Alps 97 Trocholites 78 Trois-Fontaines Formation 109, 111, 112, 114 Tubicoscinus cupulosus 35, 38, 41 Tubiphy tes 214, 243 turbidity 51 High Atlas platform 66, 67, 68
Upper Nappe System 96 Ural Mountains geology 212, 213 Permian reefs 211-216 synsedimentary tectonism 216,219, 223-225 Valisettes Shale Formation 144, 146 Vattenfallet, H6gklint patch reefs 92, 93 Verbeekina 239 Vesdre Massif 143, 144, 146, 163, 164 Alveolites-stromatoporoid biostrome 153-155, 158-160 Prayon and Fond-des-Cris sections 14~148, 153 Strunian biostromes 169-170 Vis6an reefs 10-11,173-174, 176-181 Belgium global comparison 184-185 palaeobiology 181-182 palaeoenvironment 182-184 volcanism Botoman, High Atlas 51 2, 67 and turbidity 51 volcanoclastic facies associations, Lemdad syncline 54-57
Waagenophyllum 239, 240, 246, 259 Waulsortian mounds 163, 164, 170~173, 172, 183, 184, 246 wave action, and reefs 2, 3 Wenlock, patch reef 87-93 Wentzelella 245 Wetheredella 89, 130 Zdimir cf. hercynicus 104 Zechstein facies 255, 269 Zeliskella 74, 78 Zidella? 191,203