VSIA01 9/12/08 12:58 PM Page i
VOLCANICLASTIC SEDIMENTATION IN LACUSTRINE SETTINGS
Volcaniclastic Sedimentation in La...
56 downloads
1060 Views
5MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
VSIA01 9/12/08 12:58 PM Page i
VOLCANICLASTIC SEDIMENTATION IN LACUSTRINE SETTINGS
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
VSIA01 9/12/08 12:58 PM Page ii
DEDICATION
Mario Martín Mazzoni, 1943–1999 Dr Mario Mazzoni died on Friday 1 October 1999 of a heart attack, in Quilmes, Argentina. Mario was 56 years old. He received his PhD from the Universidad Nacional de La Plata (UNLP), Argentina, and there, together with a few associates, established the Centro de Investigaciones Geologicas (Centre for Geological Investigations). He was also a senior scientist with CONICET (Consejo Nacional de Investigacion Cientifica y Tecnica: National Consortium for Scientific and Technical Investigation) and Professor of Geology at UNLP. Mario took his family to Santa Barbara, California, in 1982 to undertake a post-graduate fellowship there with R. V. Fisher. He began a long-term collaboration with R. V. at that time, and in 1988 came back to Santa Barbara to join a group of graduate students (including the editors of this volume) travelling around the western USA looking at calderas and stratovolcanoes. Mario was the premier authority on volcaniclastic rocks in Argentina, most recently beginning investigations of the Quaternary Copahue volcano and a proposed Caviahue caldera. Mario loved to travel, and welcomed colleagues from around the world to Argentina. Mario was a kind, gentle, generous person and is sorely missed by those of us who were fortunate enough to be able to work with him, or just cross paths.
VSIA01 9/12/08 12:58 PM Page iii
S P EC I A L P UBL I C AT I ON N U MB E R 30 OF T H E I NT ER NA TI O NA L A S S OCI AT I ON OF SE D I ME N T OL OG I ST S
Volcaniclastic Sedimentation in Lacustrine Settings EDITED BY JAMES D. L. WHITE AND NANCY R. RIGGS
VSIA01 9/12/08 12:58 PM Page iv
© 2001 by Blackwell Science Ltd Editorial Offices: Osney Mead, Oxford OX2 0EL 25 John Street, London WC1N 2BS 23 Ainslie Place, Edinburgh EH3 6AJ 350 Main Street, Malden MA 02148-5018, USA 54 University Street, Carlton Victoria 3053, Australia 10, rue Casimir Delavigne 75006 Paris, France Other Editorial Offices: Blackwell Wissenschafts-Verlag GmbH Kurfürstendamm 57 10707 Berlin, Germany Blackwell Science KK MG Kodenmacho Building 7–10 Kodenmacho Nihombashi Chuo-ku, Tokyo 104, Japan Iowa State University Press A Blackwell Science Company 2121 S. State Avenue Ames, Iowa 50014–8300, USA The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the copyright owner.
DISTRIBUTORS Marston Book Services Ltd PO Box 269 Abingdon, Oxon OX14 4YN (Orders: Tel: 01235 465500 Fax: 01235 465555) USA Blackwell Science, Inc. Commerce Place 350 Main Street Malden, MA 02148-5018 (Orders: Tel: 800 759 6102 781 388 8250 Fax: 781 388 8255) Canada Login Brothers Book Company 324 Saulteaux Crescent Winnipeg, Manitoba R3J 3T2 (Orders: Tel: 204 837 2987) Australia Blackwell Science Pty Ltd 54 University Street Carlton, Victoria 3053 (Orders: Tel: 3 9347 0300 Fax: 3 9347 5001) A catalogue record for this title is available from the British Library ISBN 0-632-05847-1 Library of Congress Cataloging-in-publication Data Volcaniclastic sedimentation in lacustrine settings / edited by Nancy R. Riggs and James D. L. White. p. cm. ISBN 0-632-05847-1 1. Volcanic ash, tuff, etc. 2. Sedimentation and deposition. 3. Lake sediments. I. Riggs, Nancy R. II. White, James D. L.
First published 2001 Set by Graphicraft Limited, Hong Kong Printed and bound at the Alden Press, Oxford and Northampton The Blackwell Science logo is a trade mark of Blackwell Science Ltd, registered at the United Kingdom Trade Marks Registry
QE461.V632 2001 551.48′2—dc21 For further information on Blackwell Science, visit our website: www.blackwell-science.com
00-045489
VSIA01 9/12/08 12:58 PM Page v
Contents
vii Preface 1
9
Introduction: styles and significance of lacustrine volcaniclastic sedimentation J. D. L. White & N. R. Riggs Eruptions and eruption-formed lakes Lithofacies architecture and construction of volcanoes erupted in englacial lakes: Icefall Nunatak, Mount Murphy, eastern Marie Byrd Land, Antarctica J. L. Smellie
35
Eruptive process, effects and deposits of the 1996 and ancient basaltic phreatomagmatic eruptions in Karymskoye lake, Kamchatka, Russia A. Belousov & M. Belousova
61
Eruption and reshaping of Pahvant Butte volcano in Pleistocene Lake Bonneville J. D. L. White
83
Sedimentation and re-sedimentation of pyroclastic debris in lakes Influence of magmatism and tectonics on sedimentation in an extensional lake basin: the Upper Devonian Bunga Beds, Boyd Volcanic Complex, south-eastern Australia R. A. F. Cas, C. Edgar, R. L. Allen, S. Bull, B. A. Clifford, G. Giordano & J. V. Wright
109
Sedimentology and history of Lake Reporoa: an ephemeral supra-ignimbrite lake, Taupo Volcanic Zone, New Zealand V. Manville
141
Settling and deposition of ad 181 Taupo pumice in lacustrine and associated environments J. D. L. White, V. Manville, C. J. N. Wilson, B. F. Houghton, N. R. Riggs & M. Ort
151
Post-1.8-ka marginal sedimentation in Lake Taupo, New Zealand: effects of wave energy and sediment supply in a rapidly rising lake N. R. Riggs, M. H. Ort, J. D. L. White, C. J. N. Wilson, B. F. Houghton & R. Clarkson
179
Lacustrine–fluvial transitions in a small intermontane valley, Eocene Challis volcanic field, Idaho B. A. Palmer & E. P. Shawkey v
VSIA01 9/12/08 12:58 PM Page vi
vi
199
225
Contents
Volcanic and hydrothermal influences on middle Eocene lacustrine sedimentary deposits, Republic Basin, northern Washington, USA D. R. Gaylord, S. M. Price & J. D. Suydam Lakes as sensitive recorders of eruptions and the response of distal landscapes Tephra layers in a sediment core from Lake Hestvatn, southern Iceland: implications for evaluating sedimentation processes and environmental impacts on a lacustrine system caused by tephra fall deposits in the surrounding watershed J. Hardardóttir, Á. Geirsdóttir & T. Thórdarson
247
Late Pleistocene–Holocene volcanic stratigraphy and palaeoenvironments of the upper Lerma basin, Mexico M. Caballero, J. L. Macías, S. Lozano-García, J. Urrutia-Fucugauchi & R. Castañeda-Bernal
263
Environmental and tectonic controls on preservation potential of distal fallout ashes in fluvio-lacustrine settings: the Carboniferous–Permian Saar–Nahe Basin, south-western Germany S. Königer & H. Stollhofen
285
Deposition of Mount Mazama tephra in a landslide-dammed lake on the upper Skagit River, Washington, USA J. L. Riedel, P. T. Pringle & R. L. Schuster
299
Index
VSIA01 9/12/08 12:58 PM Page vii
Preface
The editors conceived this volume in response to the growth of available literature on volcaniclastic sedimentation and on lacustrine sedimentation. Although isolated papers have addressed these themes together, this is the first attempt to assemble a group of such contributions, and the first to emphasize how volcanic eruptions can form lakes, influence or control deposition in lakes, and indeed be greatly modified themselves by occurring within lakes. In addition to support from the International Association of Sedimentologists, which provided leadership, editorial assistance, and the interface between volume editors and Blackwell Science, the volume has been sponsored by the Commission on Volcaniclastic Sedimentation (CVS) of the International Association for Volcanology and Chemistry of the Earth’s Interior (IAVCEI), through which we invited papers on a broad range of topics in the fields of lacustrine volcaniclastic sedimentation. We were very pleased with the response to our invitation, and, 3 years later, are proud to present the results in this volume. We are deeply indebted both to the authors who have contributed to the volume and patiently awaited its completion, and to the reviewers who returned manuscripts promptly and whose comments and insight invariably improved the quality of the papers included here. Vern Manville kindly provided a modified version of a diagram from his contribution as a model for the volume’s cover art. Finally, we heartily thank Guy Plint, IAS Special Publications editor, who showed phenomenal patience as we chased final authors and reviewers to the close of the process. Mike Talbot, Bergen, Norway R. V. Fisher, Santa Barbara, CA, USA Grant Heiken, Los Alamos, NM, USA David Lowe, Hamilton, NZ Christoph Breitkruz, Potsdam, Germany Peter Ballance, Auckland, NZ Richard Hanson, Fort Worth, TX, USA Chuck Landis, Dunedin, NZ Thor Thordarson, Perth, Australia Roger Suthren, Oxford, UK John Smellie, Cambridge, UK Dorrik Stowe, Southampton, UK Bruce Houghton, Wairakei, NZ Mario Mazzoni (deceased), La Plata, Argentina
The reviewers are: Beth Palmer, Northfield, MN, USA David Gaylord, Pulman, WA, USA Rebecca Dorsey, Eugene, OR, USA Vern Manville, Taupo, NZ Sarah Metcalfe, Edinburgh, UK Larry Middleton, Flagstaff, AZ, USA Patrick Pringle, Olympia, WA, USA Stephan Königer, Würzburg, Germany Nick Foit, Pulman, WA, USA William Fritz, Atlanta, GA, USA Gary Smith, Albuquerque, NM, USA Peter Kokelaar, Liverpool, UK Reinhard Werner, Kiel, Germany Jocelyn McPhie, Hobart, Australia
vii
VSIA02 9/12/08 12:57 PM Page 1
Spec. Publs int. Ass. Sediment. (2001) 30, 1–6
Introduction: styles and significance of lacustrine volcaniclastic sedimentation J . D . L . W H I T E * and N . R . R I G G S † *Geology Department, University of Otago, PO Box 56, Dunedin, New Zealand; †Geology Department, Northern Arizona University, Box 4099, Flagstaff, AZ 86011, USA
INTRODUCTION The focal point of this volume, Volcaniclastic Sedimentation in Lacustrine Settings, is the lacustrine depositional record of volcanism. Lakes are a common feature in volcanic terranes, and their deposits are distinctive and useful for interpreting landscape evolution, as well as for trying to gain a systematic understanding of the behaviour of volcanic clasts and dispersal processes. The lacustrine depositional record includes contributions from eruptions taking place far upwind of lakes and their catchment areas, from intracatchment eruptions that substantially alter lacustrine depositional systems, and from eruptions that take place in, and are partly shaped by, standing water in lakes. Pumiceous lake sediments show a range of features that do not readily fit into facies schemes developed in lakes unaffected by volcanism. Indeed, lacustrine depositional systems offer a natural laboratory for separating the effects of grain size from those of grain mass during deposition and transport of vesicular volcanic fragments. Lacustrine successions are particularly interesting components of volcanic regions for at least four reasons: 1 the formation of lakes in many cases is a direct or indirect effect of volcanic eruptions, which can result in obstruction of streams, melting of ice, or creation of deep topographic depressions such as calderas and maars; 2 standing water strongly modifies the style of terrestrial eruptions and dispersal of tephra, and can result in pyroclastic deposits unique among terrestrial volcanic sequences; 3 the aqueous sedimentation of volcanic clasts, which vary strongly in density and thus settling behaviour, is of general sedimentological interest and most directly investigated in deposits from standing water; 4 lakes include uniquely low-energy terrestrial sedimentary environments, in which the most detailed and distal records of volcanic fall can be preserved.
VOLCANIC LAKES Lakes form as a result of volcanic eruptions in a variety of ways. Perhaps the most direct is by the conversion of ice to water by volcanic heat, which forms englacial and subglacial lakes in response to eruptions such as those discussed in this volume by Smellie (pp. 9–35; see also Jones, 1969, 1970; Skilling, 1994; Smellie & Skilling, 1994; Smellie & Hole, 1997). In addition, because glaciers themselves can effectively impound lakes, there may be complex interrelationships among volcanism, development of lakes, and the sedimentary record of this interplay (Werner et al., 1996). Another readily envisioned process by which volcanic eruptions can form lakes is the damming of streams by lava flows. Although this process is surprisingly ineffective in many situations because of the high permeability of jointed and flow-brecciated lavas (Segerstrom, 1950; Young & Jones, 1984; White, 1991; Hamblin, 1994), large lava flows may effectively impound lakes for long periods. This process seems to be particularly effective in silicic volcanic fields (Nairn, 1989; Palmer & Shawkey, 1997). The resulting lakes commonly provide key information for unravelling the eruptive histories of surrounding volcanoes, as shown by the contribution of Palmer & Shawkey in this volume (pp. 179–199). Streams can also be dammed directly by pyroclastic eruptions, which may disrupt drainage over large areas. Small-volume eruptions may temporarily raise the level of existing lakes (White et al., 1997), but large eruptions may result in much larger-scale reorganizations of drainage patterns (Buesch, 1991). As a result of the 1.8 ka eruption of Taupo volcano in New Zealand, ignimbrite not only dammed the outlet to Taupo caldera itself (Wilson & Walker, 1985; White et al. this volume, pp. 141–151), but also impounded
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
1
VSIA02 9/12/08 12:57 PM Page 2
2
J. D. L. White and N. R. Riggs
significant temporary lakes both adjacent to (Smith, 1991) and well beyond the caldera (Manville et al., 1999; Manville, this volume, pp. 109 –141). An earlier Taupo Volcanic Zone eruption produced the Rotoiti ignimbrite from Okataina caldera (Nairn, 1989). The ignimbrite dammed the outlet to Lake Rotorua (which occupies a separate caldera), raising the lake level by 90 m and impounding it at that level for some 20 kyr (Kennedy, 1994).
OTHER LAKES IN VOLCANIC ENVIRONMENTS Not all lakes in volcanic environments owe their origin to volcanic processes, and this is particularly true in active convergent and rifting plate-margin settings, where volcanism and tectonic subsidence are separate but spatially coincident manifestations of lithosphericscale processes. In their contribution here, Gaylord et al. (pp. 199 –225) show that volcanogenic lake sedimentation was favoured in the highly extended region represented by the Republic Graben of Washington state because of a combination of rapid basin subsidence, moist climatic conditions, an abundant supply of loose volcanic detritus, and the topographically elevated and isolated nature of the Okanogan Highlands.
LACUSTRINE ERUPTIONS The first section of the volume, Eruptions and Eruption-formed Lakes, is represented by Smellie’s account (pp. 9–35) of intraglacial eruption and deposition of a volcaniclastic suite within the eruptionformed lake. He presents a new model for the ways in which glacial hydrology affects impoundment of intraglacial lakes and the resulting drainage pathways (see also Smellie, 2000a). Smellie’s model has wideranging implications for analysis of volcaniclastic successions in ice-dammed lakes of any origin. Once subglacial eruptions have melted enclosing ice to form a body of standing water, eruption processes converge with those typical of other subaqueous, ‘Surtseyan’, eruptions in which abundant water interacts with erupting magma to produce distinctive cypressoid jets and steam-laden eruption plumes (White & Houghton, 2000). Lakes in which Surtseyan eruptions occur may themselves be a result of volcanic activity, as is the case for the eruption and deposits discussed by Belousov & Belousova (pp. 35 – 61) that developed in the caldera lake occupying the edifice of
Karymskoye volcano. In 1996, seismic activity alerted the Kamchatka Volcanological Observatory to activity at Karymskoye caldera. A helicopter overflight revealed repeated bursts of ash-laden water and laterally expanding steam currents from the previously icecovered lake. Subsequent investigation showed that the eruption caused tsunami (seiches) to travel across the lake, overflowing into the Karymskaya river to form downstream lahars (debris flows and hyperconcentrated flows) and floods. A small tuff ring was built above the lake surface by the eruption, and depositional features of the ring indicate distribution of ash and other debris, including large blocks of the ice that covered the lake before eruption, by a variety of mechanisms as first the eruption and then the volcano itself shoaled above the lake. Eruptions initiated beneath non-volcanic lakes share the wide variety of eruption processes, but additionally may provide unique insights into the history of the enclosing lake itself. Pahvant Butte volcano erupted into Lake Bonneville near its highstand level (Gilbert, 1890; Oviatt & Nash, 1989; White, this volume, pp. 61–83), and ash from the eruption provides a time plane and a marker of shoreline position unparalleled in the Bonneville basin. Reworked ash on the volcano itself is a sensitive water-level indicator, and provides critical support for a late lake-level oscillation in Lake Bonneville just before its catastrophic breakout into the Snake River catchment (Gilbert, 1890; Spencer et al., 1984; Sack, 1989). Relatively deep standing water (≈ 85 m) at the eruption site produced a distinctive suite of eruption-fed clastic deposits that built up from the floor of the lake to its surface. It is inferred that material erupted from the subaqueous vent was dispersed upward into the water column, then entrained into dilute aqueous density currents, turbidity currents, which dispersed the debris laterally to form extensive subhorizontal beds of well-sorted ash. The source characteristics of such density currents are unique in their combination of intermittency and sediment dispersion. The contribution by Cas et al. (pp. 83–109) addresses deposits formed in a pre-existing lake in which predominantly rhyolitic magma erupted both subaqueously within the lake basin and subaerially, affecting the catchments beyond the lake. Eruptions included both explosive and non-explosive phases (Cas et al., 1990), with the latter dominant in the lake. Volcaniclastic deposits formed from direct volcanic fallout, from eruption-related sediment gravity flows, and from sediment gravity flows occurring between eruptions that were sourced from within the lake.
VSIA02 9/12/08 12:57 PM Page 3
Introduction These are interbedded with lacustrine suspension and density-current deposits composed of non-volcanic sediment and of remobilized volcanic debris, both carried by streams to the lake from surrounding catchment areas.
ERUPTION-IMPOUNDED LAKES Uniquely among this group of papers, Manville describes in this volume (pp. 109 –141) the deposits of a lake that formed as a direct result of a large eruption from a nearby volcano. Emplacement of the Taupo ignimbrite profoundly disrupted the Waikato River catchment area near the volcano to form Lake Reporoa, which covered a large area to depths of tens of metres yet is inferred to have formed, filled, overflowed, and been drained over the course of less than a decade. During its brief existence, a full suite of very strongly pumiceous deposits, including deltaic successions and basin-centre turbidites, was formed. Enlargement of Lake Taupo, albeit following its almost complete emptying during the course of a large-scale intralacustrine eruption (Wilson & Walker, 1985), resulted in deposition of a transgressive accumulation of highly pumiceous ignimbrite-derived pyroclastic debris at shoreline to water depths of tens of metres in a zone surrounding the present lake. White et al. (this volume, pp. 141–151) investigate distinctive features of these deposits to develop a general assessment of lacustrine pumice-deposition processes. In their contribution, Riggs et al. (pp. 151–179) detail specific depositional environments developed around the raised lake and interpret the different lithofacies developed in terms of a balance among the rate of accommodation development, wave and current energy, and sediment influx; draining of the lake from its posteruptive highstand to the present level is also briefly addressed, and Manville et al. (1999) have modelled the resulting outbreak flood.
SEDIMENTATION AND RESEDIMENTATION OF PYROCLASTS As volcaniclastic deposits became a subject of specific study, investigators having backgrounds in stratigraphy, sedimentology, and volcanology converged upon them. This fertile mix of expertise has driven rapid advances in understanding of such deposits, but has also resulted in inconsistent and overlapping sets of terminology applied by different workers. Our own
3
preference is for terminology following Fisher & Schmincke (1984), which is based upon grain origin. Hence a pyroclast is a fragment formed during an eruption, and a rock made of such fragments is named as a pyroclastic rock (e.g. ‘tuff ’) without regard to whether the pyroclasts form a ‘primary’ fall deposit or have been ‘redeposited’ by aqueous or aeolian processes. A major alternative point of view is that pyroclasts worked by water are best considered sedimentary particles, and their products named as sedimentary rocks (Cas & Wright, 1987; McPhie et al., 1993). In this volume each contributor has made clear how terms are used. Section 2, Resedimentation in Lakes, includes papers examining the dispersal and sedimentation within lakes of volcaniclastic debris from a variety of largely extralacustrine eruptive sites. In addition to rhyolitic debris supplied by intralacustrine eruptions, Cas et al. (pp. 83–109) describe here how deposits of the Bunga Beds also contain material formed by eruptions alongside the lake and then carried into the lake via fluvio-deltaic systems. Closely linked with eruptions on land are ‘continuous-feed’ turbidity currents, which are postulated to have produced some of the turbidites in response to hyperpycnal outflows that may represent the basinal extension of lahars. Climate and tectonism are considered the primary controls on sedimentation globally (e.g. Bridge & Leeder, 1979; DeCelles et al., 1991; Bettis & Autin, 1997). The contribution by Palmer & Shawkey (pp. 179–199) develops this theme by describing lacustrine products of volcanism in the context of these broader controls. Eocene deposits that accumulated in a small intermontane basin in central Idaho, USA, record with remarkable sensitivity the interplay between tectonism, climate, and volcanic activity. Interstratified lacustrine and fluvial deposits document changes in sediment supply, base level, and discharge. Gaylord et al. (pp. 199–225), in a complementary but unrelated paper, document the facies array that accumulated in a rapidly subsiding graben that formed during Eocene extension along the US Cordillera. As topography was lowered while metamorphic core complexes rose as a result of extension in western North America, basins that developed over the complexes had remarkable preservation potential. Because of the close proximity of active volcanoes to the lacustrine depositional sites, Gaylord et al. were able to document the specific contribution of hydrothermal alteration to processes of disaggregation and incorporation of a variety of products to the sedimentary systems.
VSIA02 9/12/08 12:57 PM Page 4
4
J. D. L. White and N. R. Riggs
LAKES AS TEPHRASTRATIGRAPHIC REPOSITORIES Section 3 concludes the volume with papers focused on the role of lakes as distal repositories of volcanic ash, their use in reconstructing prehistoric eruptions and environmental conditions, and the techniques used to best utilize this unique, but not always uniquely interpretable, data set. Widespread ash layers useful for such work commonly arise from extrabasinal eruptions, from which ash is spread over downwind areas that may extend across hundreds of thousands of square kilometres (Walker, 1973). Such layers may extend across both terrestrial and marine environments, and offer the most precise available means of correlating terrestrial and marine biostratigraphy (e.g. Carter et al., 1995). In contrast to this assessment of how tephra horizons can be used to provide widespread stratigraphic correlations, the contribution of Hardardóttir et al. (pp. 225 –247 ) details how normal biogenic and biochemical lacustrine processes are interrupted by fallout accumulation. Throughout Iceland, lakes provide a detailed record of volcanism, but from a detailed study of Lake Hestvatn, these researchers show how complex environmental changes normally reflected in diatom productivity and pollen populations must be interpreted through the filter of tephra depositional processes. A similar technique is applied by Caballero et al. (pp. 247–263) to understanding the evolution of Nevado de Toluca over the past 40 kyr. In this case, longer-term activity from Toluca had more longlasting effects on the size and shape of the lake than on the biological processes taking place within it. Not surprisingly, these researchers are best able to document environmental changes, in part brought on by volcanic activity, during non-volcanic periods. Erosion and sediment redistribution are ubiquitous in terrestrial settings, and lacustrine deposits containing age-diagnostic material are exceptionally useful tools for chronostratigraphic correlation. Pyroclastic deposits are not only commonly datable by isotopic means, but also form lithostratigraphic units that may extend across widely varying depositional settings (Fisher & Schmincke, 1984). Königer & Stollhofen show in this volume (pp. 263 –285) how pyroclastic material, preserved as both ‘primary’ waterlain fall deposits in lakes and ‘reworked’ ash in associated fluvial settings, can be used to develop a chronostratigraphic and lithostratigraphic framework for other-
wise disjunct terrestrial units of Permo-Carboniferous age in the Saar–Nahe basin. Ash layers formed during lacustrine transgressions, and on footwall blocks of the rift basin, are commonly primary waterlain fall deposits, whereas equivalent tephra deposited in other surroundings exists only in reworked form, in many instances mixed with non-volcanic sediment. Information derived from study of tephra produced by a single eruption and deposited in a landslidedammed lake downwind of Crater Lake, Oregon, has been used by Riedel et al. (pp. 285–299) in their contribution to infer post-eruption lacustrine accumulation rates in excess of 15 m yr–1. They also show that a regionally 2-cm-thick layer of tephra, preserved as a waterlain ash-fall bed in the lake, was eroded from catchment hillslopes and redeposited to form > 10 m of lacustrine suspension deposits and turbidites.
CONCLUSION An impressive range of geological information, deposit types, temporal duration and geographical extent, and approaches to analysis is manifest in the studies presented here. The key feature of lacustrine volcaniclastic sedimentation is that it responds, often strongly, to an allogenic forcing mechanism that does not affect lakes uninfluenced by volcanoes. Studies of the resulting successions benefit from a multidisciplinary approach, and in turn illuminate both the response of the forced lacustrine setting and aspects of the volcanic forcing mechanisms themselves.
REFERENCES Belousov, A. & Belousova, M. (2000) Eruptive process, effects and deposits of the 1996 and ancient basaltic phreatomagmatic eruptions in Karymskoye lake, Kamchatka, Russia. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 35 – 61. Blackwell Science, Oxford. Bettis, E.A., III & Autin, J.W. (1997) Complex response of a midcontinent North American drainage system to late Wisconsinan sedimentation. J. sediment. Res., 67, 740–748. Bridge, J.S. & Leeder, M.R. (1979) A simulation model of alluvial stratigraphy. Sedimentology, 26, 617–644. Buesch, D.C. (1991) Changes in depositional environments resulting from emplacement of a large-volume ignimbrite. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 45, 139–153. Caballero, M., Macías, J.L., Lozano-García, S., UrrutiaFucugauchi, J. & Casteñada-Bernal, R. (2000) Late
VSIA02 9/12/08 12:57 PM Page 5
Introduction Pleistocene–Holocene volcanic stratigraphy and palaeoenvironments of the upper Lerma basin, Mexico. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 247–263. Blackwell Science, Oxford. Carter, L., Nelson, C.S., Neil, H.L. & Froggatt, P.C. (1995) Correlation, dispersal, and preservation of the Kawakawa Tephra and other late Quaternary tephra layers in the Southwest Pacific Ocean. N.Z. J. Geol. Geophys., 38, 29–46. Cas, R.A.F. & Wright, J.V. (1987) Volcanic Successions, Modern and Ancient. Allen and Unwin, London. Cas, R.A.F., Allen, R.L., Bull, S.W., Clifford, B.A. & Wright, J.V. (1990) Subaqueous, rhyolitic dome-top tuff cones: a model based on the Devonian Bunga Beds, southeastern Australia and a modern analogue. Bull. Volcanol., 52, 159–174. Cas, R.A.F., Edgar, C., Allen, R.L., Bull, S., Clifford, B.A., Giordano, G. & Wright, J.V. (2000) Influence of magmatism and tectonics on sedimentation in an extensional lake basin: the Upper Devonian Bunga Beds, Boyd Volcanic Complex, southeastern Australia. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 83 –109. Blackwell Science, Oxford. Decelles, P.G., Gray, M.B., Ridgeway, K.D., et al. (1991) Controls on synorogenic alluvial-fan architecture, Beartooth Conglomerate (Paleocene), Wyoming and Montana. Sedimentology, 38, 567–590. Fisher, R.V. & Schmincke, H.-U. (1984) Pyroclastic Rocks. Springer, Berlin. Gaylord, D.R., Price, S.M. & Suydam, J.D. (2000) Volcanic and hydrothermal influences on middle Eocene lacustrine sedimentary deposits, Republic basin, northern Washington, USA. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 199–225. Blackwell Science, Oxford. Gilbert, G.K. (1890) Lake Bonneville. Monogr. US geol. Surv., Denver, CO, 1. Hamblin, W.K. (1994) Late Cenozoic Lava Dams in the Western Grand Canyon. Geol. Soc. Am. Mem., Boulder, CO, 183. Hardardóttir, J., Geirsdóttir, Á. & Thórdarson, T. (2000) Tephra layers in a sediment core from Lake Hestvatn, southern Iceland: implications for evaluating sedimentation processes and environmental impacts on a lacustrine system caused by tephra fall deposits in the surrounding watershed. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 225 –247. Blackwell Science, Oxford. Jones, J.G. (1969) Intraglacial volcanoes of the Laugarvatn region, southwest Iceland. J. geol. Soc. London, 124, 197–211. Jones, J.G. (1970) Intraglacial volcanoes of the Laugarvatn region, southwest Iceland II. J. Geol., 78, 127–147. Kennedy, N. (1994) New Zealand tephro-chronology as a tool in geomorphic history of the c. 140 ka Mamaku Ignimbrite Plateau and in relating oxygen isotope stages. Geomorphology, 9, 97–115. Königer, S. & Stollhofen, H. (2000) Environmetal and tectonic controls on preservation potential of distal fallout
5
ashes in fluvio-lacustrine settings: the Carboniferous– Permian Saar–Nahe Basin, SW Germany. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 263 –285. Blackwell Science, Oxford. Manville, V. (2000) Sedimentology and history of Lake Reporoa: an ephemeral supra-ignimbrite lake, Taupo Volcanic Zone, New Zealand. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 109 – 141. Blackwell Science, Oxford. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1999) Paleohydrology and sedimentology of a post-1.8 ka breakout flood from intracaldera Lake Taupo, North Island, New Zealand. Geol. Soc. Am. Bull., 111, 1435–1447. McPhie, J., Doyle, M. & Allen, R. (1993) Volcanic Textures: a Guide to the Interpretation of Textures in Volcanic Rocks. CODES Key Centre, University of Tasmania, Hobart. Nairn, I.A. (1989) Sheet V16AC Tarawera; Geological Map of New Zealand, 1 : 50 000. Department of Scientific and Industrial Research, Wellington. Oviatt, C.G. & Nash, W.P. (1989) Late Pleistocene basaltic ash and volcanic eruptions in the Bonneville basin, Utah. Geol. Soc. Am. Bull., 101, 292–303. Palmer, B.A. & Shawkey, E.P. (1997) Lacustrine sedimentation processes and patterns during effusive and explosive volcanism, Challis volcanic field, Idaho. J. sediment. Res., 67, 154 –167. Palmer, B.A. & Shawkey, E.P. (2000) Lacustrine–fluvial transitions in a small intermontane valley, Eocene Challis volcanic field, Idaho. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 99–118. Blackwell Science, Oxford. Riedel, J.L., Pringle, P.T. & Schuster, R.L. (2000) Deposition of Mount Mazama tephra in a landslide-dammed lake on the upper Skagit River, Washington, USA. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 285 –299. Blackwell Science, Oxford. Riggs, N.R., Ort, M.H., White, J.D.L., Wilson, C.J.N., Houghton, B.F. & Clarkson, R. (2000) Post-1.8-ka marginal sedimentation in Lake Taupo, New Zealand: effects of wave energy and sediment supply in a rapidly rising lake. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 151–179. Blackwell Science, Oxford. Sack, D. (1989) Reconstructing the chronology of Lake Bonneville: an historical review. In: History of Geomorphology: from Hutton to Hack (Ed. Tinkler, K.J.), Binghamton Symposia in Geomorphology, International Series, Vol. 19, pp. 223–256. Unwin Hyman, Boston, MA. Segerstrom, K. (1950) Erosion studies at Paricutin volcano, State of Michoacan, Mexico. US geol. Surv. Bull., 956-A, 1–164. Skilling, I.P. (1994) Evolution of an englacial volcano: Brown Bluff, Antarctica. Bull. Volcanol., 56, 573–591. Smellie, J.L. (2000a) Subglacial eruptions. In: Encyclopedia of Volcanoes (Eds Sigurdsson, H., Houghton, B.F.,
VSIA02 9/12/08 12:57 PM Page 6
6
J. D. L. White and N. R. Riggs
McNutt, S.R., Rymer, H. & Stix, J.), pp. 403– 418. Academic Press, San Diego, CA. Smellie, J.L. (2000b) Lithofacies architecture and construction of volcanoes erupted in englacial lakes: Icefall Nunatak, Mount Murphy, eastern Marie Byrd Land, Antarctica. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 9–35. Blackwell Science, Oxford. Smellie, J.L. & Hole, M.J. (1997) Products and processes in Pliocene–Recent, subaqueous to emergent volcanism in the Antarctic Peninsula: examples of englacial Surtseyan volcano construction. Bull. Volcanol., 58, 628– 646. Smellie, J.L. & Skilling, I.P. (1994) Products of subglacial volcanic eruptions under different ice thicknesses: two examples from Antarctica. Sediment. Geol., 91, 115–129. Smith, R.C.M. (1991) Landscape response to a major ignimbrite eruption, Taupo Volcanic Center, New Zealand. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 45, 123–137. Spencer, R.J., Baedecker, M.J., Eugster, H.P., et al. (1984) Great Salt Lake, and precursors, Utah: the last 30,000 years. Contrib. Mineral. Petrol., 86, 321–334. Walker, G.P.L. (1973) Explosive volcanic eruptionsaa new classification scheme 11. Geol. Rundsch., 62, 431– 446. Werner, R., Schmincke, H.-U. & Sigvaldason, G. (1996) A new model for the evolution of table mountains: volcanological and petrological evidence from the Herdubreid and Herdubreidarotgl volcanoes (Iceland). Geol. Rundsch., 85, 390–397. White, J.D.L. (1991) The depositional record of small,
monogenetic volcanoes within terrestrial basins. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 45, 155–171. White, J.D.L. (2000) Eruption and reshaping of Pahvant Butte volcano in Pleistocene Lake Bonneville. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 61–83. Blackwell Science, Oxford. White, J.D.L. & Houghton, B.F. (2000) Surtseyan and related eruptions. In: Encyclopedia of Volcanoes (Eds Sigurdsson, H., Houghton, B., McNutt, S., Rymer, H. & Stix, J.), pp. 495–512. Academic Press, San Diego, CA. White, J.D.L., Houghton, B.F., Hodgson, K.A. & Wilson, C.J.N. (1997) Delayed sedimentary response to the 1886 a.d. eruption of Tarawera, New Zealand. Geology, 25, 459– 462. White, J.D.L., Manville, V., Wilson, C.J.N., Houghton, B.F., Riggs, N.R. & Ort, M. (2000) Settling and deposition of ad 181 Taupo pumice in lacustrine and associated environments. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 141–151. Blackwell Science, Oxford. Wilson, C.J.N. & Walker, G.P.L. (1985) The Taupo eruption, New Zealand I. General aspects. Phil. Trans. R. Soc. London, Ser. A, 314, 199–228. Young, H.W. & Jones, M.L. (1984) Hydrologic, demographic, and land-use data for the Snake River Plain, southeastern Idaho. US geol. Surv. Water-Resour. Invest. Rep. 84-4001.
VSIC01 9/12/08 2:07 PM Page 7
Eruptions and eruption-formed lakes
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
VSIC01 9/12/08 2:07 PM Page 9
Spec. Publs int. Ass. Sediment. (2001) 30, 9–34
Lithofacies architecture and construction of volcanoes erupted in englacial lakes: Icefall Nunatak, Mount Murphy, eastern Marie Byrd Land, Antarctica J. L. SMELLIE British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK
ABSTRACT Mount Murphy is a large Miocene shield volcano flanked by several small basaltic satellite centres that were erupted beneath a thick (> 200 m) ice sheet. Three empirical models illustrating the hydraulic evolution of glacio-volcanic systems are deduced from glacier physics, with distinctly different implications in each case for the resultant lithofacies architecture. Glacier hydraulic considerations and facies analysis are used to describe the evolution of one of the satellite centres (Icefall Nunatak). The nunatak was constructed from several vents during three main stages. Each stage demonstrates different aspects of englacial volcano construction, mainly in a flooded vault or lacustrine setting. An initial mainly effusive phase was dominated by lava and cogenetic joint-block breccia, and eruption was probably confined mainly within an englacial vault or lake (stage I). Renewed activity, at a different vent and beneath a re-established ice sheet (stage II), began with coarse sediments flushed away subglacially. A subaqueous tuff cone was then constructed in an englacial lake, from explosively erupted coarse glassy tephra probably produced mainly during sustained eruptions and distributed by high-density turbidity currents. Fine detritus is common only in the basal tuff cone unit, possibly as a result of lower, denser (largely subaqueous?) eruption columns. A spectacular slope failure is represented by numerous large blocks, which were displaced to low elevations on extensively fractured tuff cone flanks, and the failure event may have initiated zones of high pore-water discharge. Stage II culminated with two phases of lava delta progradation, indicating that the volcanic edifice ultimately penetrated the entire ice-sheet thickness and that the vent became emergent. Stage III commenced with lava effusion, probably through a thin re-formed cover of permeable snow and firn. A small cinder cone was also constructed and was partially palagonitized because of its structural position on top of a water-saturated volcanic pile and likely presence of vent intrusions driving hydrothermal circulation.
INTRODUCTION (Cas & Wright, 1987) of a volcanic provenance from redeposited syneruptive tephra in lake successions can be difficult, yet it is important if we are to interpret the palaeoenvironmental and eruptive records preserved in ancient lake successions. Volcanoes erupted in lakes are also invaluable sources of information on hydrovolcanism. However, uplifted exposed sections are comparatively uncommon and few studies give details of the lithofacies present and their relationships (e.g. White, 1996). By contrast, subglacially erupted volcanoes, which are constructed largely within englacial lakes, are common and often well exposed, particularly in Antarctica (e.g. Hamilton,
Volcanoes that form as a result of explosive eruptions in lakes represent active high-energy systems with rapid episodic (linked to eruptions) input of coarsegrained tephra to the high-relief steep submerged volcano flanks (e.g. Sohn, 1995; White, 1996; Smellie & Hole, 1997). Erosional modification in exposed settings can complicate the depositional record of lacustrine volcanoes (e.g. White, 1996) and most lakes in volcanic settings are depocentres for sedimentary (volcanogenic epiclastic) detritus in addition to tephra. Identification of primary volcanic processes depends on the successful screening out of non-volcanic influences. Distinguishing purely epiclastic sedimentation
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
9
VSIC01 9/12/08 2:07 PM Page 10
10
J. L. Smellie
1972; LeMasurier, 1972; Wörner & Viereck, 1987; Smellie et al., 1988, 1993a,b; LeMasurier et al., 1994; Skilling, 1994; Smellie & Skilling, 1994; Smellie & Hole, 1997). Unlike other lacustrine systems, which receive variable influxes of sedimentary detritus from the surrounding (non-glacial) terrain, englacial lakes formed around active volcanoes are essentially isolated physically by the surrounding ice from most external (nonvolcanic) sediment sources. Englacial lakes are also very protected settings and erosional modification caused by waves or strong currents is typically minor. This paper explores the hydrodynamic background of eruptions within glaciers and the results are illustrated by examining the lithofacies architecture and construction of a small, very well exposed polygenetic volcanic centre of late Miocene age at Icefall Nunatak, near Mount Murphy, eastern Marie Byrd Land. The centre displays the varied eruptive and depositional processes characteristic of basaltic englacial lacustrine volcanoes, which are also known as table-mountain or tuya volcanoes (e.g. Jones, 1969, 1970; Allen et al., 1982). In these volcanoes, subaqueous (pillow lava, tuff cone) lithofacies are overlain by hyaloclastite deltas as a consequence of shoaling and emergence of the vents. Activity is Surtseyan during the emergent period, when explosive eruptions take place in a flooded vent (Kokelaar, 1983) and, in many aspects of their construction and eruptive characteristics, these englacial volcanoes are generally similar to Surtseyan volcanoes in marine situations (Jones, 1966; Smellie & Hole, 1997). The study also illustrates the fundamental controls of glacier hydrology on the sequence of events and depositional record of hydrovolcanic eruptions in englacial environments.
GEOLOGICAL BACKGROUND Volcanism in Marie Byrd Land extends back to Late Oligocene times (28 Ma), at least, and coincided with the development of a widespread Antarctic ice sheet (LeMasurier, 1972; LeMasurier & Rex, 1982). The volcanism is alkaline, related to the impingement of a major mantle plume beneath the stationary Marie Byrd Land crustal microplate and coincident development of a basin-and-range-like extensional province in the Ross embayment (the West Antarctic rift system) during Cenozoic times (LeMasurier & Rex, 1989; Behrendt et al., 1991; Hole & LeMasurier, 1994). The volcanic province contains at least 18 major phonolite and trachyte shield or stratovolcanoes with large (up to 10 km diameter) summit calderas, and numerous
smaller basaltic centres (LeMasurier, 1990; Hole & LeMasurier, 1994; Panter et al., 1994, 1997). The proportion of pyroclastic deposits is generally small in most of the stratovolcanoes and comprises numerous scoria cones and very rare Plinian fall and pyroclastic flow deposits (Panter et al., 1994; Wilch et al., 1999). Conversely, hydroclastic deposits (tephra and hyaloclastite autobreccias) are locally abundant and are particularly common in the basal sections of some volcanoes (‘basal sequence’ of LeMasurier, 1972; see also LeMasurier & Rex, 1982; LeMasurier, 1990). Mount Murphy is situated in eastern Marie Byrd Land (Fig. 1). It is a large dissected volcano dominated by a basanite to peralkaline trachyte shield ≈ 25 km in diameter with a postulated small (4–5 km diameter) ice-filled summit caldera breached on the south side (LeMasurier, 1990; Figs 1 & 2). The shield was constructed relatively quickly in late Miocene times (mainly 8–9 Ma) on a lithologically varied basement composed of early Cenozoic and older granitoid and gabbroid plutons, alkaline dykes, gneiss and turbidite sedimentary strata. The surface of the local basement is a prominent feature forming a conspicuous massif rising from < 400 m to > 2000 m above sea-level (a.s.l.; Fig. 1). Younger activity was entirely basaltic (basanite, alkali basalt and hawaiite) and comprised several small satellite centres erupted in latest Miocene times (6–7 Ma) and now highly dissected, and a few small Pliocene to Recent tuff and cinder cones. The Mount Murphy volcano has had a complicated history of interactions between magma and former ice (Smellie et al., 1993b; LeMasurier et al., 1994). The lower part of the shield succession (between 400 and 1500 m a.s.l.) is dominated by basaltic lavas and hydroclastic deposits formed both explosively and by autoclastic brecciation. It includes interbedded thin diamictites (tillites) resting on multiple glacially eroded striated surfaces and an association with ice coeval with eruptions is undoubted. These initial eruptions were of subglacial ‘sheet-flow’ type, comparable in many respects with those described by Walker & Blake (1966) and Smellie et al. (1993a), and formed when the slopes of the volcano were mantled by relatively thin ‘ice’ (probably mainly firn and/or snow < 100 m thick). The overlying succession consists almost solely of subaerially erupted basaltic and trachytic lavas. After construction of the shield, several latest Miocene satellite centres (including Icefall Nunatak) were constructed. Unlike the shield, they display multiple subaqueous to subaerial transitions similar to passage zones in seamounts and in table-mountain volcanoes erupted beneath thick ice sheets (greater than ≈ 200 m
; VSIC01 9/12/08 2:07 PM Page 11
Volcanoes erupted in englacial lakes
(a)
11
111 W
(b)
90 W
Hedin Nunatak
40
0m
75 15'S
Ca
ra?
Exposed volcanic rock, unvisited
0m 20 Mount Murphy
Mt Murphy
0
Marie Byrd Land
Exposed volcanic rock, visited
lde
Turtle Peak
Volcanic outcrop, obscured by ice Exposed pre-volcanic basement Pre-volcanic basement, obscured
Icefall Nunatak
Stage IIIb
II
?
00
Stage IIIa
180 W
Stage II
0
500
km
II
I
III
Stage I Vent position
III
Pre-volcanic basement fault (inferred
10
? m 400 600 m I
(c)
m
0
10
km
Fig. 1. (a) Sketch map showing location of the study area (Mount Murphy) in Marie Byrd Land. (b) Sketch map showing location of Icefall Nunatak and simplified geology of Mount Murphy. (c) Geological sketch map of Icefall Nunatak, with the stratigraphy separated into the three evolutionary stages described in this paper. The known and inferred locations of vents responsible for constructing the polygenetic volcano at Icefall Nunatak are also shown in (c).
Fig. 2. Aerial view of Mount Murphy and surrounding nunataks, looking east-south-east. Mount Murphy is a large glacially dissected stratovolcano constructed on a north-sloping fault-block massif of pre-Cenozoic ‘basement’ rocks (Fig. 1), whereas Hedin Nunatak, Turtle Rock and Icefall Nunatak are small basaltic satellite centres. The latter were erupted beneath a gently north-sloping late Miocene ice sheet, which, at times, was perhaps 100 m thicker than that present on the south side of Icefall Nunatak today, and it would also have completely covered Hedin Nunatak and Turtle Rock. US Navy photograph TMA1719 frame F31-132.
thick; see below and Jones, 1966, 1969). From c. 3.5 Ma onward, only cinder cones and rare tuff cones were formed. These young pyroclastic cones lack evidence for significant interaction with ice. Icefall Nunatak is one of the satellite centres situated on the west side of Mount Murphy. It is well
exposed in an east–west-trending cliff ≈ 0.9 km long and 200 m high (Fig. 3). The base is concealed by ice and the top corresponds to the present-day erosion surface. The volcano was erupted subglacially in late Miocene times (c. 6.5 Ma) and contains evidence for eruptions from several vents.
VSIC01 9/12/08 2:07 PM Page 12
12
J. L. Smellie
Fig. 3. Sketch of Icefall Nunatak, looking south, showing (a) the distribution of the major lithofacies, and (b) the lithofacies architecture (units described in the text). The locations and ages of dated samples are also shown, and the locations of features illustrated as photographic figures in this paper.
EVIDENCE FOR A GLACIAL SETTING FOR ERUPTIONS AT ICEFALL NUNATAK Icefall Nunatak is dominated by subaqueous volcaniclastic and volcanic lithofacies, whose characteristics (see below) indicate that they accumulated in ponded water, in either a lacustrine or a marine setting. Apart from one thin unit, the sequences lack the abundant tractional structures indicative of a shallow-marine environment, consistent with a very protected location (lake or deep water (below wave base) ). The deposits are also entirely devoid of fossils. Because even rapidly constructed submarine edifices are colonized by marine vegetation and benthic fauna soon after eruption or contain infiltrated marine pelagic fossils (Surtsey Research Society, 1970; Kokelaar & Durant, 1983; Smellie et al., 1998), a non-marine (lacustrine) environment is more likely. However, there is no known Miocene palaeotopography that could have acted as a barrier and confined a former non-glacial lake, suggesting that ice may have acted as a barrier and the lake may have been glacially confined. That a glacial setting is likely is also suggested by the history of Cenozoic glaciation in the region, which
probably commenced at c. 40 Ma in East Antarctica and became widespread throughout West Antarctica from early mid-Miocene time (e.g. Cooper et al., 1991). This is supported by field evidence from the Mount Murphy shield succession, which was erupted in association with thin ‘ice’ in late Miocene times (8–9 Ma) (see above). Moreover, the volcanic sequence at Hedin Nunatak, one of the western satellite centres situated just 6 km north-north-west of Icefall Nunatak (Figs 1 & 2), is composed of multiple superimposed hyaloclastite deltas with passage zones that vary in elevation within a vertical interval of 100 m. The passage zones reflect former water levels coeval with eruptions (Jones, 1969). The entire Hedin Nunatak sequence formed rapidly between 6.50 ± 0.06 and 6.20 ± 0.12 Ma. There is insufficient time for the variations in water level to be an effect of a fluctuating eustatic sea level. This suggests that eruptions were in an englacial lake(s), in association with a relatively thick ice sheet (> 200 m; see Smellie & Skilling, 1994; and see below). The passage zones broadly reflect elevations of the icesheet surface (Smellie et al., 1993a). Icefall Nunatak erupted between 6.80 ± 0.10 and 6.47 ± 0.08 Ma (Fig. 3). It therefore erupted between the periods of ice cover represented by the Mount Murphy shield
VSIC01 9/12/08 2:07 PM Page 13
Volcanoes erupted in englacial lakes succession and that at Hedin Nunatak. It also overlaps in age with Hedin Nunatak. A glacial eruptive environment and relatively thick ice cover are thus likely for the Icefall Nunatak sequence. From passage zone elevations at Icefall and Hedin nunataks, water levels at the two localities were up to 200 m different at essentially coincident times (≈ 6.5 Ma). These differences are attributed here to eruptions beneath a gently northsloping ice sheet rather than reflecting a common sea level followed by faulting between the two localities, although the exposures are discontinuous and effects of local tectonism cannot be wholly excluded. In summary, these various lines of evidence, none of which is independently conclusive, suggest strongly that Icefall Nunatak was formed in a glacial environment, in association with coeval ice. Within that setting, important features, such as the presence of passage zones and volcaniclastic lithofacies deposited in ponded water, are only consistent with eruptions beneath a relatively thick ice sheet and accumulation within an englacial vault or lake.
EFFECTS OF GLACIER PHYSICS ON SUBGLACIAL ERUPTIONS Eruptions are subglacial when the vents are situated beneath a glacier, whereas the resulting volcanoes are constructed englacially (i.e. surrounded by a glacier; terminology similar to that used by Wright, 1980). Because of rapid (volcano-induced) melting, the volcanoes are typically surrounded by meltwater confined in an englacial vault or lake (Jones, 1969, 1970; Smellie & Skilling, 1994). Glacier physics (particularly thermal regime, structure and hydrology) exerts a fundamental control on many aspects of subglacial eruptions. It is used below to construct three simple models for subglacial eruptions. Comprehensive recent reviews of glacier hydrology and other aspects of glacier physics relevant to this paper have been given by Paterson (1994) and Menzies (1995). The present discussion is not intended to be exhaustive and a fuller treatment will be presented elsewhere. Although empirical, the general principles can be applied to many local situations and they may be able to explain satisfactorily the principal features of most subglacial eruptions. Glacier thermal regime Glaciers or parts of glaciers (zones) can be classified according to temperature distribution into two broad categories, temperate and polar. A temperate glacier is
13
at the melting point throughout (although periods of melting alternate with refreezing), whereas temperatures in polar glaciers are well below freezing throughout the year and they are frozen to their bed. This classification is oversimplified and conditions in most glaciers are dynamic and vary from one point to another, both spatially and temporally. Despite this variation, it is useful to conduct the ensuing discussion as if only two categories of glacier thermal regime existed. The discussion is further limited to temperate glaciers (or cold glaciers in which melting point is reached at the bed) because water can migrate under these conditions, a condition that is most informative for subglacial eruptions. Glacier hydrology Large quantities of meltwater are created during subglacial eruptions, and glacier hydrology exerts a dominant influence on the sequence of events in such eruptions and on the distribution and types of lithofacies formed. In turn, glacier hydrology is controlled to a large extent by glacier structure, itself mainly determined by snow densification (see next section). Glacier hydraulics describes the motion of water within a glacier system. The most important hydraulic effect of snow densification is on glacier permeability. In simple terms, snow and firn are aquifers and ice is an aquiclude (illustrated graphically by Gore, 1992, fig. 2). Crevasses are also aquifers. Provided the ice is temperate, englacial seepage may also occur in unfractured ice along crystal boundaries or via a network of connected small tubes and veins but, because of the impermeability of ice, hydraulic flow by such seepage is almost negligible compared with flow along the underlying bedrock surface. Meltwater can also pass through an underlying permeable sediment layer (Darcian flow) but, in most instances, meltwater will flow mainly along the glacier bed. Any discussion of glacier hydrology must focus on the hydraulic gradient or gradient of the water pressure potential of a system, which determines the flow direction of any meltwater. The hydraulic gradient within ice, at the glacier bed and in any underlying permeable sediment layer is controlled predominantly by ice surface slope (Björnsson, 1988; Syverson et al., 1994). Consequently, for many glaciers, water flows in the general direction of the surface slope. During eruptions, the ice surface subsides as ice is melted over the eruption site. Initially, water may drain subglacially but distortion of the local hydraulic gradient by the depressed ice surface rapidly constrains meltwater to
VSIC01 9/12/08 2:07 PM Page 14
14
J. L. Smellie
flow towards the eruption site. Thus the site becomes effectively sealed by an encircling ice barrier, and a water-filled vault forms above the volcano, melting to the surface and forming an englacial lake (Björnsson, 1988). The lake will drain subglacially when the water becomes deep enough to float the ice barrier, resulting in catastrophic floods (known as jökulhlaups in Iceland). Glacier structure After precipitation, snow is transformed into ice by a process of densification that involves the collapse of the snow particle lattice framework and the progressive compression and elimination of trapped air. The transformation and its rate depend on temperature. It happens much more rapidly in glaciers in temperate regions compared with polar regions. Densification results in glaciers divided into different zones or layers, corresponding to snow, firn and ice, which differ from each other in their temperature and other physical characteristics. ‘Snow’ refers to particulate ice crystals that are essentially unchanged since they were precipitated; ‘ice’ comprises ice crystals and grains in which the interconnecting air passages have been sealed off; and ‘firn’ refers to the intermediate stages of transformation of snow into ice. Fundamental changes in physical properties occur during densification. Typical values for density are in the range of 50–400 kg m–3 for snow, 400–800 kg m–3 for firn, and 800–917 kg m–3 for glacier ice (Paterson, 1994). Fractures (crevasses) form in areas where glacier ice is undergoing extension (e.g. overlying subglacial topographic highs), where tensile stresses exceed the tensile strength of ice. However, crevasses can remain open down to depths of only ≈ 20 –30 m as internal deformation pressure acts to seal them at greater depths. This statement ignores the influence of meltwater in crevasses, which can cause fractures to remain open to greater depths. Thus, a layer of crevassed ice may also be present above unfractured ice in some glaciers. For simplicity, the discussion that follows will consider a layered glacier structure, comprising an upper ‘firn’ layer (including both snow and firn), an intermediate layer of crevassed ice and a lower layer of massive ice. The broad use of the term ‘firn’ follows the convention of Paterson (1994). It has no practical effect on the conclusions reached below. For purposes of discussion, the presence and hydraulic effects of dirt bands and fold structures caused by glacier flow and deformation over uneven bedrock are also ignored in favour of a simple layered sheet of glacier ice overlying an even horizontal or gently sloping bedrock (Fig. 4).
Fig. 4. Section through an idealized glacier showing the empirical layered structure used in this paper to model the hydrodynamic effects of subglacial eruptions. Although only temperate glaciers are considered in this paper, polar glaciers will show a similar structure. Not to scale.
Effects of thermal regime, glacier structure and hydrology on subglacial eruptions During eruptions beneath glaciers, distinctive sequences of lithofacies are formed whose lithofacies types and architecture are controlled by the thickness and structure of the overlying glacier (Smellie et al., 1993a; Smellie & Skilling, 1994). Thus far, two broad types of eruptive sequence have been documented, corresponding to eruptions beneath ‘thick’ and ‘thin’ glaciers. Beneath ‘thick’ glaciers (i.e. greater than ≈ 200 m thick), a basal section of pillow lavas and/or subaqueous volcaniclastic lithofacies (mainly syneruptive redeposited tephra; McPhie et al., 1993) is formed in a water-filled vault or lake and is overlain by subaerial lavas and cogenetic breccias of hyaloclastite delta(s) (Jones, 1969, 1970; Skilling, 1994; Smellie & Skilling, 1994; Smellie & Hole, 1997). This sequence type and its lithofacies are characteristic of subglacially erupted table-mountain (tuya) volcanoes in Iceland and elsewhere, and they can be difficult to distinguish from lacustrine or marine volcanoes (see Staudigel & Schmincke, 1984; Werner et al., 1996; White, 1996). However, the common draping of subaqueous sections by subaerial lithofacies in the products of a single eruption is one important way in which englacial volcanic successions can be distinguished from other subaqueous volcanic successions (Wörner & Viereck, 1987; Smellie et al., 1993a; Skilling, 1994; Smellie & Skilling, 1994; Smellie & Hole, 1997). Conversely, a vault or lake cannot form beneath ‘thin’ glaciers (i.e. composed of firn and/or crevassed ice; typically < 100 m thick). Although in theory water
VSIC01 9/12/08 2:07 PM Page 15
Volcanoes erupted in englacial lakes may be directed initially into the eruption site owing to distortion of the local hydraulic gradient, a sealed vault cannot be sustained. Thermal erosion of the firn layer by viscous heat dissipation from the meltwater will result in the meltwater rapidly becoming connected hydraulically with the rest of the glacier and flushing away. Open fractures in glacier ice will also allow water to escape. In both circumstances, any meltwater will be removed continuously at the glacier bed during the course of an eruption, resulting in a distinctive association of volcanic and volcaniclastic lithofacies that are characterized by evidence for flowing rather than ponded water, i.e. of ‘subglacial sheet-flow type’ (model 1, illustrated in Fig. 5a; Smellie et al., 1993a). Measured depths to the firn–ice transition range between 38 and 115 m, although 60 –70 m is typical (Paterson, 1994, table 2.2). With a crevasse layer present (≤ 30 m thick), this gives an approximate maximum limit of 150 to perhaps 200 m of firn and crevasses for eruptions yielding such volcanic sequences. The common occurrence of subglacially erupted table-mountain (tuya) volcanoes has been explained by a hydraulic theory in which the volcanoes are constructed within englacial water-filled vaults or lakes (Björnsson, 1988). Eruptions are initially subaqueous, becoming subaerial when the vents rise above the lake surface. The presence of laterally extensive hyaloclastite deltas implies a period of relative stability of the encircling glacial lake, with which subaerial lavas interacted and were shattered into hyaloclastite delta foreset beds along a passage zone. Meltwater is envisaged to accumulate until the hydrostatic pressure at the base of the vault (Pv) exactly equals that beneath the surrounding impermeable ice barrier (Pi). The barrier is then floated and the meltwater drains catastrophically, resulting in volcanic floods ( jökulhlaups). However, a geological problem exists in applying the conventional hydraulic theory. There is ample magmatic heat in the system to melt large volumes of ice (Allen, 1980; Gudmundsson & Björnsson, 1991). However, in a closed system, at least 10 units of ice need to be melted for every unit of magma erupted in order to create space for the magma, a consequence of the different relative densities of ice and water. Although, theoretically, additional space for meltwater could be created by doming of the ice early in the eruption, there are no observations of historical eruptions to confirm whether doming is a common effect. Moreover, melting is typically so rapid (e.g. several hundred metres of ice were penetrated by melting in only a few hours during the 1996 subglacial eruption
15
in Iceland; Gudmundsson et al., 1997) that any domed ice carapace will rapidly disintegrate over the vent site. Thus, the volume of meltwater must increase at a much faster rate than magma is emplaced. Because of the disproportionate volumes involved, Pv will equal Pi long before subaerial emergence of the volcano can take place. Important geological consequences of this observation for the construction of englacial volcanoes are that englacial vaults will very rapidly become lakes and the eruptive vents must be submerged for much of their history, becoming subaerial only when specific hydraulic conditions are met. For example, when Pv = Pi, the barrier will float and the vault will drain subglacially, thus exposing the vent (model 2, illustrated in Fig. 5b). If the eruption continues, the subaqueous lithofacies will be juxtaposed with subaerial lithofacies emplaced in the empty vault. However, in draining, Pv becomes much smaller than Pi, thus allowing the ice barrier to reinstate. The vault may then refill and the process is set to repeat. The final volcano may have an extremely complex lithofacies architecture (implicit in the fledgling models of Smellie et al., 1993a, fig. 14). It is hard to see how a period of stable water level and hyaloclastite delta progradation could occur under conditions where the ice barrier can be lifted and subglacial drainage occur. By contrast, most natural glaciers (> 100 m thick) have a layered structure, comprising (from base up) massive ice, fractured ice and/or firn (Fig. 4). This feature was not considered by previous workers, and glaciers were depicted as homogeneous ice. In model 2, the overlying permeable layer(s) was too thin relative to the underlying impermeable ice and floating of the ice barrier was an inevitable consequence of Pv = Pi. However, flotation cannot occur if the meltwater surface in the vault intersects an aquifer while Pv < Pi. As in eruptions beneath thin glaciers, meltwater will escape through the aquifer and the vault will drain by overflowing (model 3, illustrated in Fig. 5c). In this situation, the surface of the englacial lake can be relatively stable during the eruption, with the possibility of forming laterally extensive hyaloclastite deltas during subaerial effusion. In practice, however, water levels are likely to diminish slowly owing to thermal erosion at the outflow point(s), and this may lead to the coeval subaqueous lithofacies being draped by subaerial lithofacies. Overflowing may be the commonest situation responsible for most table-mountain volcanoes with extensive hyaloclastite deltas, and this is another specific hydraulic situation in which the (initially submerged) vents in englacial volcanoes can become subaerial. Calculations suggest that overflowing is likely
; VSIC01 9/12/08 2:07 PM Page 16
(a) Model 1: Eruption within thin 'ice'
Meltwater escapes through permeable firn/fractured ice
h1 < c. 150 m
h1
Bedrock
Tephra flushed by meltwater beneath glacier; overlain by lava and breccia
(b) Model 2: Eruption beneath thick ice with thin permeable layer (i) Vault-filling stage h1
h2
h1 + h2 >->> c. 150 m
h1 < (0.15–0.25(h2))
Vault sealed by ice barrier
Bedrock
Only subaqueous deposits
(ii) Vault-draining stage
Former lake level
Englacial lake drains subglacilly
Bedrock
Subaerial lava drapes subaqueous deposits
(c) Model 3: Eruption beneath thick ice; thick permeable layer(s)
h1 + h2 > c. 150 m, < c. 1000m
Englacial lake overflows through permeable firn/crevasse layer
h1 > (0.15–0.25(h2))
h1
h2
Ice barrier seals meltwater
Bedrock
Extensive lava deltas
Snow and firm
Fractured ice
Unfractured ice
Direction of meltwater flow
Fig. 5. Three empirical models for subglacial eruptions beneath glaciers of different thickness. The hydrodynamic conditions are different in each case, resulting in significantly different lithofacies architectures. h1, thickness of firn and/or fractured ice (i.e. permeable layer(s) ); h2, thickness of unfractured ice. (See text for further explanation.) Not to scale.
VSIC01 9/12/08 2:07 PM Page 17
Volcanoes erupted in englacial lakes to occur only where firn and/or fractured ice layers are greater than ≈ 10 –25% of the thickness of unfractured ice. Moreover, influence of firn and fractured ice is negligible in glaciers > 1000 m thick and meltwater generated in these situations is likely to drain only subglacially (i.e. model 2). Overflowing was observed for the first time above a subglacial eruptive fissure during the October 1996 eruption in Vatnajokull, Iceland (Gudmundsson et al., 1997), although concurrent subglacial drainage (through subglacial conduits maintained by thermal erosion) and the short duration of the eruption prevented a lava delta forming.
LITHOFACIES Icefall Nunatak consists of subaqueously emplaced lavas and volcaniclastic lithofacies dominated by sideromelane lapilli and ash (Table 1). The dominant glassy clasts in most lithofacies are angular and have cuspate to blocky shapes unmodified by sedimentary processes. These pyroclasts have shapes and variable vesicularity consistent with explosive hydrovolcanic disruption of a vesiculating magma (Houghton & Wilson, 1989). Moreover, non-juvenile clasts are very uncommon, suggesting that the fragmentation was driven by surface water rather than ground water, as otherwise clasts derived from the aquifer would be more common ( White, 1996). The presence of intact and fragmented basalt pillows in the volcaniclastic sequence, together with pillow lava, turbidites, hyaloclastite breccia and water-chilled sheet lavas (see below and Table 1) also indicates a general subaqueous environment of deposition. The lithofacies have been named using the nongenetic descriptive terminology of McPhie et al. (1993). For brevity, the use of ‘volcanic’ is omitted from the lithofacies names. They were probably formed mainly by syneruptive redeposition of pyroclastic and autoclastic particles. Initial division in the clastic lithofacies is by dominant grain size. They are mainly types of breccia (prefix ‘B’, Table 1), including stratified and massive varieties. Further subdivision is based on dominant clast type (vesicular and non-vesicular sideromelane (subscripts ‘vs’, ‘nvs’), tachylite (‘sc’ after dominant scoria clasts), crystalline (‘lith’ indicating lithic; note that these are almost invariably juvenile clasts) ). There are fewer lithofacies of sand grade (prefix ‘S’) but sandy sediments are volumetrically abundant in parts of the sequence. Muddy lithofacies are minor and always coupled genetically to sandstones (lithofacies SM). Lavas are divided into
17
blocky-jointed and columnar types (Lj and Lcol, respectively) and pillow lava (Lp). Some basalt units are intrusions and others are transitional to some of the clastic lithofacies as a result of in situ brecciation and/or redeposition (e.g. Bnvs, Blith). Disturbed and contorted strata are also distinguished as a separate lithofacies (Z).
VOLCANIC EVOLUTION OF ICEFALL NUNATAK Icefall Nunatak is a polygenetic edifice constructed from the products of at least five vents (Fig. 1). Three of the vents were located within the outcrop area. Of these, two are occupied by basalt intrusions, whereas the approximate position of the third is identified by vent-proximal lithofacies (welded Strombolian scoria in a cinder cone remnant). The remaining two vents erupted basalt magma with different mineralogical characteristics (see below), but their location is only poorly constrained, using homoclinal to crudely radial bedding orientations and lithofacies characteristics. However, by comparison with other subglacial volcanoes (e.g. Jones, 1969), the two poorly located vents are likely to have been situated < 1 km away from the preserved outcrop. The construction of the Icefall Nunatak volcano can be described in three major stages of development (I–III), each separated by a prominent unconformity, although stage II was responsible for most of the exposed deposits (Fig. 3; Table 2). Other internal erosive boundaries provide further scope for subdivision within each stage (Ia–c; IIa–h; IIIa–b). Stage I is represented by basalt lavas and breccias, which were emplaced entirely subaqueously, probably mainly in an englacial vault. These are overlain by stage II basalt lavas, breccias and gravelly sandstones, which evolved from a subaqueous (englacial vault, then lake) sequence of vertically aggraded mainly syneruptive redeposited tephra into a lacustrine lava delta having a subaerial lava topset. Stage III is the thinnest sequence present and is largely subaerial. It comprises a thick water-cooled basalt lava and breccia overlain by a small cinder cone relic. The magma erupted throughout the volcano’s history was alkali basalt in composition (Table 3). Mineralogical and textural differences are evident between basalt lava erupted in each stage but are consistent within stages. Stage I basalts are essentially aphyric, whereas phenocrysts (mainly olivine, clinopyroxene and plagioclase) are conspicuous in the other two stages. However, phenocrysts are
Ic; IIc
IIg, IIh
IIe
Ia, Ib, Ic; IIf, IIh; IIIa?
Ib; IIIb
Ic; IIb, IIIc, IId
As SGp but coarser grained; sedimentation from successive high- and low-density stages in turbidity currents under declining flow conditions; sediment input probably represents freshly erupted resedimented tephra linked directly to eruption columns or jets and slumping; mainly R2R3–TbTe turbidites of Lowe (1982)
Sideromelane clasts generated mainly by spalling at the topset–foreset ‘brinkpoint’ in a hyaloclastite (lava) delta; emplaced by either a continuous ‘rain’ of particulate debris avalanching down the steep delta foreset slope or by redeposition from density-modified grain flows; essentially hyaloclastite breccia Uncertain sedimentological interpretation but field relations suggest deposit followed major slope failure event; probably slump deposit emplaced as large cohesionless debris flow
Clasts generated by mechanical rupture during subaqueous lava emplacement; many clasts formed in situ, others underwent minor redeposition probably mainly by local avalanching in density modified grain flows
Proximal pyroclastic deposits (lapillistones) of Strombolian eruptions; subaerial cinder cone remnant (unit IIIb) and subaqueous stratum (Ib); palagonite and zeolite alteration in IIIb as a result of structural position of cone, on top of watersaturated volcanic edifice associated with dyke intrusion; pyroclasts in Ib possibly formed within steam cupola surrounding subaqueous eruption column, thus reducing contact with water until after deposition
Subaqueously and subaerially erupted hydroclastic tephra redeposited by cohesionless debris flows or high-density turbidity currents [ (S1,S2)S3 turbidites of Lowe, 1982]
Sandy gravelly coarse breccia; poorly sorted; planar stratification; often very thick beds (10 cm–3 m); mainly massive or with reverse-graded bases; often normal graded in top 10–20 cm, passing up through 10 cm of faintly planar laminated fine sandstone–siltstone into massive mudstone; 5% conspicuous large clasts (to 80 cm), mainly in massive centres to beds; erosive bases and amalgamation common; large clasts include gravelly sandstone and mudstone up to 50 cm in diameter
Bnvs Gravelly breccia, minor coarse sandstone; fines poor; indistinct planar stratification dipping at up to 30°; numerous irregular pillows, often flattened, up to 80 cm across, with wrinkled and chilled glassy surfaces; stratification may be lens-like down-dip on a decametre scale
Grey fines-poor or fines-free gravelly breccia; mainly massive; very rare indistinct and local planar stratification and one dune bedform seen; possible rare segregation pipes; mixed vesicular and (mainly) poor to non-vesicular sideromelane; locally rich in fine-grained crystalline clasts; dispersed basalt pillows with glassy rinds; drapes and intrudes fractures in units IIc and IId
Matrix-poor or matrix-free gravelly breccia; dominated by finegrained crystalline clasts; blocky shapes, non-vesicular; minor sandy matrix of sideromelane clasts variably crowded with crystallites; minor tachylite; minor complete and fragmented basalt pillows; grades laterally into blocky-jointed lava (Lj); mainly massive; rare faint stratification and planar beds; local jigsaw breccia
Gravelly matrix-poor breccia bed up to 6 m thick; grey, maroon and khaki; formed of moderate to highly vesicular, rarely incipiently vesicular lapilli, many with fluidal surfaces; blocky shapes and large ragged bombs also common; variably tachylite- or sideromelane-rich; local superficial palagonite alteration and pervasive zeolite (unit Ib); As above, plus poor discontinuous planar stratification; dense ovoid bombs common, some cored; locally welded; few thin (10– 40 cm) sand-grade beds, reverse-graded to massive, traceable over metres because of yellow coloration, contain abundant sideromelane droplets (achneliths) (unit IIIb)
Gravelly coarse sandstone to sandy fine breccia; clasts typically 1 mm to 2 cm, mainly sideromelane, variably vesicular; planar continuous stratification; beds 5 cm to a few metres thick; few
Bvs
Bm
Blith
Bsc
SGp
Stratified vesicular sideromelane breccia
Stratified nonvesicular sideromelane breccia
Massive sideromelane breccia
Massive to poorly stratified lithic breccia
Tachylite scoria breccia
Planar-stratified gravelly sandstone
Unit*
Interpretation
Description
Code
Lithofacies
Table 1. Summary and interpretation of the principal lithofacies observed at Icefall Nunatak
VSIC01 9/12/08 2:07 PM Page 18
18 J. L. Smellie
(IIf, IIh); IIIa
Ia, (Ic); (IIe), IIf, (IIg)
IIc
Columnar basalt is common in subaerial sequences but association with sideromelane breccia (Bnvs) and blockyjointed lava (Lj) indicates subaqueous environment probably corresponding to passage zone in hyaloclastite (lava) delta; suppressed colonnade and expanded entablature in upper 20m-thick lava indicate rapid cooling of a lava flooded by water; upper lava transforms to a neck at west end of outcrop Pillow lava and pillow intrusion in wet sediments
Zones of focused dewatering possibly induced by synsedimentary faulting and /or pressure release associated with sector collapse of tuff cone flanks; essentially formed in situ; rare shear-like fabric possibly induced in more consolidated sediments by faulting and /or sliding of large blocks (lithofacies Z3)
Forms irregular parts of some blocky-jointed lavas (Lj) and a 20-m-thick lava at the crest of the nunatak; the latter contains a thin (2– 4 m) basal colonnade overlain by a well-developed entablature; traced laterally, the lava changes into a sharply cross-cutting columnar intrusion
Isolated, complete and fragmented basalt pillows and irregular lens-like masses of pillow lava; locally common; individual pillows are up to 80 cm across, ovoid to flattened, with chilled glassy rinds up to 1 cm thick and rare wrinkled surfaces; no interpillow sediment
Z1: Heterogeneous mixture of sandstone clasts (dm to rarely 5 m diam.) dispersed in massive sandstone matrix; ‘swirling’ textures common in latter; clasts have diffuse rounded to relatively sharp angular margins; diffuse dewatering structures common; penetrative shear-like joint fabric rarely present; forms steep cross-cutting zones spatially associated with faults and/or upper surface of unit IIc Z2: Folded and ‘slurry-like’ beds up to 1.5 m thick; the latter closely resemble Z1 in appearance but always contain dark brown sandy mudstone matrix and are conformable beds
Lcol
Lp
Z
Columnar-jointed lava
Pillow lava
Contorted or disturbed strata
IId
Volcanoes erupted in englacial lakes
*See Table 2 and Fig. 3 for description and distribution of units.
Z3: large (up to 20 m diam.) blocks of stratified sediment (mainly SGp) with variable often steeply dipping orientations; corresponds to unit IId
Ia, Ib; IIf, IIg, IIh
Basalt lava extruded into water and intruded into watersaturated sediments; the jointing is a consequence of rapid cooling in contact with water [cf. ‘kubbaberg’ (box-jointed) lavas in Iceland described by Bergh & Sigvaldason, 1991]
< 1–20 m thick; very irregular pods, lenses and more continuous basalt lavas characterized by blocky to hackly jointing; joint surfaces commonly rust-stained; may change laterally into columnar or pillow lava, or monomict lithic breccia
Lj
Blocky-jointed massive lava
IIc
IIc
Subaqueously and subaerially erupted fine hydroclastic tephra redeposited as Tae, Tbe and Tce turbidites from lowdensity turbidity currents; some may represent residual flows which were completely detached from, and bypassed, their high-density turbidity current precursors
Thin planar beds (1–5 cm thick); form discrete sequences a few dm thick (up to 1.8 m); sandstones have sharp bases, normal grading and small-scale load structures; some climbing ripple cross-sets; pass up gradationally into massive dark brown mudstone
SM
Graded– laminated fine sandstone– mudstone
Folding as a result of synsedimentary gravity-induced sliding that may have generated slump sheets; slurry-like beds formed by density inversion (Rayleigh–Taylor instabilities) leading to gravitational collapse and mixing of sandstone and mudstone strata essentially in situ Gravity-induced detachment of blocks of indurated strata and translation down the tuff cone flanks
IIa
Syneruptive reworking and clast-by-clast deposition from traction currents
Fine sideromelane breccia to medium sandstone; dominated by traction current bedforms, mainly planar and trough crossstratification; beds lens-like, none traceable > 5 –10 m; bed thicknesses 10– 40 cm
SGt
Cross-bedded gravelly sandstone
large lithic clasts (to 40 cm, typically < 25 cm); poorly to well sorted; mostly massive or normal graded; rare reverse-graded bases; rare planar lamination; erosive bases and amalgamation common; few load structures, some containing ballistic blocks; very common lithofacies
VSIC01 9/12/08 2:07 PM Page 19
19
VSIC01 9/12/08 2:07 PM Page 20
20
J. L. Smellie
Table 2. Summary of evolutionary stages, units and characteristic lithofacies of volcanoes at Icefall Nunatak Stage
Description of main event(s)
Units*
Principal lithofacies†
III
Lava effusion into thin ice; Strombolian eruptions
a, b
Tachylite scoria breccia (Bsc), columnar-jointed lava (Lcol); possible massive lithic breccia (Blith)
II
Lateral extension by hyaloclastite
f–h
Subaerial emergence, slope failure (faulting, sector collapse) Subaqueous tuff cone (rapid planar-stratified gravelly vertical aggradation) Subglacial sediment flushing
d, e
Stratified non-vesicular sideromelane breccia (Bnvs); blocky-jointed lava (Lj); (lava) deltas; lithic breccia (Blith) Massive sideromelane breccia (Bm); large rotated blocks of SGp, slurry-like and contorted beds (Z) Stratified vesicular sideromelane breccia (Bvs); sandstone (SGp); graded–laminated sandstone – mudstone (SM) Cross-bedded gravelly sandstone (SGt)
I
Subaqueous tuff cone (in part) Subglacial ‘pillow volcano’
b, c a c a, b
Stratified sideromelane breccia (Bvs); resedimented lithic breccia (Blith) Blocky-jointed lava (Lj), pillow lava (Lp), lithic breccia (Blith)
*See Fig. 3 for distribution of units. †See Table 1 for lithofacies codes and descriptions.
generally more abundant and larger in stage III basalts, and they also have more coarsely crystalline groundmasses than in preceding stages. By contrast, compositional differences are slight and can be explained by variable accumulation of phenocrysts (particularly ferromagnesian minerals). The three growth stages can be interpreted as products of a single continuous volcanic growth cycle. Alternatively, the multiple vents involved, and evidence for breaks in the succession (unconformities and petrological differences between stages), raise the possibility that the three major parts of the succession are from unrelated vents active at very different times. However, the lack of significant geochemical differences between stages, and the age dating constraints (indicating a relatively short eruptive period), suggest that any breaks in eruption were probably short (sufficient for only minor fractional crystallization, but sufficient to create differences in the size and relative abundances of the erupted crystals). A genetic connection between the three stages is more likely. Stage I (‘pillow volcano’ stage; englacial vault; possible transition to subaqueous tuff cone) Description Unit I is restricted to the eastern basal part of the outcrop. It has an exposed maximum thickness of ≈ 60 m. Three subunits are recognized, and are mainly formed of monomict lithic breccia and blocky-jointed lava.
Unit Ia has a sloping upper contact dipping southwest at about 30° (Figs 3 & 6). It comprises 50–60 m of ( juvenile) lithic orthobreccia (Blith), which is predominantly massive but locally shows faint crude stratification dipping parallel to the upper contact. The angular basalt clasts range from fine gravel to cobbles 5–10 cm in diameter and there are dispersed basalt pillows. The unit is coarser downward, where it contains irregular lenses and lobes of blocky-jointed and pillowed lava (Lj, Lp) a few metres thick (Fig. 7). Small parts of the deposit are rich in non-vesicular sideromelane clasts and lava pillows. Sideromelane clasts rich in small crystals (commonly 50 vol. %) are ubiquitous but generally minor volumetrically, restricted to the gravelly sandy size fraction, and the majority of clasts are holocrystalline (lithic ( juvenile) ); zeolite and/or carbonate cement is locally conspicuous. Unit Ib closely resembles Ia, comprising ≈ 12 m of basal blocky-jointed lava intimately associated with irregular patches of rust-stained lithic breccia. It is overlain by 20 m of massive gravelly breccia with lenses of scoriaceous and non-vesicular lava. The shapes of some of the lava units are very irregular, and large and small lava lobes protrude into the enclosing breccia. The lavas and breccias are overlain by two thin (1–6 m) basalt sheet lavas separated by ≈ 4 m of crudely planar bedded, massive gravelly breccias. The two upper lavas have scoriaceous surfaces and parts of the uppermost lava are entirely scoriaceous. It transforms westwards into coarse scoria breccia (Bsc), and is possibly continuous with a similar vesicular
VSIC01 9/12/08 2:07 PM Page 21
21
Volcanoes erupted in englacial lakes Table 3. Chemical compositions and summary petrography of lavas at Icefall Nunatak Sample: 40 Ar/39Ar age (Ma): Unit: Lithology: Summary petrography:
SiO2 TiO2 Al2O3 Fe2O3T MnO MgO CaO Na2O K2O P2O5 LOI Total Normative Ol Normative Ne mg-number Cr Ni Cu Zn Ga Rb Sr Y Zr Nb Ba La Ce Nd Pb V
MB.48.10 6.80 ± 0.10 Ib alk. basalt lava < 5% phenocrysts of ol > cpx >> pl
MB.48.5 IIf alk. basalt lava 25% phenocrysts of ol = pl >> cpx >> sp
Crystals to 0.6 mm Fine groundmass
Crystals to 3 mm Fine groundmass
MB.48.21 6.47 ± 0.08 IIIa alk. basalt lava 25 – 40% phenocrysts of ol = pl >> cpx >> sp > ap Crystals to 5 mm Coarse groundmass
MB.48.15 IIIb alk. basalt bomb 30% phenocrysts of ol >> cpx >> pl >> sp Crystals to 3.5 mm Coarse groundmass
44.46 2.78 15.27 12.90 0.18 7.12 11.45 3.04 1.04 0.64 1.59 100.48
46.37 2.47 15.29 12.24 0.24 5.88 9.83 3.34 1.10 0.57 2.83 100.16
45.76 2.69 15.36 13.25 0.20 8.94 9.33 3.28 0.84 0.71 – 0.03 100.31
44.77 2.85 15.17 12.96 0.18 8.19 10.28 3.29 0.72 0.69 0.79 99.88
13.92 5.97 0.52
13.48 1.81 0.49
20.46 2.70 0.57
17.50 4.41 0.56
195 85 73 94 22 18 672 27 171 38 240 20 72 29 8 248
317 129 64 102 20 22 655 27 186 40 324 31 68 29 8 203
225 197 60 89 19 12 752 27 165 37 286 31 59 28 9 189
268 137 68 94 20 9 728 29 217 38 241 23 77 34 10 206
All samples analysed by X-ray fluorescence at the University of Keele, UK, using standard procedures. Fe2O3T, all iron calculated as Fe2O3. LOI, loss on ignition. Normative olivine (Ol) and nepheline (Ne) calculated using Fe2O3/FeO = 0.2. mg-number = MgO/(FeOT + MgO), using molar proportions, where FeOT is all iron calculated as FeO.
dyke nearby to the west although the exposures are discontinuous. Unit Ic is > 30 m thick. It is conformable on top of unit Ib and wedges out against the sloping surface of unit Ia. Beds within Ic terminate against Ia without thinning. The unit comprises crudely stratified, blocky and gravelly breccia beds and lenses (Blith) with dispersed basalt pillows up to 80 cm across, alternating with continuous beds of gravelly sandstone (SGp) 30 cm to 4 m thick. The latter have erosive
bases, commonly with a thin basal reverse-graded interval and thin normal-graded tops. They have abundant tachylite and sideromelane clasts with a wide range of vesicularities (incipient to high; categories after Houghton & Wilson, 1989) together with highly crystalline non-vesicular grains resembling clasts in the lithic breccias (Blith). By comparison with units Ia and Ib, the sideromelane and tachylite clasts in Ic are crystal poor (0–5%, rising to ≈ 15% in tachylite).
VSIC01 9/12/08 2:07 PM Page 22
22
J. L. Smellie Fig. 6. View of the 200-m-high rock section exposed at the east end of Icefall Nunatak. The cliff sequences were constructed during three principal eruptive stages described in this paper. Particularly prominent features include a subhorizontal surface separating stages I and II, several very large collapsed stage II blocks (lithofacies Z3, unit IId), and homoclinally sloping large-scale breccia beds and interbedded lavas of stage II lava delta(s) (units IIg, IIh). A single thick basalt lava (unit IIIa) forms the cliff edge at upper right side and the highest exposures are formed by a cinder cone relict (unit IIIb). (Note that the viewpoint for this photograph differs slightly from that used to construct Fig. 3, hence the details do not correspond exactly.)
Interpretation
Fig. 7. Blocky-jointed basalt lava (Lj) and cogenetic carbonate-cemented, coarse, fines-free lithic breccia (Blith) in unit Ia. The pencil is about 15 cm long.
In a glacial context, this lithofacies formed within an englacial vault and/or possibly lake (Fig. 8) and corresponds structurally, but not in lithofacies, to the predominantly effusive pillow volcano stage observed in many subaqueous volcanoes (e.g. Staudigel & Schmincke, 1984; Moore, 1985; Skilling, 1994; Smellie & Skilling, 1994; Smellie & Hole, 1997). Stage I is dominated by a large thickness of lithic orthobreccia and lesser blocky-jointed lava. The patchy sideromelane matrix in the breccias (Blith) formed from a basalt lava by thermal shock and spalling in a subaqueous environment, but the abundance of lithic ( juvenile) fragments and crystal-rich sideromelane indicates derivation mainly from a well-crystallized lava. Juvenile clasts in younger sedimentary beds (unit Ic) are crystal poor, suggesting that the high crystal content of clasts in the lithic breccias was achieved either by crystallization during effusion of the cogenetic basalt lavas or that the lavas were erupted from a separate batch of highly crystallized magma from a vent different from that responsible for the younger units. The field relations and similar petrographical characteristics of the lavas and breccia clasts indicate a cogenetic relationship. The overwhelming dominance of juvenile lithic clasts indicates that they are not hyaloclastite breccias sensu stricto but probably formed as joint-block deposits by mechanical breakage analogous to autoclastic brecciation (Kokelaar, 1986). The disruption was probably aided by hydrofracturing and steam explosions during subaqueous
;
VSIC01 9/12/08 2:07 PM Page 23
23
Volcanoes erupted in englacial lakes
Stage I
Unit 1a
Firn/crevasse layer
Meltwater escapes through permeable firn/crevasse layer
Ice depression
Water-filled vault
Ice
Ice barrier seals meltwater vault
Bedrock
Extrusive and intrusive blocky-jointed lavas and lithic breccias
Meltwater channel cuts into ice by thermal erosion
Unit Ib
Water level lowering
Steam cupola?
Bedrock
Effusion of blocky-jointed lavas and lithic breccias, then vesicular lavas and scoria
Unit Ic
Meltwater flow
Fig. 8. Schematic representation of the possible development of lithofacies erupted during stage I. This phase corresponds to the pillow volcano stage of other subaqueous centres, although only minor pillow lava was produced. Emplacement was probably mainly within an englacial vault and /or lake. (See text for description.) Not to scale.
Bedrock
effusion, leading to the observed association with patchy sideromelane matrix (see Smellie et al., 1998). The association of the lavas and breccias with isolated basalt pillows and pillow lava lenses suggests that the mode of formation of the lavas was close to the transition between blocky and pillowed forms. Laboratory simulations of subaqueous lava flow behaviour have suggested that pillow lavas are favoured over other forms at low magma discharge and/or gentle slopes (Griffiths & Fink, 1992; Gregg & Fink, 1995). Discharge rate is hard to estimate in
Explosive hydrovolcanic eruptions; resedimented breccias
ancient sequences but the cogenetic association with abundant lithic orthobreccia suggests relatively high strain rates (sufficient to enhance extensive breccia development over massive lava) consistent with a relatively rapid magma discharge, and the lavas in unit Ia, at least, were extruded on comparatively steep slopes (≈ 30°). The paucity of vesicles indicates that either the lavas were substantially degassed or vesiculation was suppressed by the overlying hydraulic pressure. However, the presence of highly vesicular basalts (lavas and sideromelane clasts) in units Ib and Ic
VSIC01 9/12/08 2:07 PM Page 24
24
J. L. Smellie
suggests that the magma was not degassed throughout stage I (see below). If vents were subaqueous, the presence of abundant scoria in the uppermost lava in unit Ib would appear anomalous. However, under steady-state sustained eruptions, a steam cupola may form around the eruption column, preventing the incorporation of large volumes of water and essentially isolating the hot vesiculating pyroclasts from contact with water until after deposition (Kokelaar, 1983, 1986; Kokelaar & Busby, 1992). By contrast with breccias in unit Ia and most of those in Ib, breccias (Blith) in the lower parts of unit Ic may have been formed mainly by resedimentation. They are better bedded and lack associated lava lenses, and they were probably derived by collapse of higher parts of unit Ia, which they onlap and resemble in clast content. They are interbedded with and pass up into sandy–gravelly turbidites (SGp) containing some highly vesicular sideromelane, suggesting that eruptions responsible for the clasts were explosive. The presence of tachylite in these sediments need not indicate that eruption columns had a subaerial expression (see Fisher & Schmincke, 1984), as tachylite is also found in basalt pillows (Kawachi & Pringle, 1988) and an origin by reworking unit Ia is possible. The field evidence merely indicates subaqueous deposition. However, in an englacial setting, the stage I volcano would have been constructed in a water-filled cavity (vault or lake) and vents would also have been subaqueous for a substantial period of their activity. The transition to vesicular magma in unit Ic could have been accomplished either by shoaling accompanying progressive vertical aggradation of the vent, or else by a lowering of the contemporary water level. The presence of highly vesicular lavas in unit Ib at comparable structural heights to non-vesicular lavas in unit Ia (Fig. 3) suggests that a lowered water level is more likely, corresponding to reduced hydraulic pressures over the vent. Stage II (sediment flushed subglacially; subaqueous tuff cone; englacial lake; subaerial emergence and slope failure; lava delta) Description Stage II is represented by a sequence > 150 m thick, which is subdivided into eight units (IIa–h). Most of the lithofacies distinguished in the study were formed during this stage and many are confined to it (Table 1). They are mainly bedded gravelly breccias (including hyaloclastite breccia) and gravelly sandstones. A thin
Fig. 9. Planar-stratified and channel-based trough crossstratified breccia and sandstone (SGt) in unit IIa. The crossstratification is broadly unidirectional in this unit and the sediments were probably deposited from traction currents during a brief period of subglacial meltwater flushing during the earliest eruptive phase of stage II. The pencil is about 15 cm long.
distinctive basal unit (IIa) with evidence for palaeocurrent activity is succeeded by two much thicker units (IIb, c) of stratified gravelly sandstone with abundant syndepositional faulting and disturbed beds, which are draped by similarly thick units of sideromelane (hyaloclastite) breccia (IIg, h). Other units identified include intrusive breccias and lavas (units IIe, f) and spectacular large rotated blocks of gravelly sandstone (unit IId). Unit IIa comprises ≈ 2 m of orange to greenish grey coarse gravelly sandstone and breccia formed mainly from non-vesicular juvenile clasts. It forms a tabular horizontal deposit resting with a sharp, slightly unconformable planar contact on the top of unit Ic and is overlain by units IIf and IIg. Beds are ≈ 10–40 cm thick, comprising alternating breccia, gravelly sandstone and laminated coarse to medium sandstone (SGt; Fig. 9). Planar stratified, sometimes crudely normal-graded beds alternate with channel-based trough cross-stratified lenses. Single beds are continuous up to 10 m laterally but are usually < 5 m. The cross-stratification is broadly unidirectional and indicates SSE-flowing palaeocurrents. Units IIb and IIc are very similar. They are dominated by gravelly sandstone (SGp) together with a much smaller proportion of breccia (Bvs), thin-bedded sandstone–mudstone (SM) and disturbed strata (Z). Each unit is 60–80 m thick. The base of unit IIb is obscured but it is separated from overlying unit IIc by a sharp planar discordant surface dipping southward at 30–35°, varying locally to near-vertical. The unit
VSIC01 9/12/08 2:07 PM Page 25
Volcanoes erupted in englacial lakes
25
Fig. 10. View of the western end of Icefall Nunatak. The rock face is ≈ 170 m high and shows unit IIb (predominantly dark-coloured because of abundant mud matrix, with pale interbeds offset by minor faults) in steep fault contact with unit IIc (pale-coloured). The prominent dark-coloured rock forming the upper third of the section comprises stratified hyaloclastite breccia (unit IIg) intruded by coarse lithic breccia and block-jointed lava (unit IIf ), capped by a thick columnar lava (unit IIIa). The viewpoint for this photograph differs slightly from that used to construct Fig. 3, hence the details do not correspond exactly.
IIb sequence is predominantly brown (Fig. 10) owing to the presence of significant mud matrix. Mud-poor beds are khaki yellow and they dominate unit IIc. The deposits are mainly fine gravelly sandstones and coarse sandstones, variably indurated (mud rich) and friable (mud poor; Fig. 11). Large clasts of reworked lithofacies (mainly slabby to contorted SM intraclasts) up to 40 cm in diameter (mainly < 25 cm) are relatively uncommon in IIb but locally numerous in IIc. A few also have associated impact structures, with bedding deformed around the clasts. Beds are typically 1–3 dm thick (range 5 cm to a few metres) and are mainly massive, with ill-defined gradational surfaces. There is no sequential organization (e.g. upward thickening or coarsening). The angular (cuspate to blocky) sideromelane and tachylite (up to 20%) clasts show essentially no post-eruptive shape modification, and vesicularity varies from incipient to high (Houghton & Wilson, 1989). Bed thicknesses and continuity are maintained in exposures ranging up to a few tens of metres across. Amalgamation is common in these sequences and a few beds appear to wedge out laterally. Where gravelly sandstones are interbedded with lithofacies SM, the beds have sharp, planar surfaces, which are locally erosive and may show a variety of structures (loading, grading, planar stratification). However, these more structured sections are minor in the sequence. Breccia (Bvs) is an uncommon lithofacies but is often conspicuous because of the thickness of single beds (typically 1–3 m). The basal surfaces are generally highly erosive and some disappear laterally by amalgamation with overlying beds. Large clasts (up to 80 cm) are common and include basalt pillows
Fig. 11. Planar-stratified gravelly coarse sandstones interpreted as sediment gravity flow deposits (unit IIc). (Note the ill-defined bedding surfaces, faint planar stratification in some beds and massive appearance of others.) Amalgamation is common in the section but is not well seen in this photograph. Much of the stage II subaqueous tuff-cone was constructed of this lithofacies. Rapid, essentially continuous aggradation of coarse hydroclastic tephra linked to sustained eruptions (of continuous-uprush type?) is suggested. The pencil is about 15 cm long.
and fragments of lithofacies SGp and SM. Units IIb and IIc are both characterized by abundant steep faults (Fig. 3), commonly with massive to rarely laminated sandstone and fine breccia matrix along the fault planes. The faults do not extend above unit IIc, and those in unit IIb do not penetrate unit IIc. The sense of movement shown by deformation of adjacent beds suggests that both normal and reverse faults are present. Displacements of a few decimetres to many
VSIC01 9/12/08 2:07 PM Page 26
26
J. L. Smellie
Fig. 12. Heterogeneous deposit of sandstone blocks in massive sandstone matrix (lithofacies Z1, unit IIc). The faint pale- and dark-coloured swirling textures seen in the matrixdominated lower parts of the photograph suggest that disaggregation and fluidization of poorly or unconsolidated sediments has occurred during dewatering, whereas a paler-coloured more resistant sandstone bed is preserved as dispersed relicts above. The latter frequently have conspicuous narrow (millimetres to 1 cm) brown rims as if affected peripherally by interaction with hydrothermal fluids. The field notebook is 20 cm long.
metres are typical, although the lack of distinctive marker lithologies makes precise estimates uncertain in many instances. In several places, the IIc sequence consists of a heterogeneous mixture of blocks of massive and stratified khaki yellow gravelly sandstone with a variety of orientations (including vertical), in massive sandstone ‘matrix’ (lithofacies Z; Fig. 12). The surfaces of many of the constituent blocks are gradational (indeed, large parts of some units are simply diffuse and inhomogeneous, lacking coherent blocks), there are diffuse dewatering zones and the textures are slurry-like (Fig. 12). These areas commonly cut across
a few tens of metres of section, boundaries are gradational and they are spatially associated with faults and the upper surface of unit IIc; they do not pass up into overlying units (Fig. 3). Some also show a pervasive shear-like foliation cutting indiscriminately across blocks. Also included in the lithofacies are rare, conformable, heterogeneous slurry-like beds of khaki sandstone clasts in dark brown muddy sandstone matrix (Fig. 13), and discrete folded beds (Fig. 14). Unit IId comprises several very large (up to ≈ 20 m) blocks of lithofacies SGp (Fig. 6). Stratification in these blocks is commonly tilted to very steep angles (> 50°) and they are injected by basalt dykelets rooted in lava unit IIf and by stringers of grey breccia (Bm; unit IIe). Sediments forming the blocks are indistinguishable from those seen elsewhere in units IIc, but they also include blocks of sandstone formed from aphyric highly vesicular sideromelane. Unit IIe is composed of khaki–grey orthobreccia (Bm). It is draped unconformably on top of, and occupies numerous fractures within, units IIc and IId but is not affected by the faulting observed in IIc. The breccia reaches a maximum thickness of about 15 m and is dominated by coarse, blocky, poorly to non-vesicular crystalline basalt and sideromelane clasts, about 10% khaki highly vesicular sideromelane, minor tachylite and basalt pillows, and ≈ 5–10% silt-grade matrix of blocky non-vesicular sideromelane. Clasts are cemented by zeolite. Unit IIf also drapes unconformably across other units (IIc–e) and it intrudes unit IIg (Fig. 3). It consists of polyhedral blocky- to locally columnar-jointed basalt lava and lithic ( juvenile) orthobreccia (Lj and Blith), which together extend along the entire length of the Icefall Nunatak outcrop. The lava is up to 20 m thick whereas the breccia is locally > 50 m. The lithofacies characteristics and relationships are identical to those of comparable lithofacies observed in stage I (see unit Ia, described above), although occurring on a larger scale. Units IIg and IIh are very similar conformable successions up to 100–120 m in total thickness. Unit IIh is inaccessible but was observed closely using binoculars; IIg is accessible at two localities. They are both formed mainly of polyhedral blocky-jointed and columnar basalt lava and crudely stratified breccia (Lj, Lcol and Bnvs). The two units differ in several respects: Unit IIg is dominated by lithofacies Bnvs, which has a pervasive khaki–orange coloration and few basalt lava lenses, whereas in unit IIh lithofacies (?)Bnvs is grey and is associated with abundant thin columnar lavas (Figs 3 & 6). The lavas are generally
VSIC01 9/12/08 2:07 PM Page 27
Volcanoes erupted in englacial lakes
27
Fig. 13. Heterogeneous ‘slurry-like’ mixture of relict sandstone clasts (some showing plastic deformation) dispersed in dark sandy mudstone (lithofacies Z2, unit IIc). Some textures are very similar to those described in Fig. 12, but this deposit contains conspicuous mud-rich matrix and it occurs as a conformable stratum. The bed probably formed by a combination of synsedimentary processes, including piecemeal sinking of a rapidly deposited sandstone bed into underlying mudstone and concurrent dewatering in the sandstone bed underlying the latter. The field notebook is 20 cm long.
Interpretation
Fig. 14. Upturned gravelly sandstone and sandstone–mudstone strata (SGp and SM) included in lithofacies Z2 (unit IIc). (Note rip-up clasts of thinner sandstone–mudstone beds are seen within one of the deformed beds with a prominent erosive base.) The betterdefined stratification in the lower part of the section and better internal bed organization (e.g. grading) suggest that sedimentation was more pulsed than in the massive-looking sections dominated by SGp (Fig. 11). Palaeocurrents responsible for the deformation travelled right to left. The field notebook is 20 cm long.
1–6 m thick, becoming thicker and having very uneven upper surfaces in eastern sections in which mushroom-like lava apophyses appear to intrude up into the associated massive breccia. A crude planar stratification is present in both units but bed continuity is hard to estimate. Strata dip south at ≈ 25– 30° in both units in the eastern rock face. Bedding is deformed or disappears within 10 –20 m of intrusive apophyses of unit IIf.
Stage II commenced with the deposition of a variety of subaqueous volcaniclastic lithofacies (Fig. 15). Initially, coarse sediments were deposited from traction currents (lithofacies SGt; unit IIa), indicating significant current activity, which is entirely absent elsewhere in the Icefall Nunatak succession. An abundance of non-vesicular clasts in that unit suggests that initial eruptions in stage II were probably nonexplosive. From the modelling of hydraulic activity during subglacial eruptions presented earlier, it is inferred that deposition of unit IIa took place subglacially during the brief period before an englacial vault had formed and while subglacial meltwater was able to drain freely beneath the glacier. This interpretation implies that an ice sheet had re-established itself over the stage I volcano before stage II activity. The period required for an ice sheet to re-establish over a formerly active vent may not be long owing to the speed with which (thermally softened) ice will flow into a cavity. However, because of continued localized melting over the vent, a depression may persist on the glacier surface for decades (Gudmundsson, 1996). By contrast, units IIb and IIc are dominated by a variety of turbidites (lithofacies Bvs, SGp, SM; Table 1), indicating subaqueous deposition below wave base, and inferred from reasoning outlined previously to be within a water-filled vault or lake. The dominance of angular sideromelane with a wide range of vesicularities in these deposits suggests that the clasts were formed from predominantly explosive hydrovolcanic eruptions (see Houghton & Wilson, 1989). The
; VSIC01 9/12/08 2:07 PM Page 28
28
J. L. Smellie
Stage II No surface depression
Unit IIa
Meltwater drains subglacially
No ice barrier
Gravels deposited from traction currents
Units IIb, c
Blocky-jointed lavas and lithic breccias? (inferred, not exposed)
Meltwater escaping through firn/crevasse zone aquifer
Ice barrier seals vault
Early slope failure structures corresponding to unit IIb
Explosive hydrovolcanic eruptions; tephra distributed on tuff cone flanks by sediment gravity flows
Subaerial emergence; lava deltas begin to prograde
Units IId, e
Major slope failure of tuff cone flank (units IId, IIe)
Units IIf, g, h Meltwater continues to escape supraglacially
Intrusive blocky-jointed lava with lithic breccia carapace (unit IIf)
Lava delta progradation
Fig. 15. Schematic representation of the possible development of lithofacies erupted during stage II. The lithofacies accumulated mainly within an overflowing englacial lake. The presence of blocky-jointed lavas and breccias similar to those formed in stage I is postulated but they are not currently exposed. (See text for description.) Not to scale.
VSIC01 9/12/08 2:07 PM Page 29
Volcanoes erupted in englacial lakes presence of basalt pillows in some lithofacies (Bvs) may indicate minor coeval effusion of pillow lava, or the pillows were reworked from an early stage II ‘pillow volcano’ (not now exposed) or from the stage I edifice. The relatively steeply dipping, massive-bedded, poorly sorted beds in units IIa and IIb, formed of coarse sideromelane tephra, resemble those of ventproximal parts of other subaqueous tuff cones (e.g. Staudigel & Schmincke, 1984; Moore, 1985; Skilling, 1994; Sohn, 1995; Smellie & Hole, 1997) and correspond to a period of subaqueous tuff cone growth. The sequence characteristics suggest a virtually continuous tephra supply, probably linked to sustained eruptions and rapid edifice construction (Smellie & Hole, 1997). In marine Surtseyan centres, such activity corresponds to the continuous-uprush style of eruption (Kokelaar, 1983; Moore, 1985). These are periods of relatively high rates of magma discharge and vertical aggradation of the volcanic pile. Gravelly sandstones are also locally interbedded with thick breccia beds (Bvs), thin-bedded fine sandstone–mudstone (SM) and slurry-like beds (Z; Fig. 13). In these locations, the SGp beds are better defined and sharp erosive bases are common. Sedimentation of each of these lithofacies probably took place in more widely separated pulses compared with the much thicker and more massive-bedded sections. The breccias and some sandstones contain fragments of reworked lithofacies (mainly basalt pillows, SM and SGp) suggesting an origin by the collapse of unstable volcano flanks, whereas the SM lithofacies may represent periods of temporary quiescence. Some associated gravelly sandstone beds could alternatively be related to small ventclearing explosions or discrete tephra jetting activity associated with relatively low rates of magma supply. However, the contribution of these lithofacies to the vertical aggradation of the tuff cone was minor. Unit IIb differs from IIc principally in an abundance of mud matrix in IIb. In other descriptions of subaqueous tuff cone sequences, a paucity of fines has been noted as a significant distinguishing characteristic and attributed to either the finest detritus bypassing the vent edifice (by a variety of processes) and accumulating in the surrounding basin, or to submerged vents producing generally coarser tephra (Kokelaar & Durant, 1983; Cas et al., 1989; Skilling, 1994; Kano et al., 1996; Smellie & Hole, 1997). At Icefall Nunatak, muddy fines were clearly produced during at least part of the activity, but the separation of mud-sized grains appears to have been much less efficient in unit IIb than in the younger unit IIc. Two explanations are possible, as follows.
29
Gradients on the volcano flanks may have been low during the early stages of activity (unit IIb), when the edifice may have been a broad low tephra mound. This has been noted in lacustrine tuff cones and occurs during the earliest stages of construction (White, 1996), although it has not been reported from subglacial volcanoes. The inertia of sediment gravity flows crossing a low tephra mound may be insufficient to generate the level of turbulence and shear required for fines separation from the moving flows (see Kokelaar & Busby, 1992). By contrast, unit IIc was draped on volcano flanks oversteepened by faulting (e.g. boundary between units IIb and IIc; see above), thus optimizing the inertia and turbulence in turbidity currents and leading to more efficient particle separation. Alternatively, eruption columns during unit IIb were lower and denser than for unit IIc. They may have been largely or entirely contained by the water column. This is possibly supported by the absence of accretionary lapilli in both successions, a common component in many hydrovolcanic deposits, and consistent with low, wet columns (Moore, 1985; Schumacher & Schmincke, 1991; Smellie & Hole, 1997). A consequence of lower denser columns should be less efficient separation of coarse and fine particles (unit IIb?) compared with eruption columns with a sizeable subaerial component (unit IIc?) (see Cas et al., 1989; Cashman & Fiske, 1991; Kokelaar & Busby, 1992; Kano et al., 1996). It may be significant that unit IIc is succeeded by subaerial lava deltas (units IIg and h; see below), indicating shoaling and emergence of the tuff cone. At least two major episodes of slope failure of the southern flank of the cone are indicated by the observed field relationships. Units IIc and IIb are extensively faulted internally, both are in fault contact and IIc has an uneven (faulted or collapsed?) upper surface (Fig. 3); several very large blocks of stratified gravelly sandstone are present at low elevations on the volcano (unit IId) and appear to have become detached from higher parts of the edifice and rolled downslope; and the volcano flank at the end of unit IIc time was draped by massive grey breccia (Bm) of unit IIe, which also back-injected unit IIc along a series of irregular fractures. The spatial association between steep cross-cutting zones of disturbed sediments (lithofacies Z; Fig. 12), faults and the surface of unit IIc also implies an origin for the disturbed zones and faults related to collapse of the cone flank. The abundant evidence for water escape and poorly lithified state of constituent blocks supports a syndepositional age for much (all?) of the
VSIC01 9/12/08 2:07 PM Page 30
30
J. L. Smellie
deformation. The diffuse boundaries both surrounding the lithofacies and around blocks within the lithofacies also suggest that the lithofacies was not emplaced in a dry state (an essential feature of debris avalanche deposits; Siebert, 1984) and the field relations suggest that it was mainly formed more or less in situ. The outcrops of this lithofacies may be in areas where there was high water escape, perhaps preferentially focused on coeval synsedimentary faults. Momentary pressure relief during fault formation may have caused expansion into the faults of pore water from adjacent poorly or unconsolidated waterrich sediments, thus initiating fluidization, sediment disaggregation and focused pore-water discharge. These events may be favoured in a sequence with excess pore-fluid pressures caused by rapid vertical aggradation of tephra, particularly if faulting was related to sector collapse of the volcano flanks and resulted in rapid localized pressure release. Broadly similar processes (on a smaller scale) have been advanced to explain the presence of a sedimentary matrix injected along synsedimentary fault planes (Smellie & Hole, 1997). The foliation observed within some examples of the lithofacies implies the presence of a strong shearing couple consistent with deformation effects of synsedimentary faulting on less watersaturated sections or the sliding of large blocks (unit IId) down the tuff cone flanks. Some of the large blocks have an aphyric basalt provenance, suggesting that parts of the early volcano constructed during stage I were also involved in stage II slope failure. Units IIg and IIh are largely formed of nonvesicular sideromelane breccia (Bnvs) in steeply dipping, large-scale beds corresponding to foresets in a southward-prograding lava delta (Table 1). The field relations and textural characteristics suggest that the breccias were derived from the thermal quenching and disintegration of the associated lavas, many of which penetrated far down the foreset bedding and are intimately associated with a cogenetic breccia carapace (Fig. 3). The presence of mushroom-like apophyses that intrude upward from some of the lavas into overlying breccia (Figs 3 & 6) suggests that post-emplacement inflation (and upward break-out) of those flows may have occurred (compare the formation of inflated pahoehoe sheet flows; Self et al., 1998). The formation of a lava delta is unequivocal evidence that effusion took place subaerially and the lavas entered water (an englacial lake). However, a passage zone, marking the water level contemporaneous with eruption, is not clearly preserved. An estimate for the minimum elevation of the lake surface is the base of unit IIIa near the
west end of the nunatak (i.e. ≈ 700 m a.s.l.). This height is 100 m above the maximum elevation of englacial lakes associated with multiple subglacial eruptions at nearby Hedin Nunatak, erupted during the same period. The abundance of blocky, poorly to non-vesicular sideromelane in unit IIe suggests that emergence and growth of a lava delta had already begun before the slope failure event decribed above. Although they are conformable, there is a marked colour difference between units IIg and IIh. The strong khaki–orange coloration of unit IIg is characteristic of pervasively palagonite-altered sideromelane, whereas the dark grey colour of unit IIh (inaccessible) is more characteristic of relatively unaltered sideromelane. The alteration of unit IIg is possibly associated with the intrusion of the cogenetic lava–breccia couplet that forms unit IIf. However, the very uneven upper surface of IIf is not mirrored by an uneven alteration front. The strong khaki–orange coloration is pervasive throughout, and stratigraphically confined to unit IIg, suggesting that the upper surface of that unit represents a time break (paraconformity). A less plausible alternative explanation is that different vents were responsible for units IIg and IIh, and IIg was deposited closer to (and altered hydrothermally by) its vent compared with that responsible for unit IIh. However, bedding in both units appears to have a similar homoclinal orientation consistent with a single source vent. Moreover, unit IIe, which is below and stratigraphically older than IIg, shows little alteration, yet it should have been altered similar to unit IIg if that alteration was due to a very proximal vent. However, even if two different vents were present, hydrothermal activity associated with the unit IIg vent must have ceased before effusion of unit IIh for IIh to show so little alteration, and a time break is still implied by the relationships. Stage III (lava effusion (within firn and/or snow?); Strombolian and/or Hawaiian activity) Description Only two units are present in stage III, which has a total thickness of ≈ 70 m. The oldest comprises a 20m-thick basalt lava (unit IIIa) that caps the prominent north-facing escarpment and forms the ice-scoured summit platform of the nunatak; the same lava has also been traced laterally to a small unnamed outcrop situated ≈ 1 km west of Icefall Nunatak (Fig. 1). The lava has a thin (2–4 m) basal colonnade overlain by much thicker entablature. The basal surface is inacces-
;
VSIC01 9/12/08 2:07 PM Page 31
Volcanoes erupted in englacial lakes
31
Stage III
Meltwater overflow (minor)
Fig. 16. Schematic illustration of the final phase of development of the volcanic centre at Icefall Nunatak (stage III). The lava shows evidence for substantial water cooling and was probably extruded into a superficial layer of snow and firn, thus resembling some subglacial eruptions of sheet-flow type. (See text for description.) Not to scale.
Units IIIa, b
Lava effusion within firn/snow; cinder cone eruptions
Ice seals vault
sible but in western exposures it appears to be sharp, without development of breccia. It is also steeply discordant to bedding at the west end of the nunatak, where it transforms to a cross-cutting columnar neck. The contact is more gently discordant and undulating eastwards, becoming conformable with the unit and possibly wedging out in eastern exposures. Thin lenses of grey scoriaceous clinker are patchily present within the lava. Inaccessible eastern exposures show a highly uneven upper lava surface with multiple irregular flame-like apophyses extending at least 20 m up into grey breccia-like rock, similar to relationships between lavas and breccia within unit IIh (Fig. 6). The lava is overlain (contact inaccessible) by up to 50 m of dark grey to maroon coarse scoria breccia (unit IIIb) formed of crudely planar-stratified moderately to highly vesicular sideromelane and tachylite, many with fluidal surfaces and associated with and loosely bound to welded lenses of slightly flattened bombs up to 25 cm long. Ovoid bombs up to 12 cm in diameter, with non-vesicular cores and narrow microvesicular rims, are dispersed in the deposit; some have cores of cognate lithic clasts. Also present are minor thin sand-grade beds with abundant broken sideromelane droplets (achneliths). The breccia is welded and maroon coloured on the south side of the nunatak. Parts of the deposit are khaki–yellow-coloured, related to local palagonite alteration; other parts are cemented by zeolite. Interpretation
Unit IIIa forms a continuous outcrop changing eastward from a steeply cross-cutting neck to a conformable lava flow (Fig. 16). The presence in the lava of columnar jointing with a thin basal colonnade and much thicker entablature suggests chilling by rapid convective cooling caused by water penetrating along downward-propagating cracks (e.g. Long & Wood,
1986). Such chilling is characteristic of subaerial lavas flooded by water but has also been observed in lavas in some subglacial and submarine volcanic sequences (Walker & Blake, 1966; Schmincke & Bednarz, 1990; Bergh & Sigvaldason, 1991; Smellie et al., 1993a). The association with probable cogenetic breccia in eastern parts of the outcrop also supports the suggestion of a water-rich environment. At Icefall Nunatak, the lava is situated at the top of the volcanic sequence, where it forms a mesa-like landform (compare Icelandic table mountains). The sharp basal lava contact, lacking any intervening sediments, and compositional similarities between stages II and III suggest that the gap in time between the two stages may have been small. However, in view of the speed with which ice sheets can reestablish over inactive vent sites, a thin glacier surface (mainly snow or firn?) probably re-formed over most or all of the subaerially exposed parts of the stage II volcano. Thus, the lava may have interacted with the thin glacier cover and it could be analogous to some subglacial sheet-flow sequences (see Smellie et al., 1993a; Fig. 5a). The presence of a massive lava sheet rather than pahoehoe (or pillowed) lava is suggestive of relatively high rates of magma supply. The intimate relationship between lava lobes and dark grey (?)breccia in western exposures is likely to be cogenetic and resembles similar occurrences present in stage II (unit IIh) and described elsewhere (e.g. Bergh & Sigvaldason, 1991; Smellie et al., 1993a). Unit IIIb is a cinder cone remnant. Absence of evidence for chilling and fracturing of pyroclasts by external water, the generally high vesicularity, abundance of tachylite, paucity of fines and bedding characteristics indicate relatively low-energy subaerial eruptions mainly of Strombolian type. Conversely, the palagonite alteration of sideromelane is more characteristic of hydrovolcanic sequences (subaerial and subaqueous, e.g. Jakobsson, 1978; Wohletz & Sheridan, 1983). Its occurrence in Strombolian tephra is not
VSIC01 9/12/08 2:07 PM Page 32
32
J. L. Smellie
commonly recorded (Houghton & Hackett, 1984; LeMasurier et al., 1994; Smellie & Hole, 1997) but it is probably a common feature in newly emergent volcanoes associated with local high geothermal gradients (e.g. related to dyke intrusion; Jakobsson, 1978; Smellie & Hole, 1997). The thin sand-grade beds with abundant achneliths probably formed from lava spray during very fluid eruptions (Walker & Croasdale, 1972) and the dense ovoid bombs may represent clots of degassed magma re-entrained in low-energy eruption columns. Re-entrainment is a common process in mildly explosive Hawaiian eruptions and favoured by (but not dependent on) the presence of a lava lake (Parfitt et al., 1995; Wilson et al., 1995). The distribution of normal faulting, quaversal bedding attitudes and development of welding indicate a vent locus close to the south-east side of the nunatak. The stage III vents were not coincident, and both are located south of the presumed position of the principal stage II vent.
CONCLUSIONS From a review and interpretation of the empirical effects of glacier physics (particularly hydraulics) on eruptions beneath temperate ice, three empirical volcano ‘types’ are proposed, which differ in the kinds and proportions of the lithofacies present and in their lithofacies architecture. Under conditions of thin ‘ice’ (i.e. mainly composed of snow and firn; generally < 100 m thick), corresponding to ‘sheet-flow eruptions’ previously described (Smellie et al., 1993a), meltwater generated above an eruptive site is able to escape continuously beneath the glacier and the associated volcaniclastic lithofacies show abundant evidence for deposition from traction currents (model 1; Fig. 5). By contrast, in eruptions beneath thicker ice (i.e. formed mainly of impermeable ice; glaciers generally > 200 m thick), meltwater accumulates in a vault or lake above the vent. Two volcano ‘types’ associated with thick ice are distinguished, whose formation is determined by differences in the structure of the overlying glacier. One type has no published example but is probably dominated by subaqueously emplaced lava, breccia and tuff cone deposits, which may be draped unconformably by coeval subaerial lava (model 2). The other type, characterized by a subaqueous sequence dominated by a variety of sediment gravity flow deposits in a subaqueous tuff cone, is capped by hyaloclastite breccias formed in lava delta(s) (model 3). It corresponds to the well-known volcanoes known as table mountains or tuyas (Jones, 1969, 1970; Smellie & Hole, 1997).
The sequence of lithofacies mapped at Icefall Nunatak, a remnant of a small Miocene polygenetic volcano, is interpreted in the context of these models. From the abundant evidence for deposition in ponded water, the volcano was probably erupted beneath thick ice, within englacial vaults and lakes. If a relatively stable ice-sheet surface elevation around the volcano persisted throughout the eruptive period, the ice sheet would have been at least 200 m thick based on the outcrop thickness observed today, although the base of the sequence is unexposed. Three major constructive stages are distinguished, each separated by unconformities representing periods in which the ice sheet was re-established over the eruptive site(s). During stage I, only subaqueously emplaced lavas, cogenetic breccias and sediment gravity flows were formed, suggestive of accumulation (and likely eruption) in an englacial vault or lake. Stage II commenced with the deposition of coarse volcaniclastic debris from traction currents, in an early period when meltwater was temporarily able to flow away from the eruptive site, before formation of a second englacial vault. This may be the first time such deposits have been recognized in an englacial volcano. Subsequently, a subaqueous tuff cone was constructed in a vault and lake. It underwent extensive syndepositional deformation and sector collapse. The characteristics of the tuff cone lithofacies suggest that vertical aggradation took place rapidly during sustained eruptions, possibly akin to the continuous-uprush style of activity observed in marine Surtseyan centres, and the predominant sandy–gravelly deposits are interpreted as mainly syneruptive redeposited tephra. Stage II activity culminated in two phases of subaerial effusion and lateral progradation of hyaloclastite deltas into an englacial lake, when the volcano emerged above the coeval ice sheet. Effusion of a widespread lava occurred during stage III, followed by construction of a small cinder cone. Evidence for substantial water cooling of the lava suggests that the lava interacted with a re-established thin glacial cover, possibly just snow and firn, analogous to effusion during eruptions of sheet-flow type.
ACKNOWLEDGEMENTS The fieldwork in Marie Byrd Land formed part of the West Antarctic Volcano Exploration (WAVE) project (1989–1991). WAVE was supported by the British Antarctic Survey, National Science Foundation and the New Zealand University Grants Committee. The
VSIC01 9/12/08 2:07 PM Page 33
Volcanoes erupted in englacial lakes author gratefully acknowledges efforts by W. C. McIntosh and T. Wilch (New Mexico Institute of Mining and Technology, USA) to date by 40Ar/39Ar the author’s samples from the Mount Murphy sequences described in this paper. Thanks are also due to all members of the WAVE team at Mount Murphy (W. C. McIntosh, K. S. Panter, J. A. Gamble, P. Rose and W. Atkinson) for discussions and cheerful assistance in the field, and to the US Navy VXE-6 Squadron for transport within Antarctica. Comments on this paper by Jocelyn McPhie and R. Werner are gratefully acknowledged.
REFERENCES Allen, C.C. (1980) Icelandic subglacial volcanism: thermal and physical studies. J. Geol., 88, 108–117. Allen, C.C., Jercinovic, M.J. & Allen, J.S.B. (1982) Subglacial volcanism in north-central British Columbia and Iceland. J. Geol., 90, 699–715. Behrendt, J.C., LeMasurier, W.E., Cooper, A.K., Tessensohn, F. & Damaske, D. (1991) The West Antarctic Rift Systemaa review of geophysical investigations. In: Contributions to Antarctic Research II (Ed. Elliot, D.H.), Am. geophys. Union, Antarctic Res. Ser., 53, 67–112. Bergh, S.G. & Sigvaldason, G.E. (1991) Pleistocene mass-flow deposits of basaltic hyaloclastite on a shallow submarine shelf, South Iceland. Bull. Volcanol., 53, 597–611. Björnsson, K. (1988) Hydrology of ice caps in volcanic regions. Vísindafélag Íslendinga, Societas Scientarium Islandica, 45, 139pp. Cas, R.A.F. & Wright, J.V. (1987) Volcanic Successions. Allen & Unwin, London. Cas, R.A.F., Landis, C.A. & Fordyce, R.E. (1989) A monogenetic, Surtla-type, Surtseyan volcano from the Eocene– Oligocene Waiareka–Deborah volcanics, Otago, New Zealand: a model. Bull. Volcanol., 51, 281–298. Cashman, K.V. & Fiske, R.S. (1991) Fallout of pyroclastic debris from submarine volcanic eruptions. Science, 253, 275–280. Cooper, A.K., Barrett, P.J., Hinz, K., Traube, V., Leitchenkov, G. & Stagg, H.M.J. (1991) Cenozoic prograding sequences of the Antarctic continental margin: a record of glacio-eustatic and tectonic events. Mar. Geol., 102, 175–213. Fisher, R.V. & Schmincke, H.-U. (1984) Pyroclastic Rocks. Springer, Berlin. Gore, D.B. (1992) Ice-damming and fluvial erosion in the Vestfold Hills, East Antarctica. Antarctic Sci., 4, 227–234. Gregg, T.K.P. & Fink, J.H. (1995) Quantification of submarine lava-flow morphology through analog experiments. Geology, 23, 73 –76. Griffiths, R.W. & Fink, J.H. (1992) Solidification and morphology of submarine lavas: a dependence on extrusion rate. J. Geophys. Res., 97, 19729–19737. Gudmundsson, M. (1996) Ice–volcano interaction at the subglacial Grimsvötn volcano, Iceland. CRREL Special Report 96-27, 34–40.
33
Gudmundsson, M.T. & Björnsson, H. (1991) Eruptions in Grímsvötn, Vatnajökull, Iceland. 1934–91. Jökull, 41, 21–45. Gudmundsson, M.T., Sigmundsson, F. & Björnsson, H. (1997) Ice–volcano interaction of the 1996 Gjálp subglacial eruption, Vatnajökull, Iceland. Nature, London 389, 954 –957. Hamilton, W. (1972) The Hallett Volcanic Province, Antarctica. US geol. Surv. Prof. Pap. Denver, CO, 456-C. Hole, M.J. & LeMasurier, W.E. (1994) Tectonic controls on the geochemical composition of Cenozoic mafic alkaline volcanic rocks from West Antarctica. Contrib. Mineral. Petrol., 117, 187–202. Houghton, B.F. & Hackett, W.R. (1984) Strombolian and phreatomagmatic deposits of Ohakune Craters, Ruapehu, New Zealand: a complex interaction between external water and rising basaltic magma. J. Volcanol. geothermal Res., 21, 207–232. Houghton, B.F. & Wilson, C.J.N. (1989) A vesicularity index for pyroclastic deposits. Bull. Volcanol., 51, 451–462. Jakobsson, S.P. (1978) Environmental factors controlling the palagonitization of the Surtsey tephra, Iceland. Bull. geol. Soc. Denmark (special issue), 27, 91–105. Jones, J.G. (1966) Intraglacial volcanoes of south-west Iceland and their significance in the interpretation of the form of marine basaltic volcanoes. Nature, 212, 586 –588. Jones, J.G. (1969) Intraglacial volcanoes of the Laugarvatn region, south-west IcelandaI. Q. J. geol. Soc., London, 124, 197–211. Jones, J.G. (1970) Intraglacial volcanoes of the Laugarvatn region, southwest Iceland. J. Geol., 78, 127–140. Kano, K., Yamamoto, T. & Ono, K. (1996) Subaqueous eruption and emplacement of the Shinjima Pumice, Shinjima (Moeshima) Island, Kagoshima Bay, SW Japan. J. Volcanol. geothermal Res., 71, 187–206. Kawachi, Y. & Pringle, I.J. (1988) Multiple-rind structure in pillow lava as an indicator of shallow water. Bull. Volcanol., 50, 161–168. Kokelaar, B.P. (1983) The mechanism of Surtseyan volcanism. J. geol. Soc., London, 140, 939–944. Kokelaar, B.P. (1986) Magma–water interactions in subaqueous and emergent volcanism. Bull. Volcanol., 48, 275–289. Kokelaar, B. & Busby, C. (1992) Subaqueous explosive eruption and welding of pyroclastic deposits. Science, 257, 196 –201. Kokelaar, B.P. & Durant, G.P. (1983) The submarine eruption and erosion of Surtla (Surtsey), Iceland. J. Volcanol. geothermal Res., 19, 239–246. LeMasurier, W.E. (1972) Volcanic record of Cenozoic glacial history of Marie Byrd Land. In: Antarctic Geology and Geophysics (Ed. Adie, R.J.), pp. 251–260. Universitetsforlaget, Oslo. LeMasurier, W.E. (1990) B. Marie Byrd Land. In: Volcanoes of the Antarctic Plate and Southern Oceans (Eds LeMasurier, W.E. & Thomson, J.W.), Am. geophys. Union, Antarctic Res. Ser., 48, 147–163. LeMasurier, W.E. & Rex, D.C. (1982) Volcanic record of Cenozoic glacial history in Marie Byrd Land: revised chronology and evaluation of tectonic factors. In: Antarctic Geoscience (Ed. Craddock, C.), pp. 725–734. University of Wisconsin Press, Madison, WI. LeMasurier, W.E. & Rex, D.C. (1989) Evolution of linear volcanic ranges in Marie Byrd Land, West Antarctica. J. geophys. Res., 94, 7223–7236.
VSIC01 9/12/08 2:07 PM Page 34
34
J. L. Smellie
LeMasurier, W.E., Harwood, D.M. & Rex, D.C. (1994) Geology of Mount Murphy Volcano: an 8-m.y. history of interaction between a rift volcano and the West Antarctic ice sheet. Geol. Soc. Am. Bull., 106, 265–280. Long, P.E. & Wood, B.J. (1986) Structures, textures, and cooling histories of Columbia River basalt flows. Geol. Soc. Am. Bull., 97, 1144 –1155. Lowe, D.R. (1982) Sediment gravity flows: II. Depositional models with special reference to the deposits of highdensity turbidity currents. J. sediment. Petrol., 52, 279–297. McPhie, J., Doyle, M. & Allen, R. (1993) Volcanic Textures. A Guide to the Interpretation of Textures in Volcanic Rocks. Tasmanian Government Printing Office, Hobart. Menzies, J. (Ed.) (1995) Modern Glacial Environments: Processes, Dynamics and Sediments. Butterworth–Heinemann, Oxford. Moore, J.G. (1985) Structure and eruptive mechanisms at Surtsey volcano, Iceland. Geol. Mag., 122, 649– 661. Panter, K.S., McIntosh, W.C. & Smellie, J.L. (1994) Volcanic history of Mount Sidley, a major alkaline volcano in Marie Byrd Land, Antarctica. Bull. Volcanol., 56, 361–376. Panter, K.S., Kyle, P.R. & Smellie, J.L. (1997) Petrogenesis of a phonolite–trachyte succession at Mount Sidley, Marie Byrd Land, Antarctica. J. Petrol., 38, 1225–1253. Parfitt, E.A., Wilson, L. & Neal, C.A. (1995) Factors influencing the height of Hawaiian lava fountains: implications for the use of fountain height as an indicator of magma gas content. Bull. Volcanol., 57, 440– 450. Paterson, W.S.B. (1994) The Physics of Glaciers. Pergamon– Elsevier Science, Oxford. Schmincke, H.-U. & Bednarz, U. (1990) Pillow, sheet-flow and breccia volcanoes and volcano-tectonic cycles in the Extrusive Series of the northeastern Troodos ophiolite (Cyprus). In: Ophiolites: Oceanic Crustal Analogues (Eds Malpas, J., Moores, E.M., Panayiotou, A. & Xenophontos, C.), pp. 185–206. Geological Survey Department, Nicosia, Cyprus. Schumacher, R. & Schmincke, H.-U. (1991) Internal structure and occurrence of accretionary lapilli: a case study at Laacher See Volcano. Bull. Volcanol., 53, 612– 634. Self, S., Keszthelyi, L. & Thordarson, Th. (1998) The importance of pahoehoe. Ann. Rev. Earth planet. Sci., 26, 81–110. Siebert, L. (1984) Large volcanic debris avalanches: characteristics of source areas, deposits and associated eruptions. J. Volcanol. geothermal Res., 22, 163–197. Skilling, I.P. (1994) Evolution of an englacial volcano: Brown Bluff, Antarctica. Bull. Volcanol., 56, 573–591. Smellie, J.L. & Hole, M.J. (1997) Products and processes in Pliocene–Recent, subaqueous to emergent volcanism in the Antarctic Peninsula: examples of englacial Surtseyan volcano construction. Bull. Volcanol., 58, 628– 646. Smellie, J.L. & Skilling, I.P. (1994) Products of subglacial volcanic eruptions under different ice thicknesses: two examples from Antarctica. Sediment Geol., 91, 115–129. Smellie, J.L., Pankhurst, R.J., Hole, M.J. & Thomson, J.W. (1988) Age, distribution and eruptive conditions
of late Cenozoic alkaline volcanism in the Antarctic Peninsula and eastern Ellsworth Land: a review. Bull. Br. Antarctic Surv., 80, 21– 49. Smellie, J.L., Hole, M.J. & Nell, P.A.R. (1993a) Late Miocene valley-confined subglacial volcanism in northern Alexander Island, Antarctic Peninsula. Bull. Volcanol., 55, 273–288. Smellie, J.L., McIntosh, W.C., Gamble, J.A., Panter, K.S., Kyle, P.R. & Dunbar, N.W. (1993b) Preliminary lithofacies assessment and 40Ar/39Ar ages of Cenozoic volcanic sequences in eastern Marie Byrd Land. Antarctic Sci., 5, 105 –106. Smellie, J.L., Millar, I.L., Rex, D.C. & Butterworth, P.J. (1998) Subaqueous, basaltic lava dome and carapace breccia on King George Island, South Shetland Islands, Antarctica. Bull. Volcanol., 59, 245 –261. Sohn, Y.K. (1995) Geology of Tok Island, Korea: eruptive and depositional processes of a shoaling to emergent island volcano. Bull. Volcanol., 56, 660 – 674. Staudigel, H. & Schmincke, H.-U. (1984) The Pliocene seamount series of La Palma /Canary Islands. J. geophys. Res., 89, 11195–11215. Surtsey Research Society (1970) Surtsey Research Report 5: 1968 Field Season. Surtsey Research Society, Reykjavik. Syverson, K.M., Gaffield, S.J. & Mickelson, D.M. (1994) Comparison of esker morphology and sedimentology with former ice-surface topography, Burroughs Glacier, Alaska. Geol. Soc. Am. Bull., 106, 1130 –1142. Walker, G.P.L. & Blake, D.H. (1966) The formation of a palagonite breccia mass beneath a valley glacier in Iceland. Q. J. geol. Soc., London, 122, 45– 61. Walker, G.P.L. & Croasdale, R. (1972) Characteristics of some basaltic pyroclastics. Bull. Volcanol., 35, 303–317. Werner, R., Schmincke, H.-U. & Sigvaldason, G. (1996) A new model for the evolution of table mountains: volcanological and petrological evidence from Herdubreid and Herdubreidartögl volcanoes (Iceland). Geol. Rundsch., 85, 390 –397. White, J.D.L. (1996) Pre-emergent construction of a lacustrine basaltic volcano, Pahvant Butte, Utah (USA). Bull. Volcanol., 58, 249 –262. Wilch, T.I., McIntosh, W.C. & Dunbar, N.W. (1999) Late Quaternary volcanic activity in Marie Byrd Land: potential 40Ar/39Ar-dated time horizons in West Antarctic ice and marine cores. Geol. Soc. Am. Bull., 111, 1563–1580. Wilson, L., Parfitt, E.A. & Head, J.W. (1995) Explosive volcanic eruptions aVIII. The role of magma recycling in controlling the behaviour of Hawaiian-style lava fountains. Geophys. J. Int., 121, 215 –225. Wohletz, K.H. & Sheridan, M.F. (1983) Hydrovolcanic explosions II. Evolution of basaltic tuff rings and tuff cones. Am. J. Sci., 283, 384 – 413. Wörner, G. & Viereck, L. (1987) Subglacial to emergent volcanism at Shield Nunatak, Mt. Melbourne Volcanic Field, Antarctica. Polarforschung, 57, 27– 41. Wright, A.C. (1980) Landforms of McMurdo Volcanic Group, southern foothills of Royal Society Range, Antarctica. N.Z.J. Geol. Geophys., 23, 605– 613.
VSIC02 9/12/08 2:07 PM Page 35
Spec. Publs int. Ass. Sediment. (2001) 30, 35 – 60
Eruptive process, effects and deposits of the 1996 and the ancient basaltic phreatomagmatic eruptions in Karymskoye lake, Kamchatka, Russia A . B E L O U S O V and M . B E L O U S O V A Institute of Volcanic Geology and Geochemistry, Petropavlovsk-Kamchatsky, 683006, Russia
ABSTRACT On 2–3 January 1996 an explosive eruption discharging ≈ 106 kg s–1 of basaltic magma occurred in Karymskoye lake at an initial water depth of ≈ 50 m. Characteristics of the deposits together with analyses of a videotape of several explosions have allowed us to model the eruptive events. Initial vent-clearing phreatic explosions ejected blocks of country rocks (up to 3 m diameter) to distances of up to 1.3 km. Then followed 10 –20 h of phreatomagmatic Surtseyan activity (100 –200 outbursts of water–gas–pyroclastic mixtures to heights of up to 1 km, with initial velocities of 110 m s–1). The eruption slugs collapsed back into the lake and produced base surges (runout up to 1.3 km; average velocity 12.5 m s–1). The convective eruption plume rose to a height of 3 km and deposited a thin distal fall deposit. The eruption ended with the ejection of scoria-crust bombs (specific basaltic bombs with dense core and scoriaceous crust). Pyroclasts of the eruption are mostly poorly to moderately vesicular juvenile basaltic particles shaped by a combination of magmatic vesiculation and magma–water interaction. Ninety-five per cent of the products (0.047 km3) formed an underwater tuff ring composed of parallel layers of moderately to poorly sorted lapilli ash and ash lapilli (Md – 3.9 to 0.6φ; sorting 1.5 –3.2φ), each 10 – 60 cm thick. They were deposited by water-rich base surges that originated from Surtseyan type eruption bursts. The most widespread hazards of the eruption were tsunamis and lahars. At distances < 1.3 km from the crater, base surges and ballistic clasts were very destructive. Eruptive activity in the lake before 1996 included two eruptions at c. 4800 14C yr bp. The first left deposits similar to those of the 1996 eruption and thus is interpreted as a Surtseyan eruption that occurred at the same water depth as in 1996. The second of the 4800 14C yr bp eruptions deposited extensive cross-laminated basesurge deposits and is interpreted to have occurred in very shallow water.
INTRODUCTION to fuel–coolant interaction (Kokelaar, 1983; Sheridan & Wohletz, 1983; Wohletz, 1983, 1986; Wohletz & McQueen, 1984; Zimanowski, 1998). Such eruptions are termed phreatomagmatic or hydrovolcanic. As a result of the existence of an additional mechanism of fragmentation, subaerial phreatomagmatic eruptions (in which magma contacts ground water) are always more violent, and their magma is effectively more fragmented (pyroclasts are finer grained), than if the same magma was erupted in a dry environment (Walker & Croasdale, 1972; Heiken & Wohletz, 1986). Processes of subaqueous explosive eruptions are more complicated. Violence of the explosions declines with increasing water depth because elevation of ambient pressure quickly suppresses explosive processes. Thus
The dynamics of eruptions are governed by many factors, the most important of which are properties and flux rate of magma, geometry of the conduit and vent, and environment in which magma erupts (Fisher & Schmincke, 1984; Francis, 1993). With respect to the last factor, there is much geological evidence, supported by direct observations of eruptions, that the presence of water in a place where an eruption occurs notably modifies the style of eruptive activity (e.g. Lorenz, 1973; Houghton & Schmincke, 1986). When magma erupts into a ‘wet’ environment, fragmentation can occur not only as a result of expansion of magmatic volatiles, but also by energetic interaction between magma and external water (e.g. Houghton & Wilson, 1989). This complicated process is modelled as analogous
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
35
VSIC02 9/12/08 2:07 PM Page 36
36
A. Belousov and M. Belousova
deep-water eruptions are relatively quiet and frequently purely effusive (e.g. Fisher & Schmincke, 1984). An additional factor contributing to low explosivity of subaqueous eruptions, in comparison with subaerial phreatomagmatic events, could be excess water. It was demonstrated by experiments with mixtures of water and silicate melts (Wohletz & McQueen, 1984) that the strongest explosions occur at intermediate water/melt ratios (≈ 0.3). Both higher and lower ratios lead to less violent explosions. Shallow-water explosive eruptions (depth usually < 100 m) are akin to subaerial phreatomagmatic ones, but the ejection of pyroclasts through a layer of water adds many unique features to the eruptive and depositional processes. Shallow-water (especially sublacustrine) eruptions have been only poorly studied because of the rarity of direct observations. Most published observations of shallow-water explosions were made by Thorarinsson (1964) during the eruption of Surtsey (Iceland) in 1963. Similar ‘Surtseyan’ activity was witnessed also during the eruptions of Capelinhos, Azores, in 1957 (Machado et al., 1962; Waters & Fisher, 1971). Questions arising from the scarcity of direct observations of shallow-water eruptions have not been fully addressed by investigations of their deposits. Deposits of shallow-water eruptions are rare because of their rapid reworking by water waves and /or as a result of deep burial under deposits of subsequent subaerial volcanic activity. Thus few such deposits have been studied, and their processes of formation are the subject of debate (Heiken, 1971; Kokelaar, 1983; Cas et al., 1989; Sohn & Chough, 1992; White, 1996). The aim of this paper is to document the 1996 sublacustrine eruption in Karymskoye lake and resulting deposits, as well as deposits of similar eruptions that occurred in the lake in the past. From these data we infer mechanisms of magma fragmentation, and of the deposition of resulting pyroclasts, which operate during shallow-water eruptions of basic magma.
GEOGRAPHICAL AND GEOLOGICAL SETTING Kamchatka Peninsula makes up the northern part of the Kurile–Kamchatka volcanic arc. Volcanism of the arc results from westward subduction of the Pacific plate under the Eurasian plate. About 30 active volcanoes of Kamchatka are aligned along the eastern shore of the peninsula. Together with a number of extinct volcanoes and calderas, they form the Eastern Volcanic Belt of Kamchatka (Fedotov, 1991).
Karymskoye intracaldera lake is located in an uninhabited region of the Eastern Volcanic Belt, 125 km north-east of the town Petropavlovsk-Kamchatsky (Figs 1 & 2). The lake, 4 km across with a maximum depth of 70 m, contains ≈ 0.5 km3 of fresh water. The surface of the lake normally has an altitude of 624 m above sea-level (a.s.l.). From November to June the lake is covered by ice (up to 1 m thick) and snow (up to several metres thick). The steep shores of the lake are heavily vegetated by alder bushes with rare cedar bushes and birch trees. Before the eruption there were no well-developed beaches around the lake. The lake is fed by several streams and hot springs. In the north, the lake is drained by the Karymskaya river. The first and only pre-1996 depth sketch of the lake (Ya. D. Muravyev, personal communication) displays a simple bowl-like morphology of the bottom, complicated in the northern part of the lake by a broad shoal. Around the lake there are several late Pleistocene– Holocene stratovolcanoes, of basalt to rhyodacite composition (Ivanov, 1970). Eruptions of rhyolites in the region have been widespread, connected mostly with formation of calderas. Karymskoye lake fills one of themathe Akademiya Nauk caldera, which is enclosed in an older, larger caldera. Akademiya Nauk caldera is bounded on all sides by a pronounced escarpment 50–150 m high. The caldera was formed in late Pleistocene time; the fission-track age is 28–48 ka (Masurenkov, 1980) and younger eruptions inside the caldera were not known until recently (Belousov et al., 1997). Modern volcanic activity of the region is connected mostly with Karymsky stratovolcano, which is situated 6 km to the north inside its own Karymsky caldera, 7800 14C yr old (Braitseva & Melekestsev, 1990). This volcano erupts andesite and dacite and is one of the most active volcanoes in Kamchatka. In historic time (since 1771) more than 20 prolonged eruptions are known (Gushchenko, 1979). Before 1996 the volcano had been dormant since 1982. Akademiya Nauk caldera is crossed by a major north-north-east fault (Fig. 1) which is marked by a straight-lined drainage-system pattern (Belousov et al., 1997). Its most prominent expression is the straight deep canyon that was cut by the Karymskaya river into the rim of the caldera. Rare eruptions of basic magma have occurred along the fault in late Pleistocene–early Holocene time. They built several scoria cones and maars to the north of the caldera (Fig. 1) and also an underwater tuff ring inside the caldera, which was expressed as a shoal in the northern part of Karymskoye lake. The tuff ring was formed by two eruptions, which occurred closely one after another at 4800 14C yr bp (Belousov et al., 1997).
;
VSIC02 9/12/08 2:07 PM Page 37
37
Eruptions in Karymskoye lake, Russia Tuff ring
159 25'
N
Fall
Lahar
Fissures
Karymsky volcano
ymskay a
Ballistic
Ka r
N
Rive r
Surge
54 05'
1 km
Subsidence pits
A
Karymskaya River
20 40
60
54 00'
Karymskoye Lake
Karymskoye Lake
B
A
B
0 50m
(b)
53 55'
2 km
(a)
Calderas
Monogenetic edifices
Fault
Epicentre of 1.1.1996 earthquake
Fig. 2. Oblique view of the Karymskoye intracaldera lake and the 1996 tuff ring. View from the north-west. Karymsky stratovolcano is to the left, just beyond the frame.
Fig. 1. (a) Location of Karymskoye lake and Karymsky volcano, with inset showing their location on Kamchatka peninsula. (b) Sketch map of deposits of the 1996 eruption. Boundaries are drawn schematically. Deposits of tsunamis are not shown. Line A–B shows the position of bathymetric profile (below) across the 1996 tuff ring. Dashed isobaths are drawn for 55 m and 65 m. A bump on the south slope of the 1996 tuff ring is probably a buried crater rim of the first eruption in the lake, dated at 4800 14C yr bp.
VSIC02 9/12/08 2:07 PM Page 38
38
A. Belousov and M. Belousova
The 1996 eruption in the Karymskoye lake occurred at the underwater part of the north-north-east fault, at the same site as the 4800 14C yr bp eruptions.
OBSERVATIONS OF THE 1996 ERUPTION After a strong earthquake on 1 January 1996 at 21.57 h local time (LT), there was an intense seismic swarm. Two explosive eruptions subsequently began, one after another, from two different and widely separated sources: one from the central vent of Karymsky stratovolcano and one from Karymskoye lake (Fedotov, 1998; Muravyev et al., 1998). Reports from pilots of commercial flights led to the conclusion that Karymsky stratovolcano began to erupt first, apparently early in the morning on 2 January. During the first day of the eruption, Karymsky stratovolcano emitted a steady ash plume, which drifted south-east at 200 –500 m above the summit crater (1700–2000 m a.s.l.). The brief and violent eruption in Karymskoye lake started several hours after the beginning of the eruption of Karymsky stratovolcano, at approximately 13.00 h LT. The only observation of the eruptive processes in the lake was made during a helicopter overflight at 15.20 –16.20 h LT (Muravyev et al., 1998). By the time of the overflight, the ice cover of the lake had melted, and an eruption of Surtseyan type was in progress from a vent 400 m off the northern shore. Initial water depth at the vent site was ≈ 40 –50 m. Underwater explosions occurred every 4 –12 min; in total about six explosions were observed with an average interval of 6 min. Between the explosions, water of the lake at the site of the eruption was vigorously splashing, bubbling and steaming. The explosions had different magnitudes. Analysis of videotape of the strongest explosion, recorded by V. Bahtiarov during the overflight, shows the following sequence of events (Fig. 3a). 1 The explosion initially produced a rapidly rising, dark grey, smooth-surfaced bulbous mass of expanding gas and pyroclasts, probably maintained by surface tension within a shell of water. Within several seconds the shell expanded up to 450 m high. The ejection angle of the material (between the lake surface and the outer margin of the expanding bulb) was about 60°. 2 Suddenly the shell became unstable and was pierced by multiple jets of a black mixture of pyroclasts with steam and water, of the type named ‘cock’s tail jets’ by
Thorarinsson et al. (1964). Although ejection angles of the jets varied from 40° to 90°, the strongest jets had angles > 60°. Maximum initial velocity of the jets is estimated as 110 m s–1. Simultaneously, a light grey ‘collar’ appeared around the epicentre of the explosion, which represented a nearly axially symmetric, bore-like elevation of the water surface (tsunami?) up to 130 m high, that propagated radially at about 40–20 m s–1. 3 Within ≈ 15 s from the beginning of the explosion and < 7 s after piercing the shell, the jets reached their maximum height of about 1 km, became white as a result of condensation of steam, and then collapsed back toward the lake to produce a ground-hugging base surge that moved radially from the stem of the eruptive column with a maximum velocity of about 20 m s–1 (Fig. 3b). 4 Within 48 s of the beginning of base-surge propagation, the surge travelled 600 m along the lake surface and decelerated; the average velocity of the surge was ≈ 12.5 m s–1. 5 At the end of its lateral propagation the surge strongly expanded, apparently lost most of its pyroclastic load, became lighter than the surrounding air and began to lift buoyantly. 6 Soon the surge cloud rose from the ground and ascended convectively upward. Other observed explosions were weaker than that described above. Most of them produced a single, vertical, column-like outburst of black gas–water– pyroclastic mixtures ≈ 100–150 m high (the type named ‘cypressoid jets’ by Thorarinsson et al., 1964). The columns collapsed back into the lake, producing only subtle base surges. Large concentric water waves (tsunamis) with wavelengths of ≈ 100 m were seen on the surface of the lake around the site of the eruption. Above the sites of explosions, white, pyroclastdepleted eruptive clouds convected to an altitude of ≈ 3 km and then slowly drifted downwind to the south-east. Tsunamis, generated by underwater explosions, periodically forced water from the lake into the canyon of Karymskaya river, forming pulsing lahars with a maximum discharge of 500 m3 s–1 (Muravyev et al., 1998). Four kilometres downstream from the headwaters, where the river emerges from its first deep canyon, a shallow temporary lake was formed because the next narrow canyon of the river was incapable of handling all of the lahar. By the time of the next overflight, at 11.00 h LT on 3 January, the eruption in Karymskoye lake had ceased. The precise time of the cessation is unknown because
VSIC02 9/12/08 2:07 PM Page 39
Eruptions in Karymskoye lake, Russia
39
(a)
(b) Fig. 3. Sublacustrine volcanic explosion in Karymskoye lake on 2 January 1996 at 15.42 h LT. View from the north-west. (a) Development of the explosion. Sketches drawn from the videotape. 1, Rapidly rising bulbous mass of black expanding gas and pyroclasts within a shell of water (inner contour). Within 3 s the shell has expanded to 450 m high (outer contour). White contour above is an ascending cloud from previous explosion. 2, The shell is pierced by multiple jets of a black mixture of pyroclasts, steam and water. A light grey collar around the epicentre of the explosion is believed to be a bore-like elevation of the water surface (tsunami?) up to 130 m high, that propagates radially. 3, In 7 s after piercing the shell, the jets reach their maximum height of ≈ 1 km and collapse back in to the lake to produce a base surge. 4, Within 48 s from the beginning of basesurge propagation, the surge has travelled 600 m along the lake surface. 5, The surge has lost most of its pyroclastic load, become lighter than the surrounding air, and begun to lift buoyantly. 6, The surge cloud has left the ground and convects upward. (b) Explosion in Karymskoye lake viewed from the NW. Collapse of the eruptive column has produced a base surge moving outward along the surface of the lake. Diameter of the base-surge cloud is 1.2 km. Photograph courtesy of Ya. D. Muravyev.
seismograms and barograms of the eruption were obscured by signals from the simultaneous eruption of Karymsky stratovolcano. The total duration of the eruption was 10 –20 h. Taking the average observed interval between the explosions as 6 min, we estimate that 100 –200 explosions in all occurred in the lake. By the end of the eruption, the pyroclastic deposits formed a new peninsula in the northern part of the lake and dammed the headwaters of Karymskaya river. This caused a 2.6 m elevation of water level in the lake. As a result of the eruption the pH of the lake
water had reached 3.2 and its temperature rose to 28°C (Fazlullin et al., 2000). All the fish in the lake died. On 15 May, the pyroclastic dam was breached and a second lahar rushed down the valley of Karymskaya river. A new temporary lake appeared again at the place of the previous one. The friable pyroclastic dam was quickly eroded and the level of the Karymskoye lake quickly fell to that of before the eruption. As a result, at the site of underwater explosions, a perfectly shaped tuff ring emerged from the lake water (Fig. 4a).
VSIC02 9/12/08 2:07 PM Page 40
40
A. Belousov and M. Belousova
(a)
(b)
(c)
(d) Fig. 4. (a) Tuff ring, with 600-m-diameter crater, formed by 1996 eruption. View from the north-east. (Note strong tsunami erosion on north shore.) The largest subsidence pit (site > 120 in Fig. 5) is visible on the surface of the ring (arrow). Eroded pyroclastic dam in headwaters of Karymskaya river is on the right edge of the picture. (b) Layered tuff ring deposit exposed in the deepest subsidence pit (site > 330 in Fig. 5). (Note large number of bombs littering the surface of the tuff ring.) The 1996 crater is to the right. (c) Distal facies of the 1996 tuff ring deposit (site 130 in Fig. 5). Scale is 80 cm long. Upper friable part is reworked deposits. The 1996 crater is to the left. (d) Parallel layering in the upper part of 1996 distal base-surge deposit. Scale is 25 cm long. Site 100 in Fig. 5.
After the cessation of the eruption in Karymskoye lake, the eruption of Karymsky stratovolcano continued until the end of April 2000.
TUFF RING MORPHOLOGY The eruption formed a partially submerged edifice (Figs 2 & 4a), which we refer to as a tuff ring because of its morphological similarity to classic tuff rings (Heiken, 1971). Northern parts of the tuff ring and most of its crater rim are now subaerial, but the crater itself and the southern ring slopes are under water. The highest subaerial parts of the ring lie only a few metres above the present level of the lake. The crater is 650 m across and 60 m deep with inner slopes dipping up to 12° (Ushakov & Fazlullin, 1998). The outer, southern slope of the tuff ring dips gently outward (up
to 10°), but the northern slope, which overlaps the pre-eruptive shoreline, is horizontal or even inclined inward towards the crater. The asymmetrical morphology is due to the formation of the edifice on an underwater slope, close to the shore of the lake. The northern, outer edge of the tuff ring is bounded by the steep wall of the caldera, where no deposition occurred (only strong erosion of poorly consolidated bedrock). Further northward from the crater, beyond the eroded caldera wall, deposition of tephra occurred, but the character of the deposit (which we refer to as a distal base-surge deposit) is different from that of the tuff ring deposit. On the surface of the tuff ring north-north-east of the 1996 crater there is a chain of eight pits (Figs 1b & 4a, b). Most of the pits are shallow depressions 3–5 m across, but one of them is 50 m in diameter with vertical walls 1.5 m high and a total depth of ≈ 10 m. Along
VSIC02 9/12/08 2:07 PM Page 41
Eruptions in Karymskoye lake, Russia trend from the chain of pits, to the north-north-east of the lake, a 2-km-long system of subparallel, vertical fissures was formed during the eruption (Fig. 1b). The fissures are widest close to the tuff ring (in the upper part up to 4 m wide) and narrow northward to a few millimetres. The fissures were formed as a result of west–east extension in the region (Leonov, 1998). Geodetic measurements show that the extension was 233 cm (Maguskin et al., 1998). Fissuring was connected with intrusion of a feeder dyke of the eruption ≈ 2.5 m wide (Fedotov, 1998), and we infer that the pits on the surface of the tuff ring were formed as a result of subsidence after withdrawal of magma from the feeder dyke at the end of the eruption. The radius of the base of the tuff ring is ≈ 0.8 km. Maximum height of the edifice is estimated at 40–50 m based on pre-eruptive bathymetry, and the bottom of the crater of the edifice lies close to the pre-eruptive surface, possibly slightly below it. The ratio of the height to the basal diameter of the ring is about 0.03, which is common for tuff rings (Heiken, 1971). The volume of the edifice (0.047 km3) was roughly calculated as the volume of a truncated cone (height 0.05 km, radii of the base and top 0.8 km and 0.32 km, respectively) with subtraction of the volume of the
41
crater (a cone with radius of the base 0.32 km and height 0.05 km). This volume is considered to be a maximum. A similar volume (0.04 km3) was calculated by Muravyev et al. (1998) using a water balance of the volume of the lake during the eruption.
STRATIGRAPHY AND CHARACTERISTICS OF THE 1996 PYROCLASTIC DEPOSITS Four types of ejecta were found: 1 proximal thick lapilli ash beds forming the tuff ring; 2 distal base-surge deposits; 3 deposits of distal co-surge fallout; 4 fields of bedrock blocks and juvenile bombs (ballistic material) (Fig. 1b). Tuff ring deposits The deposits forming the tuff ring were studied in the four currently available outcrops: in walls of the two subsidence pits and in the left and right banks of the Karymskaya river, where it breaches the pyroclastic dam (Figs 1b & 5). All the outcrops are arranged along
Fig. 5. Examples of grain-size histograms for the 1996 tuff ring, distal base-surge and co-surge fall deposits at various distances from the crater. Numbers on histograms are field number of site/number of layer (from bottom to top). Sketch map insert shows sampling sites: circles, tuff ring; triangles, distal base surge; square, co-surge fall. Filled symbols indicate sites for histograms shown in this figure. Sampling site of the most distal fall is not shown. Numbers on the map show thickness of deposits in centimetres.
VSIC02 9/12/08 2:07 PM Page 42
42
A. Belousov and M. Belousova
Fig. 6. Thickness of 1996 deposits versus distance from the centre of the crater. Maximum thickness of the tuff ring is estimated at 40 –50 m based on pre-eruptive depth of the lake at that location.
a straight line extending radially from the crater. In the three proximal outcrops (up to 750 m from the centre of the crater) only the upper part (2–3.5 m) of a deposit having a maximum thickness of 50 m is exposed. The deposit quickly wedges out (Fig. 6) and in the most distal outcrop (1 km from the crater) the full thickness of 1.3 m is exposed.
In the three proximal outcrops the upper part of the tuff ring is regularly medium- to thick-bedded, dark grey, fines-poor, grain-supported, basaltic ash lapilli and lapilli ash (Figs 4b, 5 & 7–9; Table 1). Dip of the beds is very gentle; in outcrop they appear subhorizontal. Bed boundaries are usually diffuse, and beds are planar and continuous. Each bed is 10– 60 cm thick, and no erosional discordances have been found. The beds appear to exhibit normal, reverse, or reverse-to-normal grading, but because many contacts between the beds are diffuse, the type of grading is difficult to determine with certainty. It appears that reverse grading is more common for the upper beds and in the lower beds normal grading is most prevalent. Some of the lowermost layers have pronounced separation of a lower lapilli layer from an upper ash layer, and represent bed couplets. The large lapilli are mostly irregular in shape and are angular to subrounded. No sag structures have been found under even the largest clasts. Grain-size distributions of the deposits are commonly unimodal with coarse- to medium-ash modes (4 to – 1φ; Figs 5 & 7, Table 1). A very low content of fines is typical of the deposits. In the upper parts of the sections, where the deposit is notably coarser, there is usually an additional lapilli mode (– 1 to – 6φ). The deposits are moderately to poorly sorted (Fig. 8).
Fig. 7. Section and grain-size data of the 1996 tuff ring deposit in the deepest subsidence pit (Fig. 4b and site 330 in Fig. 5).
VSIC02 9/12/08 2:07 PM Page 43
Eruptions in Karymskoye lake, Russia
43
Fig. 8. Relationship between sorting and median diameter (Inman coefficients) for the deposits of the first (4800 14C yr bp), second (4800 14C yr bp) and 1996 eruptions. Dashed line is ‘Surtseyan’ field after Walker & Croasdale (1972).
(a)
(b) Fig. 9. (a) Percentage of lapilli (> 2 mm), medium + coarse ash (0.063 –2 mm) and fine ash (< 0.063 mm) in the pyroclastic deposits of the first (4800 14C yr bp), second (4800 14C yr bp) and 1996 eruptions. Numbers 1, 2, 3, and 4 outline 1996 tuff ring deposits sampled at sites > 120, > 330, > 350, and 130, respectively, in Fig. 5. (b) Relationship between F1 (fraction < 1 mm) and F2 (fraction < 1/16 mm) in deposits of the 1996, and first (I) and second (II) 4800 14C yr bp eruptions in Karymskoye lake. (Note that F2 axis has a break showing very fine-grained fall deposits of the second eruption.)
Grain-size characteristics of the deposit from proximal outcrops are very similar; upper layers of the deposit commonly are more poorly sorted than lower ones. On a plot of lapilli/medium + coarse ash/fine ash (Fig. 9a) there is some inconsistent tendency for deposits to become less coarse with distance from the crater. The studied material was deposited near the water–air interface. We infer that normal grading and bed couplets of the lower beds may result from deposition in a subaquatic environment.
Each bed of the tuff ring is believed to represent the deposit of a single explosion. Taking nominal deposit thickness to be 50 m, and average proximal bed thickness to be 25 cm, it can be estimated that 200 explosions occurred, averaging ≈ 2 × 105 m3 of tephra per explosion. This is, of course, a very rough estimation, based on extrapolation of stratigraphy of the upper part of the tuff ring to the unexposed 90% of the deposit. But it coincides with the previous estimation (100–200 explosions), based on total duration of the
VSIC02 9/12/08 2:07 PM Page 44
44
A. Belousov and M. Belousova
Table 1. Granulometric characteristics of clastic deposits of eruptions in Karymskoye lake Number of samples 1996 tuff ring
24
1996 distal surge
15
1996 distal fallout
2
1996 tsunami 1996 lahar Second eruption: surge
15 3 12
Second eruption: fall
3
First eruption: tuff ring
5
Sorting (φ) 1.5–3.2 2.1 1.4–2.7 2.1 1.7–1.9 1.8 0.7–2.1 1.2 0.8–2.3 1.3 1.5–3.1 1.9 0.9–1.2 1.0 1.4–1.9 1.7
Median diameter (φ) –3.9 – 0.6 –0.9 –1.7–2.2 0.4 1.2–2.0 1.6 –1.1–2.4 0.7 –1.4–3.6 1.7 –1.9–1.3 0.1 3.8 – 4.2 4.0 –2.7– (–2.5) –2.6
Lapilli (%)
Coarse + medium ash (%)
F1 (%)
F2 (%)
20.0–87.0 44.5 2.9– 47.7 23.8 1.3–9.8 5.6 0–52.2 15.3 0–56.4 18.8 3.0–63.3 26.6 0.1–0.6 0.4 76.5–82.2 79.0
13.0–79.5 55.0 38.9–79.6 68.0 79.2–83.6 81.4 46.1–99.1 83.2 43.6–90.9 68.8 36.7–92.6 70.1 40.0–65.5 49.8 17.8–22.9 20.5
6.8–60.9 36.6 27.2–84.0 56.4 73.8–93.1 83.4 22.6–100 68.7 23.5–100 74.5 20.3–82.7 54.7 97.7–99.3 98.7 6.6–12.3 9.2
0–1.5 0.5 2.6–13.9 6.6 11.0–15.2 13.1 0.3–3.3 1.4 0–28.1 12.4 0.1–6.5 3.3 34.4–59.6 49.8 0.1–1.2 0.5
Numbers on the first line are the lowest and highest values; those on the second line give the average value. Sorting and median diameter after Inman (1952); lapilli, < − 1φ (> 2 mm); coarse and medium ash, from − 1φ to + 4φ (0.063–2 mm); F1, > 0φ (< 1 mm); F2, > 4φ (< 0.063 mm), φ = − log2(diameter in millimetres).
eruption and average interval between explosions, and thus seems reasonable. The deposit in the most distal outcrop (Fig. 4c) is different from that described above. It has smaller median diameter and better sorting. The deposit is mostly ash with discontinuous trains of coarse lapilli. Unlike the proximal deposit, this deposit lacks welldeveloped parallel bedding. Instead it has low-angle, ill-defined layering with distinct lateral grain-size variations. Laterally this deposit transforms into deposits of distal base surges. The characteristics of the tuff ring deposit are not entirely consistent with any of the depositional processes previously known to form tuff rings. Moderate sorting and lack of fines makes the deposit similar to proximal fall deposits of dry hydroclastic eruptions (e.g. LT1 facies of Sohn & Chough, 1989), but observations and videotape show that the eruption in Karymskoye lake was very wet. Deposition directly from falling tephra-finger jets of Surtseyan eruption results in poorly sorted, fines-rich deposits (e.g. facies C of Sohn & Chough, 1992) because aerodynamic sorting in the jets is minimal as a result of cohesion of wet tephra, high particle concentration and rapid emplacement. Absence of bomb sags also suggests that the tuff ring beds are not of fall origin. Moreover, proximal fall deposits of hydroclastic eruptions commonly form tuff cones with steeply dipping beds (Sohn & Chough, 1992), which is not the case for Karymskoye. The very gentle dip of the 1996 tuff ring
beds, their rather consistent thickness over large distances, and their grain-size characteristics support an origin by deposition from pyroclastic density currents. Among gas–pyroclastic density currents only some extremely energetic dry surges can deposit material strongly depleted in fines (e.g. layer B of the 1956 directed blast deposits of Bezymianny volcano; Belousov, 1996). But surges of the 1996 eruption were wet, poorly inflated and relatively weak; if they had deposited the tuff ring, the deposit would be fines rich (like, e.g. LT3 facies of Sohn & Chough, 1989). We suggest that the studied tuff ring deposit was formed by multiple (periodic) flows of a hyperconcentrated water–pyroclastic mixture, which moved radially from the crater area (Smith, 1986). The absence of scour and channelling indicates that the flows were depositional and unconfined. It is likely that the flows had different concentrations of solid particles; the more dilute flows left normally graded layers, whereas the most concentrated flows show reverse grading formed by basal shear and grain-dispersive forces (Lowe, 1982). In the distal zone the deposits of the tuff ring were formed by relatively dilute flows, which lost much of their pyroclastic load in the proximal zone. Origin of the flows is attributed to very wet base surges of the eruption. We suggest that the flows existed as peculiar water–pyroclastic underflows in the basal parts of the surges. Flows of similar origin were probably witnessed during some explosions of Surtsey (Thorarinsson, 1964; p. 43): ‘explosions were often
VSIC02 9/12/08 2:07 PM Page 45
Eruptions in Karymskoye lake, Russia followed by greyish-white cloud avalanches which rolled over the crater rims and, after the island had grown to some appreciable height, they sometimes spread a few hundred metres out over the sea. Sometimes the explosions ejected so much sea water over the crater rims that mud streams ran all the way down to the beach.’ Surface layer of the tuff ring The surface layer of the tuff ring is different from the deposits that form the upper (studied) part of the tuff ring. The layer is ≈ 1 m thick and strongly enriched in coarse lapilli and basaltic bombs up to 1 m across, which were ejected at the end of the eruption (Fig. 4b). The ‘Ballistic material’ section describes the bombs in greater detail. Aerial images of the tuff ring reveal that its surface has large-scale ripples in several places. The ripples are very gentle, almost flat patches of coarse lapilli and bombs, and areas between the patches are covered by ash. The ripples, with wavelengths of 4–6 m, are elongate parallel to the crater rim. They were formed at the end of the eruption only; ripples were found nowhere lower in the deposit. They were probably formed by tsunami waves that redistributed coarse-grained clasts on the surface of the ring (see below). The underwater surface of the tuff ring is mantled by a layer of pale grey fine vitric ash (Md 6.7φ; sorting 1.5φ) up to 30 cm thick. The layer was studied in several locations along the crater rim where it now lies onshore because the level of the Karymskoye lake fell after the eruption. This layer represents the most fine-grained pyroclastic products of the eruption, and it is inferred that when the eruption ceased, this material, along with fines eroded from the shores by the tsunami waves (see below), slowly settled in the lake to form the fine ash layer. Distal base-surge effects and deposits In the distal zone the surges attacked the steep northern shore of the lake. They sandblasted bushes as far as 1–1.3 km from the vent (Fig. 1b). Damaged bushes show that the surges climbed gradients of up to 40° on the caldera wall. The highest elevation reached by the surges is ≈ 150 m above the level of the lake. Close to the crater, all thin branches of the bushes were broken off and carried away by surges; only thick branches remain. The sides facing the crater were debarked and the wood was pitted. Most of the branches show no signs of scorching, but a few are slightly browned.
45
Hence we estimate the temperature of the surges to have been < 200°C (Banks & Hoblitt, 1981). Most of the damaged bushes died but a few resumed their growth after the eruption. The relatively low temperature of the surges explains the preservation of a thick layer of snow under the surge deposits. The presence of the snow also shows that the surges did not erode the underlying surface substantially. Near the limits of their propagation, the surges dried and broke upper thin branches of the bushes. The lower branches were not damaged at all, probably because they were protected by deep snow. Pyroclastic deposits in the area of damaged bushes are notably different from the deposits of the tuff ring. Among the bushes deposits are moderately sorted dark grey basaltic lapilli ash with abundant fines (Figs 4d, 5, 8 & 9; Table 1). The deposit comprises a sequence of two to three major, grain-supported parallel beds, each 10–35 cm thick, and several minor ones, each < 1 cm thick (Fig. 4d). Bed boundaries are usually well defined. The beds are massive or with poorly developed planar or low-angle undulatory lamination. The lowermost layer of the deposits in places contains a considerable number of uncharred fragments of plants, but in upper sections these are rare. In contrast to those of the tuff ring deposit, the deposits of distal surges are very friable. Grain size of the deposit is commonly unimodal (between – 1φ and 2φ), sometimes with an additional poorly developed finegrained mode (> 4φ). The maximum total thickness of the deposit is 1 m, but this very quickly decreases both on steep slopes and with distance from the crater (Fig. 6). The area covered by the surge deposit onshore is 0.4 km2 and its volume is ≈ 105 m3. Laterally the base-surge deposits are replaced by deposits of distal co-surge fallout. The surges’ effects and deposits show that the base surges were relatively weak, cold, and non-erosive in distal areas. Pyroclastic particles from the surges were deposited ‘grain by grain’, but with little or no traction before deposition. Distal co-surge fallout The fall deposit extends from the 1996 tuff ring to the south-east (Fig. 1b). Everywhere, the dark grey, basaltic fall of the sub-lacustrine eruption covers 1-cm-thick, light grey, andesitic ash of the first explosions of Karymsky stratovolcano. The maximum thickness of the co-surge fall deposit onshore is 5 cm at the place where the base-surge deposit transforms into the fall deposit. The transformation is marked by
VSIC02 9/12/08 2:07 PM Page 46
46
A. Belousov and M. Belousova
a disappearance of both internal lamination, characteristic of the surge deposit, and damage to bushes, which indicates high-speed lateral transport of pyroclastic particles. The thickness of the fallout deposit declines rapidly with distance (Fig. 6): on the southern shore of the lake it is only 1 cm thick. The approximate volume of the fall deposit is 0.001 km3. The fall deposit is massive, moderately sorted medium–coarse ash with a small admixture of fine ash and lapilli (Figs 5, 8 & 9; Table 1). Close to the vent, grain-size characteristics of the deposit are similar to those of the most fine-grained surge deposits. Further from the source, the fall deposit becomes finer grained and better sorted. In the fall deposit there are also a few centimetre-sized lapilli, though they do not influence the grain-size data because of their rarity. On the south-east shore of the lake, where the fall deposit consists of medium-grained ash, the diameter of the 10 largest clasts picked up in an area of ≈ 0.2 km2 ranges from 18 to 30 mm (average 21 mm). The lateral transition of the surge deposit into the fallout deposit, the similarity of grain-size distributions of the fall and the most fine-grained surge deposits, and observations of the eruption all suggest that convectively buoyant clouds of decelerated surges were the main source of the fall deposit. The carrying capacity of such clouds was very low, which explains the small volume of the resulting co-surge fallout. Although accretionary lapilli are common in phreatomagmatic deposits (Fisher & Schmincke, 1984; Schumacher & Schmincke, 1991), none were found in the surge or fall deposits. The reason could be that there was excess water in the eruption clouds and base surges (Schumacher & Schmincke, 1995). An alternative explanation is suggested by the very low air temperature during the eruption (< – 10°C). Steam droplets in the eruption cloud may have frozen immediately upon condensation to form ice, which would have prevented formation of accretionary lapilli. Several observations of the fall deposit soon after the eruption (Ya. D. Muravyev & S. M. Fazlullin, personal communication) indicate that initially the deposits were a mixture of pyroclastic particles and grains of dirty ice. Ballistic material We estimate the volume of ballistic material to be < 2% of the total erupted volume. The most distal ballistic material fell 1.3 km from the centre of the 1996 crater (≈ 1 km from the crater rim) (Fig. 1b). We distinguish four types of ballistic material: bombs of juvenile basalt (dominant); blocks of hydrothermally
altered breccia (abundant); blocks of ice (some); and bombs of remelted old rhyolite (very few). Bombs of juvenile basalt constitute > 90% of all ballistic material. They densely cover a significant part of the surface of the tuff ring (Fig. 4b) and are loosely scattered beyond its limits. In sections of the tuff ring, bombs are absent, suggesting that they were ejected only at the end of the eruption. The bombs have irregular, but approximately equidimensional shapes. Most have a scoriaceous outer layer several centimetres thick, whereas their interior is commonly poorly vesiculated, sometimes with inclusions of centimetresized xenoliths of country rocks. Such an internal structure is unusual. In most cases the outer layer of a bomb quickly solidifies, forming a glassy crust (Francis, 1993), and the internal parts solidify more slowly, providing more time for vesiculation, which breaks the outer glassy crust and forms bread-crusted bombs. The dense cores of the 1996 bombs consist of the same fresh basalt as the vesiculated outer part. We term them ‘scoria-crust bombs’, and contrast them with cored bombs, in which an accidental clast forms a core enclosed within freshly vesiculated lava. This peculiar structure, the presence of xenoliths of country rocks in the bombs, and deposition at the end of the eruption, suggest that the 1996 bombs represent solidified magma from the conduit wall, which was ripped off and enveloped by fresh magma. The previously congealed cores were unable to vesiculate, but the coating of fresh melt vesiculated strongly, thus producing the scoriaceous crust. Some of the basaltic bombs of the eruption are of cauliflower type, typical of phreatomagmatic eruptions (Fisher & Schmincke, 1984). The diameter of the basaltic bombs ranges from 10 to 60 cm, with a few up to 1 m. The largest bombs occur at intermediate distances from the vent (Fig. 10), but this distribution may not be original, because most bombs in the proximal area (on the surface of the tuff ring) are inferred to have been redistributed and broken by the tsunami waves. Thus the basaltic bombs may have originally decreased in size with distance from the vent. Such wave reworking could also be responsible for the absence of impact structures on the surface of the ring. Beyond the tuff ring (onshore) basaltic bombs left impact craters up to 1 m across. Some bombs penetrated a snow pack > 1 m thick and up to 0.7 m of frozen soil. Bushes broken and uprooted by the projectiles show no signs of scorching, showing that the bomb surfaces were not incandescent at the moment of impact.
VSIC02 9/12/08 2:07 PM Page 47
47
Eruptions in Karymskoye lake, Russia
not produce any impact structures, and, having an average density less than that of water, they probably initially fell into the lake and then floated for some time so that their present distribution is not original. Composition
Fig. 10. Maximum size of ballistic material from 1996 eruption versus distance from the crater. Blocks are hydrothermally altered country rock; bombs are juvenile basalt.
Blocks of hydrothermally altered breccia constitute < 10% of all ballistic material. The diameter of these blocks typically ranges from 10 cm to 2 m, with a few up to 3 m. The largest blocks were ejected to a maximum distance of 1.3 km (Fig. 10), where they produced impact craters up to 8 m wide and 2 m deep. Several lines of evidence suggest that these blocks were ejected by the first, vent-clearing explosion(s) of the eruption: 1 the blocks are distributed mostly beyond the limits of the tuff ring and are extremely rare on its surface and inside the tuff ring deposits — it is likely that in the proximal zone blocks are buried under the tuff ring deposit; 2 the blocks show no signs of contact with basaltic melt; 3 there is no basaltic ash of the eruption under the soil that was ejected from the impact craters. In summer 1996, when we investigated the deposits, we discovered that among the impact craters formed at the beginning of the eruption there are several craters (up to 1.8 m across and 2 m deep) that do not contain (confirmed by digging) any ‘coupled’ ballistic material, and that there are no blocks nearby that could have formed the craters and bounced away. We infer that these ‘empty’ craters were formed by large blocks of ice that later melted. Several impact craters containing large blocks of ice were observed immediately after the eruption (Ya. D. Muravyev, personal communication). The ice blocks represent fragments of an ice layer that covered the lake before the eruption. Several exotic bombs of pumice-like remelted old rhyolite contained in and/or mixed with juvenile basalt also occur. These bombs, up to 2 m across, are concentrated along the shoreline of the lake. They did
Material ejected by the 1996 eruption is composed of juvenile basalt, remelted rhyolite, and other accidental clasts. Petrography of the 1996 products was studied in detail by Grib (1998). More than 95% of the ejected material is juvenile calc-alkaline basalt (52–53% SiO2; Table 2). A small amount of white, pumice-like rhyolite (Table 2), which either forms tiny inclusions in juvenile basalt or comprises rare volcanic bombs coated and/or mixed with juvenile basalt, is interpreted as recycled materialaold rocks, remelted and vesiculated as a result of contact with hot basaltic magma. The older rocks were probably obsidian or obsidian tuff entrained by the ascending magma from the walls of the feeding fissure where it dissected rocks mantling the caldera floor. Obsidian clasts with the same composition as that of the rhyolitic bombs are common in late Pleistocene pyroclastic flow deposits formed by eruptions during the formation of Akademiya Nauk caldera (Table 2), and outcrops of the obsidianbearing pyroclastic flows are abundant along the northern shore of the lake. Our experiments with heating Table 2. Major-element analyses of ejecta from eruptions in Karymskoye lake
SiO2 TiO2 Al2O3 Fe2O3 FeO MnO MgO CaO Na2O K2O H2O P2O5 Sum
1
2
3
4
5
6
53.00 0.86 19.98 1.74 5.83 0.12 4.20 10.94 2.70 0.62 0.20 0.16 100.35
75.71 0.19 13.17 0.06 1.29 0.04 1.20 1.50 3.44 2.19 0.74 0.08 99.61
75.86 0.18 13.30 0.31 2.16 0.05 0.24 1.26 4.11 1.97 0.38 0.06 99.88
52.52 1.05 17.88 2.09 7.07 0.15 5.88 10.36 2.35 0.68 0.16 0.13 100.32
55.70 0.84 16.63 1.62 6.07 0.15 5.28 10.14 2.76 0.93 0.10 0.17 100.39
53.00 0.59 16.94 2.37 7.37 0.13 5.76 10.80 2.35 0.68 0.24 0.16 100.39
Analysed by L. A. Kartasheva, Institute of Volcanology, Petropavlovsk-Kamchatsky using wet chemistry methods. 1, Basaltic bomb of the 1996 eruption; 2, rhyolitic bomb of the 1996 eruption; 3, obsidian from pyroclastic flows erupted in the range 28 – 48 ka during formation of the Akademiya Nauk caldera; 4, 4800 14C yr bp surge deposits of second eruption; 5, 4800 14C yr bp bomb of second eruption; 6, 4800 14C yr bp tuff ring deposits of first eruption.
VSIC02 9/12/08 2:07 PM Page 48
48
A. Belousov and M. Belousova
of the obsidian have shown that at 850°C it melts and strongly froths. The resultant perlite resembles material of the rhyolitic bombs ejected by the eruption. ‘Pumice’ of similar origin, generated by heating of old rocks, was reported from eruptions of Ruapehu volcano in 1975 (Nairn et al., 1979) and from a submarine eruption off Izu Peninsula, Japan, in 1989 (Yamamoto et al., 1991). Blocks of greenish hydrothermally altered breccia are dominant among the other accidental clasts ejected by the eruption. They probably represent a tectonic breccia of the north-north-east fault along which the feeding dyke of the eruption was intruded. Vesicularity and morphology of juvenile clasts Vesicularity of the basaltic clasts from the 1996 eruption was determined according to the procedure developed by Houghton & Wilson (1989) and Hoblitt & Harmon (1993). Combined data from three samples, of 30 clasts each, collected from the tuff ring deposit (one sample) and distal base-surge deposit (two samples) are presented in Fig. 11. The deposits have vesicularity indices of 7– 63% (mean 34%). This range is appreciably larger than in most Strombolian deposits (Houghton & Wilson, 1989), thus indicating fragmentation of a variably vesiculated magma rather than wholly ‘magmatic’ fragmentation simply by bursting of bubbles. Scanning electron microscope images of sand-sized particles of juvenile basalt display different clast shapes and degrees of vesiculation (Fig. 12). Poorly vesiculated particles commonly have small (5 –90 µm), spherical vesicles that are isolated from one another by thick walls. Such particles commonly have an equant, blocky morphology, and are bounded by planar surfaces that cut vesicles (Fig. 12a). The surfaces have peculiar chip-marks, and some are intersected by thin, curved cracks (Fig. 12b & c). The character of these surfaces shows that the vesiculated glass was brittle during fragmentation. Some particles have large, irregular vesicles up to 0.3 mm, which were formed by coalescence of smaller bubbles (Fig. 12d) and can be partially bounded by smooth, curved surfaces, which were originally the inner surfaces of those large vesicles. Sometimes such smooth surfaces can form almost the whole surface of a particle (Fig. 12e). In this case the particle shape reflects mostly surface-tension moulding of liquid melt. Highly vesiculated particles have multiple, irregular bubbles of different sizes, which are separated by thin walls (Fig. 12f ). The walls are frequently broken and the bubbles are intercon-
Fig. 11. Density–vesicularity histograms for juvenile lapilli collected from deposits of the first (60 clasts), second (60 clasts) and third (90 clasts) phreatomagmatic eruptions in Karymskoye lake. Measured clasts are 1–3 cm across. Vesicularity values are derived using a dense rock equivalent value of 2.8 g cm–3.
nected. The broken walls sometimes show features of plastic behaviour where gas burst from one bubble to another towards the surface of the grain (Fig. 12g). Overall the shape of such particles is scoriaceous, determined mostly by processes of magma vesiculation. A few particles are abraded and notably rounded (Fig. 12h). Ash grains typically have micron-sized flakes adhering to their surfaces. Such flakes commonly have a blade-like shape with sharp edges. They were formed during brittle fragmentation of basaltic glass (Fig. 12b & c). Some of the flakes are demonstrably in situ, because they are not completely separated from the parent grains. The surface morphology of the 1996 juvenile particles is typical of products of phreatomagmatic basaltic
VSIC02 9/12/08 2:07 PM Page 49
Eruptions in Karymskoye lake, Russia
49
(a)
(b)
(c)
(d)
(e)
(f) Fig. 12. Scanning electron microscope images of surfaces of sand-sized juvenile basaltic particles from the 1996 tuff ring deposit (a–h) and surge deposit of the second 4800 14C yr bp eruption (i). (a) Blocky, poorly vesiculated particle formed by fragmentation of already brittle basalt as a result of water–magma interaction. (b, c) Small-scale surface features of particle in (a). Blade-like flakes of glass and marks of surface cracking and chipping formed during brittle disruption of basaltic glass. Some of the flakes were not completely separated from the parent surface. (d) Particle with large, irregular vesicles up to 0.3 mm formed by coalescence of smaller bubbles. (e) Particle of poorly vesiculated basalt formed by both forces of surface tension of liquid melt and subsequent brittle fragmentation. Probably represents droplets of basalt injected into water as a result of low-energy water–magma interaction. (f ) Highly vesiculated scoriaceous particle with multiple, irregularly interconnected bubbles of different sizes. Particles shown in (d) and (f ) were shaped mostly by vesiculation of fluid basaltic magma. (continued )
VSIC02 9/12/08 2:07 PM Page 50
50
A. Belousov and M. Belousova
(g)
(h)
Fig. 12. (continued ) (g) Small-scale feature of interbubble wall of particle illustrated in (f ) shows plastic behaviour of material of the wall when gas bursts from one bubble to another. (h) Poorly vesiculated particle, subrounded by multiple recycling in the crater. (i) Vesicles clogged by fines, a characteristic feature of deposits of the second 4800 14C yr bp eruption in the lake.
(i)
eruptions, in which fragmentation of liquid magma occurs as a result of a complex interplay between vesiculation and water–magma interaction (Heiken & Wohletz, 1986). We use here the term water–magma interaction in a broad sense, as any process of magma fragmentation induced by its mixture with water, including the mechanism proposed by Kokelaar (1983). The resulting morphology of some particles was overprinted by subsequent rounding during recycling in the crater (Houghton & Smith, 1993).
EFFECTS AND DEPOSITS OF TSUNAMIS AND LAHARS Tsunamis An area affected by tsunamis is marked around the lake by strong erosion of the shore, water-damaged bushes and distinctive deposits. The degree of shore
erosion declines with distance from the tuff ring, with the strongest erosion on the steep northern shore of the lake adjacent to the tuff ring. The slope was eroded up to 30 m above the level of the lake, with all plants and soil as much as 1.5 m thick stripped so as to expose poorly consolidated bedrock (Fig. 4a). No deposits of the eruption were found in the eroded area, other than numerous ballistic blocks pressed deeply into the slope. We infer that the shore here was eroded by both tsunamis and water-rich base surges moving outward from the crater. The tuff ring deposits adjacent to the base of the eroded slope are interbedded with layers, 10–20 cm thick, that are rich in clasts of yellow, altered rock, and pieces of plants eroded from the slope. Such layers are interpreted as backwash deposits of the tsunamis (Nishimura & Miyaji, 1995) and/or of water–pyroclastic flows formed from waterrich base surges. Along the rest of the shoreline the action of the multiple tsunamis in many places formed new beaches
VSIC02 9/12/08 2:07 PM Page 51
Eruptions in Karymskoye lake, Russia
51
(a)
(b) Fig. 13. (a) Shore of the lake affected by the 1996 tsunamis. Site with runup 5.8 m in Fig. 14a. The beach (A), cliff (B) and area with stripped vegetation (C) were formed by tsunami erosion. Limit of the tsunami invasion marked by dry bushes (D). (See text for details.) (b) Rampart formed by bushes cast ashore by tsunamis in the Karymskoye lake during the 1996 eruption. South-east shore of the lake; site with runup 3.8 m in Fig. 14a. Separate branches in the foreground mark the limit of tsunami inundation. Cone of Karymsky stratovolcano is visible in the background.
up to 50 m wide (area A in Fig. 13a), which are terminated by new cliffs up to 2–3 m high (area B). Further from the lake (inland, behind the cliff ) is an area C, up to 50 m wide, where the upper layer of soil up to 1.5 m thick was eroded and all plants were washed back into the lake. Along the outer boundary of the area inundated by the tsunamis (area D), soil was not eroded and bushes with some broken branches remained standing, but dead. It is inferred that these were killed by the warm, acidic lake water. Later they dried and serve as good markers of the inundation limit. The width of area D ranges from 1 m to 100 m. The character of the damage to the bushes in area D differs from that caused by the base surges; they are not debarked on the side facing the crater. Deposits left by the tsunamis can be divided into two main classes: those composed of non-floating material, which was deposited near the place where it was mobilized, and those representing floating objects, which before final deposition could drift considerable distances. Two types of deposits of the first class were discovered: individual blocks of poorly consolidated old tuff, and patches of sand. The blocks, 0.1–4 m across, originate from newly eroded cliffs. They are abundant on the north-east shore of the lake, and are almost entirely absent on the southern shore, which is composed mostly of hard, unerodible rocks. Blocks are scattered on the surface of the beaches or incorporated into the beach deposits below the cliffs (area A). Some blocks with volumes up to 1.3 m3 were transported by waves from their source up to 60 m inland (into areas C and D). Patches of sand are common in area
D, and occur locally in area C. The patches are usually metres to tens of metres across and up to 35 cm thick. The sand, with scattered pebbles, fragments of plants and clots of soil, is well sorted (Table 1). Sometimes up to four parallel layers, with thickness 2–6 cm, can be distinguished, and are inferred to reflect the multiple waves of tsunami runup and backwash. The composition and grain size of the deposit reflects the material mobilized by the tsunami. In the zone near the crater the deposit comprises mostly fresh basaltic pyroclasts from the tuff ring. Further from the crater the deposit is composed of sand from pre-eruption beaches. Deposits of floated material are represented in areas C and D by scattered well-rounded pebbles of pumice, 2–10 cm in diameter, and branches of bushes, which are orientated mostly perpendicular to the direction of tsunami runup. Deposits are most abundant on the south-east shore, because of the north-west wind that blew during the eruption. Here pumice pebbles form patches and bands up to 3 m across and up to 20 cm thick, with openwork fabric. These resemble deposits of the much larger-scale tsunamis that accompanied the 1883 Krakatau eruption (Carey et al., 1996). In contrast to that eruption, pumice deposited by the 1996 tsunamis was not fresh, but was eroded by the tsunamis from old pyroclastic deposits. In both cases the floating pumice formed ‘pumice rafts’, which were cast ashore. In Karymskoye lake the rafts also contained abundant floating bushes. After deposition on the south-east shore, interlaced branches of bushes formed ‘wooden ramparts’ up to 2 m high, 10–15 m wide and several hundred metres long (Fig. 13b).
VSIC02 9/12/08 2:07 PM Page 52
52
A. Belousov and M. Belousova
(b)
Fig. 14. (a) Heights of tsunami runup (in metres) around Karymskoye lake and bathymetry contours (in metres) after the eruption. (b) Tsunami runup versus distance from the crater. (Note change in trend ≈ 1.3 km from the crater.)
(a)
Branches of the ramparts are completely debarked with rounded or splintered, brush-like ends, which is inferred to indicate their prolonged beating by tsunami waves before final deposition. There are two lines of evidence suggesting that the largest of the tsunamis occurred at the end of the eruption. First, the areas eroded by tsunamis are not covered by fall deposits of the eruption. Second, the composition of floated material deposited on the south-eastern shore shows that it originated mostly from the northern shore. Given the size of the lake, several hours during the eruption were needed for the floating material washed off the northern shore to drift across the lake to be deposited on the southeastern shore. The distinctive band of devastation, and the tsunami deposits, allowed us to measure the runup height of the tsunamis at 24 points around the lake (Fig. 14). Because there were several tsunamis, the measurements represent the runup of the strongest event(s). The highest runup (20 –30 m) occurs on the shore immediately adjacent to the tuff ring, 700 m from the centre of the crater. For the proximal zone, to radial distances (r) up to 1.3 km, the runup height (R) shows rapid attenuation with distance as log R = – 0.56 log r + 5.8. For the distal zone, r > 1.3 km, R decays more slowly as log R = – 1.98 log r + 16.3. For the most distant points, runup was 2–3 m. ‘Tsunami’ is a Japanese term that usually describes extraordinary water waves generated in seas or large
lakes by earthquakes, volcanic eruptions, landslides and so on (Latter, 1981). The fast decay of the runup of Karymskoye’s tsunamis in the proximal zone is inferred to indicate that they were not only water waves, but also flows of water and/or water–pyroclastic mixtures. In the distal zone, however, evidence indicates that they were conventional tsunami waves (Belousov et al., 2000). Lahars Lahars left a broad fan of deposits where the Karymskaya river emerges from its first deep canyon in the caldera rim (Fig. 1b). The fan is 0.5 km long, 0.25 km2 in area, and 2–3 m in maximum thickness. It thins over a short distance and has a volume of 2.5 × 105 m3. The fan covered a marsh, and although most of bushes of the marsh were carried away or buried by lahars, some bushes remained standing along the outer boundary of the fan. Most bushes affected by lahars were killed, probably by acidic water. The stratigraphy of the fan was studied near its outer limit, where the deposit is thin and fine grained. The section consists of pre-eruption peat covered by two dark grey sandy layers, each ≈ 50 cm thick and capped by a finegrained light grey muddy horizon, 5 cm thick. The sandy layers are well sorted (Table 1), and have parallel or low-angle lamination defined by alternating relatively coarse- and fine-grained discontinuous laminae a few millimetres thick. The layers consist of
VSIC02 9/12/08 2:07 PM Page 53
Eruptions in Karymskoye lake, Russia redeposited products of the 1996 eruption with admixed yellow, altered pumice eroded from older pyroclastic deposits. Observations suggest that the sandy layers were deposited by the first (2 January) and second (15 May) lahars of the eruption. The muddy layers settled out from temporary lakes that formed with each lahar event. The grain size and bedding characteristic of the deposits shows that both lahars had relatively low concentrations of solid particles and were hyperconcentrated flood flows (Smith, 1986).
SUCCESSION OF EVENTS AND INTERPRETATION OF THE 1996 ERUPTION PROCESSES Ascent of magma The 1996 crater, the chain of subsidence pits and the system of fissures are all aligned along a north-northeast fault (Fig. 1); it is inferred that the feeder dyke of the 1996 eruption was intruded along the fault, as has occurred here during previous eruptions of basic magma (see below). The strong earthquake that occurred at the beginning of the pre-eruption swarm on the southern end of the fault probably was of tectonic origin, but it opened a path for magma to travel from a deep basaltic chamber toward the ground surface under the lake. There is petrological evidence that part of the basaltic magma from the dyke also intruded into the silicic magma chamber of the Karymsky stratovolcano and thus provoked its eruption (Eichelberger & Izbekov, 2000). Vent-clearing explosion(s) The pre-eruption seismic swarm was associated with intrusion of a basaltic dyke ≈ 2.5 m wide along the fault zone, which is marked by hydrothermally altered breccia. A wave of elevated pore pressure propagated through the country rock ahead of the intruding dyke (Delaney, 1982; Elsworth & Voight, 1992). As the dyke approached the ground surface, opening of an eruption fissure initially allowed rapid decompression of the country rocks in which pore pressure was elevated. Decompression-fragmentation of the country rock along the fissure may have occurred by mechanisms analogous to those proposed for fragmentation of viscous magma under rapid decompression (Alidibirov & Dingwell, 2000). The resulting fragments were transported upward by a gas stream
53
emanating from the fissure, which probably originated from both boiling of decompressed hydrothermal fluids (circulating in the fault and heated by dyke intrusion) and upward migration of magmatic volatiles from the degassing, ascending dyke. The combination of the above events led to the first phreatic, vent-clearing explosion(s) of the eruption; no fresh magma was involved. Fragments of country rock probably were carried out from relatively deep levels of the fissure, having been accelerated to high velocities; the largest fragments were ejected the greatest distances (e.g. Self et al., 1980). The phreatic explosion penetrated ≈ 50 m of water and a 1-m-thick ice sheet on the surface of the lake. Some of the ice blocks were incorporated, and together with ejected blocks of country rock left the largest impact craters. There is no evidence that this powerful explosion generated strong water waves; the ice sheet probably prevented a tsunami. The removal of rock fragments by this explosion widened the fissure, making a channel through which the following eruption occurred. Main stage of the eruption: Surtseyan activity The eruption of basic magma apparently started soon after the first vent-clearing explosion(s). The character of the basaltic eruption was directly linked to the fact that the gas-charged magma, with an average discharge ≈ 106 kg s–1, had found its way to the ground surface under Karymskoye lake. The main part of the 1996 eruption consisted of a series of ≈ 100–200 Surtseyan explosions with an average interval of ≈ 6 min. Although the magnitude of the explosions varied, they probably each resulted from the same processes, and we will consider here a model for only one explosion ‘cycle’, or burst. Fragmentation of basaltic magma Between explosions, the crater of the tuff ring was filled with water, and the conduit throat was clogged by a mixture of water and pyroclasts left from the previous explosion burst. Magma lay deeper in the conduit, where there were no significant water–melt interactions; if such interactions had occurred, fragments of country rocks would be abundant in the deposit (Fisher & Schmincke, 1984), which is not the case. As indicated by the surface morphology and vesicularity indices of pyroclasts from the eruption, the explosions were caused both by vesiculation of magma and by energetic water–magma interaction
VSIC02 9/12/08 2:07 PM Page 54
54
A. Belousov and M. Belousova
(Heiken & Wohletz, 1986; Houghton & Wilson, 1989; Zimanowski, 1998). During each explosion, the process of magma fragmentation occurred in two stages. The first was mostly magmatic: rapid growth and ascent of bubbles pushed out a portion of vesiculated and partially fragmented magma from the upper part of the vent; this resembles the early stage of a Strombolian outburst. The second stage was mostly hydrovolcanic. It occurred when the products of the first stage were injected into the water–pyroclastic mixture filling the vent and crater. Contact with water led to additional fragmentation of magma. As a result, the shape of the pyroclastic particles displays features of magmatic fragmentation overprinted by features of hydrovolcanic fragmentation. Ejection and deposition At the start of an eruption burst, a gas–pyroclast mixture expanded explosively from the vent, pushing out and bulging upward the water and water–pyroclastic slurry clogging the crater. The height of a typical bulge reached several hundred metres above the lake surface. After several seconds fingers of the gas–pyroclast mixture penetrated the upper part of the bulge and entered the atmosphere as a cock’s tail jet. Ejection of pyroclasts through the layer of water led to intermixing of jets with the water layer, forming an explosion fountain that was relatively cool, had a high content of water and steam, and was too heavy to rise buoyantly and form a sustained eruptive plume. Thus it began to descend towards the lake surface as an overloaded vertical eruption column collapsing under gravity (e.g. Wilson et al., 1980). The descending mixture formed a very wet base surge in which droplets of water, as well as pyroclasts, were a major component. We here term the flowing mixture a base surge, but add the modifier ‘water-rich’ to distinguish it from more dry, conventional, gas–pyroclast base surges. Water-rich base surge can be considered to be the end member of wet base surges. While the surge propagated radially from the crater, it became density stratified as a result of gravitational settling of pyroclasts and droplets of water (Valentine, 1987). We suggest that this process led to formation of a distinct underflow at the base of the surge. In this underflow water was the continuous phase in which pyroclastic particles and gas bubbles were suspended. Thus, the upper gas–pyroclastic surge was a transport system, which contributed material to the lower water– pyroclast underflow, which was the depositional system (Fisher, 1990). Deposition from the underflow
was similar to deposition from hyperconcentrated flood flow (Smith, 1986). Both surges and underflows moved subaerially along the surface of the tuff ring; where the surges began to climb the caldera wall, the water–pyroclastic underflow decelerated first and began to flow back downslope (topographic blocking of Fisher, 1990), eroding the steep slope of the caldera and depositing backwash material near the base of the wall. As a result of topographic blocking, the densest and wettest lower part of the surge was separated from the upper, relatively dry part of the base surge, which propagated further, leaving distal base-surge deposits similar to those formed by deposition from relatively low-energy surges (Belousov et al., 1998). The surge propagated until it lost most of its pyroclastic load, became lighter than the surrounding air, and lifted buoyantly. The above situation occurred only near the end of the eruption, when the upper part of the tuff ring had emerged from the lake. Before emergence of the ring, collapse of the eruptive column created surges that moved along the water surface. The behaviour of the water–pyroclast underflow in this situation is unclear. The density of the surge underflow is expected to have been higher than that of water, so it probably sank and then propagated subaqueously along the surface of the tuff ring as a sediment gravity flow (White, 1996). After each explosion, a part of the pyroclastic material fell back into the crater and was temporarily stored there until the next explosion remobilized it. Some part of the material may have experienced several expulsions and collapses before its final ejection from the crater (Houghton & Smith, 1993). Such recycling resulted in rounding of pyroclasts. Additional rounding may have occurred during transportation of clasts by turbulent surges. End of the eruption Near the end of the eruption we infer that the magma ascent rate declined and the feeder dyke began to freeze. The walls of the dyke were covered by a thick layer of partially solidified melt, and a waning stream of fresh magma detached clots of this semi-solid material (sometimes enclosing fragments of country rock) and transported it upward. Most of the erupting magma near the end of the eruption may have represented such clots covered by only a thin layer of freshly arrived melt. The final explosions of the eruption ejected more and more of these clots, which led to an abrupt increase in the abundance of coarse-grained material in the uppermost layers of the tuff ring, and
VSIC02 9/12/08 2:07 PM Page 55
Eruptions in Karymskoye lake, Russia to the deposition of scoria-crust bombs. The basaltic bombs decrease in size and abundance with distance from the crater and their source during the final explosions was probably shallow (e.g. Self et al., 1980). The appearance of abundant scoria-crust bombs in the course of an eruption may serve as an indicator that the feeder dyke is freezing and the eruption’s end is approaching. The largest of the tsunamis occurred at the end of the eruption. This may indicate that the explosions that took place near the end of the eruption were stronger than those of the main stage. Alternatively, these explosions, which ejected only bomb-sized clasts from shallow depths, more effectively pumped energy into the generation of water waves than did the previous explosions (which ejected a large amount of finer-grained material and were at somewhat greater depth). The waves redistributed coarse-grained material on the surface of the tuff ring to form large-scale ripples, and cast ashore rafts of floating pumice and bushes.
DEPOSITS AND STYLES OF PREHISTORIC ERUPTIONS IN THE LAKE Our studies have shown that the 1996 eruption was the third episode of Holocene volcanic activity within the Akademiya Nauk caldera (Belousov et al., 1997). Basalt–basaltic andesite tephra of the two pre-1996 eruptions crops out along the coast of the lake, being thickest in its northern part (Fig. 15). The eruptions were closely spaced in time and occurred at c. 4800 14C yr bp.
Fig. 15. Thickness (in metres) of pyroclastic deposits of the first (I) and second (II) eruptions that occurred in Karymskoye lake at 4800 14C yr bp, with grain-size histograms for representative samples. Numbers on the sketch map show thickness of second eruption/thickness of first eruption. Filled circle indicates site of section 6, where samples represented by the histograms were collected.
55
Deposit of the first eruption The deposit of the first eruption is in general similar to the deposit of the 1996 tuff ring. The exposed thickness of the deposit is up to 10 m, and the deposit is composed of monotonous, rhythmically intercalated parallel, normally graded, grain-supported, dark grey beds of moderately sorted ash lapilli (Figs 8, 9, 15 & 16; Table 1). Inclination of the beds is very gentle; in outcrop they appear horizontal. The coarsest, lower part of each 0.3–1-m-thick bed commonly has an openwork fabric. The beds are well defined and laterally continuous; no erosional discordances have been found. Contacts between the beds commonly are diffuse. In some places, the unit is completely structureless, and may represent syneruptive slump deposits. Grain-size distributions of samples from the deposit are commonly unimodal (– 4φ to – 2φ), with a very low content of fines (Figs 9b & 15; Table 1). In the sections studied the deposit is notably coarser than the 1996 tuff ring deposit. Juvenile basaltic clasts constitute > 95% of the deposit. They have vesicularity indices of 9–49% (mean 28%) (Fig. 11), and are less vesicular than the 1996 juvenile clasts. Coarse lapilli are mostly irregular in shape and angular to subrounded. The surface morphology of sand-sized clasts from the deposit is identical to that of the 1996 deposits, and is inferred to reflect a similar phreatomagmatic character of the older eruption. Most of the studied sections are inferred to have been deposited subaqueously. Evidence is provided by a 10–20-cm-thick layer of yellowish fine ash (Md 7.2φ; sorting 1.3φ), which covers the deposit (Fig. 16b). The
VSIC02 9/12/08 2:07 PM Page 56
56
A. Belousov and M. Belousova
(a)
(b)
(c) Fig. 16. Deposits of the first and second 4800 14C yr bp eruptions in the lake. (a) Thick coarse-bedded lapilli of the first eruption tuff ring, covered above erosional contact by cross-stratified surge beds of the second eruption. Large angular clast near the contact is accidental fragment of lava. Thin, fine-grained layer in the deposit of the first eruption (at the level of person’s head) divides better-sorted, subaqueously deposited lower part of the deposit from more poorly sorted, subaerially deposited upper part. Site 1.5/10 in Fig. 15. (b) Two uppermost, normal-graded lapilli layers of the tuff ring deposit of the first eruption (TR) covered by layer of fine ash (A) settled from lake water after the eruption. Site 1.5/2 in Fig. 15. Scale is 35 cm long. (c) Intercalation of cross-bedded base-surge deposits and fine-grained fall layers of second eruption. Site 1.5/10 in Fig. 15. (Note a small bomb sag in the centre of the picture (arrow).) The surges probably moved from left to right. Shovel is ≈ 1.5 m long.
fine ash, which sometimes has ripple marks, is inferred to have been separated from the coarse fraction during deposition in water. After the eruption it settled on the bottom of the lake, just as occurred in 1996. The position of the fine ash layer shows that the level of the lake 4800 14C yr ago was at least 5 m higher than the present level. The 1996 tuff ring buried the broad shoal in the northern part of the lake: the shoal represents an older tuff ring 15 –20 m high with a crater 1350 m in diameter (Belousov et al., 1997; Ushakov & Fazlullin, 1998). The rim of this crater is still expressed on the bottom of the lake as an underwater bar at a depth of 50 m surrounding the 1996 crater (Fig. 1b). We assign this old edifice to the first eruption in the lake, 4800 14C yr bp. The similarity of the 4800 14C yr bp and 1996 tuff
ring deposits indicates similar eruption styles and depositional processes. Deposits of the second eruption The deposits of the second eruption either directly overlie the deposit of the first eruption above an erosional contact, or are separated from it by deposits, up to 2 m thick, of ephemeral streams, which accumulated during a short period between eruptions (Fig. 16a). The deposit of the second eruption did not form a new, individual edifice, but covered the tuff ring formed by the first eruption. The deposit is a 2-m-thick lapilli–ash succession consisting of dark grey to greenish layers of crossbedded base-surge deposits intercalated with finegrained fall layers up to 20 cm thick (Figs 8, 9, 15 &
VSIC02 9/12/08 2:07 PM Page 57
57
Eruptions in Karymskoye lake, Russia 16c). There are rare clasts of basaltic andesite up to 10 cm with impact-sag structures (Fig. 16c) that were deposited ballistically on the wet, cohesive surface of fall and base-surge deposits. The surge deposits have well-developed duneforms, with an amplitude of 10 – 40 cm and a wavelength of 1–5 m. The stoss side has an inclination of 10 –20°; the lee side is always more gentle, dipping 0 –12°. Evidence for stoss-side deposition and lee-side erosion, indicating upstream migration of dune crests, is common. In several outcrops the deposit is massive, possibly as a result of syneruptive slumping. Both the surges and intercalated fall layers have numerous rim-type accretionary lapilli up to 1.5 cm across (Schumacher & Schmincke, 1991). The surges commonly contain armoured lapilli (Fisher & Schmincke, 1984). The surge deposits are moderately to poorly sorted (Fig. 8). The deposits commonly have unimodal grainsize distributions, and are notably fines-poor relative to the deposits of 1996 distal base surges (Figs 9b & 15; Table 1). Fall deposits are very fine grained and well sorted (Figs 8 & 9; Table 1). In contrast to the deposits of the first (4800 14C yr bp) and third (1996) eruptions, clasts of the second (4800 14C yr bp) eruption are covered by a thin layer of fine ash (light mud coating of Houghton & Smith, 1993). This makes it difficult to estimate a percentage of true juvenile clasts in the deposit, but locally it is clear that some layers of the surge deposits are strongly enriched in coarse-grained accidental clasts from underlying strata. Such layers probably mark episodes of intensive widening of the vent, possibly as a result of energetic water–magma interaction in the conduit. Juvenile clasts have vesicularity indices of 18 – 65% (mean 52%)asignificantly larger than the vesicularity of juvenile clasts from the first and 1996 eruptions (Fig. 11). Coarse lapilli are irregular in shape and mostly angular. The morphology of the sand-sized particles that constitute the deposit is similar to that of the 1996 deposit, but there is more abundant adhering dust, which commonly completely clogs vesicles (Fig. 12i). The end of the eruption is marked by a 1-m-thick layer of large, thoroughly vesiculated, scoriaceous bombs overlying the sequence of surge and fall deposits. Deposits of water-rich base surges like those that formed the tuff rings of the first and 1996 eruptions were not found. The character of the deposit of the second eruption is very different from that of the first and of the third (1996) eruptions. The well-developed, cross-bedded base-surge deposits of the second eruption resemble
base-surge deposits of the 1965 Taal eruption, in which volcanic explosions occurred in very shallow water (Moore et al., 1966; Moore, 1967; Waters & Fisher, 1971). Thus, we infer that the vent of the second eruption was situated in very shallow water or even onshore, penetrating water-saturated products of the first eruption. In this situation there was neither a significant layer of water above the vent, nor an abundance of water during water–melt interaction, as took place during the first eruption and in 1996. The reduced interaction with water led to more violent explosions that produced a large amount of finegrained particles. During propagation of the surges of the second eruption much of their fine fraction was elutriated into overriding convective clouds, from which were deposited most of the fall deposits of the eruption, which are much finer grained than fallout of the 1996 eruption. Hence we infer that the surges of the second eruption were more inflated and drier than surges of the 1996 eruption. Base surges of the second eruption were also much more energetic than surges of 1996 eruption; they were able to travel > 4 km and to affect the whole lake. In contrast to the 1996 surges, particles from the surges of the second eruption came to rest after protracted basal traction, which led to formation of excellent cross-bedding. Scoriaceous bombs ejected at the end of the second eruption probably marked its transition from a phreatomagmatic eruption to a purely magmatic one.
CONCLUSIONS The eruptions in Karymskoye lake show the influence that a shallow layer of water above a vent can have on the style of basaltic eruptions. If the 1996 eruption had occurred in a dry environment the inferred discharge rate suggests that it probably would have been a vigorous Strombolian or even sub-Plinian eruption, resembling that of Mihara Yama in Japan on 21– 22 November 1986 (Earthquake Research Institute, Tokyo, 1988). Instead, several tens of metres of standing water above the vent drastically changed the character of the 1996 eruption. Rather than forming a sustained eruptive plume, unstable eruptive columns with a high water and steam content immediately collapsed back into the lake. In this situation > 95% of all pyroclastic products of the eruption were deposited from a specific type of base-surge that spread radially from the crater. To indicate their inferred origin from the collapse of very wet Surtseyan slugs, we term
VSIC02 9/12/08 2:07 PM Page 58
58
A. Belousov and M. Belousova
them water-rich base surges. A water-rich base surge can be considered to be the end member of wet base surges. Gravitational settling of pyroclasts and droplets of water in a water-rich surge forms a water–pyroclast underflow at its base, which serves as the depositional system for the surge (Fisher, 1990). Deposition from the underflow is akin to deposition from hyperconcentrated flood flow (Smith, 1986). The deposits of waterrich surges have an extremely low content of fines, and have none of the wave laminations that are characteristic of many base-surge deposits. Instead, they form well-graded parallel layers. The type of grading probably depends on the concentration of pyroclastic particles in each particular underflow of the water-rich base surges. As a result of elevation of ambient pressure (50 m of water depth above the vent) and rapid chilling in the water, pyroclastic particles of the eruption are notably denser than regular basaltic scoria. The inferred abundance of water during the water–magma interaction is probably the reason why the resulting deposits are not as fine grained as the deposits of the subaerial phreatomagmatic eruptions. An unusual feature of the lapilli-sized clasts of the eruption is their significant rounding, uncommon for most types of primary pyroclastic deposits. Apparently, a layer of water above the vent prevented many pyroclasts from being ejected from the crater by the explosion in which they formed. Instead, they experienced several episodes of recycling in the crater before final ejection, which produced their unusual roundness. The presence of several tens of metres of standing water above the vent also strongly influenced the hazards associated with the eruption. At distances exceeding 1.3 km from the vent, only tsunamis and lake-outflow lahars were dangerous. Closer to the crater base surges and ballistic material were very destructive. The first of the 4800 14C yr bp eruptions in the lake was very similar to the 1996 eruption, probably because of similar water depths above the vent. The second of the 4800 14C yr bp eruptions probably occurred in very shallow water; its deposits resemble those of subaerial phreatomagmatic eruptions. Absence either of abundant water during water–melt interaction or of high ambient pressures resulted in violent explosions that produced a large amount of fine-grained particles. Water-rich base surges were not generated. Instead, gas–pyroclastic base surges (much more energetic and dry than surges of the 1996 eruption) and related ash clouds transported most of the material of the second eruption and were the main sources of hazard associated with the eruption.
Although the studied eruptions occurred in a lake, most of the results obtained are also applicable to shallow submarine eruptions of basic magma.
ACKNOWLEDGEMENTS The idea of this publication was strongly supported by James White. We are very grateful to Yaroslav Muravyev, Sergei Fazlullin, Alexei Ozerov, Sergei Ushakov, Elena Grib, and Evgenii Gordeev, who participated with us in the field at Karymskoye lake and in very helpful discussions. We acknowledge Vilorii Bakhtiarov, who kindly provided us with a videotape of the eruption. Scanning electron microscope analysis of pyroclasts and preparation of the manuscript were carried out during our postdoctoral fellowship at Pennsylvania State University, supported by an NSF grant to Barry Voight. The English text of the paper was improved by Amanda Clarke and Barry Voight. Careful critical reviews by James White, Bruce Houghton and an anonymous reviewer have helped this paper considerably. The research described in this publication was made possible in part by Alexander von Humboldt Foundation and by Grant RG1-172 from the Civilian Research and Development Foundation and the Russian Government.
REFERENCES Alidibrov, M. & Dingwell, D.B. (2000) Three fragmentation mechanisms for highly viscous magma under rapid decompression. J. Volcanol. geotherm. Res., 100, 413 – 423. Banks, N.G. & Hoblitt, R.P. (1981) Summary of temperature studies of 1980 deposits. In: The 1980 Eruptions of Mount St Helens, Washington (Eds Lipman, P.V. & Mullineaux, D.R.), US geol. Surv. Prof. Pap., Denver, CO, 250, 295 –313. Belousov, A. (1996) Deposits of the 30 March 1956 directed blast at Bezymianny volcano, Kamchatka, Russia. Bull. Volcanol., 57, 649 – 662. Belousov, A.B., Belousova, M.G. & Muravyev, Ya. D. (1997) Holocene eruptions in the Akademiya Nauk caldera and the age of the Karymsky stratovolcano, Kamchatka. Trans. Russ. Acad. Sci., Earth Sci. Sect., 335, 653 – 657. Belousov, A., Belousova, M. & Voight, B. (1998) Powerful pyroclastic surge at Bezymianny volcano on May 9 –10, 1997 (Kamchatka, Russia). EOS Trans. Am. geophys. Union, 79, 936 –937. Belousov, A., Voight, B., Belousova, M. & Muravyev, Y. (2000) Tsunamis generated by subaquatic volcanic explosions: unique data from 1996 eruption in Karymskoye lake, Kamchatka, Russia. Pure appl. geophys., 157, 1135–1143. Braitseva, O.A. & Melekestsev, I.V. (1990) Eruptive history of Karymsky volcano, Kamchatka, USSR, based on tephra stratigraphy and C14 dating. Bull. Volcanol., 53, 195 –206.
VSIC02 9/12/08 2:07 PM Page 59
Eruptions in Karymskoye lake, Russia Carey, S., Sigurdsson, H., Mandeville, C. & Bronto, S. (1996) Pyroclastic flows and surges over water: an example from the 1883 Krakatau eruption. Bull. Volcanol., 57, 493–511. Cas, R.A.F., Landis, C.A. & Fordyce, R.E. (1989) A monogenetic, Surtla-type, Surtseyan volcano from the Eocene–Oligocene Waiareka–Deborah volcanics, Otago, New Zealand: a model. Bull. Volcanol., 51, 281–298. Delaney, P.T. (1982) Rapid intrusion of magma into wet rock: ground water flow due to pore pressure increases. J. geophys. Res., 87, 7739 –7756. Earthquake Research Institute, Tokyo (1988) The 1986–87 Eruption of Izu–Oshima Volcano. Earthquake Research Institute, Tokyo. Eichelberger, J.C. & Izbekov, P.E. (2000) Eruption of andesite triggered by dyke injection: contrasting cases at Karymsky volcano, Kamchatka and Mt. Katmai, Alaska. Phil. Trans. Royal. Soc. London A, 358, 1465–1485. Elsworth, D. & Voight, B. (1992) Theory of dike intrusion in a saturated porous solid. J. geophys. Res., 97, 9105–9117. Fazlullin, S.M., Ushakov, S.V., Shuvalov, R.A., et al. (2000) The 1996 subaqueous eruption at Academii Nauk volcano (Kamchatka) and its effects on Karymsky lake. J. volcanol. geotherm. Res., 97, 181–193. Fedotov, S.A. (Ed.) (1991) Active Volcanoes of Kamchatka. Nauka, Moscow. Fedotov, S.A. (1998) Study and mechanism of the simultaneous 1996 Karymsky volcano and Akademii Nauk caldera eruptions in Kamchatka. Volcanol. Seismol., 19, 525–566. Fisher, R.V. (1990) Transport and deposition of a pyroclastic surge across an area of high relief: the 18 May 1980 eruption of Mount St. Helens, Washington. Geol. Soc. Am. Bull., 102, 1038–1054. Fisher, R.V. & Schmincke, H.U. (1984) Pyroclastic Rocks. Springer, Berlin. Francis, P. (1993) Volcanoes: a Planetary Perspective. Clarendon Press, Oxford. Grib, E.N. (1998) Petrology of ejecta produced by the Akademiya Nauk caldera eruption of January 2–3, 1996. Volcanol. Seismol., 19, 605 – 635. Gushchenko, I.I. (1979) Eruptions of World Volcanoes (Catalogue). Nauka, Moscow (in Russian). Heiken, G.H. (1971) Tuff rings: examples from the Fort Rock–Christmas Lake valley basin, south-central Oregon. J. geophys. Res., 76, 5615 –5626. Heiken, G.H. & Wohletz, K.H. (1986) Volcanic Ash. University of California Press, Berkeley. Hoblitt, R.P. & Harmon, R.S. (1993) Bimodal density distribution of cryptodome dacite from the 1980 eruption of Mount St. Helens, Washington. Bull. Volcanol., 55, 421–437. Houghton, B.F. & Schmincke, H.-U. (1986) Mixed deposits of simultaneous strombolian and phreatomagmatic volcanism: Rothenberg volcano, East Eifel volcanic field. J. Volcanol. geothermal Res., 30, 117–130. Houghton, B.F. & Smith, R.T. (1993) Recycling of magmatic clasts during explosive eruptions: estimating the true juvenile content of phreatomagmatic volcanic deposits. Bull. Volcanol., 55, 414 – 420. Houghton, B.F. & Wilson, C.J.N. (1989) A vesicularity index for pyroclastic deposits. Bull. Volcanol., 51, 451–462. Inman, D.L. (1952) Measures for describing the size distribution of sediments. J. sediment. Petrol., 22, 125 –145.
59
Ivanov, B.V. (1970) Eruption of the Karymsky Volcano in 1962– 65 and the Karymsky Group Volcanoes. Nauka, Moscow (in Russian). Kokelaar, B.P. (1983) The mechanism of Surtseyan volcanism. J. geol. Soc. London, 140, 939 –944. Latter, J.H. (1981) Tsunamis of volcanic origin: summary of causes, with particular reference to Krakotoa, 1883. Bull. Volcanol., 44, 467– 490. Leonov, V.L. (1998) Ground surface breaks produced by an earthquake and volcanic eruptions in the Karymsky volcanic center on January 1–2, 1996. Volcanol. Seismol., 19, 655– 674. Lorenz, V. (1973) On the formation of maars. Bull. Volcanol., 37, 183–204. Lowe, D.R. (1982) Sediment gravity flows: II. Depositional models with special reference to the deposits of highdensity turbidity currents. J. sediment. Petrol., 52, 279–297. Machado, F., Parson, W.H., Richards, A.F. & Mulford, J.F. (1962) Capelinhos eruption of Fayal volcano, Azores, 1957–58. J. geophys. Res., 67, 3519 –3529. Maguskin, M.A., Fedotov, S.A., Levin, V.E. & Bakhtiarov, V.F. (1998) Ground surface deformation caused by seismic and volcanic activity in the Karymsky volcanic center during January 1996. Volcanol. Seismol., 19, 637– 654. Masurenkov, Yu. P. (Ed.) (1980) A Volcanic Center: Structure, Dynamics and Rock Compositions in the Karymsky Structure. Nauka, Moscow (in Russian). Moore, J.G. (1967) Base surges in recent volcanic eruptions. Bull. Volcanol., 30, 337–363. Moore, J.G., Nakamura, K. & Alcaraz, A. (1966) The 1965 eruption of Taal volcano. Science, 151, 15 –20. Muravyev, Ya. D., Fedotov, S.A., Budnikov, V.A., et al. (1998) Volcanic activity in the Karymsky center in 1996: summit eruption at Karymsky and phreatomagmatic eruption in the Akademii Nauk caldera. Volcanol. Seismol., 19, 567– 604. Nairn, I.A., Wood, C.P. & Hewson, C.A.Y. (1979) Phreatic explosions of Ruapehu: April 1975. N.Z. J. Geol. Geophys., 22, 155 –173. Nishimura, Y. & Miyaji, N. (1995) Tsunami deposits from the 1993 Southwest Hokkaido earthquake and the 1640 Hokkaido Komagatake eruption, Northern Japan. Pure appl. geophys., 144, 719 –733. Schumacher, R. & Schmincke, H.-U. (1991) Internal structure and occurrence of accretionary lapilliaa case study at Laacher See Volcano. Bull. Volcanol., 53, 612– 634. Schumacher, R. & Schmincke, H.-U. (1995) Models for the origin of accretionary lapilli. Bull. Volcanol., 56, 626 – 639. Self, S., Kienle, J. & Huot, J.-P. (1980) Ukinrek maars, Alaska, II. Deposits and formation of the 1977 craters. J. Volcanol. geothermal Res., 7, 39 – 65. Sheridan, M.F. & Wohletz, K.H. (1983) Hydrovolcanism: basic considerations and review. J. Volcanol. geothermal Res., 17, 1–29. Smith, G.A. (1986) Coarse-grained nonmarine volcaniclastic terminology and depositional process. Geol. Soc. Am. Bull., 7, 1–10. Sohn, Y.K. & Chough, S.K. (1989) Depositional processes of the Suwolbong tuff ring, Cheju Island (Korea). Sedimentology, 36, 837–855. Sohn, Y.K. & Chough, S.K. (1992) The Ilchulbong tuff cone, Cheju Island, South Korea: depositional processes
VSIC02 9/12/08 2:07 PM Page 60
60
A. Belousov and M. Belousova
and evolution of an emergent, Surtseyan-type tuff cone. Sedimentology, 39, 523 –544. Thorarinsson, S. (1964) Surtsey: the New Island in the North Atlantic. Almena Bokafelagid, Reykjavik. Thorarinsson, S., Einarsson, T., Sigvaldason, G. & Elisson, G. (1964) The submarine eruption off the Vestmann Islands 1963 – 64. Bull. Volcanol., 27, 435 – 445. Ushakov, S.V. & Fazlullin, S.M. (1998) Morphometric characteristics of lake Karymskoye after an underwater eruption. Volcanol. Seismol., 19, 675 – 683. Valentine, G.A. (1987) Stratified flow in pyroclastic surges. Bull. Volcanol., 49, 616 – 630. Walker, G.P.L. & Croasdale, R. (1972) Characteristics of some basaltic pyroclastics. Bull. Volcanol., 35, 303 –317. Waters, A.C. & Fisher, R.V. (1971) Base surges and their deposits: Capelinhos and Taal volcanoes. J. geophys. Res., 76, 5596–5614. White, J.D.L. (1996) Pre-emergent construction of a lacustrine basaltic volcano, Pahvant Butte, Utah (USA). Bull. Volcanol., 58, 249 –262. Wilson, L., Sparks, R.S.J. & Walker, J.P.L. (1980)
Explosive volcanic eruptions, IV. The control of magma properties and conduit geometry on eruption column behaviour. Geophys. J. R. astron. Soc., 63, 117–148. Wohletz, K.H. (1983) Mechanism of hydrovolcanic pyroclast formation: grain size, SE microscopy, experimental studies. J. Volcanol. geothermal Res., 17, 31– 63. Wohletz, K.H. (1986) Explosive magma–water interactions: thermodynamics, explosion mechanisms, and field studies. Bull. Volcanol., 48, 245 –264. Wohletz, K.H. & McQueen, R.G. (1984) Experimental studies of hydromagmatic volcanism. In: Explosive Volcanism: Inception, Evolution and Hazards Studies in Geophysics, pp. 158–169. National Academy Press, Washington, DC. Yamamoto, T., Soya, T., Suto, S., et al. (1991) The 1989 submarine eruption off eastern Izu Peninsula, Japan: ejecta and eruption mechanisms. Bull. Volcanol., 53, 301– 308. Zimanowski, B. (1998) Phreatomagmatic explosions. In: From Magma to Tephra (eds. A. Freundt & M. Rosi) pp. 25 –54. Elsvier, Amsterdam.
VSIC03 9/12/08 2:06 PM Page 61
Spec. Publs int. Ass. Sediment. (2001) 30, 61–80
Eruption and reshaping of Pahvant Butte volcano in Pleistocene Lake Bonneville J. D. L. WHITE Geology Department, PO Box 56, University of Otago, Dunedin, New Zealand
ABSTRACT Pahvant Butte, an isolated volcano that erupted in an arm of Lake Bonneville, is an outstanding example of the product of an emergent eruption, its interaction with water-level, erosion and deposition during eruption, and the volcano’s response to post-eruptive water-level changes. Pahvant Butte consists of three distinct elements: a steep-sided tuff cone that consists of tephra deposited under damp and dry conditions, breached to the south-west, and having wave-cut cliffs on the north and west; a nearly flat-topped platform extending from the cone to the south and east; and a mound of generally well-bedded, shallow-dipping, subaqueously deposited tephra underlying and coring the platform. The morphology of Pahvant Butte is dominated by the cone and platform, with the mound truncated along the wave-cut cliffs and buried beneath the platform. The Pahvant Butte eruption began subaqueously, and the mounded strata formed largely before the lake surface was breached. Upon emergence, deposition became more localized around the vent, and the tuff cone began to develop. Both subaerial and subaqueous facies of steeply dipping cone strata are present; the latter provide a rare example of subaqueous fall-fed grainflow deposits. Erosion of the growing cone by waves and runoff rapidly shifted tephra to form the platform, which consists of lacustrine volcaniclastic beach spit, lagoonal, and local deltaic deposits. Such syn-eruptive platforms are characteristic of emergent eruptions, although their specific form and constitution depend on the interplay between eruptive activity and processes of erosion and redeposition. The Pahvant Butte eruption took place during the first Bonneville highstand; afterwards the lake level fell and rose again during the Keg Mountain oscillation, which involved a substantial fall in the level of the lake between the time it first reached the Bonneville shoreline and the final occupation of that shoreline before the Bonneville breakout flood. This rise was recorded at Pahvant Butte by stacked platform foresets.
INTRODUCTION not well known. Elsewhere remnants of shallow subaqueous volcanoes have been described (e.g. Cas & Landis, 1987; Cas et al., 1989; Godcheaux et al., 1992; Werner et al., 1996; Smellie & Hole, 1997; Kano, 1998), but their original morphology is not well known. Gilbert (1890) was probably the first to argue that Pahvant Butte volcano initially erupted beneath Lake Bonneville, an interpretation accepted by numerous later workers (Condie & Barsky, 1972; Wohletz & Sheridan, 1983; Oviatt & Nash, 1989; Farrand & Singer, 1992). Water depth at the site of eruption was ≈ 85 m (Oviatt & Nash, 1989; see further discussion below). The aims of this paper are: 1 to detail the development of the volcano’s deposits and morphology, which reflect interaction of the lake
Pahvant Butte is one of a small number of volcanoes known to have erupted into the vast Pleistocene pluvial lakes such as Lake Bonneville (Fig. 1) that covered parts of the Basin and Range province of the western USA. These lakes were large bodies of water several tens of metres deep that have now disappeared or diminished, exposing both subaqueous and subaerial products of volcanoes in arid landscapes scarcely altered since the lakes’ retreat. The subaqueous to emergent volcanoes exposed by declining lake levels, such as Pahvant Butte, are ideal sites for investigation of subaqueous to emergent eruption processes and the interaction of waves, currents, and the eruption. Modern subaqueous to emergent volcanoes are difficult to study and their internal facies, particularly subaqueously developed ones, are
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
61
VSIC03 9/12/08 2:06 PM Page 62
62
J. D. L. White Morphology and structural elements of Pahvant Butte Pahvant Butte consists of three elements: 1 a complex, asymmetrical steep-sided tuff cone, breached to the south-west and having wave-cut cliffs on the north and west; 2 a nearly flat-topped platform extending from the cone to the south and east (Fig. 2), the margins of which consist of steeply outward-dipping beds of sideromelane tephra; 3 a broad mound of subhorizontally bedded tephra that is largely hidden beneath the platform (Fig. 3). Several lines of evidence support the contention that Pahvant Butte erupted into the waters of Lake Bonneville, and was not simply reshaped by the lake waters subsequent to eruption (Gilbert, 1890; Oviatt & Nash, 1989). The most direct of these relate to the depositional features of the mound, which formed during eruption and below the inferred palaeolake surface (White, 1996). The following sections address these elements in their general order of development, from subaqueous mound growth during eruption, through emergent eruption and formation of the cone, to platform and beach ridge development by lacustrine waves and currents. The ordering is a general one, however, because there was significant overlap among phases during the eruption.
THE MOUND
Fig. 1. Index map showing extent of Pleistocene Lake Bonneville at its highstand (patterned; after Currey, 1982). GSL is Great Salt Lake; similarly shaded bodies are other present-day lakes. Dashed lines indicate state boundaries of Idaho (ID), Nevada (NV), and Utah (UT). Pahvant Butte is located at 39.10°N, 112.55°W.
with eruptive and depositional processes both during and after eruption; 2 to evaluate the role of the lake’s water in development of palagonitization; 3 to briefly address potential implications of the volcano’s shoreline deposits for the history of Lake Bonneville. A related paper (White, 1996) provided a detailed discussion of the subaqueous phase of the eruption and its products.
The earliest growth phase of Pahvant Butte is recorded by mound deposits largely hidden beneath the platform and cone. Defining characteristics of the mound deposits are shallow bedding dips (< 10°), and location low in the edifice, typically beneath steeply dipping cone strata. Mound strata consist of greenish sideromelane ash grains showing the wide range of vesiculation typical of phreatomagmatically disrupted vesiculating magma (Houghton & Wilson, 1989). The following summary of mound deposits is drawn from the more extended account of White (1996), in which the strata were assigned to four lithofacies associations and a detailed assessment of their origin was provided. Lithofacies M1: well-bedded, broadly scoured coarse ash and lapilli Lithofacies M1 (lithofacies association M1 of White,
; VSIC03 9/12/08 2:06 PM Page 63
Pahvant Butte eruption
63
Deltaic platform topset
0
505
Northern gully
3
A
0 525
2
54
2
P
Platform
4
Upper Cone unit 4 (UCd facies)
3
Upper Cone unit 3 (UCw facies)
2
Upper Cone unit 2 (UCw facies)
1
Upper Cone unit 1 (earliest, UCw facies)
0
555
4
50
P
1
P
A'
P
3
0 535 50 5250 51 0 505
4950
Lower Cone and M2, subaqueous near-vent facies
P
4850
Pre-emergent mound (facies M1, M3)
4760
General viewing direction of Fig. 3 sketch by Gilbert (1890)
Fig. 2. Map of Pahvant Butte showing distribution of units described in text, and four upper cone assemblages separated by syn-eruptive slip surfaces (UC1, UC2, UC3, UC4). The deltaic platform topset grades into beach-spit deposits, both linked with platform foresets. Palagonite affects all units to varying degrees, and palagonitized zones are not mapped separately from fresh or cemented sideromelane tephra. A modern gully incised through deltaic topset and other platform deposits exposes a strip of mound deposits. Elevations are in feet, topography from USGS Pahvant Butte North and Pahvant Butte South 7.5′ topographic maps (1971; scale 1 : 24 000).
1996) was deposited directly on the lake floor ≈ 85 m beneath the lake surface, and consists largely of thin beds of coarse ash to fine lapilli. Broad scours, subtle bed lenticularity and very low-angle cross-stratification typify the M1 lithofacies (Fig. 4a), yet the thin beds are laterally continuous over tens of metres. Large clasts lie at or near the bases of beds without impact sags. Together these features suggest deposition from unchannelled, intermittently undercapacity and erosive (probably dilute), unsteady aqueous currents. These flows were capable of transporting a subpopulation of lapilli and small blocks and depositing them together with coarse to fine ash in a traction-dominated depositional regime.
Lithofacies M2: massive to weakly bedded lapilli ash with cauliflower clasts and armoured lapilli Intercalated coarse ash beds and lapilli ash breccia beds make up lithofacies M2 (lithofacies association M2 of White, 1996), which is of limited extent and interfingers laterally with M1 and M3 beds. Lithofacies M2 differs from M1 in having definite sags beneath large particles (Fig. 4b), more diffuse bedding contacts, abundant armoured lapilli, and intercalated thick beds of coarse lapilli ash breccia bearing many clasts of deformed lake sediment (Fig. 4c). The bedding sags suggest ballistic emplacement of the larger blocks. Lapilli lenses above the block sags indicate
; VSIC03 9/12/08 2:06 PM Page 64
64
J. D. L. White
West
East
1772m
2
1640m
1
5400
3
v vv v v v v vv v v
4
5000
Cross-section line A-A', Fig. 2
– Syn-eruptive slip surfaces
2
Upper Cone unit 2 (UCw facies)
1
Upper Cone unit 1 (earliest, UCw facies)
Platform
v v v Chaotic (?) intra-vent deposits (not exposed)
Lower Cone and M2, subaqueous, near-vent facies
4
Upper Cone unit 4 (UCd facies)
Pre-emergent mound (facies M1, M3)
3
Upper Cone unit 3 (UCw facies)
Fig. 3. Slightly modified version of Gilbert’s (1890) field sketch, and interpretive cross-section of Pahvant Butte. View directions differ slightly, with cross-section plane chosen to emphasize platform deposits. Units are as in text and Fig. 2 except for addition of inferred chaotic intravent deposits, not exposed at Pahvant Butte. The M2 mound facies grade into lower cone deposits and are shown together. Upper cone units are separated by syn-eruptive slip surfaces, with younger cone units onlapping older ones across the slips. Lower cone and mound units are locally displaced along slips where exposed to the west and north.
lateral transport and emplacement of surrounding coarse ash, as do subtle bedding deflections around blocks and local erosion. Deposition was from sediment-gravity flows into which ballistic blocks were emplaced during flow. Diffuse internal contacts within thick beds are inferred to indicate sedimentation from high-sediment-concentration bases of sustained currents (Kneller & Branney, 1995). Lapilli lenses infilling above bomb sags are consistent with current deposition during bomb emplacement. Lithofacies M3: broadly cross-stratified ash with local steep-foreset lenses and ripple-lamination
Moderately vesicular coarse sideromelane ash with intercalated lapilli ash and lapilli beds make up lithofacies M3, which generally overlies M1. Lithofacies M3 is overlain by ripple-marked beds that are, in turn, capped by primary subaerial tuff cone deposits (Fig. 3; see below). Very broad (≈ 1 m), low-amplitude (≈ 15 cm) dune-like bedforms (Fig. 4d) characterize M3.
Cross-stratification, duneforms and rippleforms are well developed at higher levels, and extend to the uppermost exposures of the mound at palaeodepths of ≈ 20 m below syneruptive lake level. At greater palaeodepths, ≈ 50 m or more below lake level, facies M3 grades downward into M1. This facies developed by interaction between aqueous sediment-gravity flows and oscillatory flow beneath surface waves (White, 1996), and is present in strata that underlie rippled shoreline deposits and subaerial tuff cone beds; the facies formed before emergence of the volcano. Wave-generated oscillatory currents, superimposed on the outward flows, produced the formdiscordant dune-like bedforms (White, 1996). Lithofacies M4: thick-bedded, faintly stratified ash with rotated tuff blocks
Lithofacies M4 comprises thick beds of structureless to subtly stratified sideromelane ash and lapilli ash with reversely graded bases. It commonly contains
VSIC03 9/12/08 2:06 PM Page 65
Pahvant Butte eruption
65
Fig. 4. Mound lithofacies. (a) Lithofacies M1, showing typical subhorizontal, subtly lenticular, bedding in the M1a lithofacies (pencil is 15 cm long). Dark beds are clean sideromelane ash, pale beds are carbonate-coated ash. (Note broad scour-and-fill near centre, irregular clast of lacustrine mud (now mudrock), and absence of sags beneath cauliflower ‘bombs’ near bottom of photograph.) (b) Lithofacies M2 showing horizon with cauliflower clasts, ash-rich layers containing armoured lapilli (arrow), white pieces of lacustrine mudrock (e.g. in and above broken cauliflower clast), and a sag structure infilled with lapilli (after White, 1996). (c) Lithofacies M2 lapilli breccia with scattered sediment clasts; some with lightly baked margins, others strongly deformed. (Note clast-supported texture, paucity of fine ash, and weak layering.) Cauliflower-textured juvenile clasts are present, characterized by rugged-surfaced glassy chill-rinds dissected by contraction-induced cracks. Left-hand bar on scale is 10 cm long (after White, 1996). (d) Lithofacies M3 showing broad, open dunes among subhorizontal beds with good lateral continuity and lenses of high-angle cross-strata (arrow). Bedding surfaces descend very gently to the left (to north-north-east, outward and downcurrent from cone).
rotated blocks of bedded ash, and is in contact with M1 and M3 lithofacies. Lithofacies M4 is the least characteristic of the mound lithofacies, and very similar deposits are also present in spit– deltaic foreset units of the platform, on the subaerial Pahvant Butte cone and on other emergent cones (Sohn & Chough, 1992). The rotated blocks of bedded tephra suggest partial disaggregation and flow of earlier ash deposits.
THE CONE The cone consists largely of steeply dipping layers of poorly to well-bedded, originally glassy basaltic tuff,
mostly altered to palagonite, and is interpreted as a fairly typical tuff cone. Subaerial parts of such cones have been widely described (Hamilton & Myers, 1963; Fisher, 1977; Leys, 1983; Wohletz & Sheridan, 1983; Verwoerd & Chevallier, 1987; Sohn & Chough, 1992, 1993), and are generally considered to form where magma erupts in the presence of surface water or abundant ground water (Heiken, 1971; Fisher & Schmincke, 1984). Form of the cone The strongly asymmetrical cone of Pahvant Butte is commonly described as having been breached in the
; VSIC03 9/12/08 2:06 PM Page 66
66
J. D. L. White
3
2
2
LC
M1, M3
LC
10 m
Fig. 5. Slightly oblique line drawing and photograph, taken from west-north-west of Pahvant Butte, showing anticlinal rim features of western rim segment. Prominent foreground cliffs were cut at Provo lake level. Left-dipping beds at upper left are UC3, in contact with UC2 across slip surface. Lower cone outcrops in cliff have more massive aspect.
south, where the platform is broadest and no cone wall is present (Gilbert, 1890; Wohletz & Sheridan, 1983; Oviatt & Nash, 1989). The western part of the cone rises and broadens northward toward the summit, from which it descends to a low saddle in the northeast before rising again to a subconical hill that terminates the cone to the east-south-east (Fig. 3). Two low anticlinal features underlie the north and south ends of the segment and are spectacularly draped by overlying cone beds (Fig. 5). The shape of the cone is not easily explained in terms of westerly winds documented as prevalent during eruption (Oviatt & Nash, 1989), or by lacustrine erosion of the edifice (evaluated in more detail below). The low anticlinal features may represent part of an earlier rim, the western edges of which have been eroded and eastern ones buried beneath the larger cone. To the south, the absence of a cone wall probably results from a combination of weak initial development (unfavour-
able winds or significantly inclined vent), slip failure, and erosion and burial as the platform developed. Depositional features Lower cone
Lower parts of the cone are locally exposed in cliffs below the 5050 ft (1540 m) contour (Fig. 3), and have relatively steep bedding dips (≈ 20°). Beds vary in thickness from ≈ 2 to 20 cm, characteristically have diffuse contacts, and are lensoid and discontinuous (Fig. 6). The reverse-graded beds and lenses in steeply dipping strata composing the lower cone suggest emplacement by grainflow processes (Nemec, 1990; Sohn & Chough, 1993), but the beds are not nearly as lenticular, nor as steep, as those constructed subaqueously by avalanche processes at angles of repose (Buck,
VSIC03 9/12/08 2:06 PM Page 67
Pahvant Butte eruption
67
Fig. 6. Subaqueously formed lower cone lithofacies showing typical ≈ 20° outward dip, discontinuous, lenticular layering (a, b). Rough-surfaced vesicular lapilli are clustered in lenses with few fines (c), whereas surrounding layers contain abundant medium to coarse ash with scattered lapilli (a, b). Shovel handle in (a) is 15 cm wide, pencil in (b) and (c) is 15 cm long.
1985; White, 1992). This facies is inferred to have been deposited by grainflows that evolved from fall sedimentation (Sohn & Chough, 1993), but on the lower cone this process occurred subaqueously. The subaqueous setting, in conjunction with vesicular pyroclasts, has two important effects: 1 particle fall velocity is reduced relative to subaerial settings; 2 stability of depositional slopes is reduced, as a result of the loose packing and low density contrast between particles and interstitial fluid. It is inferred that these effects counteract each other in forming subaqueous suspension-fed grainflows; slower settling greatly reduces initial saltation effects compared with subaerial settings, but the slope instability results in an intermittently mobile surface layer into which incoming particles are entrained. Upper cone Beds of the upper cone dip steeply (typically 20–30°,
but up to 35° locally), and are strongly overprinted by palagonitization (see below). Two lithofacies are distinguished, each comprising a genetically related group of beds. They form parts of four slip-bounded assemblages that developed as a result of instability during construction of the subaerial cone. Lithofacies UCw Lithofacies UCw is characterized by block and bomb sags, and by accretionary and armoured lapilli (Fig. 7). Beds typically have impact sags, often contain armoured lapilli, and show generally poor bedding continuity and definition. Duplex structures and structureless tuff with rotated bedded tuff blocks locally fill concavities formed by slope failure. Pronounced rill erosion locally produced small deeply incised channels, typically 10 cm wide and as much as 50 cm deep, with inter-rill spacings of less than 1 m. The rills are most commonly infilled with lapilli tuff.
VSIC03 9/12/08 2:06 PM Page 68
68
J. D. L. White
Fig. 7. Subaerial cone lithofacies UCw, with sag and accretion features. (a) Mixed fine- and coarse-grained horizons with white lacustrine mudrock clasts, and pronounced bedding sags beneath cauliflower texture bombs. (b) Backset-stacked lapilli tuff, formed syneruptively by duplex slip faulting. (c) Cored accretionary lapilli from same site as (b). Coin is 2 cm in diameter.
UCw interpretation
Lithofacies UCd
Features of UCw suggest transport and deposition of tephra in the presence of water. The damp ash was cohesive, and coated coarser clasts during their transport from the vent to the depositional site to form armoured lapilli. Layers of damp ash were sufficiently cohesive to produce well-developed sags beneath ballistically emplaced blocks, and to maintain local bedding coherence during failure (Fisher & Schmincke, 1984); the beds were partially disaggregated during flow to form rotated blocks in debris-flow deposits (Lorenz, 1974). Local duplex bedding reflects slip failure of damp ash, which stacked against the downslope margins of slip surfaces (Sohn & Chough, 1992). The localized deep rilling resulted from the downslope flow of water, either released from very wet tephra or thrown from the vent as ‘free’ water during eruption (Verwoerd & Chevallier, 1987). It is inferred that the association was produced largely by deposition from cock’s tail plumes or jets, which deposit material by a combination of ballistic emplacement and subsynchronous rapid fallout (Thorarinsson, 1967; Sohn, 1996), with a lesser contribution from low-energy pyroclastic density currents.
Lithofacies UCd is characterized by thinly bedded, interstratified layers of framework-supported finespoor coarse ash and lapilli together with thin ash layers (Fig. 8). Reverse grading is locally prominent in the openwork lapilli layers. Local scours are filled by broadly concave-upward lenses of fines-poor ash. In places, decimetre-thick beds of coarse ash contain 2–5-cm basalt lithic clasts, either concentrated in small lenses at bed tops or scattered randomly within beds. These are interlayered with thin, continuous ash layers a few millimetres thick. Swarms of small ventward-dipping normal faults occasionally cut the ≈ 25–30° dipping beds. Beds of UCd lack accretionary lapilli, bomb sags, and other evidence of dampness upon deposition. UCd interpretation Grainflow and fallout processes are inferred to have produced features observed in UCd. The lack of particle cohesion suggests that little or no water was present upon deposition or during transport. Lithic clasts without accompanying sags, within or at the tops of
VSIC03 9/12/08 2:06 PM Page 69
Pahvant Butte eruption
69
Fig. 8. Subaerial cone lithofacies Ucd, formed in late stages of the eruption, is characterized by (a) steep dips and sharply defined bedding, and (b) conspicuous lenses of openwork lapilli with laminae of fine tuff. Thin beds of fine to coarse tuff in places contain isolated lapilli; separate layers and lenses of fines-poor lapilli show both normal and reverse grading; there are clast-cluster fabrics and local down-dip coarsening.
sandy tephra beds, suggest emplacement by grainflow. Thinly laminated ash is suggestive of settling from suspension. Isolated larger clasts in thinly laminated strata rolled or bounced into place. During deposition of UCd the vent was probably at or near the surface of the lake, and protected from direct water influx by a tephra dam. The resulting eruption ejected hydroclasts with little early condensing steam or water, allowing relatively effective sorting in the eruptive column and /or fallout plume. By analogy with observed processes at Surtsey, the most likely eruptive style associated with these ‘drier’ deposits might be one of continuous uprush. This fails, however, to explain the closely spaced ash laminations and generally thin bedding. It is more likely that reduced access of water to the vent in this case did not coincide with increased eruptive rates, and the result was a simple transition toward ‘dry’ phreatomagmatic activity. In addition to fall-fed grainflow, deposition of some thinner and less lenticular beds is inferred to have occurred from bases or tails of surges moving down the steep volcano flanks (Crowe & Fisher, 1973; Sohn & Chough, 1989). The finest-grained laminae are interpreted as fall deposits. The restriction of recognized UCd strata to the lastformed cone deposits suggests that ‘dry’ phreatomagmatism was significant during only the closing stages of the eruption. There is no a priori reason to expect that UCd strata would be preferentially removed, and, despite the pernicious effects of palagonitization (see below), bomb sags, armoured lapilli and other dampness-related features are ubiquitous in earlier beds.
Syn-eruptive cone deformation Three extensive slip surfaces, each overlain and onlapped by younger tephra beds, demarcate facies assemblages and help define the shape of the Pahvant Butte cone (Figs 2 & 3). The oldest recognized slip surface truncates the UC1 assemblage of upper cone tuffs and is inferred to displace some lower cone deposits. Another slip surface dips inward and encircles much of the inner slope of the cone. The youngest dips outward and truncates outward-dipping beds along the north-eastern segment of the cone. All are mildly upward-concave, with very steep upper surfaces that shallow only slightly before passing out of view beneath the platform. In addition to these large slip surfaces, early failure has locally produced chaotic beds up to 5 m thick with irregular, scoop-shaped, discordant lower contacts and rotated blocks of bedded tephra in a structureless tephra matrix. Locally, small-scale, outward-dipping subvertical reverse faults offset beds by < 10 cm; much of the offset is taken up along sharp flexures, and the faults die out upward within UC4 beds. Repeated slip-failure is a characteristic accompaniment to growth of emergent Surtseyan volcanoes (Lorenz, 1974; Kokelaar, 1983); it is facilitated by depositional oversteepening as wet, jetted material is deposited by fall, and by the effects of shaking and surface impact of the jetted material (Sohn, 1996). Pervasive small-scale faulting is absent at Pahvant Butte (see Sohn & Chough, 1992). Localization of deformation along a few large slips involving UCw deposits suggests that the damp deposits were
VSIC03 9/12/08 2:06 PM Page 70
70
J. D. L. White
coherent upon or soon after deposition (Sohn & Chough, 1992), and rafts of bedded tephra within the chaotic beds similarly suggest early stratal coherence. Palagonite Strong palagonitization is characteristic of the cone deposits except for an impersistent, decimetres-thick upper crust held together by a finely crystalline, partly geopetal, pore-lining carbonate cement. In places palagonite formed preferentially within specific horizons, but in many sites palagonite zones steeply crosscut bedding boundaries. Over the whole of the cone, bed-specific palagonitization is strongly subordinate to more general alteration that is independent of both bedding and depositional facies. None of the slip surfaces consistently separate palagonitized from non-palagonitized zones, nor are rotated blocks of bedded tephra in chaotic beds preferentially palagonitized. In parts of the lower cone deposits, and locally in the inner part of the platform deposits where their contact with lower and upper cone beds is exposed (see below), palagonite zones steeply cross-cut bedding and grade from almost fresh glass, to palagonite rims, to palagonitized particles, to fully palagonitized tephra in which a pervasive pseudomatrix of zeolites and clays has completely infilled the original pore space. Differences in appearance between palagonitized and nonpalagonitized parts of individual beds are pronounced, with the latter appearing much more matrix rich in outcrop, hand specimen and thin section (see Hay & Iijima, 1968). The effect is so strong that originally openwork, virtually ash-free beds of lapilli appear, where palagonitized, to consist of coated lapilli supported by an ashy matrix. Where clasts are moderately or strongly vesicular, palagonite often forms broad rims grossly similar in appearance to accretionary coatings. Pahvant Butte has been cited by some workers as an example of palagonitization controlled by depositional processes and restricted to beds formed by subaerial deposition of wet ash (Wohletz & Sheridan, 1983; Farrand & Singer, 1991, 1992; Wohletz & Heiken, 1992). The cross-cutting palagonite zones that affect the cone, however, indicate that the palagonitization is not systematically related to depositional facies. The distribution and characteristics of Pahvant Butte palagonite and the surficial unpalagonitized layer closely match those of Hawaiian examples (Hay & Iijima, 1968) that are inferred to have formed by ground-water alteration of sideromelane tuffs with-
out regard to specific depositional facies. Moreover, steeply orientated zones of palagonite that locally cut across lower cone deposits show that the process also took place below the lake surface, in more interior regions, almost certainly as a result of hydrothermal activity (Jakobsson & Moore, 1986). Absence of palagonitization in most of the platform deposits (below) probably results, as in Hawaii, from a position below the Pleistocene water table (lake level) and beyond zones of hydrothermal activity, and from the relatively arid post-pluvial climate of the Pahvant Butte area.
PLATFORM The Pahvant Butte platform is asymmetrically arranged about the tuff cone, and its surface lies at ≈ 5050 ft (1540 m) elevation (Figs 2 & 3). No platform is present at the north-west side of the cone, where high, wave-cut cliffs extend to the base of the cone. To the south-east the platform extends over the presumed site of the major vent. An elongate tongue of the platform extends to the south-east along its north-eastern edge, and is capped by a pronounced beach ridge. The elongation of the tongue, the capping beach ridge, and the down-tongue dip of large-scale (> 20 m) foreset beds suggest that the tongue formed as a drift-aligned beach spit (Gilbert, 1890; Carter et al., 1991). Both the large-scale distribution of ash from the eruption (Oviatt & Nash, 1989) and the high northwest–south-east semicircular cone rim suggest winds from the south to south-west during the eruption and growth of the cone. Wave-cut cliffs on the north-west of the cone, and the drift-aligned spit extending to the south-west, however, suggest that waves from the west to north-west were predominant during platform development. This might result from seasonal wind shifts, with strongest winds, and thus maximum wave energy for reworking and platform growth, occurring during winter as storms moved in from the north-west. Late-stage lacustrine erosion, which probably occurred at the 4750 ft (1450 m) ‘Provo’ shoreline (Gilbert, 1890; Oviatt & Nash, 1989), significantly steepened the margins of the platform and in the process truncated the beds forming its lower part. Progradational beach deposits, local delta The platform is partly encircled by a capping beach ridge, and strong drift alignment of the north-eastern tongue (Fig. 3) suggests that this part of the platform
VSIC03 9/12/08 2:06 PM Page 71
Pahvant Butte eruption is a spit complex, formed primarily by downdrift beach progradation (Nielsen et al., 1988). South-west within the breached cone, however, the marginal topset beds show outward-dipping, metre-scale, planar– tabular cross-beds that record progradation of microdeltaic transverse bars at the mouths of feeding streams, and indicate that progradation at that site was not controlled by the direction of beach drift. This south-western part of the platform was abundantly supplied with sediment from the erosion of the cone, and it prograded southward as a delta with comparatively minor downdrift redistribution of the sediment by beach processes. The platform foresets consist of both progradational beach and deltaic deposits, but the subaqueous foresets are identical in both spit and deltaic deposits (see Nielsen et al., 1988), a reflection of the very limited transport and reworking in both beach and delta systems. In recognition of this kinship, the platform is here considered to include the large-scale foreset beds of both deltaic and prograding beach origin. It is topped by deltaic and /or beach (spit) topset deposits (see Nielsen et al., 1988; Chough & Hwang, 1997).
71
Main platform topset Delta The deltaic topset that forms much of the southwestern, and part of the north-eastern, platform is exposed only in scattered outcrops along the platform periphery. Along the western to southern edge it consists of ≈ 2-m-thick, tangentially based planar–tabular cross-beds overlying a few-metres-thick sequence of flat-bedded tuff (Fig. 9), which locally contains ripple marks, rootlet casts, and subhorizontal burrows. The foreset bedding is steep (25–30°), and locally contains lenses of backset beds that typically consist of openwork lapilli. The cross-beds in marginal topset strata generally dip toward the platform margin, but in one, more inward location near the expected location of the now-absent southern wall of the volcanic cone (Fig. 2), foreset strata dip down the axis of a large ravine that evidently follows an earlier subaqueous sediment pathway. Sediment for the delta was presumably provided by small-scale rills and shallow gullies cut into the originally unconsolidated ash
Fig. 9. Platform features: (a) west-dipping deltaic foreset beds (≈ 1 m tall) on south-western platform, hammer for scale; (b) larger-scale deltaic foreset beds dipping outward along south-western gully, fine-grained beds more resistant; R. V. Fisher for scale; (c) typical foreset beds showing subtle lenticularity as a result of shallow scouring, lenses and layers of openwork lapilli, and good general continuity of intercalated medium- to fine-grained ash beds (pencil 15 cm long); (d) close-up of scour infilled by cross-laminae in foreset bed (hammer is 30 cm long).
VSIC03 9/12/08 2:06 PM Page 72
72
J. D. L. White
(Segerstrom, 1950; Ollier & Brown, 1971), but none are preserved; larger gullies may have formed, but are not distinguishable from modern topography on the lithified remnant of the cone. Beach and spit The deltaic topset beds grade eastward into elongate gravel mounds that top two protuberances along the eastern margin of the platform. The ridges are interpreted as beach spits, one that advanced northward around the eastern margin, and another that extended eastward. They are separated by a deep gully that funnelled sediments from the narrow eastern platform down the flank of the platform (Fig. 2). Limited exposure along the upper edges of the spits shows steeply dipping (≈ 25°) beds, 2 –10 cm thick, that dip outward and strike parallel to the platform margin. Within the re-entrant between the spits, thin beds of fine-grained, intensely rippled ash exposed a few metres below the platform margin represent deposition in a protected re-entrant between the prograding spits. Platform foreset The platform foreset is exposed mostly around its periphery. Along the northern edge, cliffs have been cut nearly to the base of the platform, and a > 30-mtall foreset is partially exposed. Two types of beds make up the foreset, as follows. 1 Thick beds (0.5–1.5 m) of coarse ash and lapilli have subtle internal layers defined by discontinuous, gradational to slightly erosive, planar to weakly scoured contacts, and elongate ‘stringers’ of lapilli. The bases of some thick beds show liquefaction features, and isolated cobble-sized clasts are steeply imbricated (see Fig. 9). These beds are similar to those of mound lithofacies M4. 2 Medium beds (1–50 cm) of interlayered coarse- and finer-grained ash have bedding surfaces that are generally flat to slightly undulose, commonly with eroded bases. The beds typically have very low-angle internal cross-stratification defining broad, low bedforms a few centimetres in amplitude and with wavelengths of a metre or so. Multiple foreset units can be recognized along the southern part of the platform. Two upper units are exposed along the southern margins, and each is 10– 15 m high and tangential based, with dips as steep as 25° near its crest (see Fig. 9). This large-scale foreset bedding unit generally dips toward the platform margin, but is locally diverted, in the same sense as
the topset-bed cross-strata, along earlier subaqueous sediment pathways. The upper parts of the foresets are steep (25–30°), and discontinuous layers of openwork lapilli locally form backset beds. The foreset units are separated by locally ripple-laminated topset beds. The upper foreset unit is stacked atop the lower foreset unit (see Discussion). Foreset sedimentary features and sedimentation Bedforms are similar throughout the platform foresets. Broad, low-angle dunes and swales, with subtle internal stratification defined by thin, lenticular lapilli horizons, are characteristic, and indicate ubiquitous erosion and downslope traction transport on the foresets. High-concentration turbidity currents, cohesionless debris flows, and occasional debris falls (Nemec, 1990; White, 1992; Kim et al., 1995) are inferred to have been the primary means of sediment transport down the platform foresets. Bedform suites resulting from these processes may represent both lowerbeachface sediments (Massari & Parea, 1988) and delta-front deposits (Nemec, 1990; Kim et al., 1995), depending on the nature of the sediment delivery to the upper foresets. The Pahvant Butte platform foresets are inferred to have been supplied episodically with sediment, by storm-wave entrainment and downslope density flow (Nielsen et al., 1988), hyperpycnal streamflow (Prior & Bornhold, 1989; Orton & Reading, 1993; Mulder & Syvitski, 1995) resulting from rainfall on the cone, or by slumping of the upper foreset induced by depositional oversteepening or wave loading (Massari, 1984; Postma, 1984; Prior & Bornhold, 1990). Platform contact relationships Although the lower contact of platform strata with underlying and adjacent mound and cone strata is largely obscured, a section exposed in the northern gully (Fig. 2) offers important insight into the nature of the relationship. In the deep gully, a complex relationship between cone, mound and platform strata is revealed (Fig. 10). The lower levels of the gully expose undulose-bedded M3 mound strata. Upward, bedding becomes more complex, with high-angle cross-stratification locally developed among the undulose beds. Slightly more upsection, and somewhat closer to the vent, strong palagonitization sets in at approximately the same level that very complex bedsets of ripple cross-laminated strata, sharp erosional troughs, and an onlapping unit of more steeply
VSIC03 9/12/08 2:06 PM Page 73
Pahvant Butte eruption
73
Fig. 10. Contact relationships from northern gully (see Fig. 2). Schematic, vertically exaggerated diagram (a) summarizes relationship of mound, upper cone, and platform strata. The thick black line separates palagonitized tuff to the left from non-palagonitized tuff and tephra to the right. Photograph (b) looks toward the cone and shows topset–foreset transition near 5100 (1555 m) ft elevation; tuff of the UC4 assemblage in the background dips toward the viewer, and palagonitized tuff of the UC3 assemblage in the foreground is intercalated with rippled syn-eruptive shore-zone platform deposits.
dipping, thin-bedded strata appear. The latter are in turn sharply incised and overlain by a unit of a few metres thickness that consists largely of broken-up bedded tuff blocks. The erosional contact between inferred cone strata
and overlying ripple-laminated and trough crossbedded units strongly suggests that this complex zone represents a syn-eruptive shoreline. Syn-eruptive development is demonstrated by the offlapping relationship of the overlying, also truncated, steeply dipping strata,
VSIC03 9/12/08 2:06 PM Page 74
74
J. D. L. White
which are taken to be part of the UCd lithofacies cone deposits of the youngest depositional assemblage (UC4). At 5120 ft (1560 m) elevation, the topset–foreset break of a deltaic foreset is exposed in the northwestern gully (Fig. 9) at a level higher than the top of the north-eastern beach-spit berm (5085 ft (1550 m)). Unlike the gravelly spit-top berm, the topset beds must have formed subaqueously, and the topset– foreset break is thus a sensitive indicator of water level preserved in what would have been a sheltered reentrant along the nor-thern shore. Because it is several metres higher than the zone of intercalated UC4 and platform deposits that represent a syn-eruptive shoreline, it apparently records a (brief ? ) post-eruptive higher lake stand. The nature of the basal contact between platform strata and underlying mound strata is obscure. Dips of platform foreset strata shallow downwards, and the foresets are inferred to have tangential basal contacts and thus a conformable contact with subhorizontally dipping mound strata. Bonneville lake levels and platform evolution The local stacking relationship of multiple tangentialbased foresets, separated by locally ripple-laminated topset beds (Fig. 10, and see Fig. 11, below), indicates deposition of the observed platform-margin units under rising lake levels (see Chough & Hwang, 1997), recording stages of normal constructional regression (shoreline outbuilding) and rapid, non-accretionary transgression (drowning; Helland-Hansen & Martinsen, 1996). The stacked foreset units record small-scale lake-level changes, and may have formed as Lake Bonneville rose after a brief fall termed the Keg Mountain oscillation (Fig. 12), which is inferred to have taken place not long before initiation of the Bonneville Flood and resultant rapid lowering of the lake to the Provo level (Bills & May, 1987; Sack, 1989; Oviatt et al., 1992, 1994). Local incision and eastward longshore drift occurred along the south-western margin of Pahvant Butte during the fall and subsequent lake-level rise. Foreset beds of the upper foreset unit advance along this gully, indicating that it was cut in part before deposition of the unit. Sediment carried down and deposited at the mouth of this gully forms a poorly exposed but morphologically distinct wedge, clearly illustrated as an eastward-sloping ridge in front of the main platform in Gilbert’s (1890) engraving (top frame of Fig. 3).
Indications from sites other than Pahvant Butte that the Bonneville water level continued to rise after the eruption (Oviatt et al., 1992, 1994) are based on the best estimates of the ages of eruption and catastrophic lake lowering, and on stratigraphy of ash deposits in the Sevier Desert. It is important that the syn-eruptive shoreline features exposed along the northern gully of Pahvant Butte require that the top of the platform was approximately at or several metres below the syn-eruptive lake level. This indicates that the platform originated as a syn-eruptive feature, rather than forming long after eruption by prolonged reworking at the Bonneville highstand. A syn-eruptive origin for the platform (at a lake level somewhat below the Bonneville maximum) reduces a Pahvant Buttecentred 17-m deflection from the regional elevations predicted to result from isostatic rebound that followed removal of the Bonneville lake-water load (Bills & May, 1987). There is no field evidence for ‘post-eruptive volcanic collapse’ (Bills et al., 1994), so several metres of deflection from predicted levels remains unexplained. Overall, the shoreline record at Pahvant Butte supports the curve shown in Fig. 11, with eruption taking place approximately at the Bonneville maximum, followed by lake lowering and renewed rise (Keg Mountain oscillation?) that were associated with stacked platform foresets, and a last brief rise to a maximum recorded by the small deltaic topset–foreset contact before the final lake-level drop caused by the Bonneville flood.
DISCUSSION Even large and deep lakes are commonly short-lived bodies of water (Talbot & Allen, 1996), with less severe wave regimes than most oceanic settings. The result is that volcanoes erupted subaqueously within lakes are commonly preserved and exposed to view with only minor modification (e.g. Gilbert, 1890; Christensen & Gilbert, 1964; Tazieff, 1972). Pahvant Butte is an exceptionally well-preserved example of an emergent volcano (Kokelaar, 1986), erupted at a water depth equivalent to those of typical marine shelf settings. It records a sequence of eruptive and sedimentary events (Fig. 11) typical of that which shapes marine volcanoes of island arcs (Hoffmeister et al., 1929), shallow spreading ridges (Richards, 1959; Thorarinsson, 1967), and coastal continental intraplate settings (Coombs et al., 1986).
VSIC03 9/12/08 2:06 PM Page 75
75
Pahvant Butte eruption (a)
UC4 W
E 5400 c
5000
Late subaerial eruption stage, deposition of UCd 'dry' lithofacies, shoreline reworking and platform drift.
W
E UC3
UC2
5400
UC1
b
5000
Main subaerial eruption stage, multiple slips, deposition of UCw 'wet' lithofacies, shoreline erosion.
W
E 5400 a
5000
LC Subaquaeous development of Pahvant Butte; deposition of mound and some lower cone deposits.
Fig. 11. Summary illustration of Pahvant Butte’s history in Lake Bonneville: (a) Eruption begins under ≈ 85 m of water, producing a low mound of tephra before eruptive emergence (White, 1996), and development of M2 and lower cone deposits (depending on nature of eruptive column) at vent margins. Multiple vent sites are inferred during this stage. (b) The volcano emerges above lake level, but full access of water to the eruption site continues, feeding UCw ‘wet’ phreatomagmatic deposits fed largely by tephra jets (cock’s tail plumes). Parts of the cone fail along broad slip surfaces, some soling deeply enough to displace mound and lower cone deposits. Slip surfaces are mantled and onlapped by continuing tephra jet deposition. (c) In the late stages of eruption, lessened access of water to the vent is indicated by development of UCd ‘dry’ phreatomagmatic deposits, fed by grainfall and deposition from the base of expanded density currents carrying little or no condensed water. (d) This and later frames show only the eastern spit area (box at top; Figs 2 & 3). Early spit growth occurs at the syn-eruption lake level, probably very rapidly. (e) Following the close of eruption and early platform building, lake level drops during the Keg Mountain oscillation. A topset unit develops at a lower elevation than the syn-eruptive lake level. ( f ) Renewed rise of lake level in the latter part of the oscillation causes backstepping of the topset–foreset series, with stacking of additional foreset–topset pairs resulting from continued deposition of drift and stream-supplied tephra. (g) Highest lake level is recorded by topset–foreset break of small spit illustrated in Fig. 9. The Bonneville flood takes place as the outlet gives way and is eroded during an immense outbreak flood, which lowers the lake to the Provo level (Oviatt et al., 1992). continued on p. 76
VSIC03 9/12/08 2:06 PM Page 76
76
J. D. L. White
Fig. 11. (continued )
Eruption, emergence and construction of Pahvant Butte volcano The Pahvant Butte eruption began beneath ≈ 85 m of water (Oviatt & Nash, 1989), and gradually built a broad mound of subaqueous tephra toward the lake surface (White, 1996). Immediately adjacent to vent sites, M2 mound facies accumulated during continuousuprush eruptive phases (Kokelaar, 1983; White, 1996) and grade upward into steeply dipping beds of a
subaqueous cone rim (lower cone facies). The cones were fed by emergent tephra jets, which penetrated the water surface and fell back, carrying ash, lapilli and bombs directly to the lake surface to feed fallavalanche grainflows (Sohn & Chough, 1993). Vent sites were not fixed during this stage and, as active venting shifted, M2 and lower cone accumulations were onlapped by other subaqueous lithofacies. Wave erosion during this time, although limited in comparison with overnight shoreline retreat of ≈ 100 m at
;
VSIC03 9/12/08 2:06 PM Page 77
77
Altitude (feet)
Pahvant Butte eruption
5100'
Bonneville highstand
5000'
Pahvant Butte eruption
4900'
Keg Mountain oscillation
20
15
5000'
4800'
Altitude (feet)
Fig. 12. Diagram showing changes in level of Lake Bonneville with time. Altitudes are in feet above sea level. (Note eruption of Pahvant Butte ash just before Lake Bonneville attained its outlet-controlled highstand level, and later variations including the significant drop and rise of the Keg Mountain oscillation (Oviatt et al., 1992) before the catastrophic Bonneville Flood permanently lowered the lake by erosion of the outlet level to the so-called Provo shoreline level (prominent step c. 4750 ft (1448 m) ).)
Provo shoreline
Height of Lake Bonneville from 30 ka to ~ 10 ka
4600'
4400'
Pahvant Butte eruption
4200'
30
Surtsey (Thorarinsson, 1967), simultaneously provided material to build the platform by littoral drift. Development of the lower cone and platform gradually increased the area of the volcano lying in shallow water, and upper cone facies began to form a subaerial tuff cone. The early deposits are predominantly of UCw facies, with abundant evidence of free water in the tephra upon deposition. This indicates that water continued to reach the vent interior in sufficient quantities to maintain a ponded Surtseyan slurry (Kokelaar, 1986) generating repeated tephra jets. Emplacement of water-laden tephra, associated with rills and failure of saturated deposits to form debris flows, is inferred to have occurred largely as direct fall from tephra jets, such as those described by Thorarinsson (1967; p. 18) to have ‘ejected so much sea water over the crater rims that mud streams ran all the way down to the beach’. Thin, continuous ash beds, however, probably reflect deposition from damp, low-energy, wind-directed density currents of the sort experienced and described by Richards (1959; p. 106): ‘a large black-appearing
25
20
15
10
Age (103 yrs B.P.)
cloud poured over the graben area of the crater rim, rushed down the cone . . . particles ranged from 0.1 to 3 mm . . . dust, ash, and water mixed to form a light rain of large muddy drops, somewhat like a hailstorm . . . speed of the avalanche was about 30 knots or more.’ Efficient distribution of ejecta by energetic base surges, such as occurred at Taal in 1965 (Moore et al., 1966; Waters & Fisher, 1971), did not occur, with the result that a cone developed closely enclosing the vent area (Sohn, 1996). As the Pahvant Butte eruption continued, the cone suffered slips of various magnitudes, the larger ones forming mappable discontinuities. Slip surfaces separate units UC1, UC2, and UC3 (Fig. 3), and all slips involved material entirely or predominantly of lithofacies UCw. Repeated failure along inward-dipping slips was facilitated by ejection of material from the vent area; this removed support from the base of the inner cone walls and caused material to slump into the vent, where it disaggregated and was recycled by subsequent eruptive bursts (Kokelaar, 1983). Major
VSIC03 9/12/08 2:06 PM Page 78
78
J. D. L. White
slips along outward-dipping surfaces are inferred to have taken place as a result of wave erosion along the shoreline of the emergent volcano; slips resulting from wave erosion at Surtsey progressively truncated the highest part of the cone (Thorarinsson, 1967; p. 28). As waves truncated the outer flank of the Pahvant Butte cone and planed off adjoining parts of the underlying tephra platform, tephra was reworked to extend the platform outward. Both the cone and the platform margin were thus oversteepened, resulting in slips that involved both cone and platform strata. Such syn-eruptive outward slips are likely to typify emergent tuff cones; not only do they have shorelines exposed to waves, but the uncompacted subaqueous tephra upon which these cones are constructed is generally unstable (Skilling, 1994; Smellie & Hole, 1997). In addition to slip failure, the mound and cone were subject to major syn- and early post-eruptive reworking and redistribution, as is typical of subaqueously formed tephra edifices (Godcheaux et al., 1992; Skilling, 1994; Smellie & Hole, 1997). The last phases of the eruption seem to have produced relatively ‘dry’ deposits (lithofacies UCd; unit UC4, Figs 2 & 3) that show no signs of having contained water upon deposition. Lithofacies UCd was deposited from lowenergy density currents by traction, and as deposits of avalanches and fall-fed grainflows (Sohn & Chough, 1993). These currents crossed the northern syn-eruptive shoreline, producing complex interfingering relationships with platform strata (Fig. 10). The nature of this sort of dry, low-energy density current at Surtsey was revealed by Thorarinsson (1967; p. 23): ‘After each large explosion and the following bomb shower, a brown pumice-laden cloud enveloped . . . clouds were warm and cosy, the pumice grains being so light that they did not hurt . . . a peculiar circular motion whisked the pumice grains from one side to another.’ A critical aspect of this account is that the ballistic showers were followed by descending ‘clouds’ of dry ash. The phenomenon is analogous to tephra-finger jets in the simultaneous ejection of large blocks that pass through associated ash that subsequently follows the blocks to the ground, yet condensed water droplets are absent. It is inferred that the dry ash follows the ballistic blocks as weak vertical density flows (convective settling) rather than sedimenting as by fall of individual grains, and that this produces the high particle-delivery rates needed to form fall-fed grainflows such as those represented by UCd deposits at Pahvant Butte. Continued tephra redistribution following the eruption further built out the flat-topped platform,
which surrounds and offlaps the cone and mound facies. A combination of products from surficial cone erosion and tephra that drifted around the volcano from wave-cut cliffs to the west provided the material for platform construction. Despite an extremely limited hinterland, Gilbert-type deltaic topset deposits formed locally from sediment carried off the cone in rills and shallow gullies that are no longer preserved. The topsets are related to a large foreset–clinoform sequence fed both by deltaic and beach drift processes, each shedding sediment episodically to deeper levels via sediment-gravity flows. Characteristics of these platform foresets, including relatively coarse grain size, loose packing, abundant debris-flow deposits and variable foreset dips, reflect the adjacent highrelief source of unconsolidated sand and gravel-grade tephra. Because the tephra produced by the Pahvant Butte eruption is in general highly vesicular, the grains have low densities and are easily transported relative to dense siliciclastic grains (Smith & Smith, 1985; Oehmig & Wallrabe-Adams, 1991). This resedimentation of coarse tephra could take place under even moderate-energy wave and current conditions.
ACKNOWLEDGEMENTS Funding for this study was provided by NSF grant EAR91-05456 to R. V. Fisher, whose work on Pahvant Butte in the 1960s provided impetus and background for this study. Alec Tsongas and Stephanie Petralia assisted in the field, and Chuck Landis, Gary Smith, and Peter Kokelaar provided helpful, informative and stimulating reviews of the manuscript.
REFERENCES Bills, B.G. & May, G.M. (1987) Lake Bonneville: constraints on lithospheric thickness and upper mantle viscosity from isostatic warping of Bonneville, Provo and Gilbert Stage shorelines. J. geophys. Res., 92, 11493 –11508. Bills, B.G., Currey, D.R. & Marshall, G.A. (1994) Viscosity estimates for the crust and upper mantle from patterns of lacustrine shoreline deformation in the Eastern Great Basin. J. geophys. Res., 99, 22059– 22086. Buck, S.G. (1985) Sand-flow cross strata in tidal sands of the Lower Greensand (Early Cretaceous), southern England. J. sediment. Petrol., 55, 895– 906. Carter, R.M., Abbott, S.T., Fulthorpe, C.S., Haywick, D.W. & Henderson, R.A. (1991) Application of global sea-level and sequence-stratigraphic models in Southern Hemisphere Neogene strata from New Zealand. In: Sedimentation, Tectonics and Eustasy: Sea-Level Changes at
VSIC03 9/12/08 2:06 PM Page 79
Pahvant Butte eruption Active Margins (Ed. Macdonald, D.), Spec. Publs int. Assoc. Sediment., No. 12, pp. 41– 65. Blackwell Scientific Publications, Oxford. Cas, R.A.F. & Landis, C.A. (1987) A debris-flow deposit with multiple plug-flow channels and associated slide accretion deposits. Sedimentology, 34, 901– 910. Cas, R.A.F., Landis, C.A. & Fordyce, R.E. (1989) A monogenetic, Surtla-type, Surtseyan volcano from the Eocene – Oligocene Waiareka–Deborah volcanics, Otago, New Zealand: a model. Bull. Volcanol., 51, 281– 298. Chough, S.K. & Hwang, I.G. (1997) The Duksung fan delta, SE Korea: growth of delta lobes on a Gilbert-type topset in response to relative sea-level rise. J. sediment. Res., 67, 725 –739. Christensen, M.N. & Gilbert, C.M. (1964) Basaltic cones suggests constructional origin of some guyots. Science, 143, 240–242. Condie, K.C. & Barsky, C.K. (1972) Origin of Quaternary basalts from the Black Rock Desert region, Utah. Geol. Soc. Am. Bull., 83, 333 –352. Coombs, D.S., Cas, R.A.F., Kawachi, Y., Landis, C.A., McDonough, W.F. & Reay, A. (1986) Cenozoic volcanism in north, east and central Otago. In: Cenozoic Volcanism in New Zealand (Ed. Smith, I.E.M.), Bull. R. Soc. N.Z., 23, 278–312. Wellington. Crowe, B.M. & Fisher, R.V. (1973) Sedimentary structures in base-surge deposits with special reference to crossbedding, Ubehebe Craters, Death Valley, California. Geol. Soc. Am. Bull., 84, 663– 682. Currey, D.R. (1982) Lake Bonneville: selected features of relevance to neotectonic analysis. Open-file Rep. 82-1070. US geol. Surv., Denver, CO. Farrand, W.H. & Singer, R.B. (1991) Spectral analysis and mapping of palagonite tuffs of Pavant Butte, Millard County, Utah. Geophys. Res. Lett., 18, 2237– 2240. Farrand, W.H. & Singer, R.B. (1992) Alteration of hydrovolcanic basaltic ash: observations with visible and nearinfrared spectrometry. J. geophys. Res., 97, 17393 –17408. Fisher, R.V. (1977) Erosion by volcanic base-surge density currents: U-shaped channels. Geol. Soc. Am. Bull., 88, 1287–1297. Fisher, R.V. & Schmincke, H.-U. (1984) Pyroclastic Rocks. Springer, Berlin. Gilbert, G.K. (1890) Lake Bonneville. Monogr. US geol. Surv., Reston, VA, 1. Godcheaux, M.M., Bonnichsen, B. & Jenks, J.D. (1992) Types of phreatomagmatic volcanoes in the western Snake River Plain, Idaho, USA. J. Volcanol. geothermal Res., 52, 1–25. Hamilton, W.H. & Myers, W.B. (1963) Menan Buttes, cones of glassy basalt tuff in the Snake River Plain, Idaho. US geol. Surv. Prof. Pap., Reston, VA, 450-E, 114 –118. Hay, R.L. & Iijima, A. (1968) Nature and origin of palagonite tuffs of the Honolulu Group on Oahu, Hawaii. Geol. Soc. Am. Mem., 116, 331–376. Heiken, G.H. (1971) Tuff rings: examples from the Fort Rock–Christmas Lake Valley basin, south–central Oregon. J. geophys. Res., 76, 5615–5626. Helland-Hansen, W. & Martinsen, O.J. (1996) Shoreline trajectories and sequences: description of variable depositional-dip scenarios. J. sediment. Res., 66, 670 – 688. Hoffmeister, J.E., Ladd, H.S. & Alling, H.L. (1929) Falcon Island. Am. J. Sci., 18, 461– 471.
79
Houghton, B.F. & Wilson, C.J.N. (1989) A vesicularity index for pyroclastic deposits. Bull. Volcanol., 51, 451–462. Jakobsson, S.P. & Moore, J.G. (1986) Hydrothermal minerals and alteration rates at Surtsey volcano, Iceland. Geol. Soc. Am. Bull., 97, 648– 659. Kano, K. (1998) A shallow-marine alkali-basalt tuff cone in the Middle Miocene Jinzai Formation, Izumo, SW Japan. J. Volcanol. geothermal Res., 87, 173–191. Kim, S.B., Chough, S.K. & Chun, S.S. (1995) Bouldery deposits in the lowermost part of the Cretaceous Kyokpori Formation, SW Korea: cohesionless debris flows and debris falls on a steep-gradient delta slope. Sediment. Geol., 98, 97–119. Kneller, B.C. & Branney, M.J. (1995) Sustained highdensity turbidity currents and the deposition of thick massive sands. Sedimentology, 42, 607– 616. Kokelaar, B.P. (1983) The mechanism of Surtseyan volcanism. J. geol. Soc. London, 140, 939–944. Kokelaar, B.P. (1986) Magma–water interactions in subaqueous and emergent basaltic volcanism. Bull. Volcanol., 48, 275–289. Leys, C.A. (1983) Volcanic and sedimentary processes during formation of the Saefell tuff-ring, Iceland. Trans. R. Soc. Edinburgh: Earth Sci., 74, 15 –22. Lorenz, V. (1974) Studies of the Surtsey tephra deposits. Surtsey Research Progress Report 7, 72 –79. Surtsey Research Society, Reykjavik. Massari, F. (1984) Resedimented conglomerates of a Miocene fan-delta complex, southern Alps, Italy. In: Sedimentology of Gravels and Conglomerates (Eds Koster, E.H. & Steel, R.J.), Mem. Can. Soc. petrol. Geol., Calgary, 10, 259–277. Massari, F. & Parea, G.C. (1988) Progradational gravel beach sequences in a moderate- to high-energy, microtidal marine environment. Sedimentology, 35, 881–913. Moore, J.G., Nakamura, K. & Alcaraz, A. (1966) The 1965 eruption of Taal Volcano. Science, 151, 955 –960. Mulder, T. & Syvitski, J.P.M. (1995) Turbidity currents generated at river mouths during exceptional discharges to the world oceans. J. Geol., 103, 285 – 299. Nemec, W. (1990) Aspects of sediment movement on steep delta slopes. In: Coarse-Grained Deltas (Eds Colella, A. & Prior, D.B.), Spec. Publs int. Assoc. Sediment., No. 10, pp. 29–73. Blackwell Scientific Publications, Oxford. Nielsen, L.H., Johannessen, P.N. & Surlyk, F. (1988) A late Pleistocene coarse-grained spit–platform sequence in northern Jylland, Denmark. Sedimentology, 35, 915–937. Oehmig, R. & Wallrabe-Adams, H.-J. (1991) Hydrodynamic properties and grain-size characteristics of volcaniclastic deposits on the mid-Atlantic ridge north of Iceland (Kolbeinsey Ridge). J. sediment. Petrol., 63, 140 – 151. Ollier, C.D. & Brown, M.J.F. (1971) Erosion of a young volcano in New Guinea. Z. Geomorphol., 15, 12 –28. Orton, G.J. & Reading, H.G. (1993) Variability of deltaic processes in terms of sediment supply, with particular emphasis on grain size. Sedimentology, 40, 475–512. Oviatt, C.G. & Nash, W.P. (1989) Late Pleistocene basaltic ash and volcanic eruptions in the Bonneville basin, Utah. Geol. Soc. Am. Bull., 101, 292 –303. Oviatt, C.G., Currey, D.R. & Sack, D. (1992) Radiocarbon chronology of Lake Bonneville, Eastern Great Basin, USA. Palaeogeogr. Palaeoclimatol. Palaeoecol., 99, 225–241.
VSIC03 9/12/08 2:06 PM Page 80
80
J. D. L. White
Oviatt, C.G., McCoy, W.D. & Nash, W.P. (1994) Sequence stratigraphy of lacustrine deposits: a Quaternary example from the Bonneville basin, Utah. Geol. Soc. Am. Bull., 106, 133–144. Postma, G. (1984) Mass-flow conglomerates in a submarine canyon: Abrioja fan-delta, Pliocene, southeast Spain. In: Sedimentology of Gravels and Conglomerates (Eds Koster, E.H. & Steel, R.J.), Mem. Can. Soc. petrol. Geol., Calgary, 10, 237– 258. Prior, D.B. & Bornhold, B.D. (1989) Submarine sedimentation on a developing Holocene fan delta. Sedimentology, 36, 1053–1076. Prior, D.B. & Bornhold, B.D. (1990) The underwater development of Holocene fan deltas. In: Coarse-Grained Deltas (Eds Colella, A. & Prior, D.B.), Spec. Publs int. Assoc. Sediment., No. 10, pp. 75–90. Blackwell Scientific Publications, Oxford. Richards, A.F. (1959) Geology of the Islas Revillagigedo, Mexico, 1. Birth and development of Volcan Barcena, Isla San Benedicto. Bull. Volcanol., 22, 73–123. Richards, A.F. (1965) Geology of the Islas Revillagigedo, 3. Effects of erosion on Isla San Benedicto 1952– 61 following the birth of Volcan Barcena. Bull. Volcanol., 28, 381– 403. Sack, D. (1989) Reconstructing the chronology of Lake Bonneville: an historical review. In: History of Geomorphology: from Hutton to Hack (Ed. Tinkler, K.J.), Binghamton Symposia in Geomorphology, International Series, 19, 223–256. Segerstrom, K. (1950) Erosion studies at Paricutin volcano, State of Michoacan, Mexico. Bull. US geol. Surv., Denver, CO, 956-A. Skilling, I.P. (1994) Evolution of an englacial volcano: Brown Bluff, Antarctica. Bull. Volcanol., 56, 573–591. Smellie, J.L. & Hole, M.J. (1997) Products and processes in Pliocene–Recent, subaqueous to emergent volcanism in the Antarctic Peninsula: examples of englacial Surtseyan volcano construction. Bull. Volcanol., 58, 628– 646. Smith, G.A. & Smith, R.D. (1985) Specific gravity characteristics of recent volcaniclastic sediment: implications for sorting and grain size analysis. J. Geol., 93, 619– 622. Sohn, Y.K. (1996) Hydrovolcanic processes forming basaltic tuff rings and cones on Cheju Island, Korea. Geol. Soc. Am. Bull., 108, 1199–1211.
Sohn, Y.K. & Chough, S.K. (1989) Depositional processes of the Suwolbong tuff ring, Cheju Island (Korea). Sedimentology, 36, 837– 855. Sohn, Y.K. & Chough, S.K. (1992) The Ilchulbong tuff cone, Cheju Island, South Korea: depositional processes and evolution of an emergent, Surtseyan-type tuff cone. Sedimentology, 39, 523 –544. Sohn, Y.K. & Chough, S.K. (1993) The Udo Tuff Cone, Cheju Island, South Koreaatransformation of pyroclastic fall into debris fall and grain flow on a steep volcanic cone slope. Sedimentology, 40, 769–786. Talbot, M.R. & Allen, P.A. (1996) Lakes. In: Sedimentary Environments: Processes, Facies and Stratigraphy (Ed. Reading, H.G.), pp. 83–124. Blackwell Science, Oxford. Tazieff, H. (1972) About deep-sea volcanism. Geol. Rundsch., 61, 470 – 480. Thorarinsson, S. (1967) Surtsey: the New Island in the North Atlantic. Viking Press, New York. Verwoerd, W.J. & Chevallier, L. (1987) Contrasting types of Surtseyan tuff cones on Marion and Prince Edward Islands, southwest Indian Ocean. Bull. Volcanol., 49, 399– 414. Waters, A.C. & Fisher, R.V. (1971) Base surges and their deposits: Capelinhos and Taal volcanoes. J. geophys. Res., 76, 5596 –5614. Werner, R., Schmincke, H.-U. & Sigvaldason, G. (1996) A new model for the evolution of table mountains: volcanological and petrological evidence from the Herdubreid and Herdubreidarotgl volcanoes (Iceland). Geol. Rundsch., 85, 390 –397. White, J.D.L. (1992) Pliocene subaqueous fans and Gilberttype deltas in maar crater lakes, Hopi Buttes, Navajo Nation (Arizona), USA. Sedimentology, 39, 931– 946. White, J.D.L. (1996) Pre-emergent construction of a lacustrine basaltic volcano, Pahvant Butte, Utah (USA). Bull. Volcanol., 58, 249–262. Wohletz, K. & Heiken, G. (1992) Volcanology and Geothermal Energy. University of California Press, Berkeley. Wohletz, K.H. & Sheridan, M.F. (1983) Hydrovolcanic explosions II. Evolution of basaltic tuff rings and tuff cones. Am. J. Sci., 283, 384 – 413.
VSIC04 9/12/08 2:05 PM Page 81
Sedimentation and re-sedimentation of pyroclastic debris in lakes
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
VSIC04 9/12/08 2:05 PM Page 83
Spec. Publs int. Ass. Sediment. (2001) 30, 83 –108
Influence of magmatism and tectonics on sedimentation in an extensional lake basin: the Upper Devonian Bunga Beds, Boyd Volcanic Complex, south-eastern Australia R. A. F. CAS, C. EDGAR, R. L. ALLEN*, S. BULL†, B . A . C L I F F O R D ‡ , G . G I O R D A N O § and J . V . W R I G H T ¶ Department of Earth Sciences, Monash University, Clayton, Vic. 3168, Australia
ABSTRACT The Bunga Beds outlier of the Upper Devonian Boyd Volcanic Complex, south-eastern Australia, preserves a shallow- to relatively deep-water terrigenous sedimentary rock succession, associated syn-depositional rhyolites, minor basalts, and a variety of volcaniclastic facies, all interpreted to have formed in an extensional lake basin. Basin-margin facies associations are dominated by grey mudstones, with interbedded terrigenous and volcaniclastic sandstones and conglomerates, but facies intervals dominated by tractional sedimentary structures are limited. Local horizons are significantly bioturbated. These facies associations are interpreted as fan-delta (southern association) and delta (northern association) successions. They display initial abrupt (southern association) and gradual (northern association) upward fining representing trangression, then coarsening (representing progradation of deltas) and finally fining (deepening). They pass upwards and laterally into the deep-water basin-centre facies association, which is dominated by a classic turbidite facies association, including deep-water, black, pyritic shales, graded turbidites, slump and slide deposits and debris-flow deposits. The rhyolites were high-level, syn-depositional cryptodomes and partly emergent basin-floor domes; minor dykes also occur. The basalts are exclusively dykes and irregular intrusions, some with quenchfragmented and peperitic margins. Two groups of rhyolites are identified: a crystal-poor (< 12% crystals) group, which is commonly quench-fragmented, and a relatively crystal-rich variety (12 – 40% crystals), which is not quench-fragmented. Massive, unbedded, jigsaw-fit to clast-rotated hyaloclastite and autobreccia are common facies of the rhyolites, and peperite margins also occur. Because bedded, resedimented, dome margin autoclastic deposits are a minor facies, the majority of the rhyolites were shallow, syndepositional cryptodomes. Dome-top intralacustrine explosive activity produced two preserved, bedded, dome-top tuff cone successions dominated by diffusely bedded, water-settled fall, crystal-tuffs, and pumice lapilli-stones. The provenance of the basin-fill succession is complex. Sandstones and conglomerates in both the basinmargin and basin-centre facies associations contain dominantly basement-derived detritus, including plutonic and vein quartz, and metasedimentary lithic fragments, representing an extralacustrine terrigenous epiclastic provenance. Varying fractions of variably reworked volcanic detritus, including quartz, plagioclase and lithic debris, are mixed with the terrigenous sediments and represent an extralacustrine volcanic epiclastic provenance. One resedimented facies, however, consists of cuspate shards, and another of basaltic scoria suggesting an extralacustrine syn-eruptive resedimented pyroclastic provenance. Hyaloclastites, autobreccias and lava-associated peperites are attributed to an intralacustrine volcanic autoclastic provenance, whereas the intrusion-associated peperites have an intralacustrine syn-depositional intrusive autoclastic provenance. The tuff and pumice cone facies have an intralacustrine pyroclastic provenance. Resedimented autoclastic facies have an intralacustrine syn-eruptive resedimented autoclastic provenance, whereas resedimented pyroclastic debris has an intralacustrine syn-eruptive resedimented pyroclastic provenance. *Present address: Volcanic Resources Limited, Morteveien 57, 4085 Hundvåg, Stavanger, Norway. †Present address: CODES Department of Geology, University of Tasmania, GPO Box 252c, Hobart, Tas. 7001, Australia. ‡Present address: Normandy Ltd, Tennant Creek, N.T. 0860, Australia. §Present address: Dipartemento di Scienze Geologiche, Universita di Roma III, Largo S. Muraldo I, 00149, Roma, Italy. ¶Present address: Hoogstraat 50, 1381 VV Weesp, Netherlands. Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
83
VSIC04 9/12/08 2:05 PM Page 84
84
R. A. F. Cas et al.
INTRODUCTION The Bunga Beds outlier of the Upper Devonian Boyd Volcanic Complex (BVC) of south-eastern Australia provides an opportunity to assess the relationships between sedimentation, magmatism and tectonics in a lake basin of extensional tectonic origin. Although the original extent of this basin is not preserved, the excellent coastal exposure preserves a cross-section of the basin-fill, including two basin-margin sedimentary domains, an intervening basin-centre sedimentary facies domain and the products of syn-depositional bimodal rhyolite–basalt magmatism. This study documents the relationship between volcanism, sedimentation, tectonics and a complex array of sediment provenances on lacustrine sedimentation processes in an extensional basin.
GEOLOGICAL SETTING AND AGE The Upper Devonian Boyd Volcanic Complex (Fergusson et al., 1979; Cas & Bull, 1993) is a north– south-trending belt of bimodal rhyolitic and basaltic volcanic rocks and associated sedimentary rocks exposed along the south coast of New South Wales, Australia (Fig. 1), in the eastern margin of the Cambrian to Lower Carboniferous Lachlan Fold Belt of south-eastern Australia. The BVC forms the southernmost part of the regionally extensive, north–south-trending Yalwal–Comerong–Eden Rift Zone (McIlveen, 1974), which was part of the more widespread Basin-and-Range-like continental extensional terrane of the Lachlan Fold Belt in the Late Devonian (Cas, 1983). This generally extensional terrane formed during the late stages of, and after, the compressional Midde Devonian Tabberabberan Orogeny (O’Halloran & Cas, 1995), which terminated most marine sedimentation in the Lachlan Fold Belt before Middle Devonian time (Cas, 1983). The BVC was in turn moderately deformed and metamorphosed to lower greenschist facies during the Early Carboniferous Kanimblan Orogeny, which terminated the active geological history of the Lachlan Fold Belt. The BVC unconformably overlies, or is in faulted contact with, the Ordovician metasedimentary turbidite succession of the Mallacoota Beds and the Lower Devonian Bega Batholith (Fig. 1). In the main Fig. 1. (right ) Regional geological setting of the Boyd Volcanic Complex. (a) Location within Australia. (b) Regional geology of the south coast of New South Wales. Boxed area represents map area of Fig. 2.
VSIC04 9/12/08 2:05 PM Page 85
Sedimentation in an extensional basin
85
Fig. 2. Geological map of the coastal exposure of the Bunga Beds outlier extending from Goalen Head in the north to Picnic Point in the south. (See Fig. 1 for location.) Letters A and B refer to section lines in Fig. 11. Rose diagrams summarize available palaeocurrent data at localities along the coastal exposure, numbers beside circles corresponding to circled locality numbers along coastline.
VSIC04 9/12/08 2:05 PM Page 86
86
R. A. F. Cas et al.
outcrop belt around Eden and Merimbula (Fig. 1), the BVC consists of coherent rhyolite lavas, intrusions and ignimbrites, basaltic lavas, intrusions and localized tuff cone successions, and intercalated fluvial and lacustrine sedimentary rocks (Fergusson et al., 1979; Cas & Bull, 1993). It passes upwards into the Merimbula Group, a continental red-bed fluvial sedimentary succession with minor marine intercalations (Steiner, 1972, 1975; Fergusson et al., 1979; Fig. 1). The rhyolites of the rift zone are relatively alkaline potassic rhyolites (Dadd, 1992), whereas the mafic volcanic rocks comprise quartz normative and olivine normative tholeiites (Dadd, 1984). The Bunga Beds outlier of the BVC (Figs 1 & 2), also consists of rhyolites, basalts and sedimentary rocks, but lacks continental red beds and ignimbrites (Scott, 1972; Fergusson et al., 1979; Cas & Bull, 1993; Edgar, 1996). Rhyolites predominate over basalts, and both are represented by coherent bodies and a diversity of volcaniclastic rocks. Fossil fish and plant fragments indicate a late Givetian to early Frasnian age (latest Middle to earliest Late Devonian; Fergusson et al., 1979) and a lacustrine to brackish-water environment (J. Long, personal communication). Short-lived marine incursions in the Upper Devonian continental red-bed successions of the Lachlan Fold Belt are marked by shallow marine, shelly, fossiliferous, coarse, clastic deposits and green mudstones. However, the Bunga Beds are very different, do not contain marine invertebrates and are considered to be of lacustrine origin. Outcrop of the Bunga Beds is poor except along the coast, where cliffs up to 80 m high and rock platforms provide nearly continuous exposure for 8 km. The Bunga Beds extend only 2 –3 km inland, and an unknown part of the original basin has been removed by coastal erosion. The size and extent of the original basin are therefore unknown.
TERMINOLOGY The use of terminology for the volcaniclastic facies follows the suggestions of Cas & Wright (1987), Cas (1990, 1991) and McPhie et al. (1993). Unless the facies contain clear characteristics of both pyroclastic frag-
mentation and final deposition by primary pyroclastic processes (fallout, pyroclastic flow, pyroclastic surge) terminology commonly applied to primary pyroclastic deposits (e.g. tuff, lapilli-tuff, agglomerate, etc.) is not used. The final facies depositional characteristics are used as the primary guide in facies nomenclature. Similarly, in assessing the provenance of different lithofacies, distinctions are made between volcaniclastic rocks that are deposited syn-eruptively by primary volcanic processes, those that are redeposited syn-eruptively by surface sedimentary processes, and those that are derived from older sources and variably reworked or resedimented by sedimentary mass-flow processes, which are called epiclastic. The first and second groups will generally contain a homogeneous, first-cycle juvenile clast component but will be distinguished by their depositional facies characteristics. The third category will generally contain input of detritus from several sources, will show signs of textural modification, and is therefore considered to be epiclastic.
SEDIMENTARY FACIES OF THE BUNGA BEDS AND THEIR SIGNIFICANCE FOR ERUPTION CONDITIONS AND SETTING Two main sedimentary facies associations occur in the Bunga Beds. A basin-margin facies association extends southward from Goalen Head and also crops out northwards from Picnic Point (Fig. 2). The terms ‘northern’ and ‘southern’ basin margins are used to denote the position of these two facies belts along the present coastal exposures, not the original geographical position of the host environments relative to basin geometry, because the original basin geometry is not preserved. The basin-centre facies association, with which most of the rhyolites are associated, crops out between these two belts (Fig. 2). Basin-margin facies association Northern basin-margin facies association The northern basin-margin facies association (Figs 2–4) consists of interbedded sandstones and grey
Fig. 3. (opposite) (a) Generalized measured section through the northern basin-margin facies association (see Fig. 2 for location). Note the initial upward-fining, then -coarsening, and then -fining grain-size trends marked by the arrows. (b) Selected measured sections of distinctive facies intervals of the northern basin-margin facies association exposed along Bunga Beach. Sections are arranged in stratigraphic order, oldest at the left (northernmost), youngest at the right (southernmost), and correlate with the identified intervals in (a). Column (i), bioturbated sandstone facies; column (ii), tabular, laminated to crosslaminated vitric siltstone and fine sandstone facies; column (iii), tabular, massive to laminated to cross-stratified sandstone facies with interbedded mudstone; column (iv), cross-bedded sandstone facies; columns (v) and (vi), grey mudstone facies with interbedded sandstones and siltstone. ( Note the variations in scale and palaeocurrent directions.)
VSIC04 9/12/08 2:05 PM Page 87
Sedimentation in an extensional basin
87
VSIC04 9/12/08 2:05 PM Page 88
88
R. A. F. Cas et al.
(a)
(b)
(d)
(e)
(c) Fig. 4. Selected facies characteristics of the northern basin-margin facies association: (a) bedding in massive to laminated grey mudstone, Bunga Beach; (b) fossil Lepidodendron sp. tree log in mudstone facies, Bunga Beach; (c) tabular, laminated to crosslaminated vitric siltstone –fine sandstone showing grading, Bunga Beach; (d) tabular, massive to laminated to cross-stratified sandstone facies showing grading, Bunga Beach; (e) cross-bedded sandstone facies, Bunga Beach.
mudstones, is in excess of 100 m thick, is openly folded, and is intruded by contemporaneous rhyolites, basalt dykes, the Jurassic gabbroic diorite at Goalen Head, and a (?)Jurassic microdiorite dyke (Fig. 2). The top appears to be conformably overlain by about
30–40 m of black mudstones and sandstones of the basin-centre facies association. The most pervasive facies is a grey, massive to laminated mudstone and siltstone facies, which represents the ambient conditions in the depositional
VSIC04 9/12/08 2:05 PM Page 89
Sedimentation in an extensional basin environment. Although intervals of this facies are up to 20 m thick, beds are rarely more than 50 cm thick and are interbedded with other facies (Figs 3 & 4a). The mudstone is massive to faintly laminated whereas intercalated siltstone laminae and thin beds are planarlaminated, variably graded, undulating to symmetrical ripple-laminated, or cross-laminated. Strata are tabular except in the upper part of the succession where, together with other facies of the association, this facies forms channel-like bodies, or wedging, faultbounded slide bodies. Local intraformational slump folds also occur. The channel-like structures have low relief, and their bases are downwarps of underlying strata, not erosional. Infilling beds are thickest in the axis, thin on to the margins, and upwards become planar and more uniform in their thickness. Horizons containing short subvertical to horizontal burrows a few millimetres in length (?Chondrites), bedding surface grazing trails, and carbonaceous plant fragments, including large logs of Lepidodendron sp., are found (Fig. 4b). The mudstones consist mostly of sericite and silt-sized quartz, indicating a terrigenous origin. Fossil soils, rootlet horizons and mudcracks are lacking. The depositional surface was generally flat except for the occasional channel-like structures, which are interpreted as syn-depositional down-sags of the depositional surface. The grey mudstones were probably the deposits of a density-stratified, clay and fine silt-rich water column, originating from overflows or underflows from the mouths of rivers in flood (e.g. Sturm & Matter, 1978). Laminae indicate semipersistent current flow, whereas undulating to symmetrical rippled beds indicate the occasional influence of wave activity. The grey mudstones suggest moderately oxidizing conditions compared with the anoxic black mudstones of the basin-centre association (see below). A deltaic setting, with facies corresponding to interdistributary, grading up to prodelta is consistent with the characteristics of this facies and the overall association of facies (see below). Slumps, slide bodies and down-sag structures represent slope instability or syn-depositional basin tectonic activity (discussed further below). Tabular, laminated to cross-laminated vitric siltstones and fine sandstones (Fig. 3b(ii) & 4c) are interbedded with the mudstones and occur in a 12 m interval, low in the stratigraphy of the northern basinmargin succession. Beds are tabular at the outcrop scale, and up to 40 cm thick. Bases are generally sharp but tops are gradational into mudstone horizons millimetres to several centimetres thick. Internally the beds contain random alternations of planar, wavy and
89
asymmetrical ripple cross-laminae. The facies contains > 90% cuspate and straight-walled glass shards, and minor angular volcanic quartz and feldspar fragments, all of silt to fine sand size. A source of unconsolidated rhyolitic ash, derived from an unknown volcanic centre at or beyond the lake basin margin, is indicated. However, the facies is indicative of turbidites, indicating syn-eruptive to immediately post-eruptive resedimentation by sedimentary processes. The random order of the internal sedimentary structures indicates fluctuating flow velocities, rather than steady decelerating flow as in most deep-water turbidity currents. This facies is interpreted to be the product of short-lived, continuousfeed, turbulent underflows, with pulsing, inconsistent flow velocities, originating from river mouths in flood (e.g. Lambert & Hsu, 1979). It may represent slope deposits associated with a basin-margin delta. Crosslaminae indicate palaeoflow to the north-west (Figs 2 & 3b(ii) ). Bioturbated fine sandstone occurs in one small outcrop near the preserved base of the succession on Bunga Beach, and contains a high density of cross-cutting, subhorizontal to subvertical burrows (?Rhizocorallium). The intensity and style of burrowing suggest a relatively shallow environment. The facies consists of > 85% straight-walled and cuspate shards, with minor quartz and feldspar fragments, indicating that it represents a bioturbated example of the vitric siltstone–sandstone facies discussed above, which it underlies (Fig. 3b(i) ). Also interbedded with the mudstones are tabular, massive to laminated to cross-stratified sandstone beds (Fig. 3b(iii) & 4d). Beds are up to 1 m thick, and amalgamated intervals of beds may be several metres thick. Beds are tabular, their bases are sharp and conformable or slightly erosional, and the tops are sharp or gradational into siltstone or mudstone. Individual beds occur in various combinations of massive, diffusely layered, diffuse low-angle cross-stratified, horizontally laminated and graded textural–structural divisions. Some graded beds pass upwards from a basal massive interval into cross-bedding, then horizontal lamination, cross-lamination and then mudstone. Concentrations of mudstone intraclasts define layering in some beds. This facies consists of both basement-derived and volcanic detritus. In one sample subangular to subrounded, monocrystalline volcanic quartz makes up 44% of the framework, and volcanic lithics altered to sericite and chlorite constitute 17%. Basement-derived metasedimentary slate fragments (10%), variably rounded (?granitic) potassium feldspar
VSIC04 9/12/08 2:05 PM Page 90
90
R. A. F. Cas et al.
(10%), chert (10%), and polycrystalline quartz (5%) are minor framework components; rare tourmaline and detrital mica also occur. Silty matrix constitutes up to 20% of the rock. This facies, like the previous one, records subaqueous mass-flow events. Beds with simple internal features and sharp tops and bases represent high concentration, rapidly decelerating turbidity currents. The internally more complex beds, with multiple structural divisions, represent steadily decelerating, and in some cases, fluctuating velocity flows. These flows could also represent river-fed density currents, involving coarser sediment and higher-energy, longer-lived floods. The only facies indicating any substantial, sustained tractional reworking in the Bunga Beach section is the 4 m-thick cross-bedded sandstone facies interval (Figs 3b(iv) & 4e). The base of the facies is slightly
erosional and internally it consists of broad, scouring sets of angle-of-repose to low-angle trough crossbedded sandstone with cross-beds indicating flow to the north-north-west. This facies interval is interpreted as a distributary channel fill or channel mouth bar. The sandstones are moderately to well sorted and contain angular fragments of basement-derived pelite, quartz-rich siltstone, minor vein quartz and chert, together making up 26% of the framework in one sample. Texturally immature volcanic quartz, altered vitric lithic fragments, and minor plagioclase and cuspate shards comprise 74% of the framework, and indicate a probable rhyolitic pyroclastic source. The cross-bedding and polymictic nature of the detritus suggests, however, that like the previous sandstone and siltstone facies, the volcanic detritus of this facies was resedimented, probably from a subaerial source.
Fig. 5. Selected measured sections of distinctive facies intervals from the southern basin-margin facies association exposed from Picnic Point to the south of Bengunnu Point. Sections are in stratigraphic order from oldest (left) to youngest (right): (i) massive conglomerate –breccia facies consisting of basement-derived clasts overlying the unconformable contact with the Ordovician metasedimentary basement, and passing up into basement-derived massive to planar-laminated tabular lithic sandstones; (ii) transition from massive and laminated grey mudstone and siltstones, with minor black mudstone intercalations, up into tabular siltstones and fine sandstones, some graded, and some showing random successions of plane, wavy and cross-lamination; (iii) thickly bedded tabular, massive to laminated to cross-laminated, sandstones, most of which are also graded; (iv) coarser facies association of coarse sandstone, pebbly sandstone and pebble conglomerate, with abundant scoured contacts and cross-bedding, incised and /or faulted into diffusely layered medium– coarse sandstones; (v) massive to planar-laminated tabular sandstones. Legend as for Fig. 3.
VSIC04 9/12/08 2:05 PM Page 91
Sedimentation in an extensional basin
91
Locally, clastic dykes of pebbly sandstone also occur along Bunga Beach close to a 15-m-wide rhyolite dyke and an irregular basalt dyke, one or both of which are interpreted to have caused fluidization and upward intrusion of unconsolidated, coarse, watersaturated sediment through which the magma passed. This sediment has the same provenance as the coarse sandstones discussed above, and was presumably remobilized from a deeper level. The total facies association is therefore representative of subaqueous sedimentation in quiet water, probably a basin-margin delta or fan- or delta-slope system. Sediment was derived from both uplifted basement and acidic pyroclastic volcanic source(s) at or beyond the basin margin. Southern basin-margin succession The southern basin-margin succession (Fig. 5) extends northwards from Picnic Point (Fig. 2). It is also openly folded and contains a facies association generally similar to that in the Bunga Beach succession. At Picnic Point, the only known contact between the Bunga Beds and the tightly folded Ordovician metasedimentary basement is exposed (Fig. 6). At the contact, the basal facies of the Bunga Beds is a massive conglomerate –breccia facies consisting of basementderived metasedimentary clasts, which are clast- to matrix-supported in a matrix of coarse lithic sandstone (Fig. 5i). At the low-tide mark the contact is locally a highly irregular unconformity with hollows filled in with angular metasedimentary blocks up to 0.5 m in diameter, some of which form an in situ to clast-rotated colluvial breccia. Nearby is a fault separating Ordovician bedrock and sheared conglomerate (Fig. 6). Highly irregular basaltic dykes and pillowlike intrusive lobes also occur within the conglomerates, and cross-cut the underlying basement. The basal conglomerate–breccia facies is wedge shaped, varying from about 1 m up to 4 m thick, although its top is not exposed. Its texture is mostly matrix-supported, and it occurs in several massive beds up to 1 m or more thick. This facies is then interbedded with beds of massive to diffusely laminated tabular lithic sandstone facies up to 40 cm thick, forming an interval 4 m thick (Fig. 5i). The sandstones are amalgamated but upwards have thin shaley partings between successive beds. The conglomerates appear to be of debris-flow origin, whereas the tabular sandstones are high-concentration turbidites. The unconformable contact with the basement probably had high relief, as recorded by the conglomerate and breccia, which represent a small talus slope or fan-delta.
Fig. 6. Contact between Bunga Beds (right) and truncated turbidite beds of the Ordovician metasedimentary basement (left). The planar fabric in the conglomerate is due to shearing along the contact. Laterally, this contact is an undeformed unconformity.
This coarse basal facies package is overlain by 20 m of openly folded, laterally continuous, planar-bedded, massive to laminated grey and black mudstones and siltstones (Fig. 5ii). Beds are up to 20 cm thick and although the mudstones are massive and structureless, the siltstones locally preserve planar lamination, and at the tops of beds, asymmetrical ripple crosslamination. Tabular-bedded sandstones (see below), containing various Bouma turbidite divisions, become abundant upwards. The grey mudstones are interpreted as prodelta suspension fallout deposits, like the northern basin-margin association. The black mudstone appears to represent short-lived periods of anoxic, offshore sedimentation. Minor cross-laminae in the siltstones indicate flow directions in a broad arc from north-west to north-east. This basal facies association is faulted against tabular, massive to laminated to cross-laminated
VSIC04 9/12/08 2:05 PM Page 92
92
R. A. F. Cas et al.
Fig. 7. Measured section of the basin-centre facies association south of Bunga Head, at location 4 (Fig. 2). (Note the diversity of classic deep-water mass-flow facies and the packaging of coarse clastic facies.)
sandstones (Fig. 5iii), which appear to be a downfaulted unit overlying the basal facies association. This facies interval is at least 20 m thick, beds are up to 4 m thick, and beds are amalgamated. The composition is quartzo-feldspathic –lithic, as in the northern basin-margin association. Minor ripple cross-laminae indicate flow to the north. This sandstone facies represents subaqueous slope turbidites. This sandstone facies is also in fault contact with a coarser pebbly sandstone and pebble conglomerate
facies north of Picnic Point. Beds are up to 1 m thick and commonly have scoured basal contacts. Internally, they are massive to diffusely planar-bedded to cross-bedded (Fig. 5iv). Trough cross-bed sets up to 0.5 m in amplitude also indicate palaeoflow to the north. This facies consists of basement-derived detritus only, including chert, slate, psammite and vein quartz, and is classified petrographically as metasedimentary lithic arenite. The facies was deposited under high-energy flow conditions, including upper, transi-
VSIC04 9/12/08 2:05 PM Page 93
Sedimentation in an extensional basin
93
(a)
(b)
(c)
(d) Fig. 8. Selected facies of the basin-centre facies association south of Bunga Head, location 4 (Fig. 2): (a) graded highconcentration turbidites, some with basal load casts, and interbedded black mudstone (note slump folding in upper thinly bedded sequence); (b) coherent bedded slump; (c) disrupted, chaotic slide deposit; (d) massive, matrix-supported fabric of basaltic debris-flow deposit. The irregular margins of the scoriaceous fragments (light colour) should be noted.
tional and upper lower flow regime. This succession suggests that a coarse clastic wedge prograded from the basin margin, forming a coarse-grained delta or fan-delta with a braided, conglomeratic distributary channel system. Above the conglomerates there is a subtle upwardfining succession to feldspatho-lithic cross-bedded and massive to planar-laminated tabular sandstones (Fig. 5iv & v), representing deepening. Near Bengunnu Point black shales associated with rhyolites occur, suggesting rapid lateral and /or vertical transition into the deeper-water, basin-centre facies association. Basin-centre facies association This facies association extends from the northern margin of Bunga Head, to south of Bengunnu Point (Fig. 2), and is distinguished from the basin-margin facies association in that the mudstones are black, the coarser clastic deposits form a classic subaque-
ous turbidite facies association (e.g. Walker, 1978; Pickering et al., 1989) and there is no evidence for any prolonged tractional reworking. The association is best preserved just south of Bunga Head (location 4, Fig. 2) in an extensive platform exposure which preserves at least 100 m of section (Figs 7 & 8). It is openly to tightly folded, faulted, locally cleaved and intruded by domes and dykes of rhyolite and basalt. Elsewhere the facies association is exposed over small areas between rhyolite bodies. The distinctive black mudstones are massive to laminated, occur in intervals or beds centimetres to many metres thick, are variably pyritic and are interpreted as muds from suspension deposited in a relatively deep, quiet, anoxic environment. The laminae are pale silty layers, locally graded, and represent small-volume distal turbidites. Variably graded, massive to laminated sandstones to pebbly sandstones are also prominent (Fig. 8a). They are up to 2 m thick, are locally amalgamated,
VSIC04 9/12/08 2:05 PM Page 94
94
R. A. F. Cas et al.
and have sharp bases with classic turbidite sole structures such as load casts, flame structures (Fig. 8a) and ball-and-pillow structures. Reverse-graded bases also occur. The beds are usually massive, diffusely layered or laminated, with only minor cross-laminae. They are interbedded with black mudstones and commonly contain abundant black mudstone intraclasts that are variably disaggregated. Most beds consist of Bouma a, ae or abe divisions (Bouma, 1962; Walker, 1978; Pickering et al., 1989). Bouma c division crosslamination is rare, and only a few 1– 2 m thick, granule to pebbly sandstone beds show signs of traction carpet deposits at the bases of beds (Lowe, 1982). This facies represents turbidites deposited from poorly expanded, high-concentration turbidity currents. Disaggregation of mudstone intraclasts during flow may have modified the flow behaviour of the turbidity currents by mixing of intraclast-derived clay and silt into the matrix fluid. In the platform exposure south of Bunga Head, there is some packaging of sandstone turbidite beds, suggesting the existence of subaqueous channels and fans or lobes. The sandstones are moderately to poorly sorted litharenites. One point-counted sample consists of vein quartz (49% of framework), metasedimentary lithics (quartz sandstones, minor slate and chert, 25%), and granite rock fragments, derived from the metasedimentary and granitic regional basement. Black mudstone intraclasts also occur and were deformed during transport and deposition, and by subsequent compaction when still soft. The sandstones are texturally immature to submature. At locations 4 and 8 (Fig. 2) limited palaeocurrent data from cross-laminae indicate a general northerly flow, but also some southward flow, suggesting bi-directional flow along a basin axis, and input from several point sources. Much of the sedimentary succession at location 4 (Fig. 2) has been subjected to soft-sediment deformation and down-slope slumping. In situ coherent slumps (Fig. 8b) of bedded intervals of the previous facies, chaotic slide deposits (Fig. 8c), consisting of disrupted soft-sediment deformed bedded clasts up to several metres long, and crudely layered, clast-supported intraclast breccias consisting dominantly of mudstone and siltstone intraclasts, and rare basaltic and rhyolitic debris in a matrix of sandstone, are well developed. Intervals of slumps and slides are up to 12 m thick. They reflect slope instability and downslope gravitational collapse of basin lithofacies. Possible causes include syn-depositional intralacustrine tectonics, such as basement-normal or strike-slip growth faulting, and contemporaneous intrusion of rhyolitic
and basaltic cryptodomes and basaltic dykes and sills, which crop out nearby (see Kokelaar et al., 1985). Slump folds and ramp structures indicate a palaeoslope to the east to east-south-east. Some slump horizons are bounded above and below by planar contacts that truncate structures in both the slumps and bedded successions above and below. This suggests that the surfaces are intraformational detachment surfaces and that some of the slumping is intraformational and occurred post-depositionally but pre-consolidation. The last facies of this association is dominated by basaltic debris and is a matrix-supported pebbly, basaltic sandstone facies (Figs 7 & 8d) occurring in four internally massive, tabular beds up to 6.6 m thick at locality 4 (Fig. 2). The basalt clasts vary from highly fluid bomb shapes to scoria fragments with highly irregular delicate margins. The scoria fragments are up 3 cm in diameter, angular and have cuspate margins formed by truncated vesicles. The scoria contains aligned plagioclase crystals in a microlitic, vesicular groundmass that is variably altered to chlorite– carbonate. Vesicles rarely exceed 1 mm in diameter, represent up to 60% of clast volumes and are filled with chlorite ± carbonate. Irregular intraclasts of sandstone and mudstone, as well as carbonaceous debris, also occur. The muddy sandstone matrix consists of fine scoria, basaltic shards, plagioclase fragments, basement-derived metamorphic and plutonic quartz and siltstone lithics, and black silt and clay mixed in from the basinal black mudstone. This facies is the only indication of the eruption of basaltic magmas in the Bunga Beds, as all other occurrences of basalt are intrusive bodies (see below). The scoriaceous clasts and thin-walled shards indicate magmatic explosive eruptions or phreatomagmatic explosive eruptions of highly vesiculated magma. The four units are interpreted as debris-flow deposits resulting from the collapse of a small scoria cone or tuff ring within or at the margin of the basin, perhaps contemporaneous with eruption. Basement-derived and intraclast debris was probably incorporated by erosion of basinal sediments over which the debris flows moved. Summary of the depositional environment and basin character The depositional environment into which magma was erupted or intruded was a small, quiet, restricted basin, or an embayment of a larger basin that is no longer preserved. Preserved basin-margin facies associations indicate a relatively shallow-water setting
VSIC04 9/12/08 2:05 PM Page 95
Sedimentation in an extensional basin as shown by the cross-bedded sandstone facies, the grey, ?oxidized, prodelta mudstones and local intense bioturbation. The basin-margin environments were probably deltaic and /or fan-delta and foreslope deltaic environments. The deeper part of the basin was anoxic, below the influence of reworking agents, as indicated by the black pyritic shales and turbidite facies associations. The fish fossils, which occur in the deep-water association at locality 4 (Fergusson et al., 1979), appear to be of fresh- to brackish-water affinity (J. Long, personal communication). The contact between the Bunga Beds and the Ordovician metasedimentary basement at Picnic Point is a variably faulted unconformity mantled by palaeocolluvial bedrock-derived debris and conglomerate, fining upward into grey mudstones. The faulted contact may represent a reactivated basin-margin normal fault. Upward fining, then coarsening and finally fining trends in both basin-margin successions (Figs 3 & 5) are viewed as responses to basin tectonics involving initial extension and subsidence, then delta progradation, and then subsidence to anoxic water depths. The influence of syn-depositional tectonics is also evident from the load structures and slide packages present around Picnic Point and along Bunga Beach. The original basin geometry is unclear, but syn-depositional rhyolite and basalt dykes have a consistent northnorth-west strike, and using these as a palaeostress field indicator during basin formation (cf. Eaton, 1984), an east–west component of extension or transtension is indicated.
INTRABASINAL MAGMATISM: RHYOLITIC FACIES Rhyolites are intercalated with the sedimentary sequences from Bunga Beach in the north to just north of Picnic Point in the south (Fig. 2). The rhyolites range from coherent rhyolite bodies to massive disorganized breccias and stratified successions of rhyolitic debris. Coherent and clastic facies are volumetrically approximately equal in abundance (Fig. 2). Hydrothermal alteration and low-grade metamorphism have affected all the rhyolites, producing a pyrophyllite– quartz–sericite alteration assemblage which imparts a strong yellow–green colour to the rhyolite, but in most cases primary textures are well preserved. The three rhyolite units in the northern half of the area are almost entirely coherent, whereas the extensive rhyolites in the southern half consist of coherent domains enveloped by breccia of similar
95
composition. These two rhyolite groups can also be distinguished petrographically. The three coherent bodies in the north have a relatively high phenocryst content with 12–40% of coarse phenocrysts of quartz, plagioclase, K-feldspar and minor amphibole 2–4 mm long (rhyolite type 1, Fig. 2). The rhyolites south of the Bunga Head area are almost all poorly porphyritic, with 2–8% quartz, K-feldspar and plagioclase phenocrysts up to 1 mm in length (rhyolite type 2, Fig. 2). Coherent rhyolite bodies: lavas, cryptodomes and feeder dykes The three relatively phenocryst-rich bodies in the north have varying forms reflecting different modes of emplacement. The body at Bunga Head is a single unit at least 180 m thick and 1300 m long, with subvertical columnar jointing, diffuse subhorizontal flowlayering throughout (Fig. 9a), and a homogeneous, coarsely porphyritic texture. Small fluidal-shaped basalt inclusions are common at the southern margin and contain xenocrysts of quartz from the rhyolite, suggesting mingling of contemporaneous rhyolite and basalt magmas. The base of the rhyolite has a subhorizontal dip, parallel to flow-layering, and is completely concordant with underlying strata of the deep-water basincentre facies association. Although the underlying black mudstones have accommodated some shear during regional deformation (Fig. 2, location 3, & Fig. 9b), they are relatively undisturbed except for minor soft-sediment deformation of sandy beds, presumably as a result of the loading effect of the rhyolite. The top is not preserved. The rhyolite is coherent throughout and has a basal 30-cm-thick zone containing spherulites and lithophysae (Fig. 9b). Flow banding occurs almost to the base, and includes discordant ramp structures (Fig. 9a). This contact therefore records the remarkably passive emplacement of a thick lava onto, or a sill into, sediments of the basin floor. The lack of quench fragmentation and hyaloclastite perhaps indicates relatively late-stage emplacement, when the sediments had already been compacted and dewatered to some degree. A second relatively phenocryst-rich, coherent rhyolite body crops out lower in the stratigraphy, at Bunga Beach (Fig. 2, location 1). At its southern margin, polygonal joints converge radially inwards from a smooth convex surface, which is discordant to strike in the nearby sedimentary sequence and is interpreted to be the margin of an intrusive cryptodome that
VSIC04 9/12/08 2:05 PM Page 96
96
R. A. F. Cas et al.
(a)
(b)
(c)
(d)
(e)
(f) Fig. 9. Facies characteristics of the igneous facies of the Bunga Beds: (a) low-angle ramping in flow banding near base of Bunga Head rhyolite; (b) sharp, planar, conformable base of Bunga Head rhyolite (note lower nodular zone and lack of erosional interaction with underlying black mudstone; also note planar tectonic fabric in mudstones below contact, indicating minor shear along the contact); (c) subvertical flow banding and small-scale flow folding in weakly porphyritic rhyolite, locality 7, Fig. 2; (d) autobreccia, north of Bengunnu Point, with rotated clasts showing internal flow banding; (e) close packed, largely in situ hyaloclastite breccia, north of Bengunnu Point; (f ) clast-rotated hyaloclastite breccia with abundant fine hyaloclastite matrix at the margins of a rhyolite, north of Bengunnu Point; (g) peperite texture defined by angular quench-fragmented blocks of rhyolite dispersed in a 6-m sedimentary intraclast with a homogenized, fluidized texture; the intraclast is enclosed within a rhyolite cryptodome (location 7, Fig. 2); (h) altered syn-depositional basalt dyke showing gradation from massive coherent basalt to in situ jigsaw-fit basalt breccia, and then to basalt breccia with sedimentary matrix (peperite; location 4, Fig. 2).
VSIC04 9/12/08 2:05 PM Page 97
Sedimentation in an extensional basin
97
(g)
(h) Fig. 9. (continued )
was probably emplaced into wet, semiconsolidated sediments. A similar but smaller rhyolite occurs at the northern end of Aragunnu Beach. In addition, a 15-m-thick, vertical dyke occurs in the north (Fig. 2, location 2), although it is less porphyritic (15% phenocrysts) than the other bodies. The dyke displays strong, vertical flow-layering parallel to the dyke margins, accentuated by a colour banding reflecting variations in devitrification textures, including 1–2 cm spherulites. Another vertical, flow-banded dyke, 4 m wide, cross-cuts the Bunga Head rhyolite about 300 m north of its southern margin. The poorly porphyritic rhyolites are complex lavas, syn-depositional, high-level cryptodomes and partly extrusive cryptodomes, which have cores of coherent rhyolite grading into massive rhyolite breccia of similar composition. Flow banding is common and usually dips vertically to subvertically, but may vary widely in orientation and is commonly flow folded (Fig. 9c). Columnar jointing is not widespread, but perlitic fracture textures are common in the groundmass, indicating that much of the rock was originally glassy. North of Picnic Point, the southernmost rhyolite is packed with spherulites and lithophysae 1– 2 cm in diameter. Rhyolite breccias Autobreccias are relatively common in the poorly porphyritic rhyolites. These breccias are clast-supported frameworks of irregular, angular rhyolite clasts which vary from 10 cm to > 10 m in diameter. Flow-layering
is common in clasts and is often parallel in adjacent clasts through the breccia, indicating in situ, mainly extensional fragmentation, with little or no rotation of clasts. Clast-rotated textures also occur, implying the influence of a flow-induced shear stress (Fig. 9d). Rhyolites commonly grade from coherent, autobrecciated, and mildly fractured interiors to coarse in situ breccia with clasts of up to 15 cm diameter to an intensely granulated breccia with clasts up to 15 cm diameter suspended in a finer rhyolitic matrix at the margins. Both the coarse, in situ breccias and finer marginal breccias are interpreted to be hyaloclastite. Hyaloclastite at the margins of rhyolite bodies commonly contains irregular intraclasts of mudstone and mudstone stringers apparently squeezed between rhyolite clasts, and showing soft-sediment deformation, producing peperite textures similar to some of those recorded by Hanson & Wilson (1993). Intraclasts are usually centimetres to tens of centimetres in diameter, but at locality 7 (Fig. 2) intraclasts up to 6 m in diameter are enclosed in rhyolite. The thickness of coarse in situ breccia varies widely from a few metres to at least 100 m and is generally thicker than the intensely granulated marginal breccia, which may in places be completely absent. At the northern end of Aragunnu Bay, lobate domains of relatively coherent rhyolite 2–3 m in diameter are enclosed within a matrix of strongly fragmented material (grain size of millimetres to centimetres) of identical composition. The small coherent bodies have brecciated, gradational contacts with the matrix. Continuous outcrop in this area shows
VSIC04 9/12/08 2:05 PM Page 98
98
R. A. F. Cas et al.
that these bodies form a 20-m-thick transition zone between coherent flow-layered rhyolite and more uniformly brecciated material. The hyaloclastite breccias are strongly altered and the fracture system that defines them is commonly filled with fine-grained pyrite or stringers of mudstone. In situ breccia fragments are composed of dense, poorly to non-vesiculated rhyolite, and fragment boundaries are mostly planar to broadly arcuate (Fig. 9e). Perlitic fractures are common within the fragments. The intensely granulated marginal breccias are poorly sorted, have abundant matrix, and consist of strongly pyrophyllite –quartz-altered, rhyolitic hyaloclastite debris (Fig. 9f ). In detail they comprise scattered, irregularly shaped patches of strongly in situ brecciated rhyolite in a more intensely fragmented and disorganized rhyolitic matrix. Matrix fragments consist of altered glassy rhyolite, vary from angular blocky to irregular shapes, and are mostly < 2 cm in size, although some clasts with irregular shapes may have been pumiceous before alteration. Features that characterize these breccias as hyaloclastite are: gradational relationship with coherent or autobrecciated rhyolite lava, general trend of increasing intensity of brecciation outward, widespread occurrence of intense in situ fracturing with little or no rotation of clasts, pervasive nature of the brecciation producing sharply angular clasts 2 –3 cm in diameter, and the abundant evidence (e.g. perlitic texture) for an original glassy nature of the breccia clasts. These features are well documented from other rhyolitic hyaloclastites (Pichler, 1965; De Rosen-Spence et al., 1980; Furnes et al., 1980; Hanson & Schweickert, 1982; Yamagishi & Dimroth, 1985; Yamagishi, 1987, 1991; Hanson, 1991; Hanson & Wilson, 1993). Pichler (1965) and Kokelaar (1986) suggested that hyaloclastite largely forms by the combination of quench fracturing and autobrecciation. In the Bunga Beds this is consistent with the presence of locally large clasts (autobrecciated) suspended in largely in situ finer-grained breccias (hyaloclastite) which show local clast-rotated textures, probably as a result of continued emplacement and expansion of the lavas or cryptodomes. Quenching may also have been accompanied by small-scale steam explosions at shallow water depths, although we have no specific evidence of this here. Rhyolite –sediment contact relationships of the weakly porphyritic rhyolite bodies: implications for emplacement In contrast to the passive, sharp contacts of the
phenocryst-rich rhyolite bodies, contacts of the weakly porphyritic rhyolites in the southern half of the area all record syn-depositional interaction between lava and sediment. At location 4 (Fig. 2) contacts between two small (15 m maximum exposed dimension) flowlayered rhyolite bodies and black mudstone are irregular, and locally show flow-layering terminating outward against mudstone. At one point a long, flame-like structure of mudstone penetrates 0.5 m into a fracture in the rhyolite surface. The margin of one rhyolite body displays fine polygonal cooling joints 5 cm apart and perpendicular to the contact, representing in situ quench fracturing. These contacts are interpreted as irregular lobate intrusive margins between rhyolite and wet sediments. At location 7, large intraclasts of sediment up to 6 m in diameter are enclosed within the rhyolite, quenchfragmented blocks of rhyolite occur dispersed within the sedimentary intraclasts adjacent to contacts, showing spectacular peperite textures (Fig. 9g), and dykes of sediment have intruded several metres into the rhyolite. In the sedimentary dykes, and where rhyolite clasts occur within the intraclasts, the original fabric of the sediment has been virtually destroyed and replaced by a homogeneous fabric of sand grains and rhyolite clasts scattered evenly in a black mud matrix. The most complex mixed contact occurs at location 8 between sediments and granular hyaloclastite breccia at the margin of a rhyolite body. Here the contact is a zone several metres wide consisting of numerous irregular patches and lobes of hyaloclastite peperite breccia dispersed in and mixed with homogenized massive black silty sediment. The irregular and mixed contacts document the intrusion of rhyolite lava or magma into wet, poorly consolidated sediments. The intraclasts were probably incorporated into the interior of the rhyolite as a multilobate intrusion engulfed domains of sediment as it intruded upwards. Emplacement of the hot rhyolite into wet sediments would have heated pore water in the sediments adjacent to contacts and caused quench fragmentation of the rhyolite. Turbulent circulation of this heated water (perhaps locally steam), and continued intrusion of the rhyolite probably mobilized the fluidized mixtures of water, steam, sediment, and rhyolite clasts into fractures and away from the rhyolite body, producing peperitic mixing. Furthermore, the sediment mixtures have been injected as dykes into fractures in the rhyolite, and have intruded outward into undisturbed sediment, transporting quenched rhyolite fragments in the process. These features are consistent with the fluidization of the wet sediments by
VSIC04 9/12/08 2:05 PM Page 99
Sedimentation in an extensional basin the hot rhyolite, and the formation of peperite, as described by Kokelaar (1982) and elsewhere. These interactive rhyolite – sediment relations occur at contacts where sediment underlies, overlies or laterally abuts rhyolite. Because there is a general lack of bedded, resedimented hyaloclastite marginal to the rhyolite bodies, interbedded with the enclosing sedimentary succession, the majority of the rhyolites are interpreted as intrusive rhyolite cryptodomes and sills (see Hanson & Schweickert, 1982; Hanson, 1991; Hanson & Wilson, 1993). Bedded, redeposited hyaloclastite, mixed with sedimentary intraclast debris, only occurs in association with the two stratified tuff cone successions described below, indicating that extrusive to semi-extrusive domes that emerged above the basin floor played only a minor role in development of the facies patterns. Stratified juvenile rhyolitic volcaniclastic facies associations In two localities within the basin-centre facies association (locations 6 and 8, Fig. 2), stratified volcaniclastic rocks overlie a coherent, autobrecciated and quenchfragmented rhyolite (Fig. 10). Details of this succession and the interpretation have been provided by Cas et al. (1990). At the headland north of Aragunnu Beach (locality 6, Fig. 2) the basal rhyolite is overlain by a rhyolite–sediment breccia with flow-banded, angular blocks of rhyolite and mudstone intraclasts in a finer silt–rhyolite fragment matrix, showing peperitic textures. This is overlain by a bed of pumice– sediment breccia with a crystal-rich matrix of volcanic quartz and feldspar, as well as finer pumice shreds (Fig. 10). Then a 10-m-thick diffusely stratified, crystal-rich sandstone and pumice breccia facies is followed conformably by a 7-m-thick thinly bedded, crystal-poor sandstone and siltstone facies (Fig. 10) that is dominated by shreds and shards of pumice. Cas et al. (1990) interpreted the succession to be the product of the upward intrusion of a rhyolite cryptodome through wet unconsolidated sediments, and the growth of a phreatomagmatic dome-top tuff–pumice cone complex, as occurred during the 1953 –1957 eruption of Tuluman Volcano in the Bismarck Sea (Reynolds & Best, 1976; Reynolds et al., 1980). As the variably autobrecciated and quenchfragmented cryptodome approached the basin floor it up-arched the sediments; this caused sliding and slumping of sediment and rhyolite debris down the steepening slopes of the rising dome, thus producing the megabreccia. The pumice breccia, crystal-rich and
99
crystal-poor facies represent short-lived, phreatomagmatic explosive eruptions (Cas et al., 1990). The crystal-rich facies formed through efficient hydraulic sorting in the water column, which resulted in a coarse water-settled crystal tuff, and the fine, crystal-poor facies represents the waning of explosive eruptions, which produced the succession of thin, variably graded fine beds of crystal-poor vitric tuffs (see Fiske & Matsuda, 1964). A poorly sorted polymictic pebbly volcanic sandstone facies erodes into the preceding succession and represents debris-flow deposits caused by posteruptive degradation of the tuff cone succession (Cas et al., 1990). The uppermost stratigraphic unit is a 10-m-thick faintly stratified rhyolite pumice breccia (Fig. 10) representing a renewed pyroclastic pumice cone-forming event. A massive, diapiric intrusive, clastic rhyolite–sediment mass, fluidized and mobilized by an intruding rhyolite cryptodome in the subsurface, has been injected up through the stratified volcaniclastic sequence (Fig. 10). On the headland north of Bengunnu Point (location 8, Fig. 2) there is a generally similar succession to that described above, although there is more pumice breccia, less crystal-rich sandstone and no crystal-poor tuff (Cas et al., 1990; Fig. 10).
BASALTIC FACIES Basalt occurs in a number of localities, has a variety of forms and origins, but volumetrically is only a very minor component of the Bunga Beds. Lavas are not known, the most common occurrence of basalt being as sills, dykes and irregular intrusive pods that commonly have peperitic margins. In addition, basalt is represented by debris-flow deposits of vesicular basaltic clasts (already described above), and as inclusions in rhyolite. Intrusions and associated marginal breccias Basalt intrusions are known from several localities in the Bunga Beds (Fig. 2). They include tabular to irregular dykes and sills, irregular branching networks of dykes, and small cryptodomal pods. The basalt intrusions generally occur in close proximity to intrusive rhyolites. Dykes and sills are rarely more than 0.5 m wide, but thicknesses vary markedly, reflecting their irregular margins. Flame-like apophyses of basalt, intruding the enclosing sedimentary succession, and complementary tongues and lobes of sediment
VSIC04 9/12/08 2:05 PM Page 100
100
R. A. F. Cas et al.
Fig. 10. Measured sections through interpreted subaqueous tuff ring successions. (a) Headland north of Aragunnu Beach; (b) headland north of Bengunnu Point (from Cas et al., 1990).
indenting the basalt margins, are common. Irregular pods and cryptodomal masses of basalt are seldom more than 5 –10 m in diameter, and their margins are discordant with the enclosing bedding. The irregular margins indicate that the enclosing sedimentary successions were unconsolidated when basalt was intruded. The margins of some intrusions are brecciated and grade from coherent to in situ incipiently fractured
basalt, to slightly dispersed jigsaw breccia and then to clast-rotated breccia. Finally, peperite breccias represent the transition into the host sedimentary rocks (Fig. 9h). The basalt clasts are angular, blocky to splintery in shape, have planar to curviplanar margins and are up to 10 cm in size. Matrix, where present, consists of mixed sediment and small basalt clasts, and appears to have been injected into fractures in the
VSIC04 9/12/08 2:05 PM Page 101
Sedimentation in an extensional basin basalt. The enclosing sedimentary rock is often structureless up to 1–2 m away from the contact. The marginal breccia zone is interpreted to be peperitic hyaloclastite resulting from the quench shattering of basalt intruding into, and mixing with, wet, unconsolidated sediment (see Kokelaar, 1982). At Picnic Point a system of basalt intrusions occurs along and near the boundary fault between the Ordovician metasedimentary basement and the basal conglomerate of the Bunga Beds. The fault, which we speculate may represent a reactivated basin-margin normal fault, probably also acted as a magma conduit. Petrographically, the basalts vary texturally from almost aphanitic with tiny cryptic plagioclase(?) laths to a coarse doleritic texture dominated by large (1–2 mm) plagioclase laths pseudomorphed by sericite and quartz. These two textural types can occur in the same intrusion in an irregularly intermingled fashion, suggesting mingling of different pulses of magma. Pyroxene crystals are notably lacking, suggesting that the true composition may be basaltic andesite. The basalts are commonly vesicular, with 10–15% vesicles filled with quartz and chlorite, commonly with a colloform layering. The quartz appears to be a replacement or recrystallization product after chalcedonic and opaline quartz. The groundmass surrounding the vesicles is extremely fine grained and indicates local chilling at the time of vesiculation, before formation of coarser-grained textures in the rest of the groundmass.
DISCUSSION Tectonics, volcanism and basin formation During the Late Devonian, south-eastern Australia was influenced by extensional or transtensional tectonics and basin formation following the compressional Middle Devonian Tabberabberan Orogeny (e.g. Cas, 1983; Powell, 1983). Volcanism, frequently bimodal in character, was widespread across the Lachlan Fold Belt during the Late Devonian. In the Eden–Comerong–Yalwal Rift Zone of the east coast belt of New South Wales, bimodal basalt–rhyolite volcanic associations are especially well developed (McIlveen, 1974; Fergusson et al., 1979; Cas, 1983; Dadd, 1992; Cas & Bull, 1993). This volcanic association occurs low in the stratigraphy of the extensional basin systems, indicating a close genetic relationship between extensional to transtensional tectonics, basin formation and bimodal volcanism (Cas, 1983).
101
In the Bunga Beds, emplacement of basalts and rhyolites was coeval. They occur in close spatial relationship, their emplacement was syn-depositional (Fergusson et al., 1979; Foley, 1986; Cas & Bull, 1993; Edgar, 1996), and in places there is clear evidence of fluidal intermingling of rhyolite and basalt, such as the irregular basalt inclusions in the Bunga Head Rhyolite. The widespread occurrence of basalts in the Lachlan Fold Belt during the Late Devonian suggests that mantle activity was probably integrally related to extensional tectonics and basin formation. The basalts probably caused crustal melting and rhyolite magma formation as they invaded and rose though the crust (see Hildreth, 1981). The regional palaeogeography of the Lachlan Fold Belt during the Late Devonian was dominated by alluvial–fluvial and lacustrine environments (Cas, 1983; Powell, 1984). The Bunga Beds are a rare example of deep-water deposits, and this poses a question about the origin of the host basin. The bimodal volcanic association is consistent with an extensional basin origin, and given the rapid lateral transition from shallow basin-margin to deep, basin-centre facies, relatively significant extension or transtension and subsidence is implied (Figs 11 & 12). Ponding of basin waters is required and this could have been produced by crossfaulting, producing a fault scarp normal to the general direction of extension, or by the damming effects of constructive volcanic topography, such as a highaspect-ratio rhyolite lava flow. The absence of voluminous ignimbrites precludes an explosive caldera lake origin. Tectonic activity persisted during the accumulation of the Bunga Beds. The load structures within the basin-margin facies association at the southern end of Bunga Beach and at Picnic Point appear to have been contemporaneous down-sag features, perhaps resulting from syn-depositional, normal faulting in the basement. Along Bunga Beach, packages of strata with basal discordant slide surfaces are consistent with active growth faults. Furthermore, most dykes in the Bunga Beds are orientated NNW–SSE, indicating a basement fracture control. Basin analogues A limitation in identifying a specific analogue for the Bunga Beds lacustrine basin is the uncertainty of the extent and scale of the original basin. The preserved north–south extent of current coastal exposure is 8 km, which is only a remnant of the original basin. A scale-specific but non-volcanic basin analogue is the
102
Fig. 11. (a) Cross-section for the Bunga Beds. Section line corresponds to that marked in Fig. 2. (b) Schematic interpretation of cross-section in (a) at the time of sedimentation.
VSIC04 9/12/08 2:06 PM Page 102
R. A. F. Cas et al.
VSIC04 9/12/08 2:06 PM Page 103
Sedimentation in an extensional basin
103
Fig. 12. Schematic representation of the possible palaeogeographical context of the Bunga Beds of the Boyd Volcanic Complex. The section line approximates the coastal exposure and that represented in Fig. 11. Shown are basin-margin (fan-)deltas, subaqueous, high-relief lava domes and cryptodomes in the basin centre, localized dome-top tuff cones, and regional extensional tectonic context of the lacustrine basin.
Miocene Rubielos de Mora Basin of north-eastern Spain, which was a fault-bounded, extensional basin that is now about 8 km × 4 km in dimension. It is filled by up to 800 m of alluvial and lacustrine facies, including basin-margin progradational fluvio-deltaic and open lacustrine facies associations (Anadon et al., 1991). More appropriate lacustrine analogues include lakes of the East Africa Rift Zone, the origins of which include tectonic ponding (e.g. Lake Malawi, Scott et al., 1991; Lake Tanganyika, Rosendahl et al., 1986) and ponding against growing volcanic edifices (e.g. Lakes Kivu and Nakaru, Grove, 1986; Lakes Nakuru and Naivasha, Baker, 1986). Lacustrine sedimentation has been influenced by both basement and volcanic sources (Williams & Chapman, 1986; e.g. Lake Turkana, Yuretich, 1986; Lake Tanganyika, Baltzer, 1991), including both older volcanic rocks and active volcanic centres. In the East Africa Rift Zone, the lakes vary from small crater lakes up to huge tectonically and volcanically ponded lakes. Although the large lakes are
inappropriate analogues for the Bunga Beds lacustrine basin in terms of scale, their tectonic setting, origins and sedimentological similarities are appropriate general analogues for the Bunga Beds basin. Lake Malawi is up to 570 km long, up to 90 km wide, and up to 730 m deep. Its margins are fault scarps of exposed Precambrian basement gneisses and Cretaceous– Tertiary sedimentary rift basin successions, drained by short steep rivers. Gentler depositional slopes, drained by longer rivers, feed sediment into deltas, fan-deltas and sublacustrine fans in the lake (Scott et al., 1991). The young Rungwe Volcanics at the northern end contribute volcanic sediment to the lake. Maximum terrigenous sediment input is from tectonically active margins. The lake is permanently chemically stratified, and is anoxic below water depths of 200 m. Deltas and fans have well-developed distributary channels, and sediment-transport processes include traction and suspension in basin margins and turbidity current, debris-flow, slumping and suspension in deeper-water fan and open basin-floor environments (Scott et al., 1991).
VSIC04 9/12/08 2:06 PM Page 104
104
R. A. F. Cas et al.
Lake Tanganyika is similar, being 673 km long, up to 80 km wide, and up to 1470 m deep (Baltzer, 1991). It has steep margins and several sub-basins separated by tectonic scarps and cross-faults. Most rivers along the steep north–south-trending margins are short and drain Precambrian metamorphic basement and granites. The main northern river, the Rusizi River, drains through a volcanic complex. The longer rivers have developed fan-deltas into Lake Tanganyika, and in deeper waters, turbidites have produced small, base-of-slope fans. The lake is permanently anoxic at 100 m and greater depths. The inferred palaeoenvironments and facies of the Bunga Beds show many similarities to these larger African rift lake basins, including the presence of basin-margin fan-deltas or deltas fed by streams that drained metasedimentary and granitic basement, as well as volcanic sources. In addition, anoxic deeperwater environments existed in ‘basin-centre’ settings. Although it is impossible to use the water depths that mark the transition into deeper anoxic waters in Lake Malawi and Lake Tanganyika to constrain the water depths in the Bunga Beds basin, they at least provide an indication of the possible depths that may have existed in the Bunga Beds basin, suggesting water perhaps 100 –200 m deep, or more. The influence of palaeoenvironment on eruption style The dominant coherent and autoclastically fragmented rhyolite lavas and high-level, syn-sedimentation intrusions indicate that intralacustrine volcanism was largely non-explosive in the Bunga Beds basin. Although the top of the Bunga Beds is not exposed, it appears from the preserved succession that all eruptive activity was subaqueous, with the possible exception of the last phases of activity that built the tuff and pumice cones. The pumice breccias of the tuff cones contain highly vesiculated tube pumice and thin bubble-wall shards, indicating explosive vesiculation and fragmentation under low ambient pressuresa either shallow water or subaerial (McBirney, 1963; Fisher & Schmincke, 1984; Cas & Wright, 1987). The abundance of ignimbrites in the subaerial succession of the Boyd Volcanics around Eden (Fergusson et al., 1979; Cas & Bull, 1993), and their absence from the subaqueous Bunga Beds, suggests that environment may have had a significant effect on eruption style, water depth being perhaps sufficient to suppress voluminous explosive eruptions. Alternatively, the limited development of pyroclastic deposits in the Bunga Beds could reflect volatile-poor magmas.
However, the occurrence of some relict hornblende suggests that the rhyolites were hydrous magmas. In addition, if ambient hydrostatic pressures (= water depth) were low enough (< 200 m), then external water, superheated by the rising magmas, should have been able to expand explosively, leading to significant phreatic–phreatomagmatic activity (McBirney, 1963; Colgate & Sigurgeirsson, 1973; Peckover et al., 1973; Fisher & Schmincke, 1984; Kokelaar, 1986; Cas & Wright, 1987; Cas, 1992). As in situ pyroclastic deposits are limited to only two localized dome-top tuff cone successions, this suggests that only where intruding dome tops reached relatively shallow water depths and/or levels in the sediment pile did explosive activity occur (Cas et al., 1990). Elsewhere water depths and sediment load (for intrusions) were high enough to suppress explosive eruptions. Peckover et al. (1973) calculated that superheated sea water would be unable to expand explosively on contact with magma at depths > 700 m, and specific volume–pressure relationships assessed by McBirney (1963) showed that water heated to magmatic temperatures would be unable to expand rapidly enough to be explosive at water depths > 500–1000 m. Furthermore, all known historical marine explosive eruptions were initiated in water depths of only ≈ 200 m or less (e.g. Anak Krakatau 1927–1928, Moore, 1967; Myojin Reef 1952–1953, Morimoto, 1960; Tuluman Islands 1954–1957, Reynolds & Best, 1976; Reynolds et al., 1980; Capelhinos 1957–1958, Moore, 1967; Waters & Fisher, 1971; Surtsey 1963–1967, Thorarinsson et al., 1964; Thorarinsson, 1967; Kavachi 1950–1982, Johnson & Tuni, 1987; Izu, Japan, Yamamoto et al., 1991). Although there is no direct method of calculating the original water depth in ancient sedimentary successions, especially for the deep-water deposits within which most of the Bunga Beds rhyolites occur, the limited occurrence of in situ pyroclastic deposits suggests water depths of at least 200 m for most of the deep-water basin floor. This is consistent with water depths indicated for anoxic conditions in African rift lakes, and by the black mudstones of the Bunga Beds. The African rift lakes also indicate that water depths of hundreds of metres, enough to affect eruption style, are realistic in lacustrine basins in extensional tectonic terranes. Provenance affinities and classification of sediments in volcanically active lakes The diverse lithofacies preserved in the Bunga Beds are mirrored by very diverse petrofacies, and this
VSIC04 9/12/08 2:06 PM Page 105
105
Sedimentation in an extensional basin provides an opportunity to summarize the possible provenance influences in volcanically active basins and their successions. In this regard, ‘extralacustrine’ is used here for sediment derived from sources exposed beyond the lacustrine palaeoenvironment preserved in the Bunga Beds outlier, whereas ‘intralacustrine’ is used for clastic sources located within the lacustrine palaeoenvironment. In such basins the following provenance influences are possible for terrigenous and volcaniclastic deposits. 1 extralacustrine: terrigenous epiclastic; volcanic epiclastic; pyroclastic; syn-eruptive resedimented pyroclastic. 2 intralacustrine: volcanic autoclastic or hyaloclastic; intrusive autoclastic or hyaloclastic; pyroclastic; syn-eruptive resedimented pyroclastic; syn-eruptive resedimented autoclastic or hyaloclastic; volcanic epiclastic; terrigenous epiclastic. In addition, in lakes in appropriate climatic settings, chemical sediments of evaporitic origin and biogenic sediments may also occur, but these do not occur in the Bunga Beds. In the Bunga Beds, basement-derived, reworked and rounded metasedimentary and granitic detritus occurs in both the basin-margin and basin-centre facies associations, representing a sediment population that is clearly extralacustrine terrigenous epiclastic in the traditional sense, derived from older basement sources. Mixed with this basement-derived sediment are variable amounts of first-cycle volcanic detritus, marked by angular to subrounded volcanic quartz and feldspar grains. Mixing with basement-derived sediment and varying degrees of rounding of volcanic detritus suggest that it has been transported and mixed over at least moderate distances, and this suggests an extralacustrine source and transport in streams draining into the basin. The rare occurrence of juvenile glassy debris suggests that the volcanic detritus was not derived from strictly contemporaneous volcanism. This admixed volcanic debris is therefore classified here as extralacustrine volcanic epiclastic. By contrast, the 15 –20-m-thick interval of continuous-feed turbidites at the base of the northern basinmargin succession consists exclusively of glass shards and angular crystal grains, and clearly appears to have been derived from a contemporaneous pyroclastic volcanic event and source. Given that this facies inter-
val appears to occur low in the stratigraphy, and the only known intralacustrine pyroclastic successions are the tuff cone successions high in the stratigraphy, the facies should be classified as extralacustrine syn-volcanic resedimented pyroclastic in origin. An inferred stream system draining into the basin margin was flooded with juvenile volcanic detritus to the exclusion of basement-derived detritus, and fed the interpreted (fan-)delta environment of the northern basin-margin facies association. Although a direct pyroclastic source is not exposed, Edgar (1996) has found that the chemistry of the Bunga Beds rhyolites has A-type affinities, similar to the high-level, subvolcanic A-type Mumbulla granite suite that crops out 10 km to the west (Collins et al., 1982). The debrisflow deposits of the basin-centre facies association that overwhelmingly consist of scoria debris also appear to represent an extralacustrine (or lake margin) syneruptive resedimented pyroclastic source. In addition to extralacustrine sources, the Bunga Beds also contain intralacustrine sources of volcaniclastic debris. Intralacustrine volcanic autoclastic or hyaloclastic deposits include autobreccias and hyaloclastites, as well as intralacustrine intrusive autoclastic or hyaloclastic deposits, represented by autobreccias, hyaloclastite and peperite. Intralacustrine pyroclastic deposits are represented by the crystal and vitric tuffs and pumice breccias of the two preserved tuff and pumice cone successions. These also have associated with them intralacustrine syn-eruptive resedimented pyroclastic deposits. Where rising cryptodomes have breached the basin floor, intralacustrine syn-eruptive resedimented autoclastic or hyaloclastic and sedimentary breccias, with a mixed juvenile autoclastic volcanic and sedimentary intraclast clast population, can be identified. Although it is possible that in some volcanically active sedimentary basins there may also be intralacustrine volcanic epiclastic sources (i.e. exposures of much older volcanic rocks), these have not been recognized in the Bunga Beds, and they may be difficult to distinguish from extralacustrine volcanic epiclastic sources, unless there are clear compositional contrasts between identifiable intralacustrine and extralacustrine older volcanic source complexes.
CONCLUSIONS 1 The Bunga Beds outlier of the Upper Devonian Boyd Volcanic Complex preserves the remains of a small lacustrine basin of apparent extensional or transtensional tectonic origin.
VSIC04 9/12/08 2:06 PM Page 106
106
R. A. F. Cas et al.
2 Volcanism, sedimentation, syn-depositional magma intrusion and extensional or transtensional tectonics occurred simultaneously during basin evolution. 3 The basin succession preserves the remains of two moderately shallow-water basin-margin facies associations and an intervening deep-water basin-centre facies association. 4 Magmatic activity was bimodal (basalt–rhyolite) in composition, and is represented by syn-depositional intrusions, often with peperitic margins, partially extrusive domes, lavas, localized pyroclastic dome-top tuff cone successions, and variably resedimented volcaniclastic deposits of both intralacustrine and extralacustrine, and syn-eruptive and post-eruptive, origin. 5 The influences on sedimentation and provenance are therefore diverse, and the preserved succession provides an opportunity to review the very large number of potential sources of clastic deposits in volcanically and tectonically active basins.
ACKNOWLEDGEMENTS We thank Richard Hanson and Chuck Landis for thoughtful and helpful reviews. The research work was supported by an Australian Research Council Grant (No. A39331414) to R.A.F.C. and facilities of the Department of Earth Sciences at Monash University.
REFERENCES Anadon, P., Cabrera, Ll., Julia, R. & Marzo, M. (1991) Sequential arrangement and asymmetrical fill in the Miocene Rubelios de Mora Basin (northeast Spain). In: Lacustrine Facies Analysis (Eds Anadon, P., Cabrera, Ll. & Kelts, K.), Spec. Publs int. Assoc. Sediment., No. 13, pp. 257–275. Blackwell Scientific Publications, Oxford. Baker, B.H. (1986) Tectonics and volcanism of the southern Kenya Rift Valley and its influence on rift sedimentation. In: Sedimentation in the African Rifts (Eds Frostick, L.E., Renault, R.W., Reid, I. & Tiercelin, J.J.), Spec. Publ. geol. Soc. London, No. 25, pp. 45 –57. Geol. Soc. London, Bath. Baltzer, F. (1991) Late Pleistocene and Recent detrital sedimentation in the deep parts of northern Lake Tanganyika (East Africa rift). In: Lacustrine Facies Analysis (Eds Anadon, P., Cabrera, Ll. & Kelts, K.), Spec. Publs int. Assoc. Sediment., No. 13, pp. 147–173. Blackwell Scientific Publications, Oxford. Bouma, A.H. (1962) Sedimentology of some Flysch Deposits. A Graphic Approach to Facies Interpretation. Elsevier, Amsterdam. Cas, R.A.F. (1983) A Review of the Palaeogeography and Tectonic Development of the Palaeozoic Lachlan Fold Belt of Southeastern Australia. Geol. Soc. Aust. Spec. Publ., Perth, 10.
Cas, R.A.F. (1990) On the use of ‘epiclastic’ and other relevant terms in volcanic successions. Newslett. IAVCEI Comm. Volcanogenic Sediments, 3. Cas, R.A.F. (1991) Nomenclature of volcaniclastics. Newslett. IAVCEI Comm. Volcanogenic Sediments, 4, 3 – 6. Cas, R.A.F. (1992) Submarine volcanism: eruption styles, products, and relevance to understanding the host-rock successions to volcanic-hosted massive sulfide deposits. Econ. Geol., 87, 511–541. Cas, R.A.F. & Bull, S. (1993) Influence of Environment on Contrasting Styles of Volcanism: the Devonian Boyd Volcanic Complex, Southeastern Australia. Conference Field Guide, General Assembly IAVCEI, Canberra, 1993, Aust. geol. Surv. Org. Rec., 1993/57. Cas, R.A.F. & Wright, J.V. (1987) Volcanic Successionsb Modern and Ancient. A Geological Approach to Processes, Products and Successions. Allen and Unwin, London. Cas, R.A.F., Allen, R.L., Bull, S.W., Clifford, B.A. & Wright, J.V. (1990) Subaqueous rhyolitic dome-top tuff cones: a model based on the Devonian Bunga Beds, southeastern Australia, and a modern analogue. Bull. Volcanol., 52, 159 –174. Colgate, S.A. & Sigurgeirsson, T. (1973) Dynamic mixing of water and lava. Nature, 244, 552–555. Collins, W.J., Beams, S.D., White, A.J.R. & Chappell, B.W. (1982) Nature and origin of A-type granites with particular reference to southeastern Australia. Contrib. Mineral. Petrol., 80, 189 –200. Dadd, K. (1984) The stratigraphy and palaeogeography of the Eden–Comerong–Yalwal Rift Zone. Seventh Australian Geological Convention, Sydney. Geol. Soc. Aust. Abstr., 12, 126 –127. Dadd, K. (1992) The Middle to Late Devonian Eden– Comerong–Yalwal Volcanic Zone of southeastern Australia: an ancient analogue of the Yellowstone–Snake River Plain region of the USA. Tectonophysics, 214, 277– 291. De Rosen-Spence, A., Provost, G., Dimroth, E., Gochnauer, K. & Owen, V. (1980) Archaean subaqueous felsic flows, Rouyn–Noranda, Quebec, Canada, and their Quaternary equivalents. Precambrian Res., 12, 43 –77. Eaton, G.P. (1984) The Miocene Great Basin of western North America as an extending back-arc region. Tectonophysics, 102, 275 –295. Edgar, C. (1996) Volcanism and sedimentation in a Late Devonian rift basin: the Bunga Beds of the Boyd Volcanic Complex, southeastern New South Wales. BSc thesis, Monash University, Clayton. Fergusson, C.L., Cas, R.A.F., Collins, W.J., et al. (1979) The Upper Devonian Boyd Volcanic Complex, Eden, N.S.W. J. geol. Soc. Aust., 26, 87–105. Fisher, R.V. & Schmincke, H.-U. (1984) Pyroclastic Rocks. Springer, Berlin. Fiske, R.S. & Matsuda, T. (1964) Submarine equivalents of ash-flows in the Tokiwa Formation, Japan. Am. J. Sci., 262, 76 –106. Foley, G.A. (1986) Volcanic and sedimentary geology of a Devonian continental rift succession, Eden. BSc thesis, Monash University, Clayton. Furnes, H., Friedliefsson, I.B. & Atkins, F.B. (1980) Subglacial volcanicsaon the formation of acid hyaloclastites. J. Volcanol. geothermal Res., 8, 95 –110. Grove, A.T. (1986) Geomorphology of the African Rift system. In: Sedimentation in the African Rifts (Eds Frostick,
VSIC04 9/12/08 2:06 PM Page 107
Sedimentation in an extensional basin L.E., Renaut, R.W., Reid, I. & Tiercelin, J.J.), Spec. Publ. geol. Soc. London, No. 25, pp. 9 –16. Geol. Soc. London, Bath. Hanson, R.E. (1991) Quenching and hydroclastic disruption of andesitic to rhyolitic intrusions in a submarine islandarc sequence, northern Sierra Nevada, California. Geol. Soc. Am. Bull., 103, 804 – 816. Hanson, R.E. & Schweickert, R.A. (1982) Chilling and brecciation of a Devonian rhyolite sill intruded into wet sediments, northern Sierra Nevada, California. J. Geol., 90, 717–724. Hanson, R.E. & Wilson, T.J. (1993) Large-scale rhyolite peperites (Jurassic, southern Chile). J. Volcanol. geothermal Res., 54, 247–264. Hildreth, W. (1981) Gradients in silicic magma chambers: implications for lithospheric magmatism. J. geophys. Res., 86, 10153–10192. Johnson, R.W. & Tuni, D. (1987) Kavachi, an active forearc volcano in the western Solomon Islands: reported eruptions between 1950 and 1982. In: Marine Geology, Geophysics and Geochemistry of the Woodlark Basin– Solomon Islands (Eds Taylor, B. & Exon, N.F.), CircumPacific Council Energy Miner. Resour. Earth Sci. Serv., 7, 89–112. Kokelaar, P. (1982) Fluidisation of wet sediments during the emplacement and cooling of various igneous bodies. J. geol. Soc. London, 139, 21–33. Kokelaar, P. (1986) Magma–water interactions in subaqueous and emergent basaltic volcanism. Bull. Volcanol., 48, 275–290. Kokelaar, B.P., Bevins, R.E. & Roach, R.A. (1985) Submarine silicic volcanism and associated sedimentary and tectonic processes, Ramsey Island, S.W. Wales. J. geol. Soc. London, 142, 591– 613. Lambert, A. & Hsu, K.J. (1979) Non-annual cycles of varvelike sedimentation in Walensee, Switzerland. Sedimentology, 26, 453 – 461. Lowe, D.R. (1982) Sediment gravity flows: II, depositional models with special reference to the deposits of high density turbidity currents. J. sediment. Petrol., 52, 279 –297. McBirney, A.R. (1963) Factors governing the nature of submarine volcanism. Bull. Volcanol., 26, 443 – 453. McIlveen, G.R. (1974) The Eden–Comerong–Yalwal rift zone and the contained mineralization. Rec. N.S.W. geol. Surv., 16, 245 –277. McPhie, J., Doyle, M. & Allen, R. (1993) Volcanic Textures. A Guide to the Interpretation of Textures in Volcanic Rocks. University of Tasmania, Hobart. Moore, J.G. (1967) Base surge in recent volcanic eruptions. Bull. Volcanol., 30, 337–363. Morimoto, R. (1960) Submarine eruption of the Myojin Reef. Bull. Volcanol., 23, 151–160. O’Halloran, G.J. & Cas, R.A.F. (1995) Evidence for syndepositional deformation from the Late Devonian of the Mansfield Basin, east–central Victoria. Aust. J. Earth Sci., 42, 581–596. Peckover, R.S., Buchanan, D.J. & Ashby, D. (1973) Fuel– coolant interactions in submarine vulcanism. Nature, 245, 307–308. Pichler, H. (1965) Acid hyaloclastites. Bull. Volcanol., 28, 293–310. Pickering, K.T., Hiscott, R.N. & Hein, F.J. (1989) Deepmarine Environments: Clastic Sedimentation and Tectonics. Unwin Hyman, London.
107
Powell, C.McA. (1983) Tectonic relationship between the Late Ordovician and Late Silurian palaeogeographics of southeastern Australia. J. geol. Soc. Aust., 30, 353 –373. Powell, C.McA. (1984) Silurian to Mid-Devonian. In: Phanerozoic Earth History of Australia (Ed. Veevers, J.J.), pp. 309 –328. Oxford University Press, Oxford. Reynolds, M.A. & Best, J.G. (1976) Summary of the 1953 – 57 eruption of Tuluman volcano, Papua New Guinea. In: Volcanism in Australasia (Ed. Johnson, R.W.), Elsevier, Amsterdam, pp. 287–296. Reynolds, M.A., Best, J.G. & Johnson, R.W. (1980) 1953–57 Eruption of Tuluman Volcano: Rhyolitic Volcanic Activity in the Northern Bismarck Sea. Geol. Surv. Papua New Guinea Mem., 7. Port Moresby. Rosendahl, B.R., Reynolds, D.J., Lorber, P.M., et al. (1986) Structural expressions of rifting: lessons from Lake Tanganyika, Africa. In: Sedimentation in the African Rifts (Eds Frostick, L.E., Renaut, R.W., Reid, I. & Tiercelin, J.J.), Spec. Publ. geol. Soc. London, No. 25, pp. 29 – 43. Geol. Soc. London, Bath. Scott, P.A. (1972) The Devonian stratigraphy and sedimentology of the Araganui area. BSc thesis, Australian National University, Canberra. Scott, D.L., Ng’ang’a, P., Johnson, T.C. & Rosendahl, B.R. (1991) High resolution acoustic character of Lake Malawi (Nyasa), East Africa, its relationship to sedimentary processes. In: Lacustrine Facies Analysis (Eds Anadon, P., Cabrera, Ll. & Kelts, K.), Spec. Publs int. Assoc. Sediment., No. 13, pp. 129 –145. Blackwell Scientific Publications, Oxford. Steiner, J. (1972) The eruptive history and depositional environment of the Devonian extrusive rocks, Eden, New South Wales. J. geol. Soc. Aust., 19, 261–271. Steiner, J. (1975) The Merimbula Group of the Eden– Merimbula area, N.S.W. J. R. Soc. N.S.W., Proc., 108, 37–51. Sturm, M. & Matter, A. (1978) Turbidites and varves in Lake Brienz (Switzerland): deposition of clastic detritus by density currents. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker, M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 147–168. Blackwell Scientific Publications, Oxford. Thorarinsson, S. (1967) Surtsey: the New Island in the North Atlantic. Viking Press, New York. Thorarinsson, S., Einarsson, T., Sigvaldson, G.E. & Ellison, G. (1964) The submarine eruption off the Vestmann Islands. Bull. Volcanol., 27, 435 – 445. Walker, R.G. (1978) Deep-water sandstone facies and ancient submarine fans: models for exploration for stratigraphic traps. Bull. Am. Assoc. petrol. Geol., 62, 932–966. Waters, A.C. & Fisher, R.V. (1971) Base surges and their deposits: Capelhinos and Taal Volcanoes. J. geophys. Res., 76, 5596 –5614. Williams, L.A.J. & Chapman, G.R. (1986) Relationships between major structures, salic volcanism and sedimentation in the Kenya Rift from the equator northwards to Lake Turkana. In: Sedimentation in the African Rifts (Eds Frostick, L.E., Renaut, R.W., Reid, I. & Tiercelin, J.J.), Spec. Publ. geol. Soc. London, No. 25, pp. 59 –74. Geol. Soc. London, Bath. Yamagishi, H. (1987) Studies on the Neogene subaqueous lavas and hyaloclastites in Southwest Hokkaido. Report No. 59, Geological Survey of Hokkaido, Japan, 5 –117.
VSIC04 9/12/08 2:06 PM Page 108
108
R. A. F. Cas et al.
Yamagishi, H. (1991) Morphological features of Miocene submarine coherent lavas from the Green Tuff basins; examples from basaltic and andesitic rocks from the Shimokita Peninsula, northern Japan. Bull. Volcanol., 53, 173–181. Yamagishi, H. & Dimroth, E. (1985) A comparison of Miocene and Archaean rhyolite hyaloclastites: evidence for a hot and fluid rhyolite lava. J. Volcanol. geothermal Res., 23, 337–355.
Yamamoto, T., Soya, T., Suto, S., et al. (1991) The 1989 submarine eruption off eastern Izu Peninsula, Japan: ejecta and eruption mechanisms. Bull. Volcanol., 53, 301–308. Yuretich, R.F. (1986) Controls on the composition of modern sediments, Lake Turkana, Kenya. In: Sedimentation in the African Rifts (Eds Frostick, L.E., Renaut, R.W., Reid, I. & Tiercelin, J.J.), Spec. Publ. geol. Soc. London, No. 25, pp. 141–152. Geol. Soc. London, Bath.
VSIC05 9/12/08 2:05 PM Page 109
Spec. Publs int. Ass. Sediment. (2001) 30, 109 –140
Sedimentology and history of Lake Reporoa: an ephemeral supra-ignimbrite lake, Taupo Volcanic Zone, New Zealand V. MANVILLE*,† *Geology Department, University of Otago, PO Box 56, Dunedin, New Zealand
ABSTRACT The emplacement of a voluminous ignimbrite at the climax of the 1.8 ka Taupo eruption buried an area of ≈ 20 000 km2 in the central North Island of New Zealand. Infilling of valleys and the blockage of drainages by pyroclastic material destroyed hydrological networks and allowed the development of numerous temporary supra-ignimbrite lakes by the ponding of runoff on the modified landscape surface. Rapid erosion of landscape-mantling eruption products generated extremely high sediment yields and hence the rapid accumulation of locally thick lacustrine deposits. Most lakes were relatively short lived, as their natural dams of unconsolidated and poorly sorted pumiceous material were prone to rapid erosion and failure. The largest supra-ignimbrite lake yet identified occupied an area of 190 km2 at its maximum in the Reporoa basin, impounding ≈ 2.5 km3 of water. Its sedimentology is largely representative of other post-Taupo 1.8 ka temporary lakes, with primary pyroclastic units being overlain by the deposits of secondary pyroclastic flows and phreatic explosions, which were then transgressed by shoreline facies and draped by lacustrine beds that accumulated during lake filling. After breaching of the natural dam, the basin-floor sequence was incised by re-establishing dendritic drainage networks. Several decades after the formation and destruction of Lake Reporoa, the basin was again partially inundated by a flood triggered by the failure of the ignimbrite dam at the outlet of the intracaldera Lake Taupo. This event re-established the Waikato River in its present course, depositing an aggradational fan at the entrance to the basin that was modified and incised by waning flow. Local base levels were lowered by the entrenchment of the Waikato River, which caused a further phase of stream incision in hinterland areas.
INTRODUCTION 1997; Palmer & Shawkey, 1997), the Miocene Peach Springs Tuff (Buesch, 1991), the Miocene Jemez volcanic field (Mack et al., 1996), the Pliocene Tokai Group of Japan (Nakayama & Yoshikawa, 1997), and the Holocene Mount Mazama eruption (Nelson et al., 1988). The Taupo Volcanic Zone (TVZ), with its unique record of frequent large explosive eruptions from rhyolitic calderas (Fig. 1), provides a natural laboratory in which to investigate the aftermath of such events. The catastrophic erosion, transport and deposition of huge volumes of volcaniclastic material represent the most enduring and widespread hazards associated with volcanism. The most recent large-scale (30 km3 dense rock equivalent (DRE)) ignimbriteemplacing eruption occurred from the Taupo caldera only c. 1850 14C yr ago. The young age and degree
Explosive eruptions from andesitic–dacitic stratovolcanoes around the world in the last century have allowed the direct, detailed documentation of volcanic impacts on hydrological systems and the processes involved at cone-type volcanoes, e.g. Tarawera (White et al., 1997), Santa Maria (Kuenzi et al., 1979), Fuego (Vessell & Davies, 1981), Irazu (Waldron, 1967), Paricutín (Segerstrom, 1960), Mount St Helens (Lehre et al., 1983; Collins & Dunne, 1986) and Pinatubo (Pierson et al., 1992; Major et al., 1996; K. M. Scott et al., 1996). Evidence of the impact of large-scale ignimbrite-emplacing rhyolitic eruptions from caldera volcanoes, however, is mostly limited to ancient examples, i.e. the Eocene Challis volcanic field (Palmer, †Wairakei Research Centre, Private Bag 2000, Taupo, New Zealand
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
109
VSIC05 9/12/08 2:05 PM Page 110
110
V. Manville 1740E
1780E
360S
TVZ
380S
ROTORUA
400S 100 km
Waikato River
KAPENGA
MANGAKINO
OKATAINA
REPOROA MAROA WHAKAMARU AREA OF FIG. 2
TAUPO
30 km
caldera margin
TVZ boundary
Tongariro Volcanic Centre
stratovolcano 1760E
of preservation of these deposits, in association with newly recognized chronostratigraphic markers, allow good constraints to be placed on the relative and absolute timing and duration of various sedimentary responses that followed the Taupo 1.8 ka eruption. This contribution addresses four main issues related to the Taupo 1.8 ka eruption that have broader applications to the field of volcanic sedimentation: 1 documentation of the impact that the emplacement of a landscape-modifying ignimbrite had beyond the caldera; 2 a refinement of the stratigraphic record related to the Taupo 1.8 ka eruption;
Fig. 1. Map showing the location of the Taupo Volcanic Zone in the central North Island of New Zealand, known rhyolitic calderas, and the area discussed in this study.
3 the sedimentology and history of a supra-ignimbrite lake that developed after the blockage of drainages and is considered representative of others mapped in the TVZ; 4 the significance of the temporary storage of large volumes of water in elevated intracaldera and ephemeral supra-ignimbrite lakes to both short- and long-term readjustments in hydrological systems to caldera volcanism. Previous studies of the sedimentary response to this eruption have concentrated on the intracaldera environment (e.g. Smith, 1991a,b; Clarkson, 1996; Riggs et al., this volume), although aggradation in all North
VSIC05 9/12/08 2:05 PM Page 111
Ephemeral supra-ignimbrite lake Island rivers draining the central volcanic plateau also occurred (Fleming, 1953; Healy, 1967; Kear & Schofield, 1978). Generally, the stratigraphic record of volcanic areas is divided into relatively brief syneruption and much longer inter-eruption periods of deposition, distinguished by the overload of depositional systems by pyroclastic material or the dominance of normal ‘background’ conditions, respectively (Smith, 1991). The increased resolution in the aftermath of the Taupo 1.8 ka eruption allows further subdivision of this chronological framework with a true ‘syn-eruption’ period covering those events that occurred during the course of the actual eruption (i.e. Smith & Houghton, 1995); a ‘post-eruption’ interval after volcanic activity ceased but during which the sedimentary environment responded to the disturbance; and an ‘inter-eruption’ period when normal background processes operated and the landscape was stable. The formation of temporary lakes or the raising of extant ones is a common phenomenon in volcanic areas as a result of the blockage of drainages by a variety of processes, with important implications for the readjustment of drainage systems to eruptions. Barriers may include dams of primary pyroclastic material (Buesch, 1991; Smith, 1991a,b); Scott et al., 1996; White et al., 1997), lava flows (Hamblin, 1994), and lahar or other aggradational channel deposits (Kuenzi et al., 1979; Pierson et al., 1992; Umbal & Rodolfo, 1996). Sudden releases of temporarily stored water from such lakes are a significant hazard (e.g. Healy, 1954; Waythomas et al., 1996; White et al., 1997). In addition to the immediate phase of floodinduced erosion and/or aggradation, these break-outs trigger renewed periods of erosion in catchment areas by deepening trunk streams and lowering local base levels, causing aggradation downstream (White et al., 1997). The overall effect is to interrupt the normal pattern of exponential decline in post-eruptive sedimentary response (as measured by sediment yield), prolonging the response period and causing drainage networks to reintegrate in a step-wise fashion.
GEOLOGICAL SETTING The Taupo Volcanic Zone The 300-km-long TVZ in the central North Island of New Zealand (Fig. 1) is a product of the northwesterly subduction of the Pacific Plate at the southern end of the Tonga–Kermadec Trench (Cole, 1990; Houghton et al., 1995). It is defined by an envelope
111
around all identified or inferred vents active since 2.0 Ma and associated with a north-east-striking structural trend (Wilson et al., 1984). The central TVZ is an area of intense Quaternary silicic volcanism associated with rapid extension and thinning of continental crust (Houghton et al., 1995; Wilson, 1996). At least 34 caldera-forming ignimbrite eruptions have occurred from eight relatively short-lived, nested, and/or overlapping volcanic centres over the past 1.6 Myr (e.g. Wilson et al., 1984; Houghton et al., 1995; Wilson et al., 1995), making it comparable to the Yellowstone system in terms of lifespan, erupted volumes and area. Taupo contrasts with Yellowstone, however, in the very high frequency but relatively small size of caldera-forming eruptions (Wilson, 1993; Houghton et al., 1995). Depocentres in the TVZ are dominated by large lakes, giving rise to thick sequences of intercalated volcanic and lacustrine units in calderas and faultblock grabens. Basement is buried beneath 2.5–3 km of low-density volcanic material (Rogan, 1982), beyond the limit of geothermal drilling. The stratigraphic record in the central region reflects the varying nature and style of volcanism and the intermittent pattern of major caldera-forming eruptions (Smith, 1991a; Smith et al., 1993), which are also recorded in distal onshore and oceanic environments (e.g. Watkins & Huang, 1977; Kyle & Seward, 1984; Moore, 1991; Shane, 1991). The largest extant lakes, such as Taupo and Rotorua, were formed by caldera collapse and have 10–100 kyr lifespans, whereas smaller lakes fill eruption craters or valleys dammed by lava or ignimbrite (Healy, 1975). The oldest lacustrine deposits in the TVZ date back to > 600 ka, whereas later units of the Huka Group persist into late Pleistocene time (Grindley, 1965; Steiner, 1977). Mapped from drill-hole data and rare surface outcrop and underlying an area of 110 km × 40 km (Grange, 1937; Grindley, 1959, 1963, 1965; Steiner, 1963, 1977; Browne, 1971, 1973, 1978; Lloyd, 1972; Briggs, 1973; Wood, 1983; Bignall, 1990), these beds may have been deposited in one continuous lake, or in a series of lakes with migrating boundaries. The 26.5 ka Oruanui eruption terminated Huka Group sedimentation and confined Lake Taupo to its present area, blockage of the outlet briefly raising the lake level to c. 465 m above sea-level (a.s.l.) (von Hochstetter, 1864; Smith et al., 1993). A smaller lake marked by a shoreline at ≈ 360 m a.s.l. developed in the Reporoa basin (Fig. 2) to the north (Pain & Pullar, 1975; Smith et al., 1993). A similar pattern followed the 1.8 ka Taupo eruption: Lake Taupo temporarily refilled to ≈ 34 m above its present (and approximately pre-eruption) level, and
VSIC05 9/12/08 2:05 PM Page 112
112
V. Manville 176010’E
176020’E
PA ER OA
RA NG
E
315 m
317 m
tap u Wa io
Lake Ohakuri
am
Stre
318 m
Reporoa 318 m
38030’S
38030’S
Orakei Korako
319 m
I
Broadlands
ING AR OA P
HIG
LA TE
H
AU
320 m
Orakonui Stream
320 m
R KU
RA
KA
HO TA Aratiatia
Te Toke 322 m
Waikato River
5 km
Pueto Stream 176020’E
176010’E
300 m ASL contour ground above 320 m ASL Lake Reporoa palaeoshoreline Lake Reporoa palaeoshoreline (inferred)
317 m
maximum elevation of lacustrine features terraces related to Taupo break-out flood watercourses location of ignimbrite dam
VSIC05 9/12/08 2:05 PM Page 113
Ephemeral supra-ignimbrite lake blockage of the outlet to the Reporoa basin by pyroclastic material resulted in the development of a shallow supra-ignimbrite lake (Pain & Pullar, 1975). The Taupo 1.8 ka eruption The 1.8 ka Taupo eruption was one of the most violent and complex rhyolitic eruptions in the last 5 kyr (Wilson & Walker, 1985). Five phases of Plinian and phreatomagmatic fall deposits and intraplinian ignimbrite were followed by the cataclysmic Taupo ignimbrite (Wilson & Walker, 1985; Talbot et al., 1994; Smith & Houghton, 1995). A total eruptive volume of 105 km3 is estimated (30 km3 DRE), distributed as fall deposits > 10 cm thick over an area of 30 000 km2 east of the vent, and as a near-circular 20 000 km2 area of ignimbrite around the vent (Wilson, 1985; Wilson & Walker, 1985). The ignimbrite is composed of vesiculated rhyolitic pumice bombs, lapilli and coarse ash, platy and cuspate glass shards, < 5% lithic fragments and < 3.5% crystals (Wilson, 1985; Wilson & Walker, 1985). There are two topographically distinct variants: the ignimbrite-veneer deposit (IVD) is generally 0.5– 1.0 m thick and drapes topographic highs; the valleypond ignimbrite (VPI) infills valleys and depressions to depths of 5–60 m (Walker et al., 1981a,b; Wilson & Walker, 1982; Wilson, 1985). Many of the Taupo ignimbrite valley ponds and their adjacent veneer deposits are overlain by up to 1 m of matrix-rich bedded deposits, interpreted as the deposits of laterally moving flows (Wilson & Walker, 1985). Basal contacts are slightly reworked and may truncate segregation pipes in the underlying ignimbrite, suggesting a brief time-gap between the two. These beds are identified as layer 2c ignimbrite, deposited by pyroclastic flows draining off steep slopes (Wilson, 1985), and/or as secondary phreatic deposits produced by large phreatic explosions where water flashed to steam on contact with the hot ignimbrite shortly after emplacement (e.g. Rowley et al., 1981). Post-eruptive environment Emplacement of the Taupo ignimbrite at the climax of the 1.8 ka Taupo eruption had three main impacts on the landscape: the burial of the land surface beneath hot, unconsolidated and unsorted pyroclastic material; the destruction of hydrological networks by the infilling and blockage of valleys and channels; and the total destruction of all vegetation (Smith, 1991a).
113
The climate was similar to modern conditions, i.e. humid temperate with prevailing south-westerly winds and rainfall distributed fairly evenly throughout the year, with most high-intensity events occurring in the late summer (Thompson, 1984; Smith, 1991b). However, short-term weather patterns were affected by the virtual emptying of Lake Taupo and the injection of ≈ 30 km3 of fine-grained ash into the atmosphere (Wilson & Walker, 1985). Local weather patterns were also affected by the creation of a 20 000 km2 highalbedo–low-evapotranspiration pumice plain. Convective rainstorms probably increased in frequency for several years after the eruption as a result of the influence on the local climate (Froggatt, 1982; Grant, 1985; Wilson & Walker, 1985), and there may have been a fall in air temperature (Taylor et al., 1980). Deposition of the ignimbrite was immediately followed by a period of intense erosion and deposition as loose, unconsolidated pyroclastic material was remobilized by fluvial and mass-flow processes. Destruction of vegetation, which reduced losses by evapotranspiration and interception (Pierce et al., 1970), and reduced infiltration rates as a result of rainbeat compaction of the ignimbrite surface (Waldron, 1967; Leavesley et al., 1989) increased surface runoff to 80–90% of incident precipitation, similar to values recorded for clear-felled catchments (Pierce et al., 1970). Combined with large volumes of unconsolidated, fine-grained pyroclastic material, and exacerbated by the local climatic pattern of high-intensity rainstorms (Thompson, 1984), this increased sediment yields by two or three orders of magnitude in upland areas (Segerstrom, 1960; Waldron, 1967; Collins et al., 1983; Collins & Dunne, 1986). Erosion of the IVD facies produced dense networks of rills and gullies on ridges, the eroded material being deposited on VPI surfaces by debris flows, where it was prone to reworking by early, high-concentration, flood-prone ephemeral streams in unstable channels. Sediment supply to these areas declined over a period of months to years as upland rill networks consolidated and stabilized (Collins & Dunne, 1986), and where catchments were open the now-underconcentrated streams were able to erode box canyons in the VPI fill. Elsewhere, where catchments were closed by pyroclastic dams, bodies of standing water developed in depressions on the VPI surface. The emptying of these ephemeral lakes, after the breaching of their temporary dams by the arrival of headward-eroding gullies and/or piping or overtopping failure, triggered renewed
Fig. 2. (opposite) Outline topographic map of the Reporoa Basin, the area covered by Lake Reporoa and the location of some of the ignimbrite dams that blocked the Waikato River.
VSIC05 9/12/08 2:05 PM Page 114
114
V. Manville
stream erosion as water-courses readjusted their gradients to the new base levels. Some years to decades later, revegetation and the establishment of stable base levels by the integration of drainage networks ended the phase of syn-eruptive sedimentary response. Reporoa basin physiography The Reporoa basin, also known as the Reporoa depression or the Taupo–Reporoa basin (Healy, 1962; Pain & Pullar, 1975; Wilson et al., 1984), lies 25 km north-east of Lake Taupo (Figs 1 & 2). The name means ‘long swamp’ in Maori, attesting to its poor drainage. The northern part of the basin is formed by the 0.24 Ma Reporoa caldera (Nairn et al., 1994), whereas the southern end opens into a fault-angle depression between the eastward tilted Paeroa Block and the westward facing Kaingaroa Fault scarp (Modriniak & Studt, 1959). Thickness of the basin fill exceeds 2.5 km and consists of > 300 m of Huka Group lacustrine sediments overlying primary pyroclastic and reworked volcaniclastic units, and rhyolitic and dacitic lavas and breccias above proximal intracaldera Kaingaroa Ignimbrite (Nairn et al., 1994). Later post-Oruanui and post-Taupo 1.8 ka temporary lakes have left deposits and terraces at altitudes of ≈ 360, ≈ 330 and ≈ 320 m a.s.l. (Grange, 1937; Pain & Pullar, 1975; Smith et al., 1993). The outline of the Reporoa basin is effectively defined by the 320 and 340 m contours (Fig. 2). The main basin is elliptical, orientated NNE–SSW, with narrow arms extending to the south-west and west. The Waikato River, the overflow drainage from intracaldera Lake Taupo, flows for 9 km through a deeply incised gorge or spillway from the outlet before descending the Aratiatia cascade and entering the Reporoa basin at the head of the south-western arm. It then flows north along the margin of the northeasterly tilted Tahorakuri high, passing a constriction in the Broadlands area, which partitions off the smaller Te Toke sub-basin from the main basin to the north, before bending westwards through 170° around the nose of the Tahorakuri high and leaving the basin through the western arm. A narrow gorge cut through Huka Group material defines the downstream limit of the basin near Orakei Korako. Streams rising on the flanks of the Paeroa Range and the Kaingaroa Plateau traverse the northern Reporoa basin, which is drained by the partly incised Waiotapu Stream. Much of the basin floor is flat and lies between 290 and 305 m, apart from a few post-caldera rhyolite domes that project through the basin-fill and some low terraced
hills composed of Huka Group sediments, and 26.5 ka Oruanui pyroclastic and volcaniclastic deposits (Schofield, 1965; Wilson et al., 1984; Self & Healy, 1987), which rise to 40 m above the basin floor in the east, south-west and south. Taupo 1.8 ka pyroclastic deposits blanket the Reporoa basin, infilling depressions and mantling the entire land surface with a 0.5–1.0-m-thick regional chronostratigraphic marker.
LITHOFACIES AND PETROFACIES Sedimentary processes acting on the diverse primary pyroclastic deposits of the Taupo 1.8 ka eruption in a range of depositional environments produced a wide variety of lithofacies and petrofacies (Smith, 1991b; Riggs et al., this volume; White et al., this volume). Part of this variability arises from the wide range of particle shapes, sizes and densities, and hence differing hydraulic behaviour of the primary pyroclastic materials, which allows the extremely efficient partitioning of components during erosion, transport, and deposition. Four major components are distinguished in Taupo 1.8 ka material: pumice, fine vitric ash, lithic fragments, and crystals. ‘Pumice’ includes all particles of vesiculated rhyolite, from > 1.0-m-diameter blocks down to porous sand-sized grains. ‘Fine vitric ash’ refers to platy and/or cuspate shards of rhyolitic glass formed by the fragmentation of vesicular rhyolite to particles smaller than the vesicle size. The distinction has sedimentological implications. The density of pumice fragments varies according to their grain size, vesicularity, and degree of water saturation (Whitham & Sparks, 1986; Houghton & Wilson, 1989; Manville et al., 1998; White et al., this volume). Dry pumice has a relatively low density, inversely proportional to clast size. If its density is < 1.0 g cm–3, pumice may float for long periods before becoming sufficiently waterlogged to sink (Whitham & Sparks, 1986; Manville et al., 1998). In contrast, fine vitric ash has the density of rhyolitic glass (≈ 2.4 g cm–3) and sinks immediately. Lithic fragments (density 2.6–2.7 g cm–3) are dominated by varieties of rhyolites, but dacite, andesite, greywacke, welded ignimbrite and mudstone clasts are also present (Ewart, 1963; Walker et al., 1981a). Crystals are mainly andesine feldspar (2.7 g cm–3) in the medium to coarse sand fraction, with minor hypersthene and magnetite (3.5 and 5.2 g cm–3), which become relatively more abundant in the fine to very fine sand fraction. Quartz is a minor component at all grain sizes (Ewart, 1963). Charcoal is a distinctive but minor component of the Taupo 1.8 ka ignimbrite,
VSIC05 9/12/08 2:05 PM Page 115
Ephemeral supra-ignimbrite lake sourced from vegetation incorporated and carbonized by the pyroclastic flow, and varying in size from the trunks of mature forest trees to sand-sized fragments. In the sedimentary environment it behaves somewhat like pumice, being extremely soft and easy to abrade, of low density, and capable of floating temporarily before waterlogging and sinking. Lithofacies descriptions and interpretations The main lithofacies identified in resedimented pyroclastic material in the Reporoa basin are summarized in Table 1, together with their corresponding petrofacies. Some lithofacies are characteristic of a single depositional process operating in a particular sedimentary setting, but others may be produced by a variety of mechanisms in a range of environments. Therefore, careful consideration of facies associations and geometries is required to develop a detailed environmental model in later sections. Granulometric data were obtained from dry sieve methods for fractions coarser than 0.063 mm and Coulter (laser collimeter) techniques for finer fractions. Non-systematic variations in the density of volcaniclastic material (Smith & Smith, 1985) makes these methods of limited applicability, as they measure the dimensions of particles, not their hydrodynamic behaviour. Attempts to measure the latter for pumiceous sediments using rapidsediment analysers (automated settling tubes) is complicated by the buoyancy of pumice and the narrow range of grain sizes that can be used. Experiments show that saturated pumice pebbles and granules have a similar settling velocity to coarse lithic sands, and hence are hydraulically equivalent (Clarkson, 1996). Consequently, what appears from sieve and visual data to be a very poorly sorted mixed-component sediment may actually be well sorted hydraulically (White et al., this volume). For the purposes of this study, dimensional grain-size data are considered adequate for the description of primary and secondary pyroclastic deposits, and are acceptable for most subaqueous lithofacies from Lake Reporoa (with the exception of coarse fluvial pumice and lithic gravels) because these sediments are sufficiently fine grained for the low-density effects of pumice to be minimized (Manville et al., 1998). Broad trends such as an increase or decrease in fines or gravel content remain evident. Statistical parameters of sorting (inclusive graphic standard deviation, σ1), and mean grain size (graphic mean, Mz) were derived from the methods of Folk & Ward (1957).
115
Pumiceous diamict The pumiceous diamict facies consists of highly pumiceous poorly or unsorted material (σ1 = 2.72) with abundant matrix fines (Mz = 2.56φ) and minor crystal and lithic fragment components. Texturally it is almost identical to primary IVD facies (σ1 = 3.05, Mz = 2.91), but may be distinguished by the absence of segregation–degassing pipes, and the presence of vesicles in some cases. Most beds lack any bedding structures or fabric, although in some units a faint stratification is marked by bands of pumice clasts with local inverse or inverse-to-normal grading (Fig. 3a). The matrix may be unconsolidated or semilithified, and in some examples discrete rounded pockets of crystal–lithic-enriched matrix are developed. Beds are 0.3–1.0 m thick and tabular or lenticular with planar or gullied contacts, which may show large-scale convolute ball-and-pillow load structures. Upper surfaces may show evidence of reworking by tractional currents, i.e. improved sorting and the density segregation of grains. The similarity in structure, texture and composition to IVD deposits suggests that this lithofacies was formed by early remobilization of pyroclastic material without significant modification. However, lack of grading, organized clast fabrics, and poor sorting are characteristic of a variety of processes including subaerial and subaqueous debris flows (e.g. Johnson, 1970; Rodine & Johnson, 1976; Johnson & Rodine, 1984; Rodolfo & Arguden, 1991), high-concentration subaqueous turbidity currents (Lowe, 1982), and hot, dry, secondary pyroclastic flows (e.g. Torres et al., 1996). All of these processes may have been operative in the post-eruptive environment. Their deposits may be distinguished by subtle features or consideration of the enclosing units. For example, debris-flow deposits may be identified by features attributable to water saturation such as vesicles with fine-grained linings representing trapped air-bubbles, and pocket-shaped crystal–lithic segregations representing dewatering structures (Postma, 1983). Crude stratification and clast fabrics probably arose from shearing within laminar flows (i.e. Hampton, 1979; Nemec & Steel, 1984; Waresback & Turbeville, 1990). The deposits of high-concentration turbidity currents (Middleton & Hampton, 1976; Lowe, 1982) may be largely ungraded with a reverse-graded lower section, and interbedded with unequivocally subaqueous units.
‘Saturation grading’, settling from floating pumice raft Winnowing of primary pyroclastic deposits or early sedimentary sequence in high-energy environment by tractional currents, i.e. beach swash-zone or waning stage of flood
Inversely graded pumice gravels, vitric matrix often absent in upper part of bed Coarse, gravelly crystal–lithic sand, ± pumice granules, ± lithic pebbles
P, L
P
P
Pv
xl ± P ± L
xl, P
P
P, xl
v
v
v
v
Normally graded pumice gravel
Openwork pumice gravel
Inversely graded pumice gravel
Cross-bedded crystal– lithic pebbly sand
Horizontally stratified– laminated crystal–lithic sand
Massive–laminated pumice sand
Rippled sand
Massive–laminated vitric silt
Pumice dropstones
Vesiculated vitric ash
Accretionary lapilli bed
Suspension settling
Very fine-grained vitric silt, pumice shards, ± rounded pumice pebbles
Fine-grained vitric unit with abundant spherical rim-type accretionary lapilli concentrated near base. Regional extent, mantles earlier deposits
Fine-grained vitric ash with abundant air-bubble vesicles
Fallout from co-ignimbrite ash (layer 3), and remnants of Plinian plume
Accumulation of wet or damp ash aggregates, phreatic explosion deposits
Suspension settling of waterlogged clasts
(a) Symmetrical wave ripples: reworking by oscillatory tractional currents in shallow water above wave base (b) Asymmetrical current ripples: lower flow regime unidirectional tractional currents (c) Climbing ripples: as for current ripples with addition of net sediment deposition
Fine to medium pumiceous ± crystal–lithic sands, symmetrical oscillatory ripples with crystal–lithic crests or unidirectional climbing current ripples
Rounded pumice granule to cobbles, matrix-supported by vitric fines
Tractional currents. Turbidites and subaqueous mass flows. Hyperconcentrated stream flow
Massive, laminated, or inverse- to normally graded fine to coarse pumiceous sands
Tractional currents, upper flow regime. Swash-zone, shallow sheetflow. Secondary phreatic explosion surge beds
Beach strandline, back-beach berm
Subaqueous reworking of saturated pumice gravels, nearshore environment
Fluvial channel and bar deposits. Shallow braided streams. Tractional currents in submarine channels
116
Laminae 1–2 mm thick of alternating crystal–lithic and /or pumiceous fine to medium sand
Very coarse matrix-free pumice gravel, often very well rounded and smoothed
Normally graded fine pumice gravel, often matrix free. Occurs in onlapping lenses
Coarse pebbles to cobbles of well-rounded pumice, matrix free, trough and planar cross-bedding. Beds of lithic gravels also occur
Early, wet, mass flow remobilization of primary pyroclastic deposits. Subaerial and subaqueous debris flows. Hot secondary pyroclastic flows
Cross-bedded pumice gravel
Highly pumiceous and unsorted with abundant matrix fines, minor crystal and lithic content. Texturally almost identical to primary IVD but segregation pipes absent and vesicles may be present. Clast fabric generally absent, may show faint stratification marked by bands of pumice clasts with local inverse or inverse-to-normal grading. Thickness 0.3–10 m, tabular or lenticular. Erosive planar or gullied contacts, may be loaded. Tops may be reworked
Pv
Interpretation
Pumiceous diamict
Description
Petrofacies
Facies
Table 1. Summary of lithofacies and corresponding petrofacies and characteristic sedimentary structures in reworked Taupo 1.8 ka pyroclastic deposits. Petrofacies: P, pumice; v, vitric fines; xl, crystals; L, lithic fragments, ( ), minor components
VSIC05 9/12/08 2:05 PM Page 116
V. Manville
VSIC05 9/12/08 2:05 PM Page 117
Ephemeral supra-ignimbrite lake
117
(b)
(c) (a)
(e) (d) Fig. 3. Photographs of field exposures of typical lithofacies developed in the post-Taupo 1.8 ka sedimentary sequence. (A) Stacked pumiceous diamict beds. (Note erosive and pillowed lower contacts and variable sorting and size distribution of pumice clasts.) Large scale divisions 10 cm. (b) Cross-bedded pumice gravels, forming part of the aggradational fan deposited by the Taupo break-out flood in the Te Toke sub-basin. Large scale divisions 10 cm. (c) Onlapping lenses of normally graded pumice gravel (1) overlying pumiceous sand beds (2), which drape an inclined depositional surface cut into horizontally laminated pumiceous and crystal–lithic sands (3) and pumiceous diamicts (4) of the early remobilization sequence in the Reporoa basin. Length of tool shaft 18 cm. (d) Matrix-free, normally graded, openwork pumice gravel draping lacustrine massive and laminated vitric silts. Large scale divisions 10 cm. (e) Inversely graded pumice gravel interbedded with lacustrine massive and laminated vitric silts. Penknife 12 cm long. (continued p. 118)
VSIC05 9/12/08 2:05 PM Page 118
118
V. Manville material and will develop similar sedimentary fabrics (White et al., this volume). Openwork pumice gravel
(f) Fig. 3. (continued ) (f ) Tabular cross-bedded crystal–lithic coarse gravelly sands. Recessional terrace tread associated with the Lake Taupo break-out flood. Large scale divisions 10 cm.
Cross-bedded gravel The cross-bedded gravel lithofacies consists of matrixfree pebbles and cobbles organized into trough and planar cross-beds (Fig. 3b). Individual sets are up to 50 cm thick, and foreset dip is steeper in smaller troughs. Broad shallow channel structures are common. Two distinct petrofacies are shown by this lithofacies: 1 a very poorly sorted (σ1 = 3.28) pumiceous facies with well-rounded pumice pebbles and cobbles, minor charcoal fragments, and lenses of coarse pumice– crystal–lithic sand; 2 subangular to subrounded lithic gravels interbedded with coarse crystal–lithic sands. Similar facies are common in braided fluvial systems (e.g. Miall, 1977, 1978; Rust, 1978) and in submarine channels (Hein & Walker, 1982; Clifton, 1984), formed by the migration of gravel bedforms and the scouring and infilling of small-scale channels by tractional currents. Normally graded pumice gravel This lithofacies is made up of generally fine to medium-sized pumice pebbles showing normal grading (Fig. 3c). A vitric silt matrix is generally concentrated towards the base of each unit. It occurs in shallowly dipping, onlapping lenses or low-angle cross-beds on gentle slopes or as flat-lying units generally < 20 cm thick interbedded with laminated or massive vitric silts. This facies is interpreted as the product of subaqueous reworking of saturated pumice clasts by tractional currents. Once denser than water, pumice clasts behave in a similar fashion to normal quartz–feldspathic
The openwork pumice gravel facies is composed of normally or ungraded beds of matrix-free coarse particles (Fig. 3d). Individual clasts are generally very well rounded, some with a distinctive, almost ‘polished’ appearance, and very large cobbles or even small pumice boulders may occur, although lithic clasts are absent at all sizes. Individual beds vary from 5 to 30 cm in thickness and clast fabrics, with the exception of some steeply imbricated oblate clasts, are not developed. This facies is interpreted to be a beach berm feature (based on its distribution and lithofacies associations), created by wave action casting the largest (and lightest) clasts furthest up the beach beyond the reach of backwash (Bluck, 1967; Postma & Nemec, 1990). Inversely graded pumice gravel The inversely graded pumice gravel facies shows almost perfect inverse grading in units 2–20 cm thick that drape underlying topography (Fig. 3e). Dense crystal and lithic fragment components are absent although minor rounded charcoal fragments may occur. Where present, fine vitric matrix material is concentrated near basal contacts, and/or occurs as thin infiltration films coating clasts towards the top of beds. Clasts out of sequence with the inversely graded distribution, and often very steeply imbricated, may occur at any level within the beds. This lithofacies is inferred to result from the waterlogging and sinking of members of a stationary floating raft of pumice clasts on the surface of a standing body of water. The smallest particles saturate and sink first (Manville et al., 1998; White et al., this volume), giving rise to the ‘saturation-graded’ fabric. Very steep imbrication of elongate clasts is due to uneven saturation producing pumices with a buoyant end, which sink in a vertical orientation and are then buttressed upright by other grains settling around them. Cross-bedded crystal–lithic pebbly sand The cross-bedded crystal–lithic pebbly sand facies is petrographically distinct, consisting of dense volcaniclastic components forming low-angle planar or trough cross-bedded units (Fig. 3f ). Grain size is dominantly in the coarse to fine sand range, with occa-
VSIC05 9/12/08 2:05 PM Page 119
Ephemeral supra-ignimbrite lake sional much larger lithic clasts (σ1 = 1.34, Mz = 0.07). Foreset amplitude varies from 5 to 50 cm. This lithofacies is interpreted to have formed from the migration of bedforms and the infilling of erosional scours in shallow braided streams (Picard & High, 1973; Miall, 1977, 1978; Rust, 1978). Low-angle cross-beds are deposited by the migration of lowamplitude bedforms under upper flow regime conditions (Bridge & Best, 1988; Paola et al., 1989; Bridge & Best, 1997). Horizontally stratified–laminated crystal–lithic sand The horizontally stratified–laminated sand lithofacies is composed of very fine to coarse, moderately to poorly sorted sand (σ1 = 1.88), arranged into crudely stratified beds or stacked packages of multiple laterally continuous laminae parallel to bedding. Lamina thickness varies from only a few to many hundreds of grain diameters, with boundaries defined by subtle variations in grain density, sorting, size, and /or grainpacking density. Formative mechanisms include the migration of low-relief bedforms whose amplitude is limited by very shallow flow depths of a few centimetres in ephemeral streams (Smith, 1971); deposition from the migration of very-low-angle bedforms or sand sheets under upper flow regime conditions (Tunbridge, 1981; Allen, 1984; Bridge & Best, 1988, 1997; Paola et al., 1989); and grainby-grain deposition from hyperconcentrated streamflow (Beverage & Culbertson, 1964; Pierson & Scott, 1985). Relatively high-energy conditions are implied by the absence of low-density pumiceous material. Massive –laminated pumice sand The massive –laminated pumiceous sand facies is composed of poorly sorted pumiceous medium sand (σ1 = 1.63, Mz = 2.76) with occasional scattered coarser pumice clasts (Fig. 4a). Beds are massive to laminated, laminae are formed by variations in grain size and composition, with crystal–lithic components appearing in the very fine sand layers. This facies may be interbedded with rippled sands and massive to laminated silts. Deposition is inferred to have occurred from a similar range of mechanisms to those listed above, but with the addition of rapid settling from high-density turbidity currents (Lowe, 1982) in a subaqueous environment as indicated by the associated lithofacies. Lower-energy conditions or a difference in the composition of the sediment supplied is implied by the very high pumice content.
119
Rippled sands The rippled sand facies is made up of very fine- to medium-grained, pumiceous sands, with a variable crystal–lithic fragment component, which may become dominant. Sorting is better than for most other pumiceous lithofacies (σ1 = 1.20, Mz = 2.94). A range of ripple types is developed, including symmetrical ripples with rounded crests, asymmetrical ripples, and climbing ripples. The symmetrical ripples are characterized by linear, sometimes bifurcating crests in plan view, rounded cross-sectional profiles, and concentrations of coarser and/or denser grains on crests (Fig. 4b). Ripple amplitude is 2 cm or less and wavelength varies from 3 to 8 cm. Internal structures are complex and vary between adjacent sets with foresets organized into bi-directional cross-laminae bundles, tangential cross-laminae, drapes, and undulatory lower surfaces (De Raaf et al., 1977). Silt is concentrated along ripple-train contacts and partings, and gives rise to linsen and flaser bedding with increasing content (De Raaf et al., 1977). Asymmetrical ripples have distinct stoss and lee sides with well-defined foresets up to 5 cm high. Climbing ripples (Fig. 4c) are distinguished by cross-laminations in which ripples climb over downstream stoss sides (Jopling & Walker, 1968; Allen, 1971). Symmetrical ripples are usually interpreted as being the product of oscillatory or orbital water currents generated by wave action in shallow-water conditions (Reineck & Singh, 1980). Asymmetric ripples are formed by the migration of small bedforms under the influence of unidirectional tractional currents under lower flow regime conditions (Allen, 1968, 1969; Collinson & Thompson, 1982). Climbing ripples result from the simultaneous migration of ripple bedforms and net sediment deposition (Allen, 1971) at the base of dense sediment-laden currents, i.e. underflows or turbidity currents (Bouma, 1962; Jopling & Walker, 1968; Sturm & Matter, 1978). The angle of climb is a function of the ratio of the vertical sediment deposition rate and the lateral sediment transport rate (Allen, 1970, 1971). Massive–laminated vitric silt The massive–laminated vitric silt facies is characterized by beds of poorly sorted, very fine pumiceous sand and vitric silt (σ1 = 1.71, Mz = 4.91). Laminae consist of alternating seams of very fine vitric sand and silt, with rare clay partings (Fig. 4d). Interbeds of coarser pumice sand beds, thin pumiceous
VSIC05 9/12/08 2:05 PM Page 120
120
V. Manville
(b)
(a)
(c)
(e) (d) Fig. 4. Photographs of field exposures of typical lithofacies developed in the post-Taupo 1.8 ka sedimentary sequence. (a) Massive and crudely stratified, planar-bedded pumiceous sand and fine gravel. Grading patterns are inverse to normal. Pencil 15 cm long. (b) Oscillatory rippled fine to medium pumiceous sand. Upper part of fine sand section shows loading by overlying planar based subaqueous mass-flow deposit. Lake Reporoa transgressive shoreline sequence. Large scale divisions 10 cm. (c) Climbing rippled fine to medium pumiceous sand. Up-section evolution in ripple geometry indicates waning flow conditions. Interpretation is of deposition from a sediment-laden dense underflow on the floor of Lake Reporoa. Large scale divisions 10 cm. (d) Laminated and massive fine to very fine pumice sand and vitric silt. Basinal lacustrine facies of Lake Reporoa. Large scale divisions 10 cm. (e) Massive vitric silt containing rounded pumice pebble and cobble ‘dropstone’ clasts. Basinal lacustrine facies. Small scale divisions 1 cm. (f ) Accretionary lapilli bed (1), overlying thinly stratified pumiceous and crystal–lithic sands (2) is interpreted as secondary phreatic explosion ejecta. Base and top of sequence (3) consists of mass-flow deposits. Small scale divisions 1 cm. (continued )
VSIC05 9/12/08 2:05 PM Page 121
Ephemeral supra-ignimbrite lake
(f ) Fig. 4. (continued )
diamict units, or pumice dropstone-bearing layers may occur. These sediments are interpreted as being the result of the suspension settling of fines through the water column. Laminated beds accumulated below wave base where the laminae produced by intermittent introduction of material were preserved. Pumice dropstones The pumice dropstone lithofacies is extremely bimodal, consisting of sub- to well-rounded pumice clasts, varying in size from granules to cobbles, which ‘float’, matrix-supported, in beds of massive to laminated vitric fines (Fig. 4e). The pumice clasts may be isolated, or organized into single-clast-thick horizons, commonly associated with a slight coarsening in the matrix material. Laminations may be slightly deformed beneath the clasts and drape their tops. These deposits are interpreted to represent the intermittent waterlogging and sinking of individual pumice clasts from a wind-blown floating raft drifting on the surface of a body of standing water. Barely waterlogged pumice clasts have an effective settling velocity very close to that of silt- and clay-sized particles (White et al., this volume), and this characteristic allows them to gently touch down on the depositional surface, where they are draped and buried by background suspension settling of fines. Vesiculated vitric ash Thin, highly vesicular, beds of vitric silt, interbedded with crystal–lithic-rich fine sands are locally present low in the post-eruptive sequence. Characterized by an abundance of vermiform and subspherical vesicles
121
with a lining of fines, individual beds are only a few centimetres thick, and the total thickness of such deposits never exceeds 20 cm. Unidirectional lenticular or low-angle cross-bedding may be developed. These units are interpreted to result from the accumulation of wet or damp ash aggregates, which trapped air or steam within the matrix (Lorenz, 1974; Rosi, 1992). Deposition from laterally moving flows is indicated by current-bedding structures (e.g. Waters & Fisher, 1971; Cas & Wright, 1987). Outcrop distribution is insufficient to identify distal thinning trends. Stratigraphically they lie either side of the regional chronostratigraphic marker formed by the accretionary lapilli bed described below, suggesting that they were produced by multiple secondary phreatic explosions, probably occurring along watercourses buried by emplacement of the hot Taupo ignimbrite. Accretionary lapilli bed This fine-grained and very poorly sorted unit (σ1 = 3.08, Mz = 4.25) contains numerous spherical, rimtype accretionary lapilli (Schumacher & Schminke, 1991) up to 15 mm in diameter (Fig. 4f ). The bed shows normal coarse-tail grading of pumice, crystals and lithic fragments in a matrix of fine vitric ash. The accretionary lapilli consist of a finer-grained rim surrounding a core of slightly coarser, more porous vitric material, sometimes with a 1–4-mm-diameter pumice fragment kernel, and are concentrated towards the base. Accretionary lapilli form by the early aggregation of ash in wet conditions such as those occurring in phreatic base surges (e.g. Moore et al., 1966; Fisher & Waters, 1970; Waters & Fisher, 1971) or rain-flushing of eruption plumes (Self & Sparks, 1978).
LITHOFACIES ASSOCIATIONS AND DEPOSITIONAL ENVIRONMENTS On the basis of vertical and lateral facies analysis, the lithofacies above were grouped into seven distinct associations, inferred to have formed by sedimentary erosion, transport and depositional processes operating in particular depositional subenvironments within an overall lacustrine and fluvial setting (Table 2). These include: 1 early remobilization units (including the accretionary lapilli bed); 2 highstand palaeoshoreline; 3 transgressive shoreline;
Stacked and interlensing units on VPI surfaces Thin drapes on VPI surface Thin drape over topography Linear band associated with topographic bench/notch at 316 –320 m a.s.l. Thin sheet onlapping sloping topography up to 316 –320 m a.s.l. Low-lying flat basin floor surfaces Basinward thinning fans and wedges at tributary stream mouths
(2) Debris flows, hyperconcentrated streamflows
(3) Secondary phreatic deposits
(4) Accretionary lapilli bed
(1) Highstand palaeoshoreline
(2) Transgressive shoreline
(3) Basinal lacustrine
(4) Tributary inflow deltas–basin-floor fans
Massive–laminated vitric silt, rippled sand, openwork pumice gravel Cross-bedded pumice gravel, rippled sand, massive–laminated silt
Incised terrace surfaces stepping down to modern Waikato River Thin drape over sub-310 m a.s.l. areas of the Reporoa basin, off-channel embayments Low terrace surfaces adjacent to modern streams
(ii) Recessional terraces
(iii) Backwaters
(2) Post-Lake Reporoa to Recent alluvial
Cross-bedded crystal–lithic pebbly sand, horizontally bedded–laminated crystal–lithic sand
Downstream thinning fan surface in Te Toke sub-basin
Cross-bedded pumice gravel, cross-bedded crystal–lithic pebbly sand, horizontally bedded–laminated crystal–lithic sand
Pumiceous diamict, cross-bedded pumice gravel, graded pumice gravel, inversely graded pumice gravel, massive–laminated pumice sand, rippled sand
Massive–laminated vitric silt, pumice dropstones, inversely graded pumice gravel
Cross-bedded crystal–lithic pebbly sand, horizontally bedded and laminated crystal–lithic sands, graded pumice gravel, rippled sand, massive–laminated vitric silt
Openwork pumice gravel, cross-bedded crystal–lithic pebbly sand, horizontally bedded and laminated crystal–lithic sands
Accretionary lapilli bed
Vesiculated vitric ash, horizontally bedded and laminated crystal–lithic sands, rippled sands
Pumiceous diamict, horizontally bedded and laminated crystal–lithic sands, massive–laminated pumice sands
Pumiceous diamict
Major lithofacies
(i) Aggradational fan
(1) Lake Taupo break-out flood
Stacked and interlensing units on VPI surfaces
Distribution and geometry
(1) Secondary pyroclastic flows
Subenvironments
122
C, Post-Lake Reporoa
B, Lake Reporoa
A, Early remobilization units
Major associations
Table 2. Lithofacies associations and inferred depositional environments
VSIC05 9/12/08 2:05 PM Page 122
V. Manville
VSIC05 9/12/08 2:05 PM Page 123
Ephemeral supra-ignimbrite lake
123
Fig. 5. Representative stratigraphic logs through the primary suite of Taupo 1.8 ka pyroclastic units and the post-eruption sedimentary sequence in the Reporoa basin.
basinal lacustrine; basin-floor fans and deltas; Lake Taupo break-out flood; post-Lake Reporoa to Present. Reconstruction of the relationships and geometries of these packages of sediment was simplified by their youth and the degree of preservation of original physiographic features. Representative stratigraphic columns through post-1.8-ka sediments in the Reporoa basin are shown in Fig. 5. Stratigraphic thickness is strongly dependent on depositional environment, and lateral and vertical facies variations are rapid (e.g. Tucker, 1978; Van Dijk et al., 1978), but correlation between outcrops permits a schematic reconstruction of stratigraphy (Fig. 6). 4 5 6 7
Early remobilization deposits The generic term ‘early remobilization deposits’ refers to a complex suite of reworked pyroclastic sediments formed very shortly after emplacement of the Taupo ignimbrite, and directly overlying the primary, eruption-emplaced pyroclastic sequence. Many of these deposits are very similar texturally and compositionally to the ignimbrite, suggesting that sedimentary processes were largely limited to the redistribution of pyroclastic material. These processes involved various ignimbrite–water interactions at a range of temperatures and water contents. Early remobilization deposits are classified into four groups: 1 deposits of secondary pyroclastic flows;
; VSIC05 9/12/08 2:05 PM Page 124
124
V. Manville
m ASL 322
Maximum elevation of lacustrine sediments
318
Maximum elevation of surface decreases from 315 m ASL to 311 m ASL northwards
314 310 306
Recessional terraces step down towards Waikato River
Terrace surface in underlying topography
302
Waikato River
298
Buried palaeovalley
294 290
Pre-Taupo 1800a sequence: Huka Group, Oruanui pyroclastics, palaeosols, airfall tephras
Lacustrine sediments: laminated and massive pumice sands and silts
Primary Taupo 1800a pyroclastic sequence
Lake Taupo break-out flood deposits: pumiceous gravels and sands, boulder beds
Early remobilisation deposits: secondary pyroclastics, debris flows, ephemeral stream beds, accretion lapilli bed Transgressive shoreline deposits
Recessional terrace deposits: crystal-lithic rich gravels and coarse sands Highstand shoreline deposits
Tributary inflow deltas, basin floor fans
2 debris and hyperconcentrated flow deposits; 3 deposits of secondary phreatic eruptions; 4 the accretionary lapilli bed. The last is not strictly speaking a remobilization unit, but is included on the basis of its stratigraphic relationship with the others. The deposits of secondary pyroclastic flows are dominated by pumiceous diamict facies, and include both layer 2c ignimbrite (Wilson, 1985; Wilson & Walker, 1985) and later deposits formed by the remobilization of still-hot ignimbrite. Identification is based on stratigraphic relations including erosive and unconformable basal contacts on VPI facies, owing to their similarity in appearance, texture, and composition to primary pyroclastic deposits. Secondary pyroclastic flows were probably generated by a combination of steep local slopes and unconsolidated hot material, and various triggering mechanisms such
Fig. 6. Schematic stratigraphy of primary pyroclastic units and sedimentary volcaniclastic deposits in the Reporoa basin. Stratigraphic thicknesses are approximate; height of deposits and topographic surfaces have been corrected for post-1.8 ka tectonic tilting to normalize the highstand level of Lake Reporoa at 318 m a.s.l.
as intense ground-shaking associated with caldera collapse (Wilson & Walker, 1985) or the undercutting of VPI deposits by early stream processes (e.g. Torres et al., 1996). Debris- and hyperconcentrated-flow deposits are found throughout the Reporoa basin with a similar distribution to secondary pyroclastic deposits, i.e. adjacent to steep slopes and on VPI surfaces. Highly pumiceous and unsorted debris-flow deposits with abundant matrix fines are almost identical in texture and composition to primary ignimbrite facies (Fig. 7), but are distinguished by the absence of segregation pipes and/or the presence of vesicles that indicate a water-saturated matrix and hence emplacement at < 100°C. Debris-flow units 0.3–1.0 m thick are commonly stacked, with planar or gullied erosive basal contacts. Subspherical to vermicular fines-lined vesicles interpreted as trapped air-bubbles, and soft-
VSIC05 9/12/08 2:05 PM Page 125
Ephemeral supra-ignimbrite lake
125
Fig. 7. Granulometric analyses of representative samples from the primary Taupo 1.8 ka ignimbrite (veneer facies, IVD), compared with samples from early remobilization beds deposited by debris flows. Insets show components of: at left, debris-flow deposit; at right, ignimbrite. IVD material and early debris-flow deposits derived from it are texturally and compositionally similar.
sediment deformation and dewatering structures, including convolute bedding, load casts, and local pockets of crystal–lithic enrichment, indicate a watersaturated matrix. Faint stratification, marked by concentrated bands of pumice clasts with local inverse or inverse-to-normal grading, is locally developed, and some beds have reworked and winnowed tops with concentrations of crystal–lithic sands. Valleys draining steep catchment areas on the Kaingaroa Plateau (Fig. 2) are partially filled by thick accumulations of debris-flow deposits. Complex stratigraphic relationships and soft-sediment deformation structures imply multiple mass-flow events sourced from IVD-mantled uplands and deposited in a subaqueous environment in which saturated substrates were loaded by rapid deposition of overlying beds (e.g. Postma, 1983). More dilute, but still sediment-laden stream flows are also inferred to have been common in the posteruptive environment (Smith, 1991b), particularly on VPI surfaces. These hyperconcentrated early streams would have been shallow, unconfined and short-lived, triggered by runoff from brief intense rainstorms. Their deposits are preserved as sheet-like units of massive to laminated pumiceous sand (Fig. 4a), or crudely stratified to laminated crystal–lithic sands forming thin sheets, basal lags in gully fill sequences,
or reworked tops to mass-flow deposits. Very high sediment concentrations, including hyperconcentrated flow (Beverage & Culbertson, 1964; Costa, 1988), are likely to have been very common in the aftermath of the Taupo 1.8 ka eruption, as a result of the low density of pumice clasts. Neutrally buoyant particles can form up to 50 vol. % of a sediment–water flow without affecting its Newtonian behaviour (Bagnold, 1954, 1955). Thin vesiculated vitric ash layers interbedded with laminated or low-angle cross-bedded pumice and crystal–lithic fine to coarse sands are found throughout the Reporoa basin directly overlying primary or secondary pyroclastic deposits and close to the course of the modern Waikato River. They are interpreted as the deposits of secondary phreatic explosions, generated when water buried by the ignimbrite flashed to steam. No associated rootless phreatic explosion craters or rims (e.g. Rowley et al., 1981; Wilson & Walker, 1985; Moyer & Swanson, 1987), have been identified in the Reporoa basin. At Mount St Helens (Rowley et al., 1981; Moyer & Swanson, 1987) and Mount Pinatubo (Torres et al., 1996), such features typically formed along buried watercourses and were destroyed by erosion when streams reoccupied their channels. The vesiculated beds indicate deposition of
VSIC05 9/12/08 2:05 PM Page 126
126
V. Manville
wet or damp ash aggregates (Lorenz, 1974; Rosi, 1992), whereas lenticular or low-angle cross-bedding is consistent with deposition from laterally moving flows (e.g. Waters & Fisher, 1971; Cas & Wright, 1987). Stratigraphically they lie both above and below the regional chronostratigraphic marker formed by the accretionary lapilli bed described below, indicating the occurrence of multiple secondary phreatic explosions over a period of some time, although interfingering relationships and multidirectional transport trends cannot be distinguished. The distinctive accretionary lapilli bed is widely distributed in the Reporoa basin, directly overlying the ignimbrite or interbedded with early remobilization units (Figs 4f & 5). This bed was first identified southwest of Lake Taupo (Wilson & Walker, 1985) and preserved only where rapidly buried, i.e. by debris-flow or lake deposits, or by collapse into ground fissures to form clastic dykes. It was previously interpreted as a local fall deposit related to secondary phreatic explosions. Mapped over an area of > 500 km2 in the Reporoa area during the course of this study, it is now interpreted as a co-ignimbrite ash bed [layer 3 of Sparks et al. (1973) and Sparks & Walker (1977)] and forms an important new chronostratigraphic marker in the post-eruptive sequence. The relatively coarse pumice and crystal–lithic components are considered unlikely to have been derived by elutriation from the ignimbrite, but rather represent fallout from the remnants of the Plinian plume. The fine vitric ash that accumulated around these nuclei in the damp lower 15 km of the atmosphere to form the accretionary lapilli was derived both from atmospheric fallout from the plume and elutriation from the pyroclastic flow. Lake Reporoa highstand palaeoshoreline Prominent terraces or benches cut into the deposits of the Taupo 1.8 ka eruption have previously been mapped around the Reporoa basin at ≈ 305 m a.s.l. and ≈ 320 m a.s.l. (Healy, 1967; Pain & Pullar, 1975). Fieldwork for this study has identified a topographic bench and associated beach and lake sediments at 316 –322 m a.s.l. around the Reporoa basin (Figs 2 & 6), marking the highstand level of Lake Reporoa. Terrace treads and risers at other altitudes are inherited from the pre-eruption landscape or related to post-drainage events. The palaeoshoreline is variably developed around the Reporoa basin, being most prominent around the northern part of the main basin (Fig. 2), where wave fetch from prevailing south-westerly and southerly storms was greatest
(Thompson, 1984). It is marked by a wave-cut bench developed in the landscape-mantling primary pyroclastic sequence, and a geomorphological boundary between the smoothed surface formed by lacustrine deposits and the gullied subaerial ignimbrite surface. A steep riser up to 1 m high may be developed at the back of the beach where the IVD has been eroded by undercutting as the unconsolidated Plinian fall bed was washed out by wave action. The highstand bench is less clear on gentle slopes, the shoaling profile attenuating wave energy and allowing more rapid transgression during lake-level rise. Crystal- and lithic-rich, coarse gravelly sands at the toe of the notch onlap the unconformity cut into or through the primary sequence, and are interpreted as wave-winnowed lag deposits formed in the swash zone (Clifton et al., 1971). Strandline deposits of well-rounded, coarse openwork pumice gravels are commonly developed, formed by wave action casting relatively light pumice clasts high up the beach (Bluck, 1967). The variation in the altitude of highstand markers around the basin reflects post-1.8-ka regional tectonic subsidence. The pattern of tilting is similar to that measured by geodetic surveys over the last 20–40 yr (Blick & Otway, 1995), but the amount is approximately half what would be attained by extrapolating recent subsidence rates over the last 1800 yr. Transgressive shoreline deposits Lithofacies associations interpreted to result from the transgression of the shoreline during the rise in level of Lake Reporoa are most strongly developed on the steeply dipping flanks of older topographic rises, where the rate of lateral transgression was lowest and wave energy was not attenuated by a long shoaling zone. Onlapping sequences of distinctive lithofacies and fining-upward sections permit reconstruction of the shoreline environment (Fig. 8). Lithofacies and petrofacies (Fig. 9) reflect the different energies and patterns of water motion developed at various depths on the sloping shoreline by wave action (Clifton et al., 1971). In an idealized vertical section, planar-bedded or -laminated coarse, pebbly, crystal–lithic sands 5–15 cm thick with a shallow offshore dip overlie an onlapping planar or scalloped erosion surface that truncates or removes the 1.8 ka Taupo pyroclastic sequence. These sediments represent a lag of dense material reworked from the pyroclastic sequence and trapped in the swash zone (Clifton et al., 1971). The dense deposits pass upwards into 30–60 cm of decimetrethick onlapping lenses of matrix-free, normally graded,
VSIC05 9/12/08 2:05 PM Page 127
Ephemeral supra-ignimbrite lake
127
Fig. 8. Geometric reconstruction of onlapping transgressive shoreline deposits showing distribution of bedforms and inferred subenvironment. Inset shows idealized complete stratigraphic section. Ornament same as for Fig. 5.
Fig. 9. Granulometric analyses of representative samples from different areas of a transgressive shoreline sequence, highlighting varying energy of depositional environment, and degree of reworking and modification of primary pyroclastic materials. Insets show components of: at left, nearshore, onlapping lenses of inversely to normally graded pumiceous fine gravel; at right, coarse pebbly sands from the swash zone.
VSIC05 9/12/08 2:05 PM Page 128
128
V. Manville
moderately to well-rounded fine pumice gravel, interpreted as being deposited in slightly deeper water where breaking waves generated unidirectional tractional currents (Fig. 8). Finer pumiceous material was transported offshore. The pumice gravel lenses fine and thin upwards, becoming overlain by 10–20 cm of pumiceous medium to very fine sands showing complex cross-lamination patterns that increase in symmetry up-section, before grading into massive and laminated very fine pumice sand and vitric silt beds. The transition from symmetrical to asymmetrical ripples reflects the change from oscillatory to unidirectional wave-driven bottom currents as incident waves are slowed and distorted by frictional interaction with the shoreface in shoaling water (Clifton et al., 1971). Massive and laminated fine-grained sediments are interpreted to have been deposited in deeper water below effective wave base, the limit of effective wave-induced currents. The scale and geometry of outcrops indicates that the transition from swash zone sands through nearshore pumice gravels into offshore laminated fines occurred through a vertical distance of 2–3 m. This shallow depth to the wave base corroborates predictions of wavelength and wave period (0.6 – 4.0 m and 0.6 –1.6 s) based on reconstructions of effective wave fetch (5–12 km for the main Reporoa basin) from lake dimensions and climatic conditions; and palaeowave estimates based on the geometry and size of oscillatory ripples using the methods of Komar & Miller (1973) and Komar (1974), as summarized by Allen (1981). The thinness of the sequence reflects rapid transgression by the shoreline, relatively lowenergy wave conditions developed in the small and shallow lake, and a lack of clastic input. Shoreline facies are weakly developed on gently shelving surfaces, being restricted to a few centimetres of winnowed crystal–lithic sand over the IVD, as a result of rapid lateral migration of the active shoreline, absorption of wave energy by the shoaling profile, and the limited flux of sediment derived from shoreface erosion. Tributary inflow deltas and basin-floor fans Post-Taupo 1.8 ka volcaniclastic sediments are thickest where large catchment areas drained into relatively constricted parts of the Reporoa basin. For example, the Orakonui Stream drains an upland area of 77 km2 into a 500-m-wide valley just upstream of the Orakei Korako blockage (Fig. 2). Thick lacustrine deposits form a terrace at ≈ 317 m a.s.l., with the re-established Waikato River occupying a central channel. Similarly,
in the Te Toke sub-basin, runoff from the 177 km2 Pueto Stream catchment constructed a surface at ≈ 322 m a.s.l. (Fig. 2). This 2.5 km2 wedge is composed of coarser pumiceous material than typical lacustrine facies (Table 2). Tabular decimetre-thick beds of rounded pumice granules and pebbles in a matrix of fine pumiceous sand dominate, interbedded with bimodal fine vitric sand–foundered pumice clast beds, and layers of massive vitric silt. Coarse-tail inverse grading is sometimes developed in poorly sorted matrix-rich diamict units. Low-angle cross-bed foresets up to 20 cm high and unidirectional current ripples in fine pumiceous sands indicate easterly palaeocurrents. Bedding is generally horizontal, but low-angle cross-cutting relationships, shallow scours, and load structures are also present. Dense crystal– lithic components are restricted to the fine sand range. These sediment packages are interpreted as tributary inflow deltas and basin-floor fans. Exposure patterns and palaeocurrent indicators suggest a fan-like geometry with basinward decrease in grain size and bed thickness, typical of subaqueous fan deposits (e.g. Nelson & Nilsen, 1984). Deposition was largely from a variety of subaqueous sediment gravity flows including debris flows (massive and structureless), and high- and low-density turbidites (poorly sorted, normally graded beds; Lowe, 1982), against a background of suspension settling of fines and foundered pumices. These dense underflows were sourced from sedimentladen streamflows entering the lake (e.g. Lambert et al., 1976; Wright, 1977; Sturm & Matter, 1978; Pharo & Carmack, 1979). Very high sedimentation rates maintained shallow water depths opposite stream mouths during lake-level rise, preventing the development of Gilbert-type deltas, and inhibiting the distal transport of incoming sediment by frictional interaction between the lake bed and the river plume (Wright, 1977). Basinal lacustrine deposits The base of the lacustrine sequence is conformable over early remobilization units and may be marked by a thin transitional zone of interbedded crystal–lithic or pumiceous sands and bimodal beds of rounded pumice clasts in a fine sandy–silty matrix. Oscillatory ripples with crystal–lithic concentrations at their crests indicate shallow water depths, possibly in incipient ponds on the undulating ignimbrite surface. These thin units pass up into massive to laminated pumiceous silts and sands deposited below wave base (Fig. 4d), interbedded with 2–12-cm-thick layers of
VSIC05 9/12/08 2:05 PM Page 129
Ephemeral supra-ignimbrite lake
129
Fig. 10. Granulometric analyses of representative samples from the deep basinal lacustrine sedimentary environment.
gravelly pumice sand (Table 2). Modal grain size of many silt beds is 5.5–6.5φ (Fig. 10). Coarse-tail inverse-grading occurs in some matrix-supported pumice gravel beds, whereas others exhibit inverseto-normal grading. Inversely graded, matrix-free pumice gravel beds are interpreted as the products of ‘saturation-grading’ in which members of a population of pumice clasts floating on the surface of the lake become waterlogged and sink in order of increasing size (Bateman, 1965; Manville et al., 1998; White et al., this volume). Solitary pumice clasts or singleclast-thick seams of pumice pebbles surrounded by massive or laminated vitric fine sand and silt are the product of intermittent foundering of floating pumices against a background of suspension settling of fines. Sediments and beds thin and fine upwards, reflecting the deepening of Lake Reporoa during filling. Basinal lacustrine deposits thin and fine rapidly away from tributary inflows, transgressive shorelines, and deeper parts of the basin, forming a 20-cm-thick subsoil ‘pan’ of compact, massive, fine-grained material over much of the main Reporoa basin (Vucetich & Wells, 1978). Close to tributary inflows, sediments coarsen to include a higher proportion of pumice gravels, and features indicative of tractional currents such as climbing rippled sands and shallow scours become common. Lake Taupo break-out flood facies Lithofacies associations inferred to result from a catastrophic release of water from intracaldera Lake Taupo some time after the draining of Lake Reporoa occur in the Reporoa basin as far downstream as
Orakei Korako. This ≈ 20 km3 flood re-established the Waikato River in its present course (Manville et al., 1999). Three depositional subenvironments are defined on the basis of lithofacies, petrofacies, and distribution (Table 2). The absence of well-developed intermediate terraces related to temporary stillstands between the + 34 m shoreline of Lake Taupo and its present level indicate that drainage occurred in a single phase (Riggs et al., this volume). Between the outlet to Lake Taupo and Aratiatia the flood is marked by erosional trimlines in the IVD, vertical-sided spillways eroded through VPI palaeovalley fills, streamlined bedforms eroded into Huka Group sediments and Oruanui pyroclastic deposits, downstreamdipping fans of coarse lithic gravel, and boulder lags above basal erosion surfaces (Manville et al., 1999). Similar features are recorded from other examples of immense floods (e.g. Malde, 1968; Baker, 1973; Kehew & Lord, 1986). Estimates of peak discharge based on instantaneous failure of the barrier (Costa & Schuster, 1988) are in the range (40–90) × 103 m3 s–1, but field data such as maximum flood heights and boulder dimensions (i.e. channel cross-sectional area and flow velocity) give more reasonable values of (17–28) × 103 m3 s–1. Below the Aratiatia constriction, aggradational flood deposits form a broad downstream-dipping fan surface composed of coarsegrained pumiceous and crystal–lithic gravels and sands (Fig. 11). Fine-grained slack-water deposits accumulated in side tributaries (e.g. Baker et al., 1983), or formed a thin silty drape, now largely incorporated into the modern soil profile, over lowlying areas of the basin. Waning flow during the falling limb of the flood hydrograph cut multiple terraces
VSIC05 9/12/08 2:05 PM Page 130
130
V. Manville
Fig. 11. Granulometric analyses of representative samples taken from sediments deposited by the break-out flood from Lake Taupo. Sediments are distinguished on the basis of composition, and whether they were deposited during the rising and peak limb of the flood hydrograph (the aggradational phase) or during the waning limb (the recessional phase). Insets show components of: at left, aggradational-phase gravel and coarse sand; at right, recessional-phase gravel and coarse sand. (Note that in the recessional deposit the majority of the coarse pumice fraction has been winnowed out.)
into the aggradational fan deposits. These recessional features are capped by thin crystal–lithic coarse sands and gravels containing boulders up to 3 m in diameter. Boulder lithologies include rhyolitic and dacitic rocks from the Taupo 1.8 ka pyroclastic deposits and various ‘basement’-derived materials such as Huka Group sediments, Oruanui pyroclastic deposits, and spherulitic rhyolite from the Aratiatia dome. A second, broader fan opens into the Te Toke sub-basin with a depositional dip of 0.5 m km–1. Waning flow again cut a series of upstream-migrating arcuate knickpoints and recessional terraces through the centre of the fan, including a 6-m-high arcuate scarp truncating the Pueto Stream lacustrine fan (Fig. 2). The modern Waikato River is underfit in these features, occupying a narrow slot incised in the lowest part of the channel. Post-Lake Reporoa to Present The Reporoa basin has remained essentially unmodified since the re-establishment of the Waikato River.
The flood entrenched the river in its present deep channel, lowering base levels and triggering a phase of erosion in catchment areas as recorded by paired terraces in tributary valleys. Hydrothermal eruptions from the Waiotapu system and its neighbours ≈ 930 yr ago (Grindley, 1963, 1965) deposited heterogeneous assemblages of altered rocks and mud matrix over small areas in the northern basin. Peat swamps developed in poorly drained depressions. Recently, clearance of scrub for agriculture has renewed erosion of the Taupo ignimbrite in some areas (Vucetich & Wells, 1978). Recent alluvium consisting of reworked Taupo pyroclastic deposits and (along the Waiotapu Stream) hydrothermal materials forms low terraces adjacent to modern watercourses (Pain & Pullar, 1975; Vucetich & Wells, 1978). Construction of dams and control structures for hydro-electric purposes in the 1940s and 1950s at the outlet from Lake Taupo, and at Aratiatia and Ohakuri, have reduced the originally fast-flowing, steep-gradient Waikato River into a slow-flowing lake through the Reporoa Basin (e.g. Lloyd, 1972).
VSIC05 9/12/08 2:05 PM Page 131
Ephemeral supra-ignimbrite lake emplacement of Taupo Ignimbrite secondary pyroclastic flows secondary phreatic explosions accretionary lapilli bed debris flows early stream flows Lake Reporoa fills Lake Reporoa break-out flood readjustment to lowered base level Lake Taupo break-out flood Lake Taupo refills readjustment to lowered base level 0
1 minute 1 hour
1 day
1 week
1 month 1 year
10 years
100 years
Fig. 12. Chronology of post-1.8-ka sedimentary responses and events. Scale on x-axis is approximately logarithmic. Tapering shaded bars indicate duration and intensity of each process. Stars indicate geologically intantaneous events such as the emplacement of the ignimbrite or the break-out flood from Lake Taupo, which can be used as chronostratigraphic markers.
CHRONOLOGY AND DEPOSITIONAL MODEL The identification of a number of new chronostratigraphic markers, such as the accretionary lapilli bed and features related to the filling and draining of temporary lakes, permits new constraints to be placed on the relative and absolute timing and duration of sedimentary processes in the aftermath of the Taupo 1.8 ka eruption (Fig. 12). Reworking of the Taupo ignimbrite began immediately after its emplacement, with the activity of secondary pyroclastic flows caused by sloughing of hot IVD material off steep slopes, possibly associated with ground-shaking during caldera collapse. During the same period, water buried by VPI deposits along stream channels flashed to steam, building up sufficient vapour pressure to trigger secondary phreatic explosions with associated cratering and the redistribution of material by base surges, ballistic fallout, and convecting plumes. A constraint on the timing of these very early processes is provided by the interbedded accretionary lapilli bed. Pumice kernels too coarse to have been elutriated from the ignimbrite are found in some accretionary lapilli, indicating a component of fallout from the remnants of the Plinian plume. Pumice kernels 0.5 cm in diameter with a density of 0.5 g cm–3 would take ≈ 40 min to fall from an altitude of 50 km (Wilson, 1972), the inferred height of the Plinian eruption column (Walker, 1980), indicating that deposition of the accretionary lapilli
131
bed occurred < 1 h after emplacement of the ignimbrite. The duration of secondary phreatic explosions triggered by the incision of still-hot pyroclastic deposits by early stream flow depends on the rate of cooling of IVD deposits up to 40 m thick. Thin IVD facies must have cooled quickly, but thick VPI units up to 40 m thick probably stayed above 100°C for several years, most cooling being accomplished by the infiltration of meteoric water. At Mount Pinatubo, thick pyroclastic deposits remained hot enough to trigger secondary pyroclastic flows and phreatic explosions several years after the 1991 eruption (W. E. Scott et al., 1996; Torres et al., 1996), although their surfaces cooled sufficiently to support fluvial activity within days (e.g. Major et al., 1996; Pierson et al., 1996; K. M. Scott et al., 1996; Umbal & Rodolfo, 1996). Early remobilization processes on slopes and ridges were dominated by debris flows and high-sedimentconcentration streams, which eroded IVD material and redeposited it on flat VPI surfaces. Most erosion was probably accomplished by short-duration highintensity storms, forming a pattern of subparallel rills and gullies on hillslopes (e.g. Horton, 1945; Schumm, 1956). Most of these gullies terminate at the Lake Reporoa highstand shoreline, suggesting that rill formation largely ceased before filling was completed. As surface runoff became concentrated on the VPI surface, early fluvial processes were initiated, forming shallow channels that became filled with mass-flow, hyperconcentrated flow (Costa, 1988), ephemeral sheetwash (Hogg, 1982) and braided stream deposits. As water began to collect in depressions on the undulating ignimbrite surface, the early streams became submerged by shallow ponds, which gradually grew and joined up to form larger lakes. Depositional processes in these very shallow water bodies were dominated by rapid influxes of pumiceous material derived from adjacent slopes, depositing a mix of subaqueous mass-flows and suspension settling beds, which were modified by tractional currents and wave action. Blockage by ignimbrite of the outlet to the Reporoa basin near Orakei Korako led to the development of a shallow lake covering 190 km2, with an estimated highstand volume of 2.5 km3. Runoff from the total catchment area of 1100 km2 is estimated at 0.02– 0.03 m3 s–1 km–2, based on modern values for the Taupo–Reporoa area (Schouten et al., 1981), with the addition of a 50% factor related to the destruction of vegetation and soil infiltration capacity (Waldron, 1967; Pierce et al., 1970; Leavesley et al., 1989). Using these values and ignoring evaporative losses, Lake Reporoa would have filled to its highstand level within
VSIC05 9/12/08 2:05 PM Page 132
132
V. Manville
Fig. 13. Sketch of depositional model for Lake Reporoa at its highstand level. Numbers refer to subenvironments or depositional processes. 1, Reintegration of dendritic drainage networks in hinterlands. Streams are still typically ephemeral and braided. 2, Rill networks developed on ridges shortly after the eruption are now stabilized and are preserved above the highstand shoreline. Below it they are filled by lacustrine sediments or smoothed out by the transgressing high-energy beach. 3, Large tributary system contributing sediment-laden stream flow. 4, Basin-floor fan. Rate of transgression is higher than rate of sediment accumulation, causing sediment packages to onlap shorewards. 5, Lacustrine sediments are coarser and thicker close to sources of abundant clastic input; accumulation rates may keep pace with the rise in lake level. 6, Dense underflow of sediment-laden water sourced from mouth of tributary stream. 7, Low-density overflow containing floating pumice clasts sourced from mouth of tributary stream. 8, Floating pumice raft blown about the surface of the lake by shifting winds. Depleted by saturation of individual clasts, replenished by buoyant material brought in by rivers or eroded from the shoreline. 9, High-energy beach, opposite maximum wave fetch and prevailing winds. Wave action erodes the unconsolidated primary Taupo 1.8 ka pyroclastic deposits, forming a notch or topographic bench at the highstand lake level. 10, Highstand beach, marked by coarse crystal–lithic pebbly sands from which all low-density material has been winnowed and carried offshore, and possibly by a strandline or berm of matrix-free coarse openwork pumice pebbles and cobbles thrown above the highstand level by wave action. 11, Transgressive shoreline sequence. Energy of depositional environment decreases offshore with increasing water depth, causing changes in componentry and grain size of sediments. Oscillatory and shoaling ripples are developed above wave base (dashed line). 12, Deep basinal environment; suspension settling of fines and occasional saturated pumice clasts from the floating raft produces massive and laminated beds. Laminations reflect fluctuations in the sediment supply. 13, Mid-basinal topographic high; reduced period of immersion and elevation above most underflow currents produces a condensed lacustrine sequence. 14, Narrow arm off the main lake; high sediment influx from tributary streams allows sedimentation to keep pace with lake-level rise, producing thick lacustrine sequences. 15, Low-energy shoreline. Limited wave fetch, shoaling profile. Limited development of shoreline features.
3 yr. Re-establishment of a water table and evaporative losses would not alter this estimate substantially. Depositional environments and processes in Lake Reporoa are summarized in Fig. 13. Clastic input was largely from major tributary streams draining upland catchments as drainage networks reintegrated, with minor contributions from shoreline erosion and aeolian transport. Extensive deltas and basin-floor fans built out from the mouths of streams draining large catchments, as sediment-laden outflows generated by storm events rapidly decelerated by turbulent mixing with lake waters and frictional interaction with the shallow lake bed (Wright, 1977). Redistribution by turbidity currents was limited by shallow depths and low gradients on the basin floor, although some dense underflows carrying fine pumiceous sands were channelled along lows to form climbing-ripple beds. Sedimentation over much of the lake was by suspension settling of fines and waterlogging pumice. Rafts of floating pumice accumulated on the surface of Lake
Reporoa; these were replenished by periodic influxes of semisaturated material, and were pushed around by the wind. These rafts contributed to the rain-out of material, including fines generated by abrasion between jostling clasts and the intermittent sinking of waterlogged fragments. Palaeowave conditions estimated using the methods of Komar & Miller (1973) and Komar (1974), as summarized by Allen (1981), give wave periods in the range 0.6–1.6 s. Such waves would produce oscillatory symmetrical ripples in water depths of 0.2–2.0 m. The maximum value corresponds to the mean wave base and is in accordance with observed stratigraphic relationships between rippled sediments and highstand shorelines, and reconstructions of lake dimensions and fetch under current climatic conditions. The weak development of shoreline features around much of the Reporoa basin and consideration of the probably very high erodibility of an unconsolidated pyroclastic dam suggests that the highstand period
VSIC05 9/12/08 2:05 PM Page 133
Ephemeral supra-ignimbrite lake was brief. A limited stillstand may have occurred if stable outflow was achieved by surface flow across and /or percolation through the barrier. No trace of this blockage remains but its approximate location may be inferred from the topography in the Orakei Korako area. Downstream of the Reporoa basin the Waikato River valley narrows to a 100-m-wide gorge cut through a 360 m a.s.l. surface composed of silicified Huka Group beds overlain by Hinuera-age sediments (Schofield, 1965; Lloyd, 1972). The depth of this slot exceeded 110 m before the creation of Lake Ohakuri for hydro-electric purposes in 1961 (Lloyd, 1972). Lake Reporoa sediments form a surface at 317 m a.s.l. immediately upstream of this constriction, whereas surfaces underlain by resedimented 1.8 ka pumice were mapped 5 km downstream of the gorge at ≈ 305 m a.s.l. at Orakei Korako (Lloyd, 1972). Once initiated, probably by lake waters overtopping the barrier at its lowest point, failure is inferred to have been rapid. The breach– outlet channel grew quickly by a combination of vertical incision and headward migration of a knickpoint. The absence of intermediate shorelines below the highstand level indicates that drainage occurred in a single phase. Multiple beach terraces are a feature of regressive lake shores, formed by wave action during storms or temporary stillstands (Donovan & Archer, 1975; Riggs et al., this volume). On the basis of analogies with artificial and natural dam failures (e.g. Costa & Schuster, 1988), breaching of the barrier would have triggered a large flood. Potential evidence for such an event is now concealed beneath Lake Ohakuri and other hydro-electric lakes, although blocks of sinter up to 1 m in diameter derived from Orakei Korako were found in now-submerged pumiceous terrace sediments at 305 m a.s.l. (Lloyd, 1972). Peak discharges may be estimated using empirical equations derived from actual dam-breach floods (Costa & Schuster, 1988). For Lake Reporoa, a peak flow of ≈ 17 000 m3 s–1 is modelled, based on the potential energy of the volume excess of water and instantaneous failure of the barrier. However, the time necessary for breach growth, and flood-routing factors, would reduce this value by a factor of at least three (Fread, 1996). This compares with a modern average discharge for the Waikato River through the Reporoa basin of ≈ 130 m3 s–1. After drainage of Lake Reporoa, tributary streams regraded to the new local base level, and became integrated into the trunk stream out of the basin established by the break-out flood. Over a period of decades drainage networks contracted still further as
133
the number of tributaries declined as a result of decreased surface runoff. This was probably caused by the re-establishment of vegetation, which intercepted rainfall, bound the substrate with roots, and increased evaporation and infiltration. Over time, the resulting fall in the water table left many channels dry. Eventually, intracaldera Lake Taupo refilled and overtopped the pyroclastic barrier at its outlet, triggering a break-out flood that re-established the Waikato River (Manville et al., 1999). Estimates of the time taken to refill to its highstand level of ≈ 390 m a.s.l. vary from 15 yr (Smith, 1991a) to a maximum of 40 yr (C. J. N. Wilson, personal communication). This latter value includes the time necessary to saturate a sublacustrine basin infilled with over 0.5 km of pumice which probably represents the 1.8 ka caldera collapse structure (Davy & Caldwell, 1998). The break-out flood eroded a major spillway between the lake outlet and Aratiatia and deposited a coarse-grained pumiceous downstream-dipping aggradational fan over lacustrine sediments in the Te Toke sub-basin and the southern part of the main Reporoa basin. Slack-water deposits accumulated in lateral tributaries along the southwestern arm of the Reporoa basin (i.e. Baker et al., 1983), and 10–20 cm of massive silt, which sometimes contains pebbles of rounded pumice, draped earlier lacustrine deposits in the main basin. This silty layer has formed the basis of the modern soil and is rarely pristine. Waning flow cut down through the aggradational fan and the underlying lacustrine and primary beds, leaving horseshoe-shaped knickpoints and a winnowed lag of coarse crystal–lithic sands and gravels on recessional terraces. The modern Waikato River is at present confined to a narrow slot cut through ‘basement’ units along its pre-eruption course.
SEDIMENTARY RESPONSES TO CALDERA ERUPTIONS The products of the 1.8 ka Taupo eruption provide a unique opportunity to study in detail the sedimentary aftermath of a large ignimbrite-emplacing eruption from a rhyolitic caldera volcano. The response to much smaller eruptions from cone volcanoes is relatively well known by comparison (e.g. Kuenzi et al., 1979; Vessell & Davies, 1981; Pierson et al., 1992). At stratovolcanoes, most material is deposited close to the volcano, often on steep topography of the cone itself where it is prone to rapid remobilization by mass-flow and ephemeral fluvial processes (e.g. Segerstrom, 1960; Waldron, 1967; Ollier & Brown,
VSIC05 9/12/08 2:05 PM Page 134
134
V. Manville
Caldera - fluvial/lake system geometry
intra-caldera lake
topographic profile
Y
supra-ignimbrite lake
area of ignimbrite
X river
Sedimentary response at X B A
D C
F E
G
H
discharge sediment yield TIME
Sedimentary response at Y B
A
H
C
discharge sediment yield TIME
Fig. 14. Sedimentary responses to ignimbrite-emplacing eruptions from rhyolitic caldera volcanoes. The sketch shows the geometry of the caldera–lake–river system. Two alternative scenarios are modelled in terms of the variation in sediment yield and fluvial discharge over time: ( X ) downstream of the outlet to the intracaldera lake with the addition of a small supraignimbrite lake developed behind a blockage in the channel; and ( Y ) along another river system uncomplicated by lake storage. (A) Pre-eruption period; background levels. ( B) Emplacement of the ignimbrite; sedimentary systems massively overloaded with fresh clastic material and fluvial system is blocked. (C) Post-eruption period; sediment yield increases by an order of magnitude or more over normal background levels as sedimentary systems respond to the overload of material. Sediment yield declines approximately exponentially with time as pyroclastic material becomes stabilized by revegetation or stable drainage systems develop. Re-formation and reintegration of drainage leads to a gradual rise in net discharge at X. (D) Supra-ignimbrite lake breaches its dam, causing a brief flood and associated spike in sediment yield as material is transported by the flood waters. (E) The sudden release of water incises channel systems downstream, lowering local base levels and triggering a renewed period of erosion as stream profiles readjust. This causes further reintegration of drainages and a period of increased sediment yield, which then declines exponentially as the system re-equilibrates. (F) Catastrophic release of water from the intracaldera lake, which has refilled to a level where it can breach the barrier of pyroclastic material at its outlet. (G) The intracaldera lake breakout flood re-establishes the overflow river and incises the trunk channel. This renewed sudden drop in local base levels again triggers a renewed phase of readjustment by tributary streams and a further peak and decline in sediment yields. (H) Drainage systems are fully integrated and vegetation is re-established. Normal background conditions resume.
VSIC05 9/12/08 2:05 PM Page 135
135
Ephemeral supra-ignimbrite lake 1971; Palmer, 1991; Palmer et al., 1993). In contrast, large-scale ignimbrite eruptions deposit many cubic kilometres of unconsolidated to welded pyroclastic debris over huge areas (e.g. Wilson, 1985; Cas & Wright, 1987; Fisher et al., 1993), mantling the landscape and causing significant topographic modifications (Buesch, 1991). The nature and duration of the post-eruptive sedimentary response to both types of eruption depend on the magnitude, geomorphological setting, and violence of the eruption, and the contemporary climate. The response to the 26.5 ka Oruanui eruption (800 km3 erupted volume) from the Taupo caldera (Wilson et al., 1984; Self & Healy, 1987), which occurred during a Pleistocene glacial maximum, lasted several thousand years, depositing the Hinuera Formation (Schofield, 1965). In contrast, the response to the 1.8 ka Taupo eruption (105 km3 erupted volume) is estimated to have lasted only a few decades (Smith, 1991a; Smith et al., 1993). Sediment yields over time are an indicator of the response of sedimentary systems to perturbations. Yields from disturbed systems typically decline in a generally exponential pattern with time (Graf, 1977; Schumm & Rea, 1995), as easily eroded material is removed and drainage networks form, integrate, and stabilize. This pattern has been observed at Mount St Helens (Collins et al., 1983; Collins & Dunne, 1986) and Pinatubo (Pierson et al., 1992; Pierson & Costa, 1994). The development of temporary lakes in blockedoff drainages or the raising of existing ones in volcanic areas is a complicating factor, altering local base levels and storing large volumes of water in elevated positions (Fig. 14). Sudden releases from such lakes are a significant hazard (e.g. Waythomas et al., 1996; White et al., 1997). In addition to the immediate phase of flood-induced erosion and/or aggradation, these breakouts trigger renewed periods of erosion in catchment areas by deepening trunk streams and lowering local base levels (White et al., 1997). The products of these erosive phases cause aggradation downstream. Consequently, the sedimentary system is subjected to renewed disturbances years, decades, or perhaps even centuries after the initial impact of the eruption, prolonging the period of post-eruptive readjustment (Fig. 14). Drainage networks reintegrate in a step-wise fashion, as break-out floods incise trunk streams and connect drainage reaches. In the Reporoa basin, two separate falls in base level were triggered as much as several decades apart, first by the emptying of Lake Reporoa and later by the break-out from Lake Taupo (Fig. 14). This is recorded in the basin by multiple terraces in incised tributary streams, each terrace
reflecting a new drop in base level and an associated period of erosion. Rapid falls in base level produce headward migrating vertical incision in unconsolidated materials, whereas slow falls or stillstands result in lateral widening of the channel (Schumm, 1993). These events will rejuvenate a drainage network and deliver increased loads of sediment downstream (e.g. Schumm et al., 1984, 1987; White et al., 1997). Downstream stratigraphic sequences should contain stacked couplets consisting of flood deposits overlain by aggradational beds and separated by erosion surfaces.
ACKNOWLEDGEMENTS Research for this study was carried out under an Otago University postdoctoral fellowship in association with the Institute of Geological and Nuclear Sciences, funded by the Foundation for Research Science and Technology, New Zealand, Contract No. C05516. The author thanks J. D. L. White, B. F. Houghton, C. J. N. Wilson, M. R. Talbot and an anonymous reviewer for constructive comments, and N. R. Riggs for careful editing.
REFERENCES Allen, J.R.L. (1968) Current Ripples and their Relation to Patterns of Water Flow and Sediment Motion. North Holland, Amsterdam. Allen, J.R.L. (1969) Some recent advances in the physics of sedimentation. Proc. Geol. Assoc., 80, 1– 42. Allen, J.R.L. (1970) A quantitative model of climbing ripples and their cross-laminated deposits. Sedimentology, 14, 5–26. Allen, J.R.L. (1971) A theoretical and experimental study of climbing-ripple cross-lamination, with a field application to the Uppsala esker. Geogr. Annal., 53-A, 157–187. Allen, J.R.L. (1984) Sedimentary Structures. Their Character and Physical Basis, Vol. I. Developments in Sedimentology 30. Elsevier, Amsterdam. Allen, P.A. (1981) Wave-generated structures in the Devonian lacustrine sediments of south-east Shetland and ancient wave conditions. Sedimentology, 28, 369–379. Bagnold, R.A. (1954) Experiments on a gravity-free dispersion of large solid spheres in a Newtonian fluid under shear. Proc. R. Soc. London, Ser. A, 225, 49– 63. Bagnold, R.A. (1955) Some flume experiments on large grains but little denser than the transporting fluid, and their implications. Proc. Inst. civ. Engnrs, 4(3), 174 –205. Baker, V.R. (1973) Paleohydrology and Sedimentology of Lake Missoula Flooding in Eastern Washington. Geol. Soc. Am. Spec. Paper, Boulder, CO, 144. Baker, V.R., Kochel, R.C., Patton, P.C. & Pickup, G. (1983) Paleohydraulic analysis of Holocene flood slack-water
VSIC05 9/12/08 2:05 PM Page 136
136
V. Manville
sediments. In: Modern and Ancient Fluvial Systems (Eds Collinson, J.D. & Lewin, J.), Spec. Publs int. Assoc. Sediment., No. 6, pp. 229–239. Blackwell Scientific, Oxford. Bateman, P.C. (1965) Geology and Tungsten Mineralization of the Bishop District, California. US geol. Surv. Prof. Pap., Denver, CO, 470. Beverage, J.P. & Culbertson, J.K. (1964) Hyperconcentrations of suspended sediment. J. Hyd. Div., Proc. Am. Assoc. Civ. Eng., 90 (HY6), 117–128. Bignall, G. (1990) Hydrology and hydrothermal alteration, Reporoa Well (1), Reporoa, New Zealand. Proceedings, 12th New Zealand Geothermal Workshop, 257–264. Blick, G.H. & Otway, P.M. (1995) Regional vertical deformation from repeated precise levelling in the Taupo Volcanic Zone. Report 95/23, Inst. Geol. Nucl. Sci., Taupo. Bluck, B.J. (1967) Sedimentation of beach gravels: example from South Wales. J. sediment. Petrol., 37, 128–156. Bouma, A.H. (1962) Sedimentology of some Flysch Deposits. A Graphic Approach to Facies Interpretation. Elsevier, Amsterdam. Bridge, J.S. & Best, J. (1988) Flow, sediment transport and bedform dynamics over the transition from dunes to upper-stage plane beds: implications for the formation of planar laminae. Sedimentology, 35, 753–763. Bridge, J.S. & Best, J. (1997) Preservation of planar laminae due to migration of low-relief bed waves over aggrading upper-stage plane beds: comparison of experimental data with theory. Sedimentology, 44, 253–262. Briggs, N.D. (1973) Investigations of New Zealand pyroclastic flow deposits, New Zealand. PhD thesis, Victoria University of Wellington. Browne, P.R.L. (1971) Petrological logs of drillholes, Ohaaki– Broadlands geothermal field. Report 52, N.Z. geol. Surv. Wellington. Browne, P.R.L. (1973) The geology, mineralogy and geothermometry of the Ohaaki–Broadlands geothermal field, Taupo Volcanic Zone, New Zealand. PhD thesis, Victoria University of Wellington. Browne, P.R.L. (1978) Petrological logs of drillholes, Kawerau geothermal field. Report 84, N.Z. geol. Surv. Wellington. Buesch, D.C. (1991) Changes in depositional environments resulting from emplacement of a large-volume ignimbrite. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 45, 139–153. Cas, R.A.F. & Wright, J.V. (1987) Volcanic Successions. Allen and Unwin, London. Clarkson, R.A. (1996) Immediate post-c. 1800a Taupo eruption secondary deposits and shorelines. MSc thesis, University of Otago, Dunedin. Clifton, H.E. (1984) Sedimentation units in stratified resedimented conglomerate, Paleocene submarine canyon fill, Point Lobes, California. In: Sedimentology of Gravels and Conglomerates (Eds Koster, E.H. & Steel, R.J.), Mem. Can. Soc. petrol. Geol., Calgary, 10, 295–320. Clifton, H.E., Hunter, R.E. & Phillips, R.L. (1971) Depositional structures and processes in the non-barred high-energy nearshore. J. sediment. Petrol., 41, 651–670. Cole, J.W. (1990) Structural control and origin of volcanism in the Taupo volcanic zone, New Zealand. Bull. Volcanol., 52, 445–459. Collins, B.D. & Dunne, T. (1986) Erosion of tephra from the 1980 eruptions of Mount St. Helens. Geol. Soc. Am. Bull., 97, 896 –905.
Collins, B.D., Dunne, T. & Lehre, A.K. (1983) Erosion of tephra-covered hillslopes north of Mount St. Helens, Washington: May 1980 –May 1981. Z. Geomorphol. N.F. Suppl., 46, 103 –121. Collinson, J.D. & Thompson, D.B. (1982) Sedimentary Structures. George Allen & Unwin, London. Costa, J.E. (1988) Rheologic, geomorphic, and sedimentologic differentiation of water floods, hyperconcentrated flows, and debris flows. In: Flood Geomorphology (Eds Baker, V.R., Kochel, R.C. & Patton, P.C.), pp. 113– 122. Wiley, New York. Costa, J.E. & Schuster, R.L. (1988) The formation and failure of natural dams. Geol. Soc. Am. Bull., 100, 1054 – 1068. Davy, B.W. & Caldwell, T.G. (1998) Gravity, magnetic and seismic surveys of the caldera complex, Lake Taupo, North Island, New Zealand. J. Volcanol. geotherm. Res., 81, 69 – 89. De Raaf, J.F.M., Boersma, J.R. & van Gelder, A. (1977) Wave-generated structures and sequences from a shallow marine succession, Lower Carboniferous, County Cork, Ireland. Sedimentology, 24, 451– 483. Donovan, R.N. & Archer, R. (1975) Some sedimentological consequences of a fall in the level of Haweswater, Cumbria. Proc. Yorks. geol. Soc., 40, 547–562. Ewart, A. (1963) Petrology and petrogenesis of the Quaternary pumice ash in the Taupo area, New Zealand. J. Petrol., 4, 392– 431. Fisher, R.V. & Waters, A.C. (1970) Base surge bedforms in maar volcanoes. Am. J. Sci., 268, 157–180. Fisher, R.V., Orsi, G., Ort, M. & Heiken, G. (1993) Mobility of a large-volume pyroclastic flowaemplacement of the Campanian ignimbrite, Italy. J. Volcanol. geothermal Res., 56, 205–220. Fleming, C.A. (1953) The Geology of Wanganui Subdivision. N.Z. geol. Surv. Bull., 52. Wellington. Folk, R.L. & Ward, W.C. (1957) Brazos River Bar: a study in the significance of grain size parameters. J. sediment. Petrol., 27, 3–26. Fread, D.L. (1996) Dam-beach floods. In: Hydrology of Disasters (Ed. Singh, V.P.), pp. 85–126. Water Science and Technology Library, 24. Kluwer Academic, Dordrecht. Froggatt, P.C. (1982) Review of methods of estimating rhyolitic tephra volumes: applications to the Taupo Volcanic Zone, New Zealand. J. Volcanol. geothermal Res., 14, 301–318. Graf, W.L. (1977) The rate law in fluvial geomorphology. Am. J. Sci., 277, 178–191. Grange, L.I. (1937) The Geology of the Rotorua–Taupo Subdivision. N.Z. geol. Surv. Bull., 37. Wellington. Grant, P.J. (1985) Major periods of erosion and alluvial sedimentation in New Zealand during the Late Holocene. J. R. Soc. N.Z., 15, 67–121. Grindley, G.W. (1959) Waiotapu. N.Z. geol. Surv. 1 : 250 000 Geol. Map Ser., Sheet N85. Grindley, G.W. (1963) Geology and structure of Waiotapu geothermal field. In: Waiotapu Geothermal Field (Ed. J. Healy) N.Z. Dept. sci. ind. Res. Bull., Wellington, 155, 10–25. Grindley, G.W. (1965) The Geology, Structure, and Exploitation of the Wairakei Geothermal Field, Taupo, New Zealand. N.Z. geol. Surv. Bull., 75. Wellington. Hamblin, W.K. (1994) Late Cenozoic Lava Dams in the Western Grand Canyon. Geol. Soc. Am. Mem., Boulder, CO, 183.
VSIC05 9/12/08 2:05 PM Page 137
Ephemeral supra-ignimbrite lake Hampton, M.A. (1979) Buoyancy in debris flows. J. sediment Petrol., 49, 753 –758. Healy, J. (1954) Submission to the Tangiwai Railway Disaster Board of Inquiry: Report of the Board of Inquiry. Government Printer, Wellington, 1–31. Healy, J. (1962) Structure and volcanism in the Taupo volcanic zone, New Zealand. In: The Crust of the Pacific Basin (Eds MacDonald, G.A. & Kuno, H.), Geophys. Monogr., Am. geophys. Union, Washington, DC, 6, 151–157. Healy, J. (1967) Recent erosion in Taupo Pumice, central North Island, New Zealand. N.Z. J. Geol. Geophys., 10, 839–854. Healy, J. (1975) Volcanic lakes. In: New Zealand Lakes (Eds Jolly, V.H. & Brown, J.M.A.), pp. 70–83. Auckland University Press, Auckland. Hein, F.J. & Walker, R.G. (1982) The Cambro-Ordovician Cap Enrage Formation, Quebec, Canada: conglomeratic deposits of a braided submarine channel with terraces. Sedimentology, 29, 309–330. Hogg, S.E. (1982) Sheetfloods, sheetwash, sheetflow, or . . . ? Earth-Sci. Rev., 18, 59–76. Horton, R.E. (1945) Erosional development of streams and their drainage basins; hydrophysical approach to quantitative morphology. Geol. Soc. Am. Bull., 56, 275–370. Houghton, B.F. & Wilson, C.J.N. (1989) A vesicularity index for pyroclastic deposits. Bull. Volcanol., 51, 451–462. Houghton, B.F., Wilson, C.J.N., McWilliams, M.O., et al. (1995) Chronology and dynamics of a large silicic magmatic system: Central Taupo Volcanic Zone, New Zealand. Geology, 23, 13–16. Johnson, A.M. (1970) Physical Processes in Geology. Freeman, Cooper, San Francisco, CA. Johnson, A.M. & Rodine, J.R. (1984) Debris flow. In: Slope Instability (Eds Brunsden, D. & Prior, D.B.), pp. 257– 361. Wiley, New York. Jopling, A.V. & Walker, R.G. (1968) Morphology and origin of ripple-drift cross lamination, with examples from the Pleistocene of Massachusetts. J. sediment. Petrol., 38, 971–984. Kear, D. & Schofield, J.C. (1978) Geology of the Ngaruawahia Subdivision. N.Z. geol. Surv. Bull., 88. Kehew, A.E. & Lord, M.L. (1986) Origin and large-scale erosional features of glacial-lake spillways in the northern Great Plains. Geol. Soc. Am. Bull., 97, 162–177. Komar, P.D. (1974) Oscillatory ripples and the evaluation of ancient wave conditions and environments. J. sediment. Petrol., 44, 169–180. Komar, P.D. & Miller, M.C. (1973) The threshold of water movement under oscillatory water waves. J. sediment. Petrol., 43, 1101–1110. Kuenzi, W.D., Horst, O.H. & McGehee, R.V. (1979) Effect of volcanic activity on fluvial– deltaic sedimentation in a modern arc –trench gap, southwestern Guatemala. Geol. Soc. Am. Bull., 90, 827–838. Kyle, P.R. & Seward, D. (1984) Dispersed rhyolitic tephra from New Zealand in deep-sea sediments of the Southern Ocean. Geology, 12, 487– 490. Lambert, A.M., Kelts, K.R. & Marshall, N.F. (1976) Measurements of density underflows from Walansee, Switzerland. Sedimentology, 23, 87–105. Leavesley, G.H., Lusby, G.C. & Lichty, R.W. (1989) Infiltration and erosion characteristics of selected tephra
137
deposits from the 1980 eruption of Mount St. Helens, Washington, USA. J. Hydrol., 34, 339–353. Lehre, A.K., Collins, B.D. & Dunne, T. (1983) Post-eruption sediment budget for the North Fork Toutle River drainage, June 1980–June 1981. Z. Geomorphol. N.F., 46, 143–163. Lloyd, E.F. (1972) Geology and Hot Springs of Orakeikorako. N.Z. geol. Surv. Bull., 85. Wellington. Lorenz, V. (1974) Vesiculated tuffs and associated features. Sedimentology, 21, 273–291. Lowe, D.R. (1982) Sediment gravity flows: II. Depositional models with special reference to the deposits of high-density turbidity currents. J. sediment. Petrol., 52, 279–297. Mack, G.H., McIntosh, W.C., Leeder, M.R. & Monger, H.C. (1996) Plio-Pleistocene pumice floods in the ancestral Rio Grande, southern Rio Grande rift, USA. Sediment. Geol., 103, 1–8. Major, J.J., Janda, R.J. & Daag, A.S. (1996) Watershed disturbance and lahars on the east side of Mount Pinatubo during the mid-June 1991 eruptions. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 895–919. University of Washington Press, Seattle. Malde, H.E. (1968) The Catastrophic Late Pleistocene Bonneville Flood in the Snake River Plain, Idaho. US geol. Surv. Prof. Pap., Denver, CO, 596. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1998) The saturation behaviour of Taupo 1800a pumice and some sedimentological implications. Sediment. Geol., 119, 5 –16. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1999) Paleohydrology and sedimentology of a post-1.8 ka breakout flood from intracaldera Lake Taupo, North Island, New Zealand. Geol. Soc. Am. Bull., 111, 1435 –1447. Miall, A.D. (1977) A review of the braided-river depositional environment. Earth-Sci. Rev., 13, 1– 62. Miall, A.D. (1978) Lithofacies types and vertical profile models in braided river deposits: a summary. In: Fluvial Sedimentology (Ed. Miall, A.D.), Mem. Can. Soc. petrol. Geol., Calgary, 5, 597– 604. Middleton, G.V. & Hampton, M.A. (1976) Subaqueous sediment transport and deposition by sediment gravity flows. In: Marine Sediment Transport and Environment Management (Eds Stanley, D.J. & Swift, D.J.P.), pp. 197–218. Wiley, New York. Modriniak, N. & Studt, F.E. (1959) Geological structure and volcanism of the Taupo –Tarawera district. N.Z. J. Geol. Geophys., 2, 654 – 684. Moore, C.L. (1991) The distal terrestrial record of explosive rhyolitic volcanism: an example from Auckland, New Zealand. Sediment. Geol., 74, 25–38. Moore, J.G., Nakamura, K. & Alcarez, A. (1966) The 1965 eruption of Taal volcano. Science, 25, 955–960. Moyer, T.C. & Swanson, D.A. (1987) Secondary hydroeruptions in pyroclastic-flow deposits: examples from Mount St. Helens. J. Volcanol. geothermal Res., 32, 299–319. Nairn, I.A., Wood, C.P. & Bailey, R.A. (1994) The Reporoa Caldera, Taupo Volcanic Zone: source of the Kaingaroa Ignimbrites. Bull. Volcanol., 56, 529–537. Nakayama, K. & Yoshikawa, S. (1997) Depositional processes of primary to reworked volcaniclastics on an alluvial plain: an example from the Lower Pliocene Ohta tephra bed of the Tokai Group, central Japan. Sediment. Geol., 107, 211–229.
VSIC05 9/12/08 2:05 PM Page 138
138
V. Manville
Nelson, C.H. & Nilsen, T.H. (1984) Modern and Ancient Deep-sea Fan Sedimentation. Soc. econ. Palaeont. Miner., Tulsa, Short Course Notes, 14. Nelson, C.H., Carlson, P.R. & Bacon, C.R. (1988) The Mount Mazama climactic eruption (~6900 yr B.P.) and resulting convulsive sedimentation on the Crater Lake caldera floor, continent, and ocean basin. In: Sedimentologic Consequences of Conclusive Geologic Events (Ed. H.E. Clifton) Geol. Soc. Am. Spec. Pap., Boulder, CO, 229, 37–57. Nemec, W. & Steel, R.J. (1984) Alluvial and coastal conglomerates: their significant features and some comments on gravelly mass-flow deposits. In: Sedimentology of Gravels and Conglomerates (Eds E.H. Koster & R.G. Steel), Mem. Can. Soc. petrol. Geol., Calgary, 10, 1–31. Ollier, C.D. & Brown, M.J.F. (1971) Erosion of a young volcano in New Guinea. Z. Geomorphol. N.F., 15, 12–28. Pain, C.F. & Pullar, W.A. (1975) Chronology of ‘palaeosurfaces’ and present land surfaces in the Reporoa Basin, North Island, New Zealand. N.Z. J. Sci., 18, 313–322. Palmer, B.A. (1991) Holocene lahar deposits in the Whakapapa catchment, northwestern ring plain, Ruapehu volcano (North Island, New Zealand). N.Z. J. Geol. Geophys., 34, 177–190. Palmer, B.A. (1997) Sedimentary record of caldera-forming eruptions, Eocene Challis volcanic field, Idaho. Geol. Soc. Am. Bull., 109, 242–252. Palmer, B.A. & Shawkey, E.P. (1997) Lacustrine sedimentation processes and patterns during effusive and explosive volcanism, Challis volcanic field, Idaho. J. sediment. Res., 67, 154 –167. Palmer, B.A., Purves, A.M. & Donoghue, S.L. (1993) Controls on accumulation of a volcaniclastic fan, Ruapehu composite volcano, New Zealand. Bull. Volcanol., 55, 176–189. Paola, C., Wiele, S.M. & Reinhart, M.A. (1989) Upperregime parallel lamination as the result of turbulent sediment transport and low-amplitude bedforms. Sedimentology, 36, 47–59. Pharo, C.H. & Carmack, E.C. (1979) Sedimentation processes in a short residence-time intermontane lake, Kamloops Lake, British Columbia. Sedimentology, 26, 523–541. Picard, M.D. & High, L.R., Jr (1973) Sedimentary Structures of Ephemeral Streams. Developments in Sedimentology 17. Elsevier, Amsterdam. Pierce, R.S., Hornbeck, J.W., Likens, G.E. & Bormann, F.H. (1970) Effect of elimination of vegetation on stream water quantity and quality. In: Symposium on the Results of Research on Representative and Experimental Basins (Ed. C. Toebes) Int. Assoc. hydrol. Sci. Publ., 96, 311–328. Pierson, T.C. & Costa, J.E. (1994) Trends in sediment yield from volcanos following explosive eruptions. Geol. Soc. Am. Abstr. Prog., 26(7), A377. Pierson, T.C. & Scott, K.M. (1985) Downstream dilution of a lahar: transition from debris flow to hyperconcentrated streamflow. Water Resour. Res., 21, 1511–1524. Pierson, T.C., Janda, R.J., Umbal, J.V. & Daag, A.S. (1992) Immediate and long-term hazards from lahars and excess sedimentation in rivers draining Mt. Pinatubo, Philippines. Report 92-4039, US geol. Surv. Water-Resour. Invest., Denver, CO, 183–203. Pierson, T.C., Daag, A.S., Delos Reyes, P.J., Regalado, M.T.M., Solidum, R.U. & Tubianosa, B.S. (1996) Flow
and deposition of posteruption hot lahars on the eastern side of Mount Pinatubo, July– October 1991. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 921– 950. University of Washington Press, Seattle. Postma, G. (1983) Water escape structures in the context of a depositional model of a mass flow dominated conglomeratic fan-delta (Abrioja Formation, Pliocene, Almeria Basin, SE Spain). Sedimentology, 30, 91–103. Postma, G. & Nemec, W. (1990) Regressive and transgressive sequences in a raised Holocene gravelly beach, southwestern Crete. Sedimentology, 37, 907–920. Reineck, H.-E. & Singh, I.B. (1980) Depositional Sedimentary Environments. Springer, New York. Riggs, N.R., Ort, M., White, D.L., Wilson, C.J.N., Houghton, B.F. & Clarkson, R. (2001) Post-1.8-ka marginal sedimentation in Lake Taupo, New Zealand: effects of wave energy and sediment supply in a rapidly rising lake. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 151–179. Blackwell Science, Oxford. Rodine, J.D. & Johnson, A.M. (1976) The ability of debris, heavily freighted with coarse clastic materials, to flow on gentle slopes. Sedimentology, 23, 213–234. Rodolfo, K.S. & Arguden, A.T. (1991) Rain-lahar generation and sediment-delivery systems at Mayon Volcano, Philippines. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 45, 71–87. Rogan, A.M. (1982) A geophysical study of the Taupo volcanic zone, New Zealand. J. geophys. Res., 87, 4073– 4088. Rosi, M. (1992) A model for the formation of vesiculated tuff by the coalescence of accretionary lapilli. Bull. Volcanol., 54, 429– 434. Rowley, P.D., Kuntz, M.A. & MacLeod, N.S. (1981) Pyroclastic-flow deposits. In: The 1980 Eruptions of Mount St Helens, Washington (Eds Lipman, P.W. & Mullineaux, D.R.), US geol. Surv. Prof. Pap., Denver, CO, 1250, 489–512. Rust, B.R. (1978) Depositional models for braided alluvium. In: Fluvial Sedimentology (Ed. Miall, A.D.), Mem. Can. Soc. petrol. Geol., Calgary, 5, 605–625. Schofield, J.C. (1965) The Hinuera Formation and associated Quaternary events. N.Z. J. Geol. Geophys., 8, 772–791. Schouten, C.J., Terzaghi, W. & Gordon, Y. (1981) Summaries of Water Quality and Mass Transport Data for the Lake Taupo Catchment, New Zealand. Water Soil Misc. Publ., Wellington, 24. Schumacher, R. & Schminke, H.-U. (1991) Internal structure and occurrence of accretionary lapilliaa case study at Laacher See Volcano. Bull. Volcanol., 53, 612– 634. Schumm, S.A. (1956) Evolution of drainage systems and slopes in badlands at Perth Amboy, New Jersey. Geol. Soc. Am. Bull., 67, 597– 646. Schumm, S.A. (1993) River response to base level change: implications for sequence stratigraphy. J. Geol., 101, 279–294. Schumm, S.A. & Rea, D.K. (1995) Sediment yield from disturbed earth systems. Geology, 23, 391–394. Schumm, S.A., Harvey, M.D. & Watson, C.C. (1984) Incised Channels: Morphology, Dynamics and Control. Water Resources Publications, Littleton, CO.
VSIC05 9/12/08 2:05 PM Page 139
Ephemeral supra-ignimbrite lake Schumm, S.A., Mosley, M.P. & Weaver, W.E. (1987) Experimental Fluvial Geomorphology. Wiley, New York. Scott, K.M., Janda, R.J., de la Cruz, E.G., et al. (1996) Channel and sedimentation responses to large volumes of 1991 volcanic deposits on the east flank of Mount Pinatubo. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 971–988. University of Washington Press, Seattle. Scott, W.E., Hoblitt, R.E., Torres, R.C., Self, S., Martinez, M.M.L. & Nillos, T., Jr (1996) Pyroclastic flows of the June 15, 1991, climactic eruption of Mount Pinatubo. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 545–570. University of Washington Press, Seattle. Segerstrom, K. (1960) Erosion and Related Phenomena at Paricutin in 1957. US geol. Surv. Bull., Denver, CO, 1104-A. Self, S. & Sparks, R.S.J. (1978) Characteristics of widespread pyroclastic deposits formed by the interaction of silicic magma and water. Bull. Volcanol., 41, 196 –212. Self, S. & Healy, J. (1987) Wairakei Formation, New Zealand: stratigraphy and correlation, N.Z. J. Geol. Geophys., 30, 73–86. Shane, P.A.R. (1991) Remobilised silicic tuffs in middle Pleistocene fluvial sediments, southern North Island, New Zealand. N.Z. J. Geol. Geophys., 34, 489– 499. Smith, G.A. (1991) Facies sequences and geometries in continental volcaniclastic sediments. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 45, 109–121. Smith, G.A. & Smith, R.D. (1985) Specific gravity characteristics of recent volcaniclastic sediment: implications for sorting and grain size analysis. J. Geol., 93, 619–622. Smith, N.D. (1971) Pseudo-planar cross-stratification produced by very low amplitude sand waves. J. sediment. Petrol., 41, 69–73. Smith, R.C.M. (1991a) Landscape response to a major ignimbrite eruption, Taupo Volcanic Centre, New Zealand. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 45, 123–137. Smith, R.C.M. (1991b) Post-eruption sedimentation on the margin of a caldera lake, Taupo Volcanic Centre, New Zealand. Sediment. Geol., 74, 89–138. Smith, R.C.M., Smith, I.E.M., Browne, P.R.L. & Hochstein, M.P. (1993) Volcano-tectonic controls on sedimentation in the Taupo Volcanic Zone, New Zealand. In: South Pacific Sedimentary Basins. Sedimentary Basins of the World, 2 (Ed. Ballance, P.F.; Series Ed. Hsü, K.J.), pp. 143–156. Elsevier, Amsterdam. Smith, R.T. & Houghton, B.F. (1995) Vent migration and changing eruptive style during the 1800a Taupo eruption: new evidence from the Hatepe and Rotongaio phreatoplinian ashes. Bull. Volcanol., 57, 432– 439. Sparks, R.S.J. & Walker, G.P.L. (1977) The significance of vitric-enriched air-fall ashes associated with crystalenriched ignimbrites. J. Volcanol. geothermal Res., 2, 329–341. Sparks, R.S.J., Self, S. & Walker, G.P.L. (1973) Products of ignimbrite eruptions. Geology, 1, 115 –118. Steiner, A. (1963) The rocks penetrated by drillholes in the Waiotapu thermal area, and their hydrothermal alter-
139
ation. In: Waiotapu Geothermal Field (Ed. J. Healy) N.Z. Dept. sci. ind. Res. Bull., Wellington, 155, 26–34. Steiner, A. (1977) The Wairakei Geothermal Area, North Island, New Zealand: its Subsurface Geology and Hydrothermal Rock Alteration. N.Z. geol. Surv. Bull., 90. Wellington. Sturm, M. & Matter, A. (1978) Turbidites and varves in Lake Brienz (Switzerland): deposition of clastic detritus by density currents. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker, M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 147–168. Blackwell Scientific, Oxford. Talbot, J.P., Self, S. & Wilson, C.J.N. (1994) Dilute gravity current and rain-flushed ash deposits in the 1.8 ka Hatepe Plinian deposit, Taupo, New Zealand. Bull. Volcanol., 56, 538–551. Taylor, B.L., Tvzi, G.-C. & Schneider, S.H. (1980) Volcanic eruptions and long-term temperature records: an empirical search for cause and effect. Q. J. R. meteorol. Soc., 106, 175–199. Thompson, C.S. (1984) The Weather and Climate of the Tongariro Region. N.Z. meteorol. Serv. Misc. Publ., 115(14), Wellington. Torres, R.C., Self, S. & Martinez, M.M.L. (1996) Secondary pyroclastic flows from the June 15, 1991, ignimbrite of Mount Pinatubo. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 665–678. University of Washington Press, Seattle. Tucker, M.E. (1978) Triassic lacustrine sediments from South Wales: shore-zone clastics, evaporites and carbonates. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 205–224. Blackwell Scientific, Oxford. Tunbridge, I.P. (1981) Sandy high-energy flood sedimentationasome criteria for recognition, with an example from the Devonian of SW England. Sediment. Geol., 28, 79–95. Umbal, J.V. & Rodolfo, K.S. (1996) The 1991 lahars of southwestern Mount Pinatubo and evolution of the lahardammed Mapanuepe Lake. In: Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines (Eds Newhall, C.G. & Punongbayan, R.S.), pp. 951–970. University of Washington Press, Seattle. Van Dijk, D.E., Hobday, D.K. & Tankard, A.J. (1978) Permo-Triassic lacustrine deposits in the Eastern Karoo Basin, Natal, South Africa. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker, M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 225–239. Blackwell Scientific, Oxford. Vessell, R.K. & Davies, D.K. (1981) Nonmarine sedimentation in an active fore arc basin. In: Nonmarine Depositional Environments: Models for Exploration (Eds Ethridge, F.G. & Flores, R.M.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 31, 34 – 45. Von Hochstetter, F. (1864) Geology of New Zealand (Transl. & Ed. Fleming, C.A., 1959). Government Printer, Wellington. Vucetich, C.G. & Wells, N. (1978) Soils, Agriculture, and Forestry of Waiotapu Region, Central North Island, New Zealand. N.Z. Dep. sci. ind. Res., Soil Bur. Bull., Wellington, 31. Waldron, H.H. (1967) Debris Flow and Erosion Control Problems Caused by the Ash Eruptions of Irazu Volcano, Costa Rica. US geol. Surv. Bull., Denver, CO, 1241-I.
VSIC05 9/12/08 2:05 PM Page 140
140
V. Manville
Walker, G.P.L. (1980) The Taupo Pumice: product of the most powerful known (ultraplinian) eruption? J. Volcanol. geothermal Res., 8, 69–94. Walker, G.P.L., Self, S. & Froggatt, P.C. (1981a) The ground layer of the Taupo ignimbrite: a striking example of sedimentation from a pyroclastic flow. J. Volcanol. geothermal Res., 10, 1–11. Walker, G.P.L., Wilson, C.J.N. & Froggatt, P.C. (1981b) An ignimbrite veneer deposit: the trail-marker of a pyroclastic flow. J. Volcanol. geothermal Res., 9, 409–421. Waresback, D.R. & Turbeville, B.N. (1990) Evolution of a Plio-Pleistocene volcanogenic–alluvial fan: the Puye Formation, Jemez Mountains, New Mexico. Geol. Soc. Am. Bull., 102, 298–314. Waters, A.C. & Fisher, R.V. (1971) Base surges and their deposits: Capelinhos and Taal volcanoes. J. geophys. Res., 76, 5596–5614. Watkins, N.D. & Huang, T.C. (1977) Tephras in abyssal sediments east of the North Island, New Zealand: chronology, paleowind velocity, and paleoexplosivity. N.Z. J. Geol. Geophys., 20, 179–198. Waythomas, C.F., Walder, J.S., McGimsey, R.G. & Neal, C.A. (1996) A catastrophic flood caused by drainage of a caldera lake at Aniakchak Volcano, Alaska, and implications for volcanic-hazards assessment. Geol. Soc. Am. Bull., 108, 861–871. White, J.D.L., Houghton, B.F., Hodgson, K.A. & Wilson, C.J.N. (1997) Delayed sedimentary response to the A.D. 1886 eruption of Tarawera, New Zealand. Geology, 25, 459–462. White, J.D.L., Manville, V., Wilson, C.J.N., Houghton, B.F., Riggs, N.R. & Ort, M. (2000) Settling and deposition of ad 181 Taupo pumice in lacustrine and associated environments. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.),
Spec. Publs int. Assoc. Sediment., No. 30, pp. 141–151. Blackwell Science, Oxford. Whitham, A.G. & Sparks, R.S.J. (1986) Pumice. Bull. Volcanol., 48, 209–223. Wilson, C.J.N. (1985) The Taupo eruption, New Zealand. II. The Taupo Ignimbrite. Phil. Trans. R. Soc. London, Ser. A, 314, 229–310. Wilson, C.J.N. (1993) Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions from Taupo volcano, New Zealand. Phil. Trans. R. Soc. London, Ser. A, 343, 205–306. Wilson, C.J.N. (1996) Taupo’s atypical arc. Nature, 379, 27–28. Wilson, C.J.N. & Walker, G.P.L. (1982) Ignimbrite depositional facies: the anatomy of a pyroclastic flow. J. geol. Soc. London, 135, 581–592. Wilson, C.J.N. & Walker, G.P.L. (1985) The Taupo eruption, New Zealand. I. General aspects. Phil. Trans. R. Soc. London, Ser. A, 314, 199–228. Wilson, C.J.N., Rogan, A.M., Smith, I.E.M., Northey, D.J., Nairn, I.A. & Houghton, B.F. (1984) Caldera volcanoes of the Taupo Volcanic Zone, New Zealand. J. geophys. Res., 89(B10), 8463–8484. Wilson, C.J.N., Houghton, B.F., McWilliams, M.O., Lanphere, M.A., Weaver, S.D. & Briggs, R.M. (1995) Volcanic and structural evolution of Taupo Volcanic Zone, New Zealand: a review. J. Volcanol. geothermal Res., 68, 1–28. Wilson, L. (1972) Explosive volcanic eruptionsaII: the atmospheric trajectories of pyroclasts. Geophys. J. R. astron. Soc., 30, 381–392. Wood, C.P. (1983) Petrological logs of Drillholes BR 26–BR 40: Ohaaki–Broadlands geothermal field. Report 108, N.Z. geol. Surv. Wellington. Wright, L.D. (1977) Sediment transport and deposition at rivermouths: a synthesis. Geol. Soc. Am. Bull., 88, 857–868.
VSIC06 9/12/08 2:05 PM Page 141
Spec. Publs int. Ass. Sediment. (2001) 30, 141–150
Settling and deposition of AD 181 Taupo pumice in lacustrine and associated environments J. D. L. WHITE*, V. MANVILLE*1, C. J. N. WILSON†, B . F . H O U G H T O N † 2 , N . R . R I G G S ‡ and M . O R T ‡ *Geology Department, University of Otago, Dunedin, New Zealand; †IGNS Ltd, Private Bag 2000, Taupo, New Zealand; ‡Geology Department, Northern Arizona University, Flagstaff, AZ 86011, USA
ABSTRACT Pumice is an unusual geological material. It is of low density, its density can vary, reversibly, through time, and it is capable of floating in water. These properties result directly from the abundance and interconnectedness of vesicles, which in sedimentary environments contain air and water in varying proportions. The variable density and sometimes positive buoyancy of pumice in water lead to unusual transport properties that complicate attempts to interpret the energy of depositional environments in which it is deposited. Experimental settling of ad 181 Taupo pumice confirms the general observation that larger clasts are the last to settle, indicating that progressive saturation and sinking of clasts from a pumice raft can produce a reverse-graded bed (saturation grading). Saturation of pumice clasts with water is mediated by inhomogeneities in the vesicle population, and in particular by more rapid transport of water through larger vesicles into the interiors of the clasts. Experiments designed to evaluate the behaviour of pumice clasts after they have become saturated show that although larger clasts retain slightly lower bulk density than smaller ones, fall velocities are nevertheless proportional to grain size. Sorting of saturated pumice by fall velocity therefore produces normally graded pumice beds (redeposition grading). Subaqueous deposition of pumice from currents results in a range of conventional styles of cross-bedding, but also produces distinctive steeply imbricated clast fabrics developed by the progressive growth of bedload cluster bedforms.
INTRODUCTION 1 stranding, in which floating pumice lodges against a depositional surface and remains as water drains away; 2 waterlogging, in which floating pumice fragments become water saturated and sink to the depositional surface; 3 saturated-clast redeposition, in which watersaturated clasts are re-entrained and deposited as nonbuoyant particles by normal sedimentary processes. We address each of these styles of deposition with reference to lacustrine deposits formed during and after the ad 181 Taupo eruption (Zielinski et al., 1994). The Taupo Volcanic Zone, in the central North Island of New Zealand (Fig. 1), is characterized by frequent large explosive eruptions from rhyolitic calderas (Houghton et al., 1995), and provides a natural laboratory for studies of pumice behaviour in the sedimentary environment (Tilly, 1987; Smith, 1991b;
It is well known that pumice clasts can float on water, that larger clasts float longer than smaller ones, and that fragments in time become water saturated and sink. These facts suggest that water-settled pumice will be deposited in reverse-graded beds as first small clasts and then larger ones saturate and sink (Bateman, 1965; Manville et al., 1998). Once fully water saturated, pumice clasts sink at rates proportional to grain size, albeit more slowly than dense lithic particles of equivalent size (Tilly, 1987; Cashman & Fiske, 1991). For pumice in general, aqueous deposition can be considered to occur in three ways:
1 Present address: IGNS Ltd, Private Bag 2000, Taupo 2730, New Zealand. 2 Present address: Geology Department, SOEST, University of Hawaii, Honolulu, HI, USA.
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
141
VSIC06 9/12/08 2:05 PM Page 142
142
J. D. L. White et al.
10 km 178 E
174 E
100 km 36 S
TVZ
40 S
Lake Taupo
Present shoreline
Lake Taupo
Highstand shoreline cut into ignimbrite or defined by older cliffs
Area covered during +34 m highstand
Low-relief or buried highstand shoreline
Fig. 1. Map showing lacustrine volcaniclastic sediments preserved around many areas of the lake’s shoreline (after Clarkson, 1996), and the location of Lake Taupo within the Taupo Volcanic Zone ( TVZ ) in the central North Island of New Zealand (inset).
Clarkson, 1996; Manville et al., 1998). We begin assessment of pumice depositional processes by addressing briefly the transition from floating to saturated pumice.
PUMICE SATURATION BEHAVIOUR Hot pumice, introduced by eruption or transport directly into water, may become immediately saturated with water as steam in the vesicles condenses to water, reducing in volume and thereby drawing external water into the unoccupied pore space (Whitham & Sparks, 1986); hot pumice is not further considered here. Most cold, dry pumice clasts float if placed on to water, and may maintain this ability for hundreds or thousands of years (pumice from the ad 181 eruption of Taupo volcano is still regularly re-entrained by streams during heavy rainfall and floated in Lake
Taupo). Cold or cooling pumice can become saturated with water in two ways; it can be picked up by streams or waves, or it can be immersed in ground water. If a pumice clast is immersed in ground water for an appropriate time (see below), as when the water table rises following deposition of tephra from an eruption, then the clast can be eroded and released into a sedimentary system already saturated. Such pre-saturated particles do not float. In contrast, pumice clasts entrained directly by surface water will initially float, and thus enjoy a period during which vesicular particles of any size have nil settling velocity (fine vitric ash lacks enclosed vesicles and begins to sink immediately, e.g. Smith & Smith, 1985). In continuous contact with water, however, pumice gradually absorbs water and sinks (Smith & Smith, 1985; Whitham & Sparks, 1986). This transition has been directly addressed by Tilly (1987) and Manville et al. (1998) for pumice produced by the ad 181 Taupo eruption.
VSIC06 9/12/08 2:05 PM Page 143
143
Settling and deposition of pumice (a) Cubes (Minoan)
(b) Rounded clasts (AD 181 Taupo)
Experimental results Investigation of pumice saturation behaviour by Whitham & Sparks (1986) demonstrated a generally log–linear ‘time to sink’ saturation trend (Fig. 2a),
2500
+
10th 25th
2000
×
50th 75th
1500
×
1000 500 00
90th tsink diameter squared × ×
××
+ 2
×
+
R2 = 0.943
×
+ 4
+
6
+
400 mm3
Time to sink (h)
Fig. 2. Two representations of pumice saturation. In (a) the square symbols in the upper right show time to sink for pumice cubes from the Minoan eruption, and the circles time to sink for rounded clasts from the ad 181 Taupo eruption. Minoan data are from Whitham and Sparks (1986) measured for individual cut cubes of pumice, typically subspherical to ellipsoid. Despite the differing populations, the overall trend supports times to sinking that are proportional to clast volume across six orders of magnitude. (b) Summary of time-to-sink data from ad 181 Taupo experiments for clasts up to 12-mm diameter (Manville et al., 1998). The symbols for 10th, 25th, etc. indicate times at which the stated percentiles of grains of the given size had sunk (e.g. ≈ 90% of 4-mm-diameter clasts had sunk by 400 h).
8
10
+
12
Pumice clast diameter (mm)
which can be compared with results of a series of recent experiments by Manville et al. (1998; Fig. 2a). Combined results from these studies indicate that time to sinking (i.e. until saturation to a mean density exceeding 1000 kg m–3) is predictably proportional to
VSIC06 9/12/08 2:05 PM Page 144
144
J. D. L. White et al.
(a)
(b) 0.9 0.8 Homogeneous diffusional saturation
Density (g/cm3 )
0.7 0.6 0.5 0.4
Reticulated diffusional saturation
0.3 0.2 0.1 Dry
10 –1
10 0
10 1
Time in water (h)
Fig. 3. Saturation of pumice clasts occurs by a combination of diffusion and reticulation, which is the pipe-like delivery of water into clast interiors through large interconnected vesicles. (a) Effects of reticulate flooding of cold Minoan pumice blocks are indicated by rapid initial change from dry densities of 0.19 – 0.32 g cm–3 to 0.48 – 0.78 g cm–3 in the first 5 min of immersion (modified from Whitham & Sparks, 1986), followed by long-term saturation toward sinking densities (see Fig. 2). (b) Representation of diffusional gradient versus reticulate network with diffusion from both interior and exterior of clasts. (c) Photograph of ad 181 Taupo pumice showing mixture of stretched large and small vesicles, the former forming the pumice reticulation system.
(c)
volume, and in the case of the rounded Taupo clasts examined by Manville et al. (1998), to the square of clast radius. Larger clasts tended to show a less uniform relationship between size and saturation time (Fig. 2c), with many sinking well before or after the times expected by projection of results from the smaller clasts. These results are inferred to indicate two distinct stages in the saturation history of larger Taupo pumices. Two-stage saturation is also strongly indicated for cold Minoan pumice (Whitham & Sparks, 1986), with rapid initial uptake of water during the first 5 min of immersion approximately doubling the density of the pumice (Fig. 3a). Interpretation The two-stage saturation behaviour shown by Taupo
and Minoan (Whitham & Sparks, 1986) pumice is likely to be commonplace because of vesicle interconnectedness (Klug & Cashman, 1996), and is inferred to result from two distinct means of water ingestion by the pumice. For small clasts or large clasts lacking large interconnected vesicles, water is absorbed along a more or less continuous front that advances inward from the clast boundaries and can be modelled as a diffusional process (Manville et al., 1998). For clasts characterized by inhomogeneous distributions of elongate or variably shaped vesicles (Fig. 3c), additional water may enter the clast by non-diffusional advance along larger, interconnected vesicle pathways, giving rise to a reticulated distribution system in which water is effectively diffused outward from internal distribution points as well as inward from the clast exterior (Fig. 3).
VSIC06 9/12/08 2:05 PM Page 145
145
Settling and deposition of pumice
Quartz settling
10.0
2.4 2.2 2.0 1.7 1.5
Saturated pumice settling
5.0 3.0
1.3
2.0
1.2 1.1 1.0 0.8 0.6 0.4
Saturated pumice density 1.0 0.7 0.5 0.3 0.2 –7φ –6φ
–5φ
–4φ
–3φ
–2φ
–1φ
0φ
1φ
2φ
3φ
4φ
5φ
Measured grain size (b) Settling tube data for saturated pumice. Histograms for phi sieve fractions
40.0 20.0 Mean settling velocity (cm/sec)
Fig. 4. Settling diagrams for watersaturated ad 181 Taupo pumice. (a) Chart constructed by Tilly (1987) for small size-classed groups of Taupo pumice. Velocities are calculated from the first decile of clasts to settle through the apparatus. Quartz settling curve for comparison, together with curve for saturated pumice density. (b) Similar chart constructed from automated settling tube data for sieved size classes of water-saturated (under vacuum) Taupo pumice sediment (histograms from – 4 to + 4 φ, by accumulated mass percentage, in half-φ steps; histogram labels show sieve size fraction; histogram headings are mean settling velocity, also plotted on thick shaded line). Mean settling velocities for smaller size classes are higher than those measured by RSA, probably indicating inclusion of small numbers of non-pumice clasts in these unpicked samples. For each size class, there is significant dispersion in settling velocities, many with departures from log-normality. The variable skewness, high variance, and bimodality shown by different size classes of what is in general appearance a homogeneous sediment suggests the difficulty of quantitatively back-calculating hydrodynamic behavioural ‘size’ classes from sieve size classes.
Settling velocity (cm/sec)
20.0
Density (gm/cm3) of water-saturated pumice
(a) Water-saturated pumice in water 40.0
10.0 5.0 3.0 2.0 1.0 0.7 0.5
30 16.2 cm/s 20 10.6 cm/s 30 10 30 8.0 cm/s 20 0 20 –4 phi 10 30 3.7 cm/s 10 0 –2 phi 0 20 0 phi 10 40 11.4 cm/s 0 2 phi 30 40 10.0 cm/s 20 30 20 5.7 cm/s 10 20 10 0 10 –3 phi 0 1 phi 0 20 1.6 cm/s –1 phi 10 0 3 phi
0.3
Tilly's line
0.2 –7φ –6φ
SATURATED PUMICE BEHAVIOUR Once water saturated, pumice behaviour much more closely resembles that of other geological materials. This is manifest both from the range of tractioninduced depositional features, such as cross-beds, climbing ripples, wave ripples and graded turbidite beds found in many subaqueously deposited pumice units, and from experimental results. Experimental results To determine representative behaviour of fully watersaturated pumice, pumice clasts collected from lacus-
–5φ
–4φ
–3φ
–2φ
–1φ
0φ
1φ
2φ
3φ
4φ
5φ
Measured grain size
trine pumice deposits formed during a highstand of Lake Taupo (Wilson & Walker, 1985) were sieved into full-φ size fractions from 3 to – 4 φ, and saturated with water under vacuum (Clarkson, 1996). Settling behaviour was investigated with an automated settling tube utilizing MacRSA© rapid sediment analysis software (Ballard, 1996). Results show that each size class exhibits a range of fall velocities, and that larger size classes have significantly higher mean settling velocities (Fig. 4). Samples subjected to both conventional sieve analysis and settling tube analysis after saturation show a marked increase in apparent sorting and a decrease in mean grain size using the latter method (Fig. 5). The increase in apparent sorting strongly
VSIC06 9/12/08 2:05 PM Page 146
146
J. D. L. White et al. OU 67339 Dry sieve data RSA data
25
Weight (%)
Freq % by weight
30
20 15 10 5 0
–4 –2 0
2
4
OU 67358
Dry sieve data RSA data Wt % components
Freq % by weight
30 25 20 15 10 5 0
–4 –2 0
2
4
70 60 50 40 30 20 10 0
14 12 10 8 6 4 2 0
RSA Sieve
–4
–2
0 2 phi size
P
–4
–3 –2 –1
4
X
0
1
L
2
Sieve size class Dry sieve data RSA data
OU 6308 Freq % by weight
30 25 20 15 10 5 0
–4 –2 0
2
4
supports the approach of measuring water-saturated clasts. It is highly unlikely that grain populations would show such good saturated hydrodynamic sorting if they had been deposited in an unsaturated state. The decrease in apparent grain size is expected from the low density of pumice. Because of the limitations of the settling tube apparatus, we did not use it to examine the behaviour of clasts larger than – 4 φ. However, similar results for mean settling velocities were obtained for a wide range of pumice grain sizes by Tilly (1987), who painstakingly assessed hydrodynamic behaviour of Taupo pumice by manual experiments with small clast populations. The implications of these data are several. First, as emphasized by Smith & Smith (1985), vesicular pyroclastic particles such as pumice do not show a proportional increase in mass with particle size. Second, a population of water-saturated pumice will settle from a well-mixed suspension or gradually from a particleladen current to form a normally graded bed (see Fisher, 1965; Cashman & Fiske, 1991). Third, settling
Fig. 5. Representative sieve versus RSA data for pumiceous (not purely pumice) sediment (RSA, Rapid Sediment Analyser, i.e. automated settling tube). (a) For three samples half-φ size classes are plotted (quartz sphere settling velocity equivalent size classes for the RSA), and show that sieve measurements systematically show larger grain sizes and greater dispersion than RSA measurements, which represent hydrodynamic behaviour. (b) A mixed pumiceous sample from the Reporoa basin (Manville, this volume) shows typical increase in sorting and decrease in equivalent grain size for RSA versus sieve data. Components for the sample are also representative, showing concentration of pumice in the larger size classes (note that ‘lithics’, L, may include vesicular or lowdensity sedimentary clasts).
tube methods are far more appropriate than sieving for determining hydrodynamic characteristics of subaqueously deposited volcaniclastic particles such as pumice (Fisher, 1965; Oehmig & Wallrabe-Adams, 1991). Experimental work with specific pumice sediment populations should allow ‘baseline’ hydrodynamic sorting characteristics for fully saturated clasts to be established, after which strong skewing, bimodality or other changes from expected distributions could potentially be interpreted in terms of incomplete saturation, mixing of saturated and unsaturated or partially saturated populations, or some combination of the two. Large clasts take a long time to become fully saturated (Manville et al., 1998) and attain negative buoyancy while only partly saturated, so it is likely that many large clasts are deposited while incompletely saturated. Field observations Widespread outcrops of lacustrine pumice deposits
VSIC06 9/12/08 2:05 PM Page 147
Settling and deposition of pumice
147
(a)
(b)
(c) Fig. 6. Pumice depositional features. (a) Two styles of pumice settling behaviour are represented in these laterally persistent, planar beds (knife is ≈ 10 cm). The lowest bed visible has a vitric ash base (arrow), which grades upward to a coarse-tail graded top dominated by centimetre-sized rounded pumice clasts, and is inferred to have formed as clasts saturated and sank from a floating pumice raft. Fines are likely to result, at least in part, from abrasion of clasts as they are jostled in the floating raft. The beds beginning above the knife are normally graded, which indicates redeposition of water-saturated pumice according to grain fall velocity. (b) Steeply imbricate, openwork pumice formed by a weak current flowing somewhat away from the viewer (lens cap 45 mm). We interpret this bed as having a cluster-type fabric, formed by upcurrent accretion of clasts against previously deposited clasts during deposition. The thickness of this bed, its lateral continuity and the upward fining suggest deposition of the pumice clasts by progressive lodgement under shallow-flow conditions in a strongly aggrading bed. Similar steeply imbricate fabrics form under oscillatory waves, but wave agitation is not a requirement for the fabric. (c) Isolated and grouped pumice clasts in finely laminated, scoured, fine vitric ash. All clasts were probably water saturated at the time of deposition, but the largest clasts probably moved only intermittently in this environment, rolling over the stable vitric ash substrate. Other clasts probably moved together with the silt, reflecting strong differences in shear velocities required to move the larger but very low density pumice versus the much finer but more dense, frequently platy, shards of ash.
around Lake Taupo (Fig. 1) represent a variety of depositional environments (see Riggs et al., this volume). We here use only a selection of strata containing a significant gravel-size pumice component to illustrate the behaviour addressed above. In deposits lacking wave-generated sedimentary structures, inferred to represent deposition below fairweather wave base, laterally continuous beds grade upward from vitric ash-rich bases to tops containing abundant rounded pumice, often itself weakly reversegraded (Fig. 6a). Far less commonly, angular, fines-
poor pumice deposits form laterally continuous beds that locally show weak reverse grading. Reverse grading is interpreted to represent deposition of pumice by sequential settling of clasts that became progressively saturated in pumice rafts. Interstitial fine ash present throughout the coarse-tail reverse-graded beds is inferred to have been produced by abrasion of pumice clasts during their residence in the pumice raft. The angular clasts and paucity of fine ash in the rare finespoor reverse-graded beds indicates sinking from rafts without significant abrasion. Any fine ash originally
VSIC06 9/12/08 2:05 PM Page 148
148
J. D. L. White et al.
travelling with the larger pumice in these cases has settled out without flotation nearer circumlacustrine supply points. Both styles of reverse grading of pumice are termed saturation grading (White et al., 1997). A variant of saturation grading that spans more than one bed may be exhibited by relatively high-density, dome-derived, pumice blocks that are present around Lake Taupo (Wilson & Walker, 1985). In a single section, smaller blocks of this type are first preserved in beds lower than those containing metres-sized megablocks derived from the same eruptive phase. This is inferred to be the result of more rapid saturation of small blocks than large ones, and hence saturation-controlled deposition. Normally graded pumice beds of different kinds are also extremely common in the lacustrine sediments around Lake Taupo that post-date ad 181. One variety occurs in the same sequences as reverse-graded beds, contains the same angular pumice lacking fines, but has subtly scoured, undulose basal contacts. Such beds are inferred to result from erosion of pumice beds at the lake floor, temporary resuspension or limited transport, followed by redeposition according to their saturated fall velocities. In this way saturation graded pumice originally arriving at the depositional surface according to the saturation time of individual clasts is converted to pumice beds showing redeposition grading (White et al., 1997). Pumice is, even when water saturated, a low-density particle; Tilly (1987) gave saturated densities of 1.5 g cm–3 for 0 φ coarse sand, and 1.15 g cm–3 for – 5 φ pebbles. This range in density probably reflects the presence of some vesicles in larger clasts that remain effectively enclosed even after sample treatment (see Smith & Smith, 1985). The systematic dependence of density on grain size suggests that such imperfect saturation is associated with thin, elongate vesicles or vesicle entrances that are not penetrated by water during simple immersion or under moderate vacuum. Because of its low density, pumice exhibits a tendency to form cluster fabrics (Brayshaw, 1984) where tabular clasts are deposited under the influence of currents. The fabrics typically comprise very steeply intermediateaxis-imbricated clasts without interstitial matrix fines, and may form either lenticular bodies fining upcurrent (Brayshaw, 1985) or thick beds with clasts approaching subvertical orientations (Fig. 6b). Small-scale examples of the former can be observed in modern roadwash, where subvertically orientated pumice grains form lenses one to a few clasts thick and a few dm2 in area where sheet flows encounter tufts of grass or other plants. The thick beds are inferred to form by progressive lodgement during rapid bed aggrada-
tion under shallow-flow conditions, as described by Seccombe (1960). Scour-filling lenses within finely laminated and cross-laminated traction-deposited fine ash may also locally show cluster fabrics, supporting the inference that the large pumice clasts were rolled into place in approximate dynamic equivalence to the surrounding ash (Fig. 6c). Large isolated, or nearly isolated, pumice clasts may also have been deposited in approximate dynamic equivalence if only partially saturated. For fully saturated clasts, vitric silt (0.032 mm; 5 φ) moves at the same critical tractive force as medium (7 mm; c. – 3 φ) pebbles (Tilly, 1987). Most other, ‘normal’, tractional bedforms are also well represented in Taupo pumice beds (e.g. Smith, 1991a,b; Clarkson, 1996; Riggs et al., this volume; Manville, this volume); steep cluster fabrics are emphasized here because they are distinctive and far more commonly developed in the circum-Lake-Taupo pumicedominated sediment than elsewhere in normal gravel deposits. Bedding structures produced by stranding vary with the style of stranding. Beach berms formed during the rapid draining and lake-level drop of Lake Taupo typically consist of very poorly organized coarse pumice (see Riggs et al., this volume), as do local pumice berms around Lake Taupo today (Clarkson, 1996; Wilson et al., 1997). Floating pumice stranded on modern beach berms typically is deposited together with plastic rubbish and bits of driftwood, with larger items carried by the momentum of breaking waves to positions further up the berm. Stranding occurs when the wave swash infiltrates into the berm (e.g. Davis, 1985; Renaut & Owen, 1991), so that there is insufficient water flowing back off the beach to transport the pumice. Stream-deposited pumice also locally shows pronounced cluster fabrics that may have formed during waning flow as water depths decrease to nil (see Stanley, 1978).
DISCUSSION AND CONCLUSIONS Most pumice deposits observed in the Taupo area exhibit bedding characteristics, such as normal grading, that are consistent with deposition according to relative fall velocity of water-saturated clasts or their subsequent reworking by currents (redeposition grading). Reverse-graded beds formed by gradual saturation and sinking of pumice (saturation graded) are far less common than normally graded ones. Cluster fabrics and other current-induced bedforms further testify to deposition of pumice clasts not solely by saturation and sinking, but under a range of con-
VSIC06 9/12/08 2:05 PM Page 149
149
Settling and deposition of pumice
Floating pumice Settling velocity
Fig. 7. Schematic summary of pumice behaviour. Clasts may initially float on water (until t1), gradually becoming water saturated (t1–t3) and sinking, smallest clasts first, to produce reverse-graded unit. Clast settling velocities reach a maximum at full saturation, at which point large clasts have higher settling velocities than small ones (e.g. t5). Complications are evident where different sized clasts are at different degrees of saturation (e.g. t2– t4). Re-entrainment and mixing of saturated pumice followed by deposition according to fall velocity produces normally graded beds. If redeposition takes place under a current, cluster fabrics of various sorts are likely to form. (Deposition by stranding not shown.)
Re-entrainment of saturated pumice Time
Suspending settling t1
t2
t3 t4
t5
t6
or Current With elongated clasts, deposition from current
ventional sedimentation regimes (Fig. 7). Recessional beach berms appear to be the only examples of clast lodgement resulting from water draining away until the clasts can no longer be carried (stranding deposits), though whether some of the larger megablocks of pumice sank, or were stranded, in shallow offshore settings is perhaps a moot point. Grain-size analysis of pumice deposits is of little qualitative use in understanding conditions at the time of deposition, but may be useful for comparison among relatively homogeneous pumice deposits. For stream deposits, where more variation in saturation should be expected as a result of bar storage and intermittent entrainment, settling tube analyses may serve as a diagnostic tool if applied to selected populations. Systematic collection of settling velocity data is an important tool for any but the most qualitative analysis of palaeohydrological conditions, and in general the assumption that clasts were water saturated at the time of deposition is acceptable. Sub-tabular clast shapes, inhomogeneous saturation, and low hydrodynamic equivalence all enhance the ability of pumice to form steeply imbricated cluster fabrics as it is deposited from bedload of slow and waning currents.
ACKNOWLEDGEMENTS This research was supported by the New Zealand Foundation for Research, Science and Technology and
a University of Otago Research Grant. We especially acknowledge the exceptional study of pumice behaviour performed by Clifton Tilly, whose MSc thesis (Tilly, 1987) was accepted posthumously and served as inspiration for further work on the systematics of pumice deposition. G. Smith and W. Fritz provided helpful reviews, which led to improvement of the paper.
REFERENCES Ballard, H.R. (1996) MacRSA: Macintosh application program for rapid sediment analysis. ©Hyram Ballard, Geology Department, University of Otago, Dunedin. Bateman, P.C. (1965) Geology and Tungsten Mineralization of the Bishop District, California. US geol. Surv. Prof. Pap., Denver, CO, 470. Brayshaw, A.C. (1984) Characteristics and origin of cluster bedforms in coarse-grained alluvial channels. In: Sedimentology of Gravels and Conglomerates (Eds Koster, E.H. & Steele, R.J.), Mem. Can. Soc. petrol. Geol., Calgary, 10, 75 – 85. Brayshaw, A.C. (1985) Bed microtopography and entrainment thresholds in gravel-bed rivers. Geol. Soc. Am. Bull., 96, 218–223. Cashman, K.V. & Fiske, R.S. (1991) Fallout of pyroclastic debris from submarine volcanic eruptions. Science, 253, 275 –280. Clarkson, R.A. (1996) Immediate post-c. 1800a Taupo eruption secondary deposits and shorelines. MSc thesis, University of Otago. Davis, R.A.J. (1985) Coastal Sedimentary Environments. Springer, New York.
VSIC06 9/12/08 2:05 PM Page 150
150
J. D. L. White et al.
Fisher, R.V. (1965) Settling velocity of glass shards. Deepsea Res., 12, 345–353. Houghton, B.F., Wilson, C.J.N., McWilliams, M.O., et al. (1995) Chronology and dynamics of a large silicic magmatic system: central Taupo Volcanic Zone, New Zealand. Geology, 23, 13–16. Klug, C. & Cashman, K.V. (1996) Permeability development in vesiculating magmas: implications for fragmentation. Bull. Volcanol., 58, 87–100. Manville, V. (2000) Sedimentology and history of Lake Reporoa: an ephemeral supra-ignimbrite lake, Taupo Volcanic Zone, New Zealand. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 109–141. Blackwell Science, Oxford. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1998) The saturation behaviour of pumice and some sedimentological implications. Sediment. Geol., 119, 5–16. Oehmig, R. & Wallrabe-Adams, H.-J. (1991) Hydrodynamic properties and grain-size characteristics of volcaniclastic deposits on the mid-Atlantic ridge north of Iceland (Kolbeinsey Ridge). J. sediment. Petrol., 63, 140 –151. Renaut, R.W. & Owen, R.B. (1991) Shore-zone sedimentation and facies in a closed rift lake: the Holocene beach deposits of Lake Bogoria, Kenya. In: Lacustrine Facies Analysis (Eds Anadon, P., Cabrera, L. & Kelts, K.), Spec. Publs int. Assoc. Sediment., No. 13, pp. 175–195. Blackwell Scientific Publications, Oxford. Riggs, N.R., Ort, M.H., White, J.D.L., Wilson, C.J.N., Houghton, B.F. & Clarkson, R. (2000) Post-1.8-ka marginal sedimentation in Lake Taupo, New Zealand: effects of wave energy and sediment supply in a rapidly rising lake. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 151–179. Blackwell Science, Oxford.
Seccombe, T.T. (1960) Early days on the Rangitaiki Swamp. 1900 –15. Whakatane hist. Soc. Rev. Smith, G.A. & Smith, R.D. (1985) Specific gravity characteristics of recent volcaniclastic sediment: implications for sorting and grain size analysis. J. Geol., 93, 619 – 622. Smith, R.C.M. (1991a) Landscape response to a major ignimbrite eruption, Taupo Volcanic Center, New Zealand. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 45, 123 –137. Smith, R.C.M. (1991b) Post-eruption sedimentation on the margin of a caldera lake, Taupo Volcanic Centre, New Zealand. In: Volcaniclastic Sedimentation (Eds Cas, R. & Busby-Spera, C.), Sediment. Geol., 74, 89–138. Stanley, D.J. (1978) Pumice gravels in the Rivière Claire, Martinique: selective sorting by fluvial processes. Sediment. Geol., 21, 161–168. Tilly, C.R. (1987) The sedimentology of the Taupo Pumice Alluvium Formation occurring in the lower region of the Hamilton Basin. MSc thesis, University of Waikato. White, J.D.L., Manville, V., Wilson, C.J.N. & Houghton, B.F. (1997) Normal pumice. In: General Assembly, International Association of Volcanology and Chemistry of the Earth’s Interior, Puerto Vallarta, Mexico, 198. Whitham, A.G. & Sparks, R.S.J. (1986) Pumice. Bull. Volcanol., 48, 209–223. Wilson, C.J.N., Riggs, N.R., Ort, M.H., White, J.D.L. & Houghton, B.F. (1997) An annotated atlas of post-1.8 ka shoreline features at Lake Taupo. In: Scientific Report, No. 97/14, p. 35. New Zealand Institute of Geological and Nuclear Sciences, Taupo. Wilson, C.J.N. & Walker, G.P.L. (1985) The Taupo eruption, New Zealand I. General aspects. Phil. Trans. R. Soc. London, Ser. A, 314, 199–228. Zielinski, G.A., Mayewski, P.A., Meeker, L.D., et al. (1994) Record of volcanism since 7000 B.C. from the GISP2 Greenland ice core and implications for the volcano– climate system. Science, 264, 948 – 952.
VSIC07 9/12/08 2:04 PM Page 151
Spec. Publs int. Ass. Sediment. (2001) 30, 151–177
Post-1.8-ka marginal sedimentation in Lake Taupo, New Zealand: effects of wave energy and sediment supply in a rapidly rising lake N . R . R I G G S * , M . H . O R T* , J . D . L . W H I T E † , C . J . N . W I L S O N ‡ , B . F . H O U G H T O N ‡ and R . C L A R K S O N † *Department of Geology, Northern Arizona University, Flagstaff, AZ 86011, USA; †Deceased. Formerly at Department of Geology, University of Otago, PO Box 56, Dunedin, 9015, New Zealand; ‡Institute of Geological and Nuclear Sciences, Wairakei Research Centre, Taupo, New Zealand
ABSTRACT The eruption of Taupo caldera, New Zealand, c. 1.8 ka, climaxed in the emplacement of the Taupo ignimbrite, which dammed the previous outlet of the caldera lake and allowed rise of the lake behind the dam to c. 34 m above current lake level. After an estimated 20 yrs of lake refilling, the ignimbrite dam was breached, and the lake drained rapidly to a position 2– 4 m above current level. From that level, the lake apparently dropped slowly to its present position. Post-1.8-ka lacustrine sediments are exposed at various levels between the highstand and the modern shoreline. The distribution of lacustrine facies at Taupo reflects the location of the mouths of rivers that drained uplands buried and/or denuded of vegetation by the erupted material, together with the position of depositional sites with respect to prevailing winds and energy systems. Four broad facies environments can be recognized, defined by combinations of high or low energy and high or low sediment influx. Sediments deposited in areas of high energy (i.e. strong wave action) and high sediment input are rich in vitric material, but contain abundant lithic material in those deposits interpreted as storm influenced. Deposits in lowenergy areas that had high sediment input are dominantly vitric, with abundant shoreface and pumice-raft deposits. High-energy, low-sediment-input areas are rich in lithic detritus, apparently reworked from ignimbrites along the pre-eruptive lake shore during storms. Deposits that characterize low-energy, lowsediment-input environments are poor in lithic detritus, reflecting the inability of currents to transport material to these sites. Terraces prominent above the modern shoreline of Lake Taupo expose transgressive sequences that range from beach face deposits at the base to below-storm wave-base deposits at the top. Many terraces are capped by deposits of well-rounded, cobble- to boulder-size pumice clasts, which are interpreted as stranded deposits of previously floating clasts, formed during rapid lake-level fall. The Lake Taupo sedimentary record provides insights into sedimentary facies and styles that may be expected during lake filling that immediately follows caldera collapse. The inferred rapid filling rate of the lake, estimated to be 5 –9 m yr–1 for the exposed section, together with the dominantly vitric nature of the source material, permitted preservation and recognition of deposits such as shoreface storm-wave deposits that are unusual in non-volcaniclastic lake settings.
INTRODUCTION In this paper, we describe young lacustrine sediments in a caldera setting at Lake Taupo, New Zealand. Intracaldera lacustrine sedimentation has not received much study because the record of sedimentation in caldera lakes is often overprinted by post-emplacement erosion or deformation and volcanism (see also Larsen & Smith, 1999). At Taupo, the short time since the last major event (1800 yr) means
The amount and timing of volcaniclastic sediment influx into volcanic lakes is controlled mostly by timing, magnitude and frequency of eruptions. Material is delivered to the lake directly by pyroclastic fall and flow processes, and via fluvial, aeolian and mass-flow remobilization of pyroclastic material. Once in the lake, the final distribution of sediment is influenced by wave-induced currents and density currents.
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
151
VSIC07 9/12/08 2:04 PM Page 152
152
N. R. Riggs et al.
that much of the post-eruption topography is preserved, and it is inferred that external climatic influences such as wind-flow patterns, longshore dispersal and seiche that act upon the lake today acted in a similar way on the refilling lake. More importantly, damming of the lake outlet by pyroclastic debris caused the lake to refill after the eruption to ≈ 34 m above its present level, with accompanying deposition of lake sediment up to the temporary highstand. After the dam was breached, lake-level fall exposed relatively deep-water sediments above the modern shoreline. Sedimentation during the filling of Lake Taupo was also unusual in that the source terrain was made up of dominantly vitric material. Additionally, the rapid rise of the lake, which probably took place within 20 yr, and even more rapid fall of the lake allow a series of snapshots of lake processes. This fortuitous combination allows reconstruction of facies associations that reflect the interaction of sediment availability and external forces. The purpose of this paper is to document sedimentary facies variations at a number of sites that best demonstrate differences in the pattern of how sediment was redistributed along the lake shore. At modern Lake Taupo, wind-driven waves represent the most important sediment transport process, and wave energy varies significantly around the lake; during post-eruptive lake filling the same was true. Another potentially important sedimentation factor is sediment influx rate. Although modern Lake Taupo is fed by a number of streams and rivers that vary greatly in discharge and sediment load, we have found that posteruptive facies distribution and character were little affected by local sediment availability. We show that low-density pumice responded to lake currents and conditions very differently at Taupo than does nonvolcanic sediment with little density variation at nonvolcanic lakes. Grain-size data presented here are estimated from outcrop observations. Non-systematic variations in the density of volcaniclastic material inhibit ready evaluation of either estimated or measured (e.g. dry sieve) dimensional grain-size data (Fisher, 1965; Smith & Smith, 1985; Oehmig & Wallrabe-Adams, 1993). For the purposes of this study, estimates of grain size and sorting are a qualitative guide to grain behaviour of the lithic and crystal components. For pumice-rich beds, size and sorting must be interpreted in terms of pumice settling dynamics (Manville et al., 1998; White et al., this volume).
Terminology Sedimentation described here for the refilling Lake Taupo was dominated by the rapid influx of finegrained material. Certain terms that are used throughout the paper must be clarified. The term ‘vitric’ is used sensu Fisher & Schmincke (1984), for juvenile, glassy fragments that comprise shards and pumice, regardless of size. Vitric fragments include vesicular pyroclastic particles, obsidian, and dense, pre-1.8-ka rhyolite. ‘Pumice’ is differentiated from ‘ash’ on the basis of size: ash may consist of bubble-wall shards derived from the disintegration of pumice, but a pumice fragment is clearly recognizable as containing discrete vesicles. Sediments are described by the following grain sizes: silt- or sand-grade ash; fine, medium or coarse crystal–lithic or pumice sand; and coarse sand, granules, pebbles, cobbles and boulders of any composition. The hydraulic behaviour of fine versus larger vitric fragments, and especially whether larger fragments are pumiceous, is of utmost importance in interpreting sedimentation patterns at Lake Taupo. Thus, care is taken in all cases to be specific about the size of the fragment (i.e. ash versus pumice) and its structure (e.g. obsidian versus pumice) while referring to it by the compositional term ‘vitric’. Taupo volcano Manville (this volume) provides a summary of the tectonic setting of Taupo Volcano and the stratigraphy of the area. The Taupo Volcanic Zone (Fig. 1) is a north-east-trending zone of cone and caldera complexes that are currently active or have been active within the last 2 Myr. Volcanism is related to subduction of the Pacific plate beneath the Australian plate along the Tonga–Kermadec System. The 1.8 ka Taupo eruption generated five fall deposits, including those from both dry Plinian and wet phreatomagmatic eruptive styles, and two ignimbrite bodies, the first (early ignimbrite flow units) coeval with the most widespread Plinian fall deposit, and the second (Taupo ignimbrite) climaxing the eruption (Wilson & Walker, 1985, and references therein; Smith & Houghton, 1995). The 1.8 ka eruption was followed after a period of a few years to decades by an effusive event represented by grey pumiceous blocks in the post-1.8-ka sediments (Wilson & Walker, 1985; see also below). The total erupted volume, including the estimated bulk of primary intracaldera material, was ≈ 35 km3 dense rock equivalent (DRE). The fall deposits were concentrated to the east of the caldera;
VSIC07 9/12/08 2:04 PM Page 153
153
Sedimentation in a rapidly rising lake
o Riv er
a ikat
Kinloch
W
Ok a i a St m
N
500
0
50
ac Ac
Kaiapo Bay
ia
Taupo
d yR a B
Taupo airport
5M
LAKE
i le Beach
50 0
0
Waitahanui
Waitahanui Bank
TAUPO
50
Te Kohaiakahu Pt. 1
em
Hin
R Iver
a ia
500
Motutere Waipehi
Kuratau W
Hi n
ui
it e
ur
ko S
tre
n
am
o
175 E
North Island New Zealand
T au
Ke
rm
ve Ri
o
40 S
a-
0 50
r
To
10
PACIFIC OCEAN
ade c
po
1
r
ia R ive r
to
Rd
a
ga
Ri ve
a
p ga
o
Ta
50 0
0
ha
n
Poukura Pa
Tongariro River delta
ta
ai
St
Kur ata u
m
W
Horomatangi Reef
W
h arero a
ng
20 km
contour interval = 100 m Fig. 1. Generalized topographic map of Lake Taupo showing principal study sites. Dashed line represents lake highstand (after Clarkson, 1996; Wilson et al., 1997); arrow shows direction of major storm winds. Inset shows generalized tectonic setting of Taupo Volcanic Zone (TVZ) and New Zealand (modified from Houghton et al., 1995; Wilson et al., 1995); TVZ stratovolcanoes are those currently active.
TAUPO VOLCANIC ZONE
Taupo Lake caldera Taupo
caldera andesitic stratovolcano 0
20
40 km
N
VSIC07 9/12/08 2:04 PM Page 154
154
N. R. Riggs et al.
the flow facies were distributed in a more symmetrical fashion around the caldera (Wilson, 1985; Wilson & Walker, 1985). The Taupo ignimbrite is subdivided into layers 1 and 2. Layer 1, in turn, is divisible into a pumice-dominated layer 1(P) and a crystal- and lithicrich layer 1(H). The thicker layer 2 comprises the ignimbrite veneer deposit (IVD) and the valleyponded ignimbrite (VPI), in reference to the tendency of deposits to mantle topographic features (IVD) or to thicken in palaeotopographic lows (VPI) (Walker et al., 1981a,b; Wilson & Walker, 1982; Wilson, 1985). At some time shortly after the explosive phases of the Taupo eruption, fragmentation of a dome forming beneath the newly re-formed lake, at or near Horomatangi reefs and/or Waitahanui Bank (Fig. 1), provided distinctive grey pumice blocks that are seen in many exposures at or near the same elevation on the north-eastern and eastern shoreline of the lake. Because there is currently no evidence for more than one dome-building event from the Horomatangi reefs, all layers that contain the grey pumice blocks are assumed to be time correlative or closely so. The effects of the 1.8 ka Taupo eruption on the North Island landscape have been extensively discussed by Smith (1991a,b). Smith (1991a) estimated that an ≈ 20 000 km2 area was denuded of vegetation for 20 –30 yr following the eruption, with an associated increase in high-sediment-concentration flows. A simple topographic map of the Taupo area (Fig. 1) shows the approximate drainage basin of the lake; mapping of lacustrine and fluvial facies indicates that current major drainages closely parallel drainages that were active immediately after the eruption. Sedimentation in and around Lake Taupo was strongly influenced by the size of available, nonwelded Taupo ignimbrite source material. The deposit of the climactic phase of the eruption, the Taupo ignimbrite, is inferred to have covered the vast majority of the catchment, to a thickness of as much as 100 m for the VPI and 15 m for the IVD (Wilson, 1985). Local exposures of the early ignimbrite flow units (i.e. in the Waitahanui drainage) contributed fines-rich material. Much of layer 1(P), and all of layer 1(H) are fines-poor. Layer 2 deposits are variably coarse- and fine-grained, but are clay poor. Overall the deposits, and hence the source of the lake sediments, comprised a wide range of poorly sorted sediment ( Wilson, 1985; Wilson & Walker, 1985). Lake-level rise The bulk of the deposits we describe formed during
the major rise in lake level after the eruption. Refilling of the lake was almost certainly a very rapid process. Smith (1991a) estimated the length of time for lake level to rise from the present lake level (inferred to closely approximate the pre-1.8-ka level) to the 34 m highstand as a few years. Manville et al. (1999) inferred that the lake was largely emptied by the eruption and estimated that the time required to raise lake level +130 m (from an emptied lake floor to presentday level plus ≈ 34 m) was c. 40 yr. Present-day Lake Taupo has a surface area of c. 620 km2 and an average depth of 97 m (Electricity Corporation of New Zealand (ECNZ), 1994), which gives a total volume of ≈ 60 km3. Present inflow into the lake is dominated by the Tongariro River, but another 30 streams also feed the lake, as does direct rainfall upon the lake. Water diversions around the Ruapehu and Tongariro volcanoes now contribute c. 20% of the water inflow to the lake, which must be subtracted for any calculations of natural inflow. Maximum mean natural inflow for a single year in the past 95 yr is about 170 m3 s–1, minimum mean annual inflow is about 80 m3 s–1, with an average of about 125 m3 s–1 (ECNZ, 1994). Following the 1.8 ka Taupo eruption, vegetation was largely destroyed in the Lake Taupo catchment area, and the white surface of the Taupo ignimbrite would have been highly reflective (Smith, 1991a; Manville, this volume), reducing evapotranspiration and fostering increased runoff. Using the maximum present-day natural inflow, it would take about 11 yr to fill the lake to its present level, assuming that there was no water left in the caldera basin after the Taupo eruption. To fill the lake another 34 m, up to the approximate highest level reached by the lake, a surface area of 700 km2 is estimated for the lake. The additional 23 km3 of lake volume would require another 4 yr to fill at maximum present-day annual inflow, or a total time to fill the lake to its highest level of about 15 yr. For reasons discussed below, this estimate must be regarded as a minimum. The calculation of the actual time to fill the lake is complicated by a number of factors. First, there is ample evidence of water having been in the lake before and during even the latest stage of the Taupo eruption (Wilson & Walker, 1985), and some probably remained in the lake basin after the eruption. In addition, a large amount of water was included in the phreatomagmatic deposits (Wilson & Walker, 1985), and probably provided a great deal of runoff during and immediately after the eruption. A second complicating factor is evaporation from the surface of the lake, as well as from the hot ignimbrite surface. A third
VSIC07 9/12/08 2:04 PM Page 155
155
Sedimentation in a rapidly rising lake complication is that it is not known how long it took to saturate the ignimbrite and fall deposits sufficiently for perennial streams to become re-established; water stored in the ignimbrite did not reach the lake quickly, and substantial storage extends the time necessary to fill the lake (Wilson et al., 1997). If a porosity value of 70% is used, and an average thickness of 4 m is assumed for the total Lake Taupo catchment area of 3312 km2, a total of 13.2 km3 more water, representing about 2.5 yr of precipitation, would need to be delivered to the surface before the lake was filled. A fourth complication is the lake inflow. If the presentday average, rather than the maximum, natural inflow is used, the length of time to fill the lake is about a third longer (about 20 yr). For the project described here, we are most concerned with rates of lake-level rise between the present lake surface and the highest level reached by the lake. We estimate a filling rate of between 5 and 9 m yr –1 for this time period, based on estimates of 4 –7 yr for this rise. Ignimbrite within the catchment basin would have been saturated and stream systems largely established by that point (Smith, 1991a,b; Manville et al., 1999), so only the second and fourth complications described above would have any significant influence, which is accounted for by the range of rise rates.
lake. We assume that similar conditions existed as the lake was refilling after the 1.8 ka eruption, and infer that wind direction affected where floating pumice ‘rafts’ came to shore, and the extent to which these rafts were reworked and agitated. Rafts trapped along the shoreline may have strongly affected suspension deposition of pumice and other vitric particles. The Taupo shoreline is marked in many places by a prominent notch that lies between c. 28 and c. 42 m above the present lake level. The break in slope behind this terrace represents the maximum lake level after the 1.8 ka eruption, and coincides with a distinct change from intensely to weakly gullied erosional surfaces (Clarkson, 1996; Wilson et al., 1997). The highstand marker lies at different elevations around the lake, apparently because of post-1.8-ka volcano–tectonic activity (e.g. Otway, 1986; Wilson et al., 1997). Transgressive lacustrine deposits are well exposed between the present lake and the highstand terrace margin in many places, and we commonly enhanced exposures by trenching. Approximately 50 sections, both in outcrop and trenches, were studied and described. Descriptions of lithofacies presented below are composites from these exposures.
LAKE TAUPO DEPOSITS Geography and geomorphology The prevailing wind direction at Lake Taupo is from the west and south-west (Timperley, 1983; Thompson, 1984). Weather systems accompanied by strong winds come from two alternative directions. Major storms that last for up to several days come from the southwest, whereas northerly winds associated with tropical depressions affect the area for shorter periods. Overall, longshore drift around Lake Taupo moves sediment from stream mouths north-north-eastward along the
Our study locations are mostly along the northern, eastern, and south-western shorelines (Fig. 1). The steep topography of the western lake shore precluded preservation of sediment in much of that area, and deposits along much of the southern shoreline have been reworked by avulsions of the Tongariro River and by formation of its delta (Wilson et al., 1997). The post-1.8-ka deposits in the study areas accumulated in offshore, shoreface, beach and lagoonal depositional environments (Fig. 2; Table 1). Below, we
rf/ sw Su
Fig. 2. Schematic shore profile illustrating depositional settings referred to in this paper. The inset shows an approximate profile from the modern shoreline to the highstand shoreline at Five Mile Beach. The small box illustrates the scale of the larger diagram: at any given time the active shoreface extends to ≈ 8 m depth, with waves shoaling over a zone perhaps 100 m wide, depending on the shoreface slope.
as h
lake-level rise 5-9 m/yr
lagoon
up wave base (8-10 m depth)
Off
sh
ore
low
er Sh
or
ef
a
pe
ce
r
highstand shoreline modern lake shoreline
~ 34 m <1.5 km
Subaerial and subaqueous debris-flow deposit
Subaqueous reworking of pumice gravel; lithic material may be storm derived Strandline/berm deposit
T–L–Sh–S
S–L S–O Sh ± S–O–L
Sh–S
O
O–L
Poorly sorted gravel; pumice clasts to 20 cm, subround to round; supported by matrix of fine ash and medium–coarse lithic sand; locally vesicular; massive Normally graded fine pumice gravel, often matrix free Poorly sorted pumice gravel, clasts to 5 cm, supported by fine to medium vitric sand with < 5% crystal/lithic sand; generally massive Pebble to cobble pumice gravel, clasts rounded to well rounded and smooth; dominantly matrix free, locally with interstitial vitric sand; graded or ungraded Pumice to lithic granules–pebbles, cobbles; beds 5–10 cm thick, fine away from lake, dip towards lake; dominantly horizontally bedded; minor low-angle cross-bedding; clasts 75% subrounded lithic, 25% subrounded pumice to 5 cm Inversely graded pumice gravel, vitric matrix often absent in upper part of bed
Fine to very coarse sand or small pebbles, variable vitric:lithic ratios, poorly sorted; pebbles and cobbles subrounded to subangular; sandsized grains angular; lithic cobbles to 5%, and 5 cm; rare rip-ups of silt to fine sand; erosional basal contacts common
Gmvxl
Ggp
Gmpx
Gmp
Ghxlv
Gg(i)p
Spxlv
Pumiceous gravel
Normally graded pumice gravel
Pumice gravel
Openwork pumice gravel
Bedded gravel
Inversely graded pumice gravel
Massive, planar, or cross-bedded crystal/ lithic sand and gravel
Storm-wave deposit; ambient sedimentary style is vitric deposition, interrupted by input of lithic debris from shoreface or fluvial environments; also occurs in lagoonal setting
Offshore ‘saturation grading’ (see White et al., this volume): pumice clasts that settled from floating pumice rafts
High-energy swash or surf-zone deposit
156
Subaqueous reworking of saturated pumice gravel
Reworked deposit derived from erosion of layer 1(H)
T
Very poorly sorted gravel; clasts subround to angular, to + 1 m in diameter in crystal–lithic sand matrix; slight normal coarse-tail grading
Interpretation
Gmxl
Environment(s)
Lithic gravel
Description
Symbol
Facies
Table 1. Summary of lithofacies in Taupo 1.8 ka lacustrine sediments
VSIC07 9/12/08 2:04 PM Page 156
N. R. Riggs et al.
Stxlv
Smv
Smxl
Shxlv
Srvx
Flv
Cross-bedded crystal or lithic pebbly sand
Massive vitric sand
Massive crystal–lithic sand
Horizontally stratified crystal /lithic sand
Rippled sand
Massive/laminated vitric silt
S–O–L
O–L
Fine to medium dominantly vitric ± crystal /lithic sand, shows ripples that are (i) symmetrical or (ii) asymmetric or (iii) climbing
Very fine-grained vitric sand to silt; generally faintly laminated, otherwise structureless; interbedded with pumice granule lenses [facies Gg(i)p ] or horizons of vitric to crystal–lithic grains or medium sand (facies Smv and/or Smxl)
Offshore or lagoonal suspension settling of fine material from streams or elutriated from mass flows, storm deposits, etc.
(i) Shoreface, above-wave-base oscillatory traction current; (ii) shoreface to offshore unidirectional traction currents; (iii) shoreface sediment deposition in unidirectional traction current; all processes may occur in lagoonal settings
Swash-zone upper plane-beds
Reworked storm deposit or storm deposit in low-energy shoreface setting
S±O
S
Low-energy environment, quiescent shoreface to offshore
High-energy environment, especially surf zone
Sh S–O
Surf-zone or shoaling wave deposit; lithic component from dense-clast fraction of Layer 1(H) and main body of ignimbrite
Sh–S
Laminae 1–2 mm thick of alternating crystal /lithic and /or pumiceous fine to medium sand
Fine- to medium-grained crystal–lithic, massive to weakly bedded sand, generally well sorted; may contain vitric component and /or as much as 5% lithic granules; Beds 1–5 cm thick
Fine- to medium-grained vitric massive sand, generally well sorted; may contain as much as 5% crystal or lithic grains that generally define partings; beds 1–5 cm thick; may contain as much as 20% pumice granules
As for Spxl, trough cross-bedded
Medium to coarse sand, variable 25 –75% lithic + crystal, remainder vitric, rare (< 10%) pumice granules; lithic gravel to small cobbles common; laminated to bedded; poorly to moderately well sorted; beds 1-10 cm thick, commonly pinch out laterally; contacts sharp and commonly erosive; grains subround to subangular
Symbols follow standard format of Miall (1978); in addition, p, pumice, v, vitric fines (ash), x, crystals, l, lithic fragments, g, graded bed, (i), inverse. Subscript letters imply decreasing amounts of component. Depositional environments: T, terrestrial; L, lagoonal; Sh, shoreline; S, shoreface; O, offshore.
Spxl
Planar to cross-bedded crystal /lithic sand
VSIC07 9/12/08 2:04 PM Page 157
Sedimentation in a rapidly rising lake 157
VSIC07 9/12/08 2:04 PM Page 158
158
N. R. Riggs et al.
describe characteristic deposits from each of the four depositional environments. In the following section, we relate sedimentation patterns to geographical position, depositional energy and sediment influx. Offshore deposits Offshore deposits, formed below normal wave base (e.g. Walker & Plint, 1992), are predominantly fine to very fine-grained vitric sand that lacks identifiable physical and biogenic sedimentary structures, apart from bedding (facies Flv, Table 1, Fig. 3). Sediment is moderately well sorted and dominantly ash. In places where sections are exposed normal to shoreline, finegrained beds tend to thicken and coarsen toward the highstand shoreline. Interbedded with fine-grained beds are thinner beds of medium to very coarsegrained sand to cobbles (facies Gmpx and Smv, Table 1). Some of these beds are pumice gravel in which some pumice is vertically orientated (facies Gg(i)p). Other beds have erosional basal contacts, and include rip-up clasts derived from the fine-grained beds and as much as 25% lithic material (facies Spxlv). Sediment in these beds is medium to coarse-grained and sorting is poor. In offshore deposits, as in the rest of the lacustrine succession, biogenic structures are absent. The offshore sediments accumulated in water deeper than 8 –10 m. As discussed below, this allowed only storm-related bursts of coarse sediment to reach the offshore environment. Clay-rich sediment that is more typical of the modern offshore environment (Nelson & Lister, 1995) was not deposited, as a result of the clay-poor source materials (Wilson, 1985). Fine-
grained beds were deposited from suspension and probably represent ambient or ‘normal’ lake-refilling sedimentation below wave base. The lack of crossstratification in the deposits could be due to either rapid sedimentation rates or bioturbation. Lack of biogenic structures and the volcanic setting strongly suggest that these structureless beds resulted from large sediment influx into the lake following the Taupo eruption (see Palmer & Shawkey, 1997). We consider it unlikely that biogenic activity had been re-established to any significant degree during the time under consideration here, which closely followed the near-total evacuation of water from the lake and devastation of flora and fauna in and around the caldera during and following the eruption (e.g. Smith, 1991a,b). Pumice gravel (facies Gg(i)p) accumulated when rafted pumice became saturated and dense enough to sink to the lake floor (e.g. Manville et al., 1998; see also White et al., this volume). Vertical orientation is inferred to be due to slight preferential weighting of one end of a pumice clast by water. Coarse-grained sand and/or beds with significant lithic material (facies Smv, Spxlv) record traction deposition during storms that were energetic enough to transport coarsegrained, dense material into offshore areas of the lake. The concentration of lithic material in the storm deposits records effective segregation of heavier lithic from silt to fine sand-grade ash in the shore region, although deposits do include some pumice. Storm waves are inferred to have set up currents that were strong enough and of sufficient duration to transport high-density sediment offshore, as well as to
Fig. 3. Offshore deposits at Five Mile Beach, at ≈ 15 m above present lake level (top of photograph). Laminated vitric sands of facies Flv (Table 1) are interbedded with clastsupported, rounded pumice-granule and -cobble beds (facies Gmpx, Table 1). The upper third of outcrop in the photograph (darker colour) records the drop of the lake level, with rounded pumice stranded as the lake fell and infilled by rippled vitric sands (facies Gmp, Srvx). Scale is 2 m.
VSIC07 9/12/08 2:04 PM Page 159
Sedimentation in a rapidly rising lake winnow and remove other components to sites farther downdrift or offshore. Shoreface deposits Deposits of the shoreface (Figs 4 & 5) differ from offshore deposits in that they are better sorted and are characterized by plane and cross-lamination. Deposits are dominantly moderately- to well-sorted vitric sediment (facies Smv, Shxlv, Table 1), although poorly
Fig. 4. Shoreface deposits at Five Mile Beach, at 10 m above present lake level. Laminated and cross-laminated vitric to crystal–lithic sand (facies Shvxl and Stxlv, Table 1) is interbedded with lithic-rich sand and gravel (facies Spxl, Table 1). Storm deposit beneath scale is shown in close-up in Fig. 6. Scale is 90 cm.
159
sorted layers of fine- to coarse-grained, lithic-rich sand and gravel to cobbles (facies Ghlxv, Table 1) are present at some locations. Lithic sand and cobble beds are commonly overlain by cross-bedded crystal–vitric (dominantly ash) fine to coarse sand (facies Smxl). Pumice fragments in shoreface deposits are coarser than those in offshore deposits, generally are better rounded and commonly occur in normally to ungraded beds with or without an ash matrix (facies Ggp, Gmpx). Fine-scale planar stratification, good sorting, and overall coarser grain size reflect deposition by shoaling waves above wave base (Reineck & Singh, 1980; Collinson & Thompson, 1982). Wave activity was sufficiently energetic to abrade pumice in rafts before settling, or on the lake floor after initial deposition. Most deposits of the shoreface environment contain well-mixed vitric, lithic, and crystal sediment grains, and the percentage of the vitric component increases with apparent water depth. Wave-driven longshore drift during fair-weather periods carried small lithic fragments and water-saturated pumice fragments (White et al., this volume) into and along the surf and shoaling zones; smaller vitric fragments were carried further offshore, where they settled from suspension. Storm influence in the shoreface is recorded by coarser and more lithic-rich beds (facies Smxl, Ghlxv). These layers record unusual events that were able to transport coarse lithic material and deposit it close to shore while transporting finer-grained lithic material further offshore. Cross-bedded, crystal–vitric beds that overlie lithic layers (Fig. 6) were deposited as these storm currents waned. Pumice gravel was deposited by reworking of saturated pumice deposits closer to shore. Shoreline deposits
Fig. 5. Shoreface and shoreline deposits at Five Mile Beach, at ≈ 15 m above present lake level. Lower ≈ 30 cm of photograph is poorly sorted lithic sand and gravel of swash and surf deposits (facies Spxl, Stxlv, Table 1). Above this lie less energetic planar-laminated shoaling-wave facies from just behind the surf zone (facies Smv, Smxl, Shvxl, Srvx.). The light-colour deposits in the upper third of the photograph are transitional from shoreface to offshore. Scale is ≈ 180 cm.
Deposits interpreted as shoreline accumulations (Figs 2 & 5) include the coarsest beds exposed, as well as those containing the largest proportions of crystal and lithic material (see Clifton et al., 1971; Table 1). Beds are predominantly granule-sized pumice grains and coarse sand to pebble-sized lithic clasts, and are moderately to poorly sorted (facies Ghxlv, Spxl, Table 1). In most places we studied, shoreline accumulations include planar-laminated swash deposits and trough cross-bedded deposits (facies Spxl, Stxlv) formed in the surf zone. Locally, however, laminated sand is interbedded with 1–5 cm-thick structureless, poorly sorted vitric–crystal–lithic sand and lithic-cobble beds (facies Ghlxv, Gmvxl).
VSIC07 9/12/08 2:04 PM Page 160
160
N. R. Riggs et al.
Fig. 6. Close-up of gravel-bed shoreface storm deposit (facies Spxlv, Table 1) at Five Mile Beach in Fig. 5. (Note scouring of underlying laminated shoreface deposits.) Lithic-rich granule bed is overlain by pumice and lithic cobbles. Vitric sands (facies Flv, Srvx) at and just above top of scale may record a calming of the lake and settling of suspended material. Scale is ≈ 60 cm.
The concentration of lithic material in these deposits, together with planar laminations and crossbeds, indicate accumulation in relatively high-energy swash and surf zone environments. Beds with planar laminations accumulated in the swash zone under upper plane bed conditions (e.g. Clifton, 1969). Deposits with cross-beds accumulated where beach steps and shallow longshore troughs were cut by breaking waves (Davis & Ethington, 1976). Structureless, poorly sorted beds (facies Ghlxv, Gmvxl) are interpreted as deposits of debris flows that resulted from erosion of ash-covered slopes, especially near stream mouths (see discussion below). Lagoonal deposits Deposits interpreted as lagoonal are nearly uniformly fine grained, often with abundant cross-laminations. Lagoonal deposits differ markedly from those described above. Grain size only rarely is larger than coarse sand, although individual pumice pebbles to small cobbles are common. Deposits comprise primarily planar- to ripple-laminated vitric–ash silt to fine sand (facies Flv, Srvx, Shxlv, Table 1), weakly planar bedded, poorly sorted coarse to fine crystal–lithic– vitric sand (facies Spxlv), and well to poorly sorted pumice-pebble gravel (facies Ggp, Gmvxl). In most cases, ripples climb away from, or parallel to, the present lake shore. Pumice granules and pebbles are
uniformly rounded to well rounded. Lithic- and crystal-rich deposits are rare, making up at most 10% of a succession. Ripple- to planar-laminated fine-sand beds (facies Srvx, Shxlv) are interpreted as having formed by winddriven currents in elongate shore-parallel lagoons, and by hyperpycnal flows from sediment-laden stream discharge into the lagoons. Material was derived either from erosion of the ignimbrite or potentially from small rafts of pumice that washed into the lagoon during storms. Well-sorted pumice-pebble gravel (facies Ggp) probably represents redeposition of watersaturated pumice by weak currents (White et al., this volume). Poorly sorted sands to pumice-gravels (facies Gmvxl) are interpreted as debrisflow deposits derived from locally eroding Taupo ignimbrite (see Clarkson, 1996). Overall, deposition is inferred to have been by weak currents, and the deposits reflect the limited range and bimodal size distribution of the available pumice sediment transported into the lagoons. Deposits characteristic of the shoreface are absent, and the position of lagoonal sites at or very close to the outer margin of the highstand terrace precludes an interpretation of offshore deposition.
LAKE TAUPO SEDIMENTATION REGIMES Wave energy and sediment influx vary with geographical location around Lake Taupo (i.e. exposure to prevailing weather systems) and with proximity to rivers, and thus we have classified our study locations into four groups based on these factors (Fig. 7). Below, we present descriptions of representative sites of accumulation under the four possible combinations of high versus low wave energy and high versus low sediment availability. The succession at each site accumulated over a very short period of time, from months to at most a few years, as the lake rose past a particular elevation. Each site is characterized by a specific facies association that reflects conditions of energy, sediment supply, topography and the rapid rate of lake-level rise. High-energy and high sediment-influx rate The study area best representing accumulation under conditions of high energy and high sediment influx is Five Mile Beach (Fig. 1). The beach has a maximum fetch of nearly 40 km from the south-west (the dominant strong wind direction), and thus bears the brunt of
VSIC07 9/12/08 2:04 PM Page 161
161
Sedimentation in a rapidly rising lake Es
5 Mile Beach
ES Es: high energy, low sediment influx
Kaiapo
ES: high energy, high sediment influx
Fig. 7. Schematic graph of energy versus sediment influx, showing sites that represent each of the categories: high energy and high sediment influx; low energy and high sediment influx; high energy and low sediment influx; low energy and low sediment influx.
increasing energy
Waipehi
eS: low energy, high sediment influx
Waitahanui Hingapo Kuratau
Acacia Bay
es: low energy, low sediment influx
Pokoura Pa
Kinloch
es
eS increasing sediment contribution
storm waves. The depositional area was long and broad, based on present-day topographic contours (Fig. 1), with a major drainage, the Waitahanui River, at its southern margin, along with several smaller streams. The total area drained is several hundred km2, indicating a large sediment input. Outcrop at Five Mile Beach is c. 1.5 km long parallel to the lake shore. The highstand terrace at Five Mile Beach is marked by lags derived from, and relics of, the ground layer (layer 1(H)) of the Taupo ignimbrite. Where preserved intact (e.g. on the Taupo airport road; Fig. 1), the proximal ground layer consists of a poorly sorted mixture of pebbles to (> 5 m) boulders in an ash matrix. This matrix generally was eroded preferentially leaving behind isolated boulders and boulder clusters. The Five Mile Beach area is also unusual in having exposed many examples of the ‘floated giant pumice blocks’ of Wilson & Walker (1985, p. 213). These blocks are incorporated in facies representing differing water depths between Five Mile Beach and Motutere, with the largest concentration in the area between Waitahanui and Five Mile Beach (Fig. 1). We infer that the grey, pumiceous-block horizon formed over a relatively short period of time, and that it thus provides an indication of lake-margin topography at the time of deposition. Transgressive deposits at Five Mile Beach, which in individual facies assemblages can be traced over 3–5 m vertically and 25 m parallel to shoreline, are characterized by fining-upward sequences. These sequences, where complete, include beach deposits at the base to offshore deposits at the top (Figs 3 –6 & 8). Surf and shoreface deposits are characterized by welldeveloped planar and low-angle cross-stratification (facies Stxlv, Shxlv, Table 1), and contain a mixture of vitric and lithic material. Lithic-rich storm layers are common in shoreface and offshore deposits (Figs 4 & 8). Lithic material in storm deposits varies in size from medium sand to cobbles. Offshore deposits are weakly
laminated to massive pumice gravel to fine vitric sand (Gg(i)p, Flv, Table 1). Overall, the prevalence of fine ash in the deposits at Five Mile Beach implies low average energy, and leads us to infer that lithic-sandrich beds, whether cross-bedded or massive/laminated, are the result of more energetic, storm-driven sediment influx. The local presence of lithic pebbles supports this interpretation. The section at Five Mile Beach is, in places, capped by pumice-gravel (Gmp), which comprises clastsupported, rounded pumice up to 20 cm in diameter with local vitric sand matrix. The deposit locally overlies other lacustrine units on an angular unconformity. Pumice gravel is interpreted as a strandline deposit that accumulated during rapid fall of the lake from its highstand. Two sections near the shoreward margin of the highstand terrace give information about the topography at the shoreline (Fig. 9). These sections have beach and talus deposits that contain > 25 cm lithic blocks derived from layer 1(H), which in this area formed a cliff at the highstand shoreline. Areas in which layer 1(H) was exposed formed much steeper beaches or beach margins than areas of shoreline cut into finergrained deposits of the main eruptive unit (layer 2). The retreat of the layer 1(H) cliff kept pace with beach formation, thus providing a steady supply of coarse lithic debris to the system. Two sections at 15 m above lake level (a.l.l.) are particularly useful in reconstructing the shoreface (Fig. 8), because the grey pumiceous blocks occur within different facies. The southern site (Fig. 8A) contains clast-supported, subangular to subrounded, grey pumiceous blocks from 5 to 35 cm in diameter, mixed with 1–5 cm white pumice and coarse vitric sand that sifted into the spaces between clasts. This deposit overlies cross-bedded crystal–lithic sand and minor gravel (facies Spxl, Stxlv, Table 1) that we interpret as surf zone deposits. The pumice blocks may represent storm beach berm or storm surf deposits; if
162
Fig. 8. Stratigraphic sections illustrating facies in high-energy, high-sediment-influx setting (ES) (15 m a.l.l., Five Mile Beach; see Fig. 2 for location). (A) Southern, and (B) northern sections 8 m apart. Tie line is at the base of units that contain Horomatangi pumice clasts. Vertical offset of 1.45 m between columns represents difference in elevation above lake level. The base of (A) is at 15 m; base of (B) is at 14 m. Explanation applies also to Figs 9 –13.
VSIC07 9/12/08 2:04 PM Page 162
N. R. Riggs et al.
VSIC07 9/12/08 2:04 PM Page 163
Sedimentation in a rapidly rising lake
163
Fig. 9. Stratigraphic sections at ≈ 34 m, Five Mile Beach. Basal deposits are in lithic-rich breccia that represents erosion of the layer 1(H) cliff along the north-east margin of the lake. The top of each section is at 34 m a.l.l. Upper 55 cm of thicker section represents lakefall stranding deposits.
the former, then rapid shoreline buildout must have occurred during the storm for berm deposits to form atop the swash zone deposits. The northern site (Fig. 8B) is about 8 m to the north, and the grey
pumiceous blocks are 1.45 m lower in elevation. Here, clast-supported, closely packed, angular, grey pumice ranges from 5 to 220 cm in diameter. Sub-rounded white pumice as much as a few centimetres in diameter
VSIC07 9/12/08 2:04 PM Page 164
164
N. R. Riggs et al.
and minor sand-grade ash partially fill the spaces between grey pumice clasts. This grey pumice layer occurs above suspension deposits (facies Ggp) and ripple-laminated, silt- and sand-grade ash deposits (facies Srvx, Smv) of the outer shoreface; the largest block in this horizon has a planar lower surface that overlies grey pumiceous blocks < 30 cm in diameter mixed into crystal–lithic and ripple-laminated sand (facies Smxl, Srvx; 215 –225 cm, Fig. 8B) that we interpret as an outer-shoreface storm deposit. In this area, within 1.5 m water depth, conditions changed from highly turbulent in the surf zone to relatively quiescent less than 10 m away on the deeper shoreface. The storm that brought in the grey pumiceous blocks is, however, recorded at both sites. The angular nature of the grey pumice clasts at the southern section may be a combination of breakage in the surf environment and progressive thermal decrepitation of large hot clasts, with fractures releasing new, hot clasts that immediately sank (see Whitham & Sparks, 1986). At the storm–surf site, the clasts rolled and broke in the waves, developing subangular to subrounded shapes. About 10 m north of this site, a small stream entered the lake. Flat-lying fine-grained deposits at the same stratigraphic level contain no grey pumiceous blocks, and represent deposits from the eroding ignimbrite on shore. The incoming stream may have deflected floating pumice to the north or south. Storm deposits within predominantly vitric ash sediments are characterized by abundant lithic material that varies in size from medium sand to cobble. Deposition of this lithic component reflects movement of shoreface and beach materials to offshore sites in strong storm-surge currents. This compares to sedimentation in other clastic offshore settings, where Reineck & Singh (1975) noted that storm layers on muddy oceanic shelves are recognized by the sandy component brought in from coastal areas. At Five Mile Beach, storm layers can clearly be recognized, lying above sediments deposited above fairweather wave base. These storm beds were not reworked by fairweather waves, suggesting that aggradation and water-level rise were rapid enough to bury shoreface deposits before normal wave action could rework them. McCubbin (1982) noted that storm layers contain abundant hydraulically light particles, generally comprising plant debris, mica, etc. At Taupo, where plant fragments and clay are absent, background suspension sedimentation is dominated by silt deposition. Coarsegrained storm beds therefore grade, without a significant break in grain size, back to the ambient silt, rather
than forming a wholly distinctive sand and silt layer enclosed within otherwise muddy sediment (Fig. 6). High energy and low sediment influx Study areas representing accumulation under highenergy and low sediment-influx conditions are at Waipehi and Kaiapo Bay (Fig. 1). Waipehi Stream is a small drainage that enters the lake north-east of Mission Bay (Fig. 1). The Waipehi site may have been protected from abundant longshore sediment influx by a promontory at Motutere (Fig. 1), but our analysis suggests that it received a good deal of wave energy. The cliff rises abruptly from the lake here, and thus we infer that the beach, although long, was very narrow. Kaiapo Bay is the smallest of a series of bays along the northern shoreline. The 20 km distance between the bay and the western shoreline of Lake Taupo provides sufficient fetch to develop substantial wave energy during storm periods, although probably less than that at Five Mile Beach. The availability of sediment at Kaiapo Bay was limited by the surrounding topography, which rises sharply from the lake; a small stream system with a catchment area of about 4 km2 drains into Kaiapo Bay. Waipehi Three sections along a road cut parallel to the modern shoreline expose shoreface to offshore deposits between 14.5 and 22.5 m a.l.l. at Waipehi; the middle section, at 15.5 m a.l.l., is illustrated in Fig. 10A. The thickness of shoreface and offshore deposits varies with elevation at Waipehi. Individual beds in the upper 25 cm of the lowest section, which comprises interbedded pumice gravel (facies Gmpx and Ggp, Table 1) and cross-bedded and ripple-laminated vitric sand of facies Shxlv and Srvx, correlate with the basal portion of the section 1 m higher, but equivalent beds in this section are crystal–lithic rich (facies Spxl; Fig. 10A). In the two upper sections, ripple-laminated, crystal–lithic sands and pumice gravel (facies Spxlv, Srvx, Gmp, Ghxlv, and Ggp) grade upward into pumicegravel suspension deposits (facies Gg(i)p) as much as 1.3 m thick. We infer that storms are represented in these deposits either as wave-rippled deposits or as lithic-rich, cross-bedded and laminated surf zone deposits. Suspension deposits (facies Ggp) contain layers of both subangular and rounded pumice, suggesting that suspension sedimentation was from pumice rafts that were augmented during storms when streamfloods
VSIC07 9/12/08 2:04 PM Page 165
Sedimentation in a rapidly rising lake
165
Fig. 10. Stratigraphic sections illustrating high-energy, low sediment-influx settings (Es) at (A) Waipehi and (B, C) Kaiapo Bay. Elevation at Waipehi is for base of section. Elevations for Kaiapo are tops of sections.
flushed unabraded pumice into the lake. Pumice fragments deposited in the shoreface are coarser than those in offshore deposits, but generally are rounder. This suggests abrasion during wave reworking of water-saturated pumice above wave base after they settled (see White et al., this volume). Kaiapo We logged four sections between 8 and 21 m a.l.l. at Kaiapo. Two are described here (8 and 16 m a.l.l.; Fig. 10) to illustrate the combined effects of highenergy, low-sediment influx and confined topography. The base of the post-1.8-ka sedimentary section is not exposed in the lower section, at 8 m a.l.l. (Fig. 10B). Approximately 70 cm of coarse-grained crystal–lithic sand and lithic-clast gravel (facies Ghxlv, Smxl, and Spxl, Table 1) are overlain in an apparent channel by pumice-gravel (facies Ggp) suspension deposits that may have formed in a lower shoreface setting under especially quiet conditions. The upper 80 cm of the section comprises interbedded fine- and coarse-grained, crystal–lithic–vitric sand to rare pebbles of facies Spxl and Flv. The coarse beds are commonly scoured, and fine-grained units are weakly bedded. One 5 cm interbed consists of cross-bedded fine ash. We interpret the 8 m a.l.l succession as dom-
inantly lithic-rich storm deposits. Fine-grained beds in the upper 80 cm section may be material winnowed from the coarse-grained storm deposits with which they are interbedded. A section at 16 m a.l.l. (Fig. 10C) exposes c. 1.5 m of cyclically interbedded pumiceous gravel and laminated ash (facies Flv, Gmv, Smv). The upper 0.5 m of the section is openwork pumice beds of facies Gmp and Ggp that dip lakeward at 12°, parallel to the present land surface. Lithic detritus is present only in rare thin beds in the section. We interpret this section as lower shoreface to offshore deposits, except for the upper layer of openwork pumice, which is inferred to represent a stranded pumice raft (see below). Lacustrine sediment is not present at the elevation of the highstand shoreline at Kaiapo Bay, where the shoreline notch is cut into Taupo ignimbrite and lacks sedimentary cover. Approximately 30 m lakeward from the highstand shoreline notch, a sedimentary section about 15 cm thick rests on primary ignimbrite and comprises massive fine to medium sand consisting of 50% pumice, 30% quartz and feldspar, and 20% mafic crystals, overlain by 10 cm of fine, faintly laminated sand. We interpret these relations to indicate that little primary pyroclastic material was supplied to the bay, and that strong wave action at the lake’s highest stand removed most of the beach material. The
VSIC07 9/12/08 2:04 PM Page 166
166
N. R. Riggs et al.
preserved sedimentary section is debris from Taupo ignimbrite or post-1.8-ka lakeshore deposits reworked by fluvial or mass-flow processes. Together with the presence of only thin highstand successions preserved at Mission Bay, Hingapo, Kuratau (see below) and elsewhere around the lake, we interpret this succession as evidence that the ‘highstand’ was not a long-term event. Low energy and high sediment influx rate Study areas representing low energy and high sediment influx are Waitahanui, Kuratau and Poukura Pa (Fig. 1). These areas are distributed around the eastern to south-western sections of the lake, but are characterized by protection from major wave energy and by locations at the mouths of major drainages or downdrift from major rivers. Waitahanui The Waitahanui site (Fig. 11A) is ≈ 7 km south of Five Mile Beach, and ≈ 1.5 km south of where Waitahanui River at present debouches into the lake (Fig. 1), although the mouth of the river was apparently much closer to the site when erosion of the ignimbrite landscape began (see below). Because this site is located at the mouth of a river draining several hundred km2, we infer that the area had high sediment input. The Te Kohaiakahu promontory, which has a steep southern face extending into deep water (Fig. 1) shielded the site from material drifting northward from the major drainages farther south (Fig. 1). Importantly, it also protected the site from the brunt of south-western storms. The lower of the two Waitahanui sections is at 9 m a.l.l. (Fig. 11A). The base of the section is in pebbly crystal–lithic sand (facies Spxl, Table 1), interpreted as surf zone sediment, that has a maximum lithic clast size of 2.5 cm. Vitric clasts makes up as much as 50% of the material in this section. These deposits are overlain by about 3 m of shoreface deposits of facies Ggp, Ghxlv, Spxlv, Spxl, and Stxlv vitric to crystal–lithic sand and pumice-gravel, with common 1 cm lithic clasts and rare soil clasts scattered throughout. The sediments are low-angle cross-bedded or plane-bedded. The presence of fine lithic lenses and horizons is noteworthy, as is the overall moderate to poor sorting of all facies. The lithic horizons (facies Ghxlv, Spxlv) are interpreted as storm deposits; a lack of strong wave-induced currents, together with the overwhelmingly vitric sediment supplied by the Waitahanui
River, tended to minimize the record of storms. The lacustrine section dips beneath a 2 m section of crossbedded and laterally discontinuous pumice-sand, -pebble, and -cobble beds that are interpreted as fluvial deposits. A second section was logged ≈ 500 m inland from the basal section (Fig. 2), close to the highstand shoreline, which is at ≈ 385 m a.s.l. or 30 m a.l.l. in this area. The exposed section consists of ≈ 3.25 m of entirely vitric, interbedded suspension (facies Gg(i)p, Ggp, Smv, Table 1) and ripple-laminated fine sand deposits (facies Srvx and Flv; (Fig. 12A) ). Pumice occurs in lenses. Clasts in these lenses coarsen upward to as much as 10 cm in diameter. Ripple cross-laminae in the upper part of the section climb to the south-southeast, i.e. away from the lake. The Waitahanui River currently enters Lake Taupo across a sand bar. Such channel-mouth bars are uncommon around the lake, and this one is probably related to the interplay of wave action behind Te Kohaiakahu promontory, and longshore sediment movement. We interpret facies observed at the inland section as similar to those now being formed behind the channel-mouth bar. In the post-eruptive setting, sediment was discharged at high rates from the Waitahanui River and accumulated at its mouth. We infer that waves refracted around Te Kohaiakahu promontory built up a low-relief bar along the lake shoreline that separated the open lake from a shallow, broad backwater area behind it. Continued arrival of sediment at a high and constant rate allowed lakeward progradation of the bar while the area behind it continued to hold a shallow backwater lagoon. Shoreward-directed ripples probably reflect washover or wind-driven wave-induced currents in the lagoon. Facies exposed at the Waitahanui sites reflect the interplay of abundant stream-supplied sediment with conditions of relatively low wave energy. The thick shoreface assemblage at the lower site (Fig. 11A) shows no indication of increasing depth upward, suggesting that the sediment supply kept pace with lake rise. At an estimated lake-level rise of ≈ 5 –9 m yr –1, this required a large and steady supply of material from the Waitahanui River. The small size of lithic fragments (1 cm) suggests that larger clasts were not locally available, or that no high-energy currents came into the area. Storm deposits are poorly developed, and the transition from gravelly surf zone facies to quieter shoreface (low-angle cross-bedded and planebedded deposits interbedded with pumice-gravel), suggests erosion of the beach swash deposits during transgression.
VSIC07 9/12/08 2:04 PM Page 167
Sedimentation in a rapidly rising lake
167
Fig. 11. Stratigraphic sections illustrating low-energy, high sediment-influx setting (eS) at (A) Waitahanui and (B) Poukura Pa. Elevations for both columns are at top of section.
Kuratau Kuratau and Poukura Pa are both on the southwestern shoreline of Lake Taupo (Fig. 1). This area is rarely affected by storms, but the storms that do affect
the area are the large, cyclonic systems that develop from the north-east and carry the strongest winds recorded in the Taupo area (Salinger et al., 1998). The area to the west of Kuratau is a long, broad plain. Abundant sediment would have drained down this
VSIC07 9/12/08 2:04 PM Page 168
168
N. R. Riggs et al.
Fig. 12. Stratigraphic sections illustrating lagoonal settings. (A) Waitahanui; (B) Kinloch. Both sections are near the lake highstand. Waitahanui site is in an abandoned quarry; elevation given is base of section. Elevation at Kinloch section is top of section.
VSIC07 9/12/08 2:04 PM Page 169
Sedimentation in a rapidly rising lake slope, and it is possible that the Kuratau River entered the lake south of its current position. We logged four sections west of the village of Kuratau over 5 m vertically. The upper three sections are within 1.5 m of one another laterally, and were dug in the riser to the 34 m terrace (here at 29 m a.l.l.) (Fig. 13). The sections are each based in a crystal– lithic sand to granule gravel, which continues down at least 50 cm in all three sections. The easternmost section (highest elevation; Fig. 13A) comprises ≈ 110 cm of facies Gmvxl, interpreted to represent a debris-flow deposit locally derived by reworking of Taupo ignimbrite. Further downslope (Fig. 13B), this material is overlain by a mixture of coarse planar-bedded crystal– lithic–vitric sand and gravel (facies Ghxlv, Spxl, Shxlv, Table 1), which are thinly interbedded. Approximately 75 cm downslope of the second section (Fig. 13C), the planar-bedded section is capped by soil-rich openwork(?) pumice (facies Gmp(?)), which is overlain by soil-rich gravel identical to that at the two upper sec-
169
tions. Overall, the sections show an interbedding of surf zone facies with deposits from mass flows off the ignimbrite. Deepening water is reflected by the increased thickness of swash deposits and the incipient development of deeper shoreface gravel. We speculate that the low frequency of high-energy storm waves permitted debris flows entering the lake to remain largely unreworked. The fourth section was logged 4 m vertically downslope. The base of this sequence (Fig. 13C) is interbedded crystal–lithic and vitric sand (facies Smv and Smxl, Table 1) in turn overlain by medium to coarse crystal–lithic sand with pockets of pumice- and lithicclast gravel (facies Ghxlv, Gmvxl, Spxl, Table 1). This succession is overlain by cross-laminated vitric sand (facies Shxlv, Srvx) and a 25 cm-thick crystal–lithic sand and gravel deposit (facies Spxlv), which we interpret as a storm deposit. Preservation of this storm deposit is an important aspect of the low-energy side of the lake: although strata below the deposit are
Fig. 13. Stratigraphic sections illustrating facies at the lake edge in the low-energy, high sediment-flux setting at Kuratau. The basal unit in all cases is interpreted as a surf-zone deposit; the overlying unit in (A) is interpreted as debris-flow deposit, and is the same in (B) and (C). This is the only site at which debris-flow deposits are interstratified with lacustrine sediments. Bases of sections are at road level at ≈ 27 m a.l.l.; the highstand terrace in this area is at 29 m a.l.l.
VSIC07 9/12/08 2:04 PM Page 170
170
N. R. Riggs et al.
laminated and cross-laminated, those above are only very slightly laminated, and are bioturbated. This suggests very quiet conditions, in which wave action was not energetic and the storm deposit was rapidly buried by lower shoreface deposits (see Elliott, 1986). Lastly, several beds or sets of lake beds mantle the terrace riser along the road exposure between the upper and lower sections. This mantling is interpreted to reflect the morphology of the underlying ignimbrite surface. Poukura Pa Poukura Pa lies ≈ 5 km north of Kuratau. Like Kuratau, the area may have received a good deal of sediment from the west. We logged a 3.8 m-thick composite section along a road cut ≈ 6 m below the lake highstand notch (Fig. 11). The basal beds comprise ≈ 0.5 m of crystal– lithic to ash sand and gravel of facies Spxl and Shxlv (Table 1) that dip 27° lakeward. This section is overlain by a thicker section (2.3 m) of very shallowly dipping ripple-laminated and planar-bedded (facies Shxlv and Srvx) to structureless, crystal–lithic-rich facies and lithic gravel of facies Ghxlv, Spxlv and Spxl and well to poorly sorted vitric facies Shxlv, Ggp and Gmv, which we interpret as swash deposits deepening to shoreface and suspension deposits. Along the road cut, the upper 1 m of suspension deposits is planed off beneath a poorly exposed, cross-bedded, crystal–lithic sand that we interpret as a wave-base storm deposit (facies Spxlv). This bed is overlain by 30 cm of planar to crosslaminated fine vitric sand of facies Shxlv and Flv. Poukura Pa has a more complete section than Kuratau and illustrates deeper-water facies associations of the low-energy environment. Surf zone deposits are dominantly fine grained to granule size, and cross-stratification is only weakly developed. The steep dip of these beds is attributed to subaqueous construction of a small step in the surf zone. The abundance of suspension deposits (224 –254 cm, 280–340 cm, Fig. 12) indicates that only weak waves or currents were available to rework deposits. Vertically orientated elongate pumice clasts are preserved in one bed. Only one storm deposit is preserved at the top of the section, probably representing an easterly storm. The shoreface sequence is nearly 3 m thick and contains sedimentary structures that vary from planar bedding, suggestive of swash zone sedimentation, to cross-bedding and ripple lamination, which we interpret as representing deeper shoreface facies. Although the succession clearly indicates deepening water, its thickness indicates that the rate of sedimentation was not substantially slower than the rate of lake-level rise.
Low-energy and low sediment-influx rate Kinloch, Hingapo and Acacia Bay represent sedimentation in the low-energy, low sediment-influx regime (Fig. 1). Kinloch township is along Whangamata Bay on the north-eastern shoreline of the lake. Although the area has a 20 km fetch from the south-west, topography is such that, at 15–34 m a.l.l., the Okaia drainage area is protected, although the south side of Whangamata Bay may receive substantial storm energy. The small catchment area of the Okaia stream results in small rates of sediment influx. The Hingapo area is north of the Tauranga–Taupo River on the south side of the lake (Fig. 1). The Motuoapa Peninsula protects the area from storm waves. Acacia Bay is on the north-eastern lake shore and has a fetch distance of 2–3 km across the bay. Three small drainages into the bay have a total catchment area of 3 km2. Acacia Bay best represents the low-energy, low sediment-influx endmember. Kinloch The section at Kinloch (Fig. 12), just below the top of the highstand terrace, is characterized by intercalated beds of fine to very fine sand- to silt-grade ash and pumice-granule gravel of facies Flv and Smv, capped by pumiceous gravel of facies Gmvxl. Overall, the section comprises fine ash, fine-grained beach and suspension deposits, and debris-flow deposits. The section completely lacks coarse-grained or lithic intervals characteristic of beach or shoreface deposition elsewhere. This section is unique in that it apparently represents calm-water conditions around the time of the lake’s highstand. We interpret this setting as lagoonal, or, possibly, as a perched pond or small lake that may have occupied a topographically low area in the Taupo ignimbrite as the lake rose (see Manville, this volume). In either case, the ignimbrite in the hinterland was being contemporaneously eroded, yielding debris flows that may have entered the lake. Debris-flow deposits are exposed uphill from this section without associated lacustrine sediments. Hingapo The sections at Hingapo (Figs 2 & 14) are thin (< 25 cm) intervals of beach and shoreface sands. The upper section contains small pockets of beach sands of facies Spxl (Table 1) cut into pumiceous gravel of facies Gmvxl that is interpreted as debris-flow deposits derived from primary Taupo ignimbrite. The lower section comprises shoreface (facies Spxl, Gmp, Srvx)
VSIC07 9/12/08 2:04 PM Page 171
Sedimentation in a rapidly rising lake
171
Fig. 14. Stratigraphic sections illustrating low-energy, low sediment-influx setting (es), at (A) Hingapo Road and (B) Acacia Bay Road. Elevation at Hingapo is at road level where sections were dug in an embankment; base of Acacia Bay section is in road cut at 27 m a.l.l. (see also Fig. 15).
and suspension deposits of facies Gmpx and Ggp that overlie primary Taupo ignimbrite. The pronounced thinness of both sections indicates a very short-lived and /or sediment-starved setting. Post-depositional erosion may have planed sediment from the top of the sections, but, as at Kaiapo, significant erosion seems unlikely, because of the consistent small thicknesses of deposits at many different elevations around the lake. The Hingapo Road sections are nearly alongside the Tauranga Taupo River and, although the sediment discharge from the river over time is inferred to have been great, the modern river is confined to a steep-walled valley that extends to within a few hundred metres of the lake and terminates ≈ 15 m above present lake level. We infer that the Tauranga–Taupo drainage was poorly integrated, and that material ponded in small tributary streams and ephemeral lakes. This, combined with a slight topographic barrier, prevented any quantity of material from accumulating at the site. The lack of accumulation is
especially noteworthy in light of the broad width and low relief of the present river mouth, as a result of the large sediment influx into the lake system subsequent to the lake’s fall from the post-1.8-ka highstand. Acacia Bay Road The Acacia Bay Road section (Figs 1, 14 & 15) is well exposed and yields information about facies development in a protected bay. Continuous exposure for > 280 m along Acacia Bay Road reveals gently lakeward-dipping strata that progressively onlap Taupo ignimbrite above a low-relief, scoured base that parallels the ignimbrite, which dips at ≈ 4° toward the lake. Individual beds can be traced for up to several metres away from the palaeoshoreline, represented by a scoured contact. The truncated surface of the ignimbrite includes both sharply scoured and subplanar segments. The basal 5–25 cm above the latter segments (Fig. 14)
VSIC07 9/12/08 2:04 PM Page 172
172
N. R. Riggs et al.
Fig. 15. Photograph of Acacia Bay Road exposures. Scale is 2 m stick with base at ≈ 27 m a.l.l. Massive unit at base is Taupo ignimbrite. (Note laterally continuous and discontinuous horizons in lacustrine sediments.)
comprise planar-laminated crystal–lithic sand, with or without lithic and pumice granules (facies Spxl, Table 1), which is interpreted as swash zone deposits (Clifton, 1988; Fritz & Howells, 1991). Individual swash bedsets can be traced lakeward into ripple-laminated fine to medium ash sand of facies Srvx, with local interbedded, 15 –20 cm-thick deposits of massive to weakly bedded or cross-bedded medium to coarse crystal– lithic sand (facies Spxl, Stxlv). These rippled vitric-sand horizons are characterized by lakeward-dipping ripple cross-laminae and are interpreted to represent waveaffected low-energy underflows (weak rip currents) escaping the confined Acacia Bay re-entrant. The thicker, coarser beds represent storm deposits. In a lacustrine, non-tidal setting, such beds typify parts of the shoreface zone (see Clifton, 1988). Upsection, the sand becomes increasingly fine, vitric, and planar laminated (facies Flv, Table 1), with only rare thin (2– 5 cm) crystal–lithic sand horizons. Pumice gravel (facies Ggp, Gg(i)p, Table 1) caps the section. Overall, the succession has a mix of crystal–lithic sand and vitric sand, which represents the inability of the lowenergy environment to fully winnow all the vitric sand out of the shoreface area. Subplanar truncation surfaces between prominent scours extend for up to 3 m laterally. These surfaces may represent gradual onlapping and burial of the shoreline by small, low-relief shoreward prograding beach berms (Renaut & Owen, 1991), which form shoreward-dipping layers. We infer that the troughs cut into the ignimbrite represent scour from washover during storms. Following storm scouring, new berms were initiated slightly shoreward of the preceding
ones, and progradation on to the ignimbrite shore was re-established. Variations in lithofacies associations around Lake Taupo Remarkably little variation in lithofacies distribution or characteristics exists among all the sections we described. Sections at all locations are characterized by fining-upward sequences that expose part or all of the onlapping beach–shoreface–offshore associations. Beach deposits at nearly all locations are characterized by planar-laminated crystal–lithic sand and gravel. Shoreface deposits are dominantly planar- and ripplelaminated or structureless vitric sand, and offshore deposits are fine to very fine-grained vitric sand and pumice gravel. There are, however, subtle variations in the distribution and sediment composition of the four major lithofacies associations. High-energy regimes differ from those in lowenergy settings primarily in that storm facies are more common and storm events also leave traces in both offshore and shoreface accumulations. For example, in the high-energy environments of Five Mile Beach (Fig. 8) and Waipehi (Fig. 10A), lithic-rich sand and gravel are found well into deposits interpreted as offshore. At Kaiapo Bay (Fig. 10B), which is confined and where storm-wave energy may have been somewhat moderated, storm deposits in offshore sediments are admixtures of crystal and lithic sand grains. Lower-energy environments record storm events, if at all, only in the shoreface environment. At Poukura Pa (Fig. 11B), lithic clasts in rare storm deposits are
VSIC07 9/12/08 2:04 PM Page 173
Sedimentation in a rapidly rising lake as large as 5 cm, indicating not only a local source of material, but wave energy occasionally sufficient to transport such clasts into the shoreface environment. Poukura Pa has as much as 3 m of shoreface sediments, with only one storm deposit, which is probably a reflection of the rarity of north-east-derived storms that would affect the area and the short time over which 3 m of section accumulated (i.e. probably as little as 1 yr). The low-energy environment at Waitahanui (Fig. 11A) is marked by storm-related lithic concentrations in the shoreface deposits that change upsection (i.e. into deeper water) into exclusively vitric sand and gravel. Differences in lithofacies distribution related to sediment supply are not as straightforward. We saw no significant differences with changing rates of sediment influx in lithofacies associations at the high-energy sites (Five Mile Beach and Kaiapo). At beaches in the low-energy regimes, storm deposits are more common in areas with a high rate of sediment influx (Kuratau (Fig. 13), Poukura Pa) than those with lower rates (Acacia Bay, Kinloch, Hingapo). Highly protected sites, such as upper Waitahanui and Kinloch (Fig. 12), were simply not subject to storm waves and currents. The noteworthy aspect of areas of high sediment supply, for example Five Mile Beach and the lower Waitahanui site, is that sedimentation rates apparently kept up with lake-level rise, in these environments, resulting in sequences more than 3 m thick that lack strong deepening-upward trends.
LAKE TAUPO SEDIMENTATION Patterns in lake sedimentation One of the most distinctive and interesting aspects of the deposits around Lake Taupo is the broad similarity of lithofacies and lithofacies associations across a large spectrum of wave-energy and sediment-discharge conditions. Apparently wave-induced currents were strong enough around almost the entire lake shore to develop a characteristic sequence of lithofacies associations related to shoaling waves, despite sedimentation rates of ≈ 1 m yr –1 or greater. The influences of depositional energy and sediment influx produced subtle changes in sedimentation conditions.
173
environment around the lake shoreline, that at Five Mile Beach, corresponds to a long, broad beach. Here, waves cut through the ignimbrite to the level of layer 1(H), and beach sediments are correspondingly rich in large lithic clasts. Conversely, at Acacia Bay wave energy was so low that steps only centimetres high were developed in the easily eroded ignimbrite (Fig. 15). An intermediate example may be represented at Kaiapo Bay, where a very small area may have experienced focused wave energy but where layer 1(H) was not reached during erosion as the lake filled. At Kaiapo Bay, the confining topography may have refracted wave energy enough to move sediment, but not to severely erode the underlying ignimbrite. Wave energy and processes on the beach Overall, wave energy was the important factor shaping sediment distribution and accumulation. At all locations except Waitahanui, wave action was strong enough to produce planar-laminated deposits that also characterize zones above fairweather wave base in many marine and lacustrine settings (Clifton, 1976; Fritz & Howells, 1991). Variations in sediment influx rate did not affect development of swash zones on beaches. The major variant among beach deposits described in this study is the presence of intercalated debris-flow deposits only at low-energy shorelines. Topography, or the extent of reworking, may explain this difference in lithofacies distribution. The presence of debrisflow deposits at rugged, low-energy shorelines would indicate that such areas were prone to mass-wasting processes. Topography and presence of debris-flow deposits around the lake do not correlate in any way, however. For example, debris-flow deposits are well developed at Kuratau, where a long gentle plain meets the lake. It is likely therefore that debris flows were common along the entire shoreline, and that the distribution of their deposits is a function of preservation potential. Storms and perhaps currents related to fairweather waves along shorelines in the high-energy regimes were energetic enough to completely rework debris-flow deposits. In contrast, storm and fairweather wave currents along shorelines in the low-energy regimes were unable to rework the debris-flow deposits, which are relatively thick in comparison with enclosing beds. These deposits were preserved by rapid lake rise in areas protected from storms.
Transgressive surface The characteristics of transgressive surfaces around Lake Taupo depend on the energy and the topography of the environment. The highest-energy
Wave energy and shoreface–offshore sedimentation Sediment availability was apparently an important factor in the degree of development of storm deposits
VSIC07 9/12/08 2:04 PM Page 174
174
N. R. Riggs et al.
at shorelines in the low-energy regimes. In general, storm beds can be destroyed by persistent fairweather wave action. Preservation of storm deposits in the shoreface therefore requires rapid sedimentation rates and /or rapid increase in accommodation. Storm deposits are more common at Kuratau and Waitahanui, where sedimentation rates were higher than at Acacia Bay, Hingapo or Kinloch, which accumulated under conditions of low sediment influx. We conclude that sediment supply was the most important factor in whether or not storms left identifiable deposits in low-energy parts of the Taupo shoreline. This is very different from the situation in high-energy settings and reflects, perhaps, differences in the ability of storms to erode and transport dense lithic sediment from local sources. More energetic storm waves affecting shorelines in the high-energy regimes were able to erode sediment from the shoreline and transport it offshore. This sediment, together with any sediment supplied by rivers, was enough to produce a storm record along nearly all the high-energy shorelines, and rapid burial preserved the record. In contrast, storms along the low-energy shorelines may not have been large or strong enough to erode and transport dense sediment to the depositional sites examined. Typical deepening succession Filling of the post-eruption lake probably took c. 20 yr, or 4 –5 yr to rise from the current lake level to the highstand shoreline (i.e. present areas of exposure). In many places, e.g. Waitahanui and Poukura Pa (Fig. 11), it appears that sedimentation kept pace with lake-level rise and thick successions accumulated but lacked evidence for substantial deepening of the water. Elsewhere, however, sections show continuous deepening. In both cases, it is apparent that, because of the rapidity of lake rise, erosion and reworking of accumulating sediments was minimized. Areas in which sedimentation kept pace with lake rise had substantial sediment input and reasonably steep topography. For example, at Waitahanui, the amount of lithic material in storm beds decreases only slowly upsection over nearly 2 m. This is probably due to the large amount of material debouching from the Waitahanui drainage, which in a stable or fallingwater setting would cause shallowing and progradation. The contrasting situation, in which persistent deepening is represented, is generally associated with gentle shoreline topography. Five Mile Beach is an excellent example of this situation: the sediment input was
clearly substantial, but the accumulating material was dispersed along the broad shoreline. Thus although the section is as much as 5 m thick, a linear deepening of water at the depositional site is indicated by the facies progression. Lake draining Lake draining is inferred to have occurred at a rate of 1.5 m day–1 (Manville et al., 1999). The effects of abrupt lake draining on sedimentation are largely limited to the presence of coarse pumice surface layers at many sites around the lake. These deposits (facies Gmp, Table 1) are clast supported and range from little or no matrix to matrix rich, locally including small areas of matrix support with cross-bedded pumiceous sand between clasts. We interpret these zones of matrix as the result of subaerial reworking that brought in ash and vitric sand to fill the interstices of earlier deposits. Clarkson (1996) interpreted some of the deposits as beach berms. Surficial coarse pumice deposits are found along the northern and north-eastern margins of the lake at heights between ≈ 10 and ≈ 30 m a.l.l. These areas may represent areas where the dominant winds drove pumice rafts aground or held them against the shore during the rapid draining of the lake. The stranded pumice rafts (see White et al., this volume) were then only locally reworked and/or infilled with pumiceous sand.
DISCUSSION AND CONCLUSIONS Lake Taupo was a highly unusual sedimentary system because of the inhomogeneous yet uniform source material throughout the post-eruption refilling of the lake. Records of lake sedimentation in calderas are not common (Nelson, 1967; Newhall et al., 1987; Larsen, 1994; Nelson et al., 1988; Larsen & Crossey, 1996), and to our knowledge, the effect of a dominantly vitric source on sedimentary styles and facies in a major caldera-related lake has not been addressed. Perhaps surprisingly, the foremost characteristic of the Taupo deposits is the lack of lithofacies diversity across a spectrum of depositional energy and sediment influx rates. Although upward-deepening successions were recognized at nearly every site, we found no distinctive lithofacies ‘fingerprint’ related to variations in wave energy or sedimentary input. Instead, wave energy was strong enough to rework sediment at all locations around the lake shore. Very high rates of
VSIC07 9/12/08 2:04 PM Page 175
Sedimentation in a rapidly rising lake sediment influx into the lake, recorded by sedimentation rates on the scale of ≈ 1 m yr –1, were not sufficient to prevent effective reworking and sorting of sediment at the beaches and shoreface. The ability of the lake wave processes to rework eruption-derived sediment in the sites examined is even more noteworthy given the very short (c. 4 –5 yr) period in which the sediment was deposited. In addition, the distinctly different rilling patterns in terrain above and below the highstand (Smith, 1991a; Clarkson, 1996) suggest that most erosion of the countryside occurred during the c. 20 yr of lake rise, not after lake drainage. We believe that the rather uniform development of lithofacies associations is the result of the consistent mixture of sediment delivered to the lake after the 1.8 ka eruption. Shoreline deposits are well developed at all locations because the vitric sediment was very easy to rework, even in the low-energy regimes. Thus, the post-eruption sedimentary record at Lake Taupo is characterized by recognizable lithofacies associations and sequences related to sediment reworking by shoaling and breaking waves. The only immediately identifiable signature left by volcanism in the Lake Taupo lacustrine sediments, apart from the grey rafted pumiceous blocks, is sediment composition; even the huge volumes of sediment delivered into a rapidly enlarging lake had little qualitative effect on shoreline processes. A corollary signature results from the particular composition of this sediment; pumice behaves very differently from other types of sedimentary grains. The very high rates of aggradation recorded with specific lithofacies assemblages, however, are likely to be unique to volcanic lakes; such rates require an abundant supply of sediment and a rapidly rising lake that can accommodate the sediment without rapid shoreline progradation or shoaling. Wave energy and storm periodicity largely controlled the abundance of storm deposits in the Taupo lithofacies associations. As expected, storm layers are more common in deposits from the high-energy settings than in those from the low-energy settings. Sediment supply was also an important variable and, coupled with increasing accommodation resulting from the rising lake, allowed preservation of storm beds above fairweather wave base in low-energy shorelines. Lastly, the phenomenally rapid lake rise and burial of accumulating deposits probably led in great part to the homogeneity of lithofacies. A lake-level rise rate of 5 –9 m yr–1 means that the snapshot of sedimentation preserved is not a view of a mature lake system, but rather of short-term examples of lake processes. Had the lake stabilized for long periods, the effects
175
of energy, sedimentation and topography would be considerably more apparent in preserved lithofacies. In particular, repeated reworking of deposits above fairweather wave base would have greatly lessened the preservation potential of fine-grained or pumice-rich beds; the natural repository of such sediment is below wave base in offshore settings. Deposits in the offshore of the stable modern lake have much better developed low-energy, deep-water characteristics (Nelson & Lister, 1995). Early post-eruptive lacustrine deposits at Taupo suggest that the interaction between volcanism and lacustrine sedimentation at other intracaldera settings will be recorded by variations in sediment composition and accumulation rate, rather than by unique changes in the associations that characterize depositional environments. Mapping of subtle variations in sediment composition and lithofacies, however, can allow reconstruction of palaeoshoreline configurations in ancient accumulations. These extremely rapid accumulation rates are, however, unlikely to be determinable in ancient deposits. The presence of debrisflow deposits and the abundance of storm layers and their position in shoreface versus offshore deposits provide information about wind direction and accumulation in sheltered versus open areas. Lake Taupo deposits differ most strongly from nonvolcanic ones in two ways. First, our study shows that storm activity was a major contributor to the sedimentary record of shoreface deposits. Although storm deposits are fairly distinctive in marine offshore accumulations (Johnson & Baldwin, 1986), few have been identified from shoreface deposits. Their preservation at Taupo is due to a combination of high and persistent sediment influx to the nearshore, and the rapid rise in lake level, which submerged these deposits below wave base before they could be eroded or reworked by waves between storms. Second, the density of pumice and fine vitric material in the Taupo depositional environment varied from 200 to 2400 g m–3 as a result of varying grain size and amounts of water saturation (Tilly, 1987; Clarkson, 1996; Ort et al., 1997; Manville et al., 1998; White et al., this volume). This influenced the characteristics of deposits in irregular ways not mirrored in non-volcanic lakes. For example, suspension deposits in non-volcanic environments are dominated by mud and clay fractions, whereas at Taupo pumice clasts up to pebble size are common in graded to non-graded beds interpreted as suspension deposits of abrasion debris and watersaturated pumice from floating rafts. In shoreface deposits, fine sand- to silt-grade vitric sands, which
VSIC07 9/12/08 2:04 PM Page 176
176
N. R. Riggs et al.
behave like normal quartzofeldspathic sand and silt (see White et al., this volume, for discussion of settling rates of vitric material under conditions of saturation), are interstratified with pumice coarse-sand to granule lenses. In these cases, coarse but low-density pumice was hydrodynamically equivalent to fine sand.
ACKNOWLEDGEMENTS This study was funded by FRST/NZ contract (all authors), Otago University Research Grant (J.D.L.W.), and an Otago University summer bursary (R.C.). We are grateful to those people whose land we trenched and to those who kindly granted us access to exposures. Beth Palmer provided a thorough review of an earlier version of the manuscript and helpful criticism throughout its writing. Very helpful critical reviews of the present manuscript were provided by Peter Ballance, William Fritz, and Vernon Manville.
REFERENCES Clarkson, R.A. (1996) Immediate post-c. 1800a Taupo eruption secondary deposits and shorelines. MSc thesis, University of Otago, Dunedin. Clifton, H.E. (1969) Beach lamination: nature and origin. Mar. Geol., 7, 553 –559. Clifton, H.E. (1976) Wave-formed sedimentary structures; a conceptual model. In: Beach and Nearshore Sedimentation (Eds Davis, R.A.J. & Ethington, R.L.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 24, 126 –148. Clifton, H.E. (1988) Transgressive embayment deposits; another type of coarsening-up sequence in coastal successions. Annual Mid-Year Meeting, Society of Economic Paleontologists and Mineralogists, Abstracts, 5, 11. Clifton, H.E., Hunter, R.E. & Phillips, R.L. (1971) Depositional structures and processes in the non-barred high-energy nearshore. J. sediment. Petrol., 41, 651– 670. Collinson, J.D. & Thompson, D.B. (1982) Sedimentary Structures. George Allen & Unwin, London. Davis, R.A.J. & Ethington, R.L. (Eds) (1976) Beach and Nearshore Sedimentation. Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 24. Electricity Corporation of New Zealand (ECNZ) (1994) Lake Taupo lake levels. Public information brochure. ECNZ, Hamilton. Elliott, T. (1986) Siliciclastic shorelines. In: Sedimentary Environments and Facies (Ed. Reading, H.G.), 2nd edn, pp. 155–188. Blackwell Scientific Publications, Oxford. Fisher, R.V. (1965) Settling velocity of glass shards. Deepsea Res., 12, 345 –353. Fisher, R.V. & Schmincke, H.-U. (1984) Pyroclastic Rocks. Springer-Verlag, Berlin. Fritz, W.J. & Howells, M.F. (1991) A shallow marine volcaniclastic facies model: an example from sedimentary
rocks bounding the subaqueously welded Ordovician Garth Tuff, North Wales, U.K. Sediment. Geol., 74, 217–240. Houghton, B.F., Wilson, C.J.N., McWilliams, M.O., et al. (1995) Chronology and dynamics of a large silicic magmatic system: Central Taupo Volcanic Zone, New Zealand. Geology, 23, 13–16. Johnson, H.D. & Baldwin, C.T. (1986) Shallow siliciclastic seas. In: Sedimentary Environments and Facies (Ed. Reading, H.G.), 2nd edn, pp. 229 –282. Blackwell Scientific Publications, Oxford. Larsen, D. (1994) The Creede Formation and other late Oligocene intracaldera sedimentary sequences in the San Juan Mountains, Colorado. Geol. Soc. Am. Abstr. Prog., 26, 25. Larsen, D. & Crossey, L.J. (1996) Depositional environments and paleolimnology of an ancient caldera lake: Oligocene Creede Formation, Colorado. Geol. Soc. Am. Bull., 108, 526 –544. Larsen, D. & Smith, G.A. (1999) Sublacustrine-fan deposition in the Oligocene Creede Formation, Colorado, USA. J. sediment. Res., 69, 675 – 689. Manville, V. (2000) Sedimentology and history of Lake Reporoa: an ephemeral supra-ignimbrite lake, Taupo Volcanic Zone, New Zealand. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 109 –141. Blackwell Science, Oxford. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1998) The saturation behaviour of pumice and some sedimentological implications. Sediment. Geol., 119, 5 –16. Manville, V., White, J.D.L., Houghton, B.F. & Wilson, C.J.N. (1999) Paleohydrology and sedimentology of a post-1.8 ka breakout flood from intracaldera Lake Taupo, North Island, New Zealand. Geol. Soc. Am. Bull., 111, 1435 –1447. McCubbin, D.G. (1982) Barrier island, strand-plains. In: Sandstone Depositional Environments (Eds Scholle, P.A. & Spearing, D.), Mem. Am. Assoc. petrol. Geol., Tulsa, 31, 247–279. Maill, A.D. (1978) Lithofacies types and vertical profile models in braided river deposits: a summary. In: Fluvial Sedimentology (Ed. A.D. Miall) Mem. Can. Soc. petrol. Geol., Calgary, 5, 597– 604. Nelson, C.H. (1967) Sediments of Crater Lake, Oregon. Geol. Soc. Am. Bull., 78, 833 – 848. Nelson, C.H., Carlson, P.R. & Bacon, C.R. (1988) The Mount Mazama climactic eruption (~6900 yr B.P.) and resulting convulsive sedimentation on the Crater Lake caldera floor, continent, and ocean basin. In: Sedimentologic Consequences of Convulsive Geologic Events (Ed. Clifton, H.E.), Geol. Soc. Am. Spec. Pap., 229, 37–58. Nelson, C.S. & Lister, G.S. (1995) Surficial bottom sediments of Lake Taupo, New Zealand: texture, composition, provenance, and sedimentation rates. N.Z. J. Geol. Geophys., 38, 61–79. Newhall, C.G., Paull, C.K., Bradbury, J.P., et al. (1987) Recent geologic history of Lake Atitlán, a caldera lake in western Guatemala. J. Volcanol. geothermal Res., 33, 81–107. Oehmig, R. & Wallrabe-Adams, H.-J. (1993) Hydrodynamic properties and grain-size characteristics of volcaniclastic
VSIC07 9/12/08 2:04 PM Page 177
Sedimentation in a rapidly rising lake deposits on the mid-Atlantic ridge north of Iceland (Kolbeinsey Ridge). J. sediment. Petrol., 63, 140 –151. Ort, M.H., Riggs, N.R. & Wilson, C.J.N. (1997) Characteristics of pumice-rich intra-caldera sedimentation following the 181 ad eruption of Taupo Caldera. Geol. Soc. Am. Abstr. Prog., 26, 114. Otway, P.M. (1986) Vertical deformation association with Taupo earthquake swarm, June 1983. In: Recent Crustal Movements of the Pacific Region (Eds Reilly, W.I. & Harford, B.E.), R. Soc. N.Z., 24, 187–200. Wellington. Palmer, B.A. & Shawkey, E.P. (1997) Lacustrine sedimentation processes and patterns during effusive and explosive volcanism, Challis volcanic field, Idaho. J. sediment. Res., 67, 154–167. Reineck, H.-E. & Singh, I.B. (1975) Depositional Sedimentary Environments. Springer, New York. Renaut, R.W. & Owen, R.B. (1991) Shore-zone sedimentation and facies in a closed rift lake; the Holocene beach deposits of Lake Bogoria, Kenya. In: Lacustrine Facies Analysis (Eds Anadon, P., Cabrera, L. & Kelts, K.), Spec. Publs int. Assoc. Sediment., No. 13, pp. 175 –195. Blackwell Scientific Publications, Oxford. Salinger, M.J., Daw, G.M., Snelder, T.H., Burgess, S.M., Porteous, A.S. & Belberova, D.Z. (1998) Meteorological hazards in the Taupo district. Report No. AD98079. New Zealand National Institute of Water and Atmospheric Research. Wellington. Smith, R.C.M. (1991a) Landscape response to a major ignimbrite eruption, Taupo Volcanic Centre, New Zealand. In: Sedimentation in Volcanic Settings (Eds Fisher, R.V. & Smith, G.A.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 45, 123 –137. Smith, R.C.M. (1991b) Post-eruption sedimentation on the margin of a caldera lake, Taupo Volcanic Centre, New Zealand. Sediment. Geol., 74, 89 –138. Smith, R.T. & Houghton, B.F. (1995) Vent migration and changing eruptive style during the 1800a Taupo eruption: new evidence from the Hatepe and Rotongaio phreatoplinian ashes. Bull. Volcanol., 57, 432– 439. Smith, G.A. & Smith, R.D. (1985) Specific gravity characteristics of recent volcaniclastic sediment: implications for sorting and grain size analysis. J. Geol., 93, 619 – 622. Thompson, C.S. (1984) The Weather and Climate of the Tongariro Region. Misc. Publ. N.Z. meteorol. Serv., 115 (14). Wellington. Tilly, C.R. (1987) The sedimentology of the Taupo Pumice
177
Alluvium Formation occurring in the lower region of the Hamilton Basin. MSc thesis, University of Waikato. Timperley, M.H. (1983) Climate and hydrology. In: New Zealand Department of Scientific and Industrial Research Information Series (Eds Forsyth, D.J. & HowardWilliams, C.), 158, 55 – 62. Walker, G.P.L., Self, S. & Froggatt, P.C. (1981a) The ground layer of the Taupo ignimbrite: a striking example of sedimentation from a pyroclastic flow. J. Volcanol. geothermal Res., 10, 1–11. Walker, G.P.L., Wilson, C.J.N. & Froggatt, P.C. (1981b) An ignimbrite veneer deposit: the trail-marker of a pyroclastic flow. J. Volcanol. geothermal Res., 9, 409 – 421. Walker, R.G. & Plint, A.G. (1992) Wave- and stormdominated shallow marine systems. In: Facies Models: Response to Sea Level Change (Eds Walker, R.G. & James, N.P.), pp. 219 –238. Geological Association of Canada, St Johns, NF. White, J.D.L., Manville, V., Wilson, C.J.N., Houghton, B.F., Riggs, N.R. & Ort, J.H. (2000) Settling and deposition of 181 A.D. Taupo pumice in lacustrine and associated environments. In: Volcaniclastic Sedimentation in Lacustrine Settings (Eds White, J.D.L. & Riggs, N.R.), Spec. Publs int. Assoc. Sediment., No. 30, pp. 141–151. Blackwell Science, Oxford. Whitham, A.G. & Sparks, R.S.J. (1986) Pumice. Bull. Volcanol., 48, 209 –223. Wilson, C.J.N. (1985) The Taupo eruption, New Zealand. II. The Taupo Ignimbrite. Phil. Trans. R. Soc. London, Ser. A, 314, 229 –310. Wilson, C.J.N. & Walker, G.P.L. (1982) Ignimbrite depositional facies: the anatomy of a pyroclastic flow. J. geol. Soc., London, 139, 581–592. Wilson, C.J.N. & Walker, G.P.L. (1985) The Taupo eruption, New Zealand. I. General aspects. Phil. Trans. R. Soc. London, Ser. A, 314, 199–228. Wilson, C.J.N., Houghton, B.F., McWilliams, M.O., Lanphere, M.A., Weaver, S.D. & Briggs, R.M. (1995) Volcanic and structural evolution of Taupo Volcanic Zone, New Zealand. J. Volcanol. geothermal Res., 238, 1–28. Wilson, C.J.N., Riggs, N.R., Ort, M.H., White, J.D.L. & Houghton, B.F. (1997) An annotated atlas of post-1.8 ka shoreline features at Lake Taupo, New Zealand. Special Report No. 97/19. Institute of Geology and Nuclear Science, Taupo.
VSIC08 9/12/08 2:02 PM Page 179
Spec. Publs int. Ass. Sediment. (2001) 30, 179 –198
Lacustrine–fluvial transitions in a small intermontane valley, Eocene Challis volcanic field, Idaho B . A . P A L M E R * and E . P . S H A W K E Y † *510 Fourth Street East, Northfield, MN 55057, USA: †Missimer International 8410 College Parkway, Suite 202, Fort Myers, FL 33919, USA
ABSTRACT Eocene stratigraphy in the East Fork Salmon River area is characterized by alternating lacustrine and fluvial lithofacies that record abrupt shifts in lakes and rivers with time. Additionally, fluvial deposits are characterized by marked lithofacies diversity. Five systems were distinguished by variations in channel geometry and planform, degree of flood basin development and vegetation, and depositional processes. The rivers included: (i) multichannel, sandy bedload river with wide and shallow channels and frequent lahars; (ii) single-channel gravel bedload river with narrow and deep channels; (iii) sandy bedload, multichannel river characterized by rapid lateral migration and avulsion of the channel belt; (iv) single- or multichannel sandy mixed load river with well-developed, vegetated flood basin; and (v) multichannel gravel bedload river, with wide and deep channels. Lithofacies diversity in the study area indicates that the East Fork depositional system was very sensitive to changes in extrinsic variables, which occurred frequently enough to keep the system close to threshold values for lake development and shoreline position, channel cross-sectional geometry and planform, avulsion and lateral migration, and aggradational and degradational behaviour. Evolution of the East Fork landscape was influenced primarily by volcanic activity, which included lava eruptions and explosive eruptions from a stratovolcano and cauldron complex. Volcanism influenced basin drainage patterns, sediment influx (size and rate), and sediment transport process (lahar versus streamflow). The most dramatic result of eruptions was the formation of a lake in the palaeovalley and changes in lake level. Lava eruptions downstream of the study area, to the south, blocked the palaeodrainage, producing a long-term change in basin palaeophysiography. Apparent lake-level fall resulted when caldera-forming pyroclastic eruptions emplaced thick (to 20 m total) deposits in the lake, causing the shoreline to retreat southward. Volcanism was also the major influence on development of fluvial style for all but the multichannel, gravel bedload river, and the only major episode of laharic activity occurred during a rare period of stratovolcano volcanism north of the study area.
INTRODUCTION accumulated in a small, intermontane valley. Rivers and lakes in this valley were particularly sensitive to variations in sediment supply, base level and discharge. Geomorphological changes related to these variables produced a sedimentary record characterized by alternating fluvial and lacustrine deposits. The Eocene succession is also characterized by an incredible variation in fluvial style. We show how changing eruption style interacted with non-volcanic sedimentation controls to produce the sedimentary succession in the East Fork area. The sedimentary succession in our study area is
Our understanding of how climatic variation and tectonism control fluvial and lacustrine stratigraphy has improved markedly in recent years (e.g. Bridge & Leeder, 1979; Alexander & Leeder, 1987; Blair, 1987; Middleton & Kraus, 1987; Blair & Bilodeau, 1988; DeCelles et al., 1991; Gordon & Heller, 1993; Blair & McPherson, 1994; Bestland, 1997; Bettis & Autin, 1997; Ryang & Chough, 1997). In this paper, we contribute to this growing body of information by adding volcanism to the list of possible sedimentation controls. We describe Eocene volcaniclastic deposits in the area of the East Fork Salmon River (Fig. 1) that
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
179
VSIC08 9/12/08 2:02 PM Page 180
180
B. A. Palmer and E. P. Shawkey
Fig. 1. Palaeogeography of the study area. Locations of the Eocene mountains are taken from Fisher et al. (1992). Fisher et al. used the term cauldron when referring to the entire volcano, and the term caldera when referring to specific landforms related to single eruptions. We retain this usage in this paper. The drainage system of the modern Salmon River is superposed on the palaeogeography to show locations of the study area.
> 350 m thick. In this paper, we focus on intervals that record major changes in depositional setting. These transition intervals include shifts between fluvial and lacustrine environments, as well as variations in sedimentation within the Eocene lake itself.
THE EAST FORK AREA Volcanic history The Challis volcanic field in central Idaho was a major centre of Eocene volcanism in the north-western USA (Moye et al., 1988). In the area of Challis, c. 7 Myr of volcanism produced lava fields and a large cauldron complex (Fig. 1: McIntyre et al., 1982; Fisher et al., 1992). Changes in eruption style make it possible to divide accumulation history into three phases. The first phase began c. 51 Ma when lava erupted from scattered vents (McIntyre et al., 1982; Fisher et al., 1992). The second phase of volcanism was characterized by contemporaneous effusive and caldera-
forming activity. Effusive activity continued until c. 47 Ma, and the first caldera-forming eruptions occurred 48.4 Myr ago with the eruption of the tuff of Ellis Creek (Table 1; McIntyre et al., 1982; Ekren, 1985; Fisher et al., 1992). Eruption of the tuff of Eightmile Creek caused a second collapse event at c. 47 Ma. The third phase of eruption history was characterized by explosive volcanism as activity at the Van Horn Peak cauldron complex continued. The final calderaforming eruption occurred 45 Myr ago with eruption of the tuff of Challis Creek (McIntyre et al., 1982; Ekren, 1985). Basin palaeogeomorphology Palaeophysiography in the Challis area was rugged during Challis volcanism (Ross, 1937; McIntyre et al., 1982; Moye et al., 1988). The study area was a valleyconfined system defined by mountain ranges orientated north–south (Fig. 1; Fisher et al., 1992; Palmer & Shawkey, 1997). The Van Horn Peak cauldron complex north of the study area was part of the watershed
VSIC08 9/12/08 2:02 PM Page 181
181
Lacustrine–fluvial transitions
Table 1. Major pyroclastic-flow deposits in the study area; isotopic dates from McIntyre et al. (1982), Fisher et al. (1992), and Ekren (1985) Unit Tuff of Red Ridge Tuff of Challis Creek
Age
Source
~ 45 Ma. Isotopic age data not available 45.5 + 0.3 to 46.5 + 0.1 Ma
West of the Challis volcanic field Final caldera-forming eruption at the Van Horn Peak cauldron complex, formed the Twin Peaks caldera Van Horn Peak cauldron complex
Tuff of Eightmile Creek
46.9 + 1.6 Ma
Tuff of Ellis Creek
48.2 + 1.4 Ma
First caldera-forming eruption at the Van Horn Peak cauldron complex
Unnamed tuff unit at Road Creek
> 48 Ma. Isotopic age data not available
?
for the river that flowed through the East Fork study area (Palmer, 1997). Although the river valley set the palaeophysiographical framework, volcanic eruptions altered the landscape. Construction of small lava fields ranging from 17 to 35 km2 altered drainage patterns and caldera-forming eruptions produced unusually large sedimentation events that also changed the slope and drainage network in the area. Climate during accumulation of deposits in the study area was temperate and wet enough to support upland forests of Metasequoia (Wing & Greenwood, 1993; Wing & Wolf, 1993). Rivers and lakes in the East Fork area received sediment from a variety of sources. Sediment reworked from pyroclastic deposits is characterized by abundant glass, fractured crystals, crystals with glassy jackets, and large crystal : lithic ratio (Palmer, 1997; Palmer & Shawkey, 1997). Sediment was also eroded from the lava fields and upland areas where Palaeozoic quartzite and granite of the Cretaceous Idaho Batholith were exposed. Sandstone derived from these sources contains less pumice and has smaller crystal : lithic ratio than sandstone derived from reworked pyroclastic deposits. Methods This study is based on 29 sections measured in the Road Creek, Sand Hollow, and Spar Canyon areas near the East Fork Salmon River (Figs 1 & 2). The
Comments Caps ridges in the Sand Hollow area Two distinct flow deposits at Spar Canyon and Sand Hollow, the two units are separated by a reworked interval at Sand Hollow Reworked by slope failure and production of major debris flow in the study area Precursor eruptions recorded by growth of turbidite ramp into the study area; flowed into lake in the study area and transformed into turbidity currents; at least 3 separate surges recorded in total turbidite succession 14 m thick unit at Road Creek; thins to east; erosion by fluvial processes removed unit in eastern part of Road Creek
stratigraphic framework is based on direct correlation between closely spaced sections and marker beds. Sandstone composition was measured by point counting 400 grains per sample. Samples were counted for crystals, volcanic rock fragments (VRF), non-volcanic rock fragments, and glass. Composition of gravelsized material was noted in the field.
SEDIMENTOLOGY Volcaniclastic deposits in this study comprise 14 lithofacies that include diamictite, conglomerate, sandstone, mudrock, tuff and carbonaceous deposits (Table 2). We used lithofacies association, deposit geometry, and vertical sequences to divide the deposits in the transition intervals into six lithofacies assemblages representing accumulation in lacustrine and fluvial environments (Table 3). Key features for distinguishing lacustrine and fluvial assemblages in the Challis area are deposit thickness and lateral extent, and sedimentary features of mudrock. Intervals of lacustrine mudrock are as much as 27 m thick and extend 7.3 km across the study area. Geological mapping in the area (Fisher et al., 1992) and results of our study suggest that thick mudrock intervals can be correlated across the entire Eocene palaeovalley (Fig. 1). The mudrock is characterized by monotonous intervals of claystone or mudstone. In contrast, fluvial mudrock in the study area is coarser
VSIC08 9/12/08 2:03 PM Page 182
182
B. A. Palmer and E. P. Shawkey
Fig. 2. Detailed maps of the Spar Canyon (A), Sand Hollow (B), and Road Creek (C) parts of the study area. Locations of sections measured for this study and the line of each section are shown on the maps.
grained (mudstone and siltstone), thinner, commonly contains abundant plant debris and root marks, and is associated with carbonaceous deposits in many locations. The following discussion emphasizes key sedimentological features and our interpretations. (See Tables 2 & 3 for complete sedimentological descriptions.) Lacustrine deposits Lacustrine deposits in the study area are the mudrock and wedge-sandstone lithofacies assemblages (Table 3). The mudrock assemblage includes the thickest and most laterally extensive deposits in the study area (Palmer & Shawkey, 1997). In contrast, the
wedge-sandstone assemblage is restricted to sections in the western part of Sand Hollow (Fig. 1, Fig. 3). Mudrock assemblage The mudrock assemblage consists predominantly of alternating beds of structureless and laminated mudrock (lithofacies Fl, Fm; Tables 2 & 3). Sandstone interbeds (lithofacies Swr, Sh; Table 2) are thin and rare, making up < 5% of the assemblage. Lateral extent, thickness, unifrmly fine grain size, and presence of interbedded sandstone with wave ripples (Tables 2 & 3) record accumulation from suspension settling in offshore parts of a large lake. Sandstone beds with wave ripples show that this association was, at times,
Rounded, moderately sorted beds of granules to boulders. Matrix of angular, moderately sorted coarse sand. Clasts dominantly volcanic (lava, tuff ), but quartzite and mudrock clasts and intraclasts are present.
Horizontal-bedded (Gh)
Angular, moderately sorted, medium to verycoarse sand. Non-graded or normally graded beds caused by a decrease in the abundance of sand in the very-coarse and coarse size fractions.
Angular, moderately sorted, fine to medium sand. Glass shards in the very-fine sand- to siltsize fractions are concentrated in some beds. Rounded pumice is concentrated in thin cmscale layers. Some beds are normally graded.
Angular, moderately sorted, fine to very-fine sand. Normally graded. Abundant plant debris in some beds.
High-angle cross bedded (St)
Low-angle crossbedded (S1)
Cross-laminated (Sr, Swr)
Sandstone
Rounded, moderately sorted beds of pebbles to boulders. Matrix of angular, moderately sorted coarse sand. Clasts dominantly volcanic (lava, tuff ), but quartzite and mudrock clasts and intraclasts are present.
Matrix supported. Dmg has < 10% clasts and cm-sized clasts. Dmm has as much as 30% clasts with a maximum clast size of 0.5 m. Clasts are subrounded to subangular and all are volcanic in origin (lava, tuff ). Matrix is angular, poorly sorted silt to very coarse sand.
Grain-size distribution and composition
Cross-bedded (Gt)
Conglomerate
Matrix-supported (Dmm, Dmg)
Diamictite
Lithofacies
Table 2. Volcaniclastic lithofacies in the transition intervals
Lacustrine–fluvial transitions continued on p. 184
Sr was deposited by migration of ripples in unidirectional flow and Swr by development of wave currents in standing water.
Scour and fill, low amplitude bedforms.
Sets < 0.3 m thick that overlie shallow scour surfaces. Tabular or lenticular beds 0.1– 0.5 m thick with sharp erosive or nonerosive basal contacts. Upper contacts are sharp or gradational into overlying mudrock. Sr: unidirectional sets 0.5 –2.0 cm thick with sharp basal contacts. Upper contacts sharp or gradational; root marks and bioturbation common. Beds tabular or lenticular and 0.1–2.5 m thick. Swr: two-directional cross strata in the same set. Symmetrical ripple form preserved. Beds 0.1–5.0 cm thick with sharp basal and upper contacts.
Dunes
Bar top and gravel flats.
Lateral accretion beds (epsilon cross strata), bar-front strata associated with transverse bars (planar cross beds), and scour-pool fill (strata filling concave troughs).
Dmm deposited by debris flows and Dmg by dilute debris flows or hyperconcentrated flows.
Interpretation
High-angle sets 0.3 – 0.5 m thick forming lenticular beds to 1.7 m thick. Erosive basal contacts and sharp upper contacts.
Structureless beds 0.2– 0.5 m thick. Basal contacts sharp; upper contacts sharp or gradational with overlying sandstone. Tabular geometry and beds extend as much as 100 m laterally.
Epsilon cross strata to 3.0 m thick drape large erosion surfaces with cross strata that dip to the west. Planar cross strata in single sets dip to the south with sets to 3.5 m thick. Cross strata fill concave-up troughs with sets that dip into the centre of the tough from both sides (in east–west exposure). Basal contacts are erosive. Upper contacts are sharp or gradational into overlying sandstone.
Non-graded and mostly structureless beds, but some Dmg beds have poorly developed horizontal stratification. Fossil wood is abundant in some beds. Basal contacts are non-erosive and planar and upper contacts are sharp. Beds are tabular and range from 0.2–10.0 m thick.
Sedimentary features and bedding
VSIC08 9/12/08 2:03 PM Page 183
183
Angular, moderately sorted, very-fine to fine sand with abundant plant debris. Normally graded.
Bioturbated (Sb)
Accumulation and concentration of organic material in poorly drained areas. Pyroclastic fall from the atmosphere with some postdepositional reworking by organisms.
Most beds thoroughly mixed and lacking recognizable trace fossils; burrows and stratification preserved locally. Root marks common in some beds and absent in others. Beds with root marks also have abundant plant debris. Beds dominantly tabular and < 0.1–1.2 m thick. Basal and upper contacts sharp or gradational. Abundant root marks and recognizable stems. Tabular to discontinuous lenses 2.0 –10.0 cm thick. Sharp or gradational contacts with adjacent mudrock. Mostly structureless, tabular deposits < 1 to 50 cm thick. Some beds burrow mottled or root marked. Many beds continuous over outcrop exposure (5–375 m) with little change in thickness. Basal contacts sharp and upper contacts sharp or gradational.
Clay to silt ± very-fine sand. Claystone, mudstone, and siltstone all common. Silt and sand fractions include minerals, glass shards. Plant debris abundant in some beds and absent in others.
Coal and carbonaceous mudstone and siltstone.
Angular, well-sorted lapilli- to ash-sized material. Most beds combination of glass (individual shards and pumice), and crystals. Lithic grains common in some beds.
Bioturbated (Fm)
Carbonaceous Deposits (C)
Tuff (T)
Rapid settling from suspended load and fallout ash from atmosphere. Reworking by organisms.
Lamina ≤ 1 mm thick and marked by variations in grain size, abundance of organic material, or color. Beds tabular and < 0.1–1.2 m thick. Basal and upper contacts sharp or gradational.
Clay to silt ± very-fine sand. Most beds claystone. Silt and sand fractions include minerals, glass shards. Rare normal grading. Plant debris abundant in some beds and absent in others.
Settling from suspended load and fallout ash from atmosphere.
Sand reworked by insects, roots, and probably animals.
Upper plane bed.
Interpretation
Laminated (Fl)
Common root marks and partially preserved stratification in places. Tabular beds 0.1–1.2 m thick with sharp, non-erosive bases and sharp or gradational tops.
Lamina developed by alternating sand sizes at 1–5 mm scale. Beds tabular or lenticular 0.1– 6.4 m thick with sharp erosive or non-erosive basal contacts. Upper contacts sharp or gradational into overlying mudrock.
Sedimentary features and bedding
184
Mudrock
Angular, poorly to well sorted, very-fine to very-coarse sand ± granules and pebbles. Poorly sorted beds have as much as 48% muddy matrix and abundant glass shards in silt and very fine sand fraction. Non-graded or normally graded. Well-rounded pumice concentrated in layers a few cm thick.
Grain-size distribution and composition
Planar-laminated (Sh)
Lithofacies
Table 2. (continued )
VSIC08 9/12/08 2:03 PM Page 184
B. A. Palmer and E. P. Shawkey
VSIC08 9/12/08 2:03 PM Page 185
185
Lacustrine–fluvial transitions Table 3. Sedimentology of lithofacies assemblages in the transition intervals Assemblage
Major lithofacies
Minor lithofacies
Description
Mudrock
Claystone Fl and Fm
Mudstone Fl and Fm, Swr, T
Thick claystone intervals (3.0 –27 m) correlated 600 m across large outcrops. Intervals also correlated over 7 km across the study area.
Wedge-sandstone
Sh, Sl
St, Sr, Swr Mudstone and siltstone F1 and Fm
Wedge-shaped sandstone body ≈ 600 m in length (east to west). Sandstone body coarsens upward from underlying mudrock assemblage through a 0.5 –1.0-m-thick interval characterized by decreasing proportion of mudstone and siltstone. At most locations, the sandstone body fines upward into overlying mudstone of mudrock assemblage.
Diamictite assemblage
Dmm, Dmg,
Gt, Gh, Sh, Sl, St, Sr, Fm
Sheets of diamictite and interbedded Sh–Sl 15 –21.5 m thick. Sheets extend 1.2 km east–west (outcrop limit). Channel forms in sheets not well defined and have width : thickness ratios of 28 : 1. Lenticular conglomerate bodies separate diamictite sheets at one location. These conglomerate bodies have width : thickness ratios of 5 : 1 to 7 : 1. Conglomerate grades laterally into sandstone lithofacies Sh and Sl.
Conglomerate
Gh, Gt, Sh, Sl
St, Dmg, Sr, Fm
Lenticular conglomerate bodies 1.5 – 6.8 m thick extend outcrop distances of 0.7 km east–west. Conglomerate interbedded assemblage with lithofacies Sh, Sl, and St and each conglomerate body fines upward to lithofacies St. Two conglomerate bodies stacked in the section at Sand Hollow and separated by lithofacies mudstone that thins from 2.0 to 0.4 m thick in less than 100 m
Sheet sandstone
Sh, Sl
St, Sr, Fl, Fm, T
Irregular sandstone sheets 3.8–17.0 m thick with thickness variation caused by marked relief (2–20 m) on underlying assemblage regional erosion surface. Sheets can be correlated over outcrop distance of 1.5 km east–west and over 4 km among outcrops north–south. Width : thickness ratios of sandstone bodies range from 18 : 1 to 42 : 1. Channel forms within sandstone bodies are poorly developed and deposit geometry is markedly tabular over outcrop distances. Sandstone bodies fine upward through a decrease in proportion of coarse sand to gravel sized grains.
Mudrock–sandstone
Sh, Sl, Fm
Dmg, St, Sr, Sb, C, T
Broadly lenticular sandstone bodies 0.7– 4.6 m thick can be correlated distances of 100 m across outcrops. Width : thickness assemblage ratios of sandstone sheets 18 : 1–25 : 1. Sandstone sheets lithofacies St, and low-amplitude lithofacies Sh–Sl. Interbedded tabular intervals of lithofacies Sb–Fb also extend over outcrop distances. Broadly lenticular sandstone bodies make up 26% of the sections (by thickness); tabular intervals make up 74% of the sections.
within wave base. The scarcity of these deposits, however, indicates that wave-driven events energetic enough to transport bed material to offshore parts of the lake were infrequent. Most sedimentation was below wave base. Metre- to millimetre-scale variations in abundance of organic material and pyroclastic material in the lacustrine mudrock (Table 2, lithofacies Fl) record episodic changes in sedimentation rate produced by floods and volcanic activity (Palmer & Shawkey, 1997). Pyroclastic-fall deposits are concentrated in
parts of the succession and record periods when explosive volcanic activity was occurring at the caldera. Wedge-sandstone assemblage The wedge-sandstone assemblage at Sand Hollow is a planar-laminated sandstone body ≈ 600 m long that thins to the east (Fig. 3, sections 1–5; Table 3). Sandstone is well sorted and fine to very fine grained. The planar-laminated sandstone coarsens upward
186
Fig. 3. Stratigraphic cross-section of the Sand Hollow area. Deposits in this area include the fluvial mudrock–sandstone, sheet-sandstone, and conglomerate assemblages. Lacustrine deposits include the mudrock and wedge-sandstone assemblages. Sections in this area include transition surfaces 1–7. Section locations are shown in Fig. 2B.
VSIC08 9/12/08 2:03 PM Page 186
B. A. Palmer and E. P. Shawkey
VSIC08 9/12/08 2:03 PM Page 187
187
Lacustrine–fluvial transitions from underlying lacustrine mudrock and, in most sections, fines upward into overlying lacustrine mudrock. At section 2 (Fig. 3), however, the top of the wedgesandstone body includes a lenticular sandstone that is ≈ 5 m wide. The sandstone is cross-stratified, with high-angle cross-beds (lithofacies St) at the base and cross-laminations (lithofacies Sr) at the top. The association with lacustrine mudrock, coarsening-upward grain-size trend, and abundance of planar lamination record accumulation in a foreshore setting along a prograding shoreline. The fining-upward trend and gradation into overlying mudrock represents the retreat or drowning of the shoreline and reestablishment of offshore sedimentation conditions. Sandy shoreline deposits are not common in the study area, and reconnaissance study suggests that shoreline deposits are rare across the entire area southeast of Challis (Fig. 1). The wedge-sandstone assemblage at Sand Hollow is the only shoreline sandstone body we observed in our study of transition deposits. The small size of the sandstone body suggests that it was a local sand spit that probably developed near the mouth of a stream. The lenticular sandstone at the top of section 2 (Fig. 3) is probably the incised deposit of the creek, and the fine grain size of the sand reflects relatively low energy at the lake shoreline and the nature of bedload transported by the creek. Fluvial deposits Fluvial deposits in the East Fork area display a remarkable amount of variation among the lithofacies assemblages (Table 4), reflecting major differences in sedimentation process through time. In this paper, we focus on the diamictite and conglomerate assemblages. Palmer (1997) described the sheet-sandstone and mudrock–sandstone assemblages in detail.
Diamictite assemblage The diamictite assemblage crops out along Road Creek (Fig. 1) and is the only lithofacies assemblage in the study area with a significant proportion of diamictite (Tables 2–4). In this assemblage, diamictite and sandstone are organized into laterally continuous sheets that extend over 1 km across the outcrop distance of Road Creek (Table 3; Fig. 4). Diamictite (mostly lithofacies Dmg) makes up the basal parts of the sheets. Basal diamictites are large units that thin and thicken along the Road Creek outcrop area (see Fig. 5, below). Thickness changes in the diamictite and underlying erosive surfaces record the old channel forms, which were broad and shallow, with width : thickness ratios of 28 : 1. Channel forms within the sheets are not well developed. Diamictite grades vertically and laterally into sandstone with scour and fill style stratification (lithofacies Sh, Sl). The lateral edges of the sheets also include thinly bedded diamictite (Fig. 4). Separating the diamictite sheets is lenticular conglomerate (Table 2; Fig. 4). The conglomerate is usually structureless or horizontally bedded (lithofacies Gh), but conglomerate with lateral accretion deposits (lithofacies Gt, Table 2) are present locally. Width : thickness ratios of the conglomerate bodies range from 5 : 1 to 7 : 1. These sedimentary bodies are neither as thick nor as laterally extensive as the diamictite sheets, and are restricted to the eastern part of Road Creek (Fig. 4). The diamictite sheets accumulated in wide, shallow channels that migrated across channel belts that were at least 1.2 km wide. The river at this time carried mostly sandy bedload. The large percentage of diamictite in the sheets indicates that most deposition resulted from lahars (hyperconcentrated flows and debris flows). Very large debris flows filled channels
Table 4. Lithofacies composition of fluvial assemblages in the transition intervals*
Assemblage Diamictite Conglomerate Sheet sandstone Mudrock– sandstone
Diamictite (Dmm, Dmg) (%)
Conglomerate (Gh, Gt) (%)
High-angle cross-bedded sandstone (St) (%)
Low-amplitude stratified sandstone (Sh, Sl) (%)
Crosslaminated sandstone (Sr) (%)
Bioturbated sandstone (Sb) (%)
Mudrock (Fl, Fm) (%)
Tuff (T) (%)
75 9 2
1 47 6
6 15 3
15 22 86
1 1 1
0 0 0
2 6 0
† 0 2
2
0
18
36
2
10
24
8
*Percentages are calculated by summing the total thickness of beds’ particular lithofacies in each assemblage. A 14-m-thick pyroclasticflow deposit is present in most sections in the diamictite assemblage. The percentages reported in this table were calculated without this deposit, to more accurately represent the sedimentary lithofacies.
188
Fig. 4. Stratigraphic cross-section of the Road Creek part of the study area. Fluvial deposits in this area are the diamictite assemblage. Lacustrine deposits are the mudrock assemblage. These deposits are separated by transitional surface 1. Section locations are shown in Fig. 2C; scale and explanation of symbols as for Fig. 3.
VSIC08 9/12/08 2:03 PM Page 188
B. A. Palmer and E. P. Shawkey
VSIC08 9/12/08 2:03 PM Page 189
Lacustrine–fluvial transitions and spilled out into areas between channels. The deposits of small debris flows and hyperconcentrated flows were preserved primarily in areas between active channels. In contrast, conglomerate accumulated in a river that was confined to the eastern part of the study area. Channels were deeper and better developed than those that deposited the diamictite sheets. Sediment transport in the gravel bedload system was mostly from normal streamflow processes. Lateral accretion beds indicate that at least parts of the channel system migrated laterally across the flood basin. Lenticular geometry of the deposit, however, indicates that lateral migration was limited. Conglomerate assemblage Deposits of the conglomerate assemblage are thicker, more laterally extensive, and characterized by greater lithofacies diversity than the conglomerate described above (Tables 3 & 4). The conglomerate assemblage is as much as 18.3 m thick and can be correlated over 150 m normal to palaeoflow (east–west, limited by outcrop at Sand Hollow) and over 2 km along palaeoflow between Sand Hollow and Spar Canyon. Deposits in this lithofacies assemblage form two lenticular conglomerate bodies from 1.4 to 6.8 m thick. Internally, the conglomerate bodies are mostly horizontally bedded conglomerate (lithofacies Gh, Tables 2 & 3) with scattered cross-bedded conglomerate (lithofacies Gt). Cross-bedded conglomerate includes isolated sets of planar cross-beds to 3.5 m thick, and strata that fill concave troughs ( Table 2). Cross-bedded sandstone (lithofacies St) and sandstone with low-amplitude stratification ( lithofacies Sh, St) are also interbedded with conglomerate lithofacies. Both conglomerate bodies fine upward through an interval marked by an increasing proportion of sandstone. The best-developed high-angle cross-beds are in the sandy upper portions of the lenticular conglomerate bodies. The two conglomerate bodies are separated by an interval of mudrock (lithofacies Fm) that thins markedly over a distance of < 100 m (Table 3). One very unusual feature of the conglomerate assemblage at Sand Hollow is the presence of an intensely altered sandstone near the base of the lower conglomerate body (sections 7 and 8; see Fig. 3, below). The altered appearance of this sandstone is the result of pervasive replacement of volcanic-rock fragments, pumice, and glass shards by clay minerals. Clay cement and replaced glass are very difficult to distinguish, and the alteration is extensive enough to nearly obliterate sedimentary structures. A similar degree of
189
alteration is not present in any other sandstone we examined for this study. The conglomerate assemblage records development of a river characterized by wide and deep channels with transverse bars. The thickness of cross-beds suggests that water was at least 3.5 m deep in places. The concave trough-fill units probably accumulated in deep scour pools that developed in areas where channels converged (Siegenthaler & Huggenberger, 1993). Planar-bedded intervals accumulated on bar tops and in deep parts of channels where sufficient shear strength was developed to move cobbles and boulders. Gravel was deposited in primary channels, and flow in secondary channels carried a mixture of sand and gravel. Mud accumulated in abandoned channels during floods or in small flood-basin lakes. Fossil wood concentrated in the lower parts of some beds suggests that much cobble and boulder transport occurred during large floods. We believe that the altered sandstone at the base of the section was produced during a period of unusually intense chemical weathering. The sandstone accumulated in an abandoned channel where sedimentation rates were very low. The rate of weathering was greater than sediment influx, and soil began to develop in the sand. Sedimentation in other parts of the channel belt, in contrast, was fast enough to prevent intense alteration of glassy detritus. Sheet-sandstone and mudrock–sandstone assemblages Sheet-sandstone and mudrock–sandstone assemblage deposits were described in detail by Palmer (1997). In this paper, we describe the key features of the assemblages, to provide a sense of how they differ from those described above (Tables 2 & 3; Fig. 5a). The sheet-sandstone assemblage is characterized by lowamplitude cross-stratification and markedly tabular geometry of the beds in sections both normal and parallel to palaeoflow (Tables 3 & 4). Eighty-six per cent of the sections (by thickness) measured in this lithofacies assemblage are planar-laminated and low-angle cross-bedded sandstone, making the sheet-sandstone assemblage the lowest diversity lithofacies assemblage we describe in this study (Table 4). Deposits in the sheet-sandstone assemblage form irregularly shaped bodies 3.8–17.0 m thick that can be correlated over 4 km from Sand Hollow to Spar Canyon (Table 3). Lateral extent in east–west exposures, normal to palaeoflow, ranges from 0.75 to 1.5 km. The fining-upward trend in the sheets is produced by a progressive decrease in the proportion
VSIC08 9/12/08 2:03 PM Page 190
190
B. A. Palmer and E. P. Shawkey
(a)
(b) Fig. 5. Fluvial and lacustrine deposits in the study area. (a) Sheet-sandstone and mudrock–sandstone assemblages at Sand Hollow. The mudrock–sandstone section is ≈ 14 m thick. (b) Surface 3 at Sand Hollow. Offshore mudrock rests directly on tuffs without any intervening evidence of erosion. Jacob staff for scale, with 0.1 m divisions. The transition is just above the 0.3 m mark.
of coarser sand grades and loss of the gravel fraction upward through the sections. Sheets are stacked to form multistorey units. The mudrock–sandstone assemblage, in contrast to the sheet-sandstone assemblage, is characterized by greater lithofacies diversity and smaller sandstone bodies (Tables 3 & 4). In fact, the mudrock–sandstone assemblage has more mudrock (24% of sections by bed thickness) than the other fluvial assemblages (Table 4). Sandstone bodies in the assemblage are 0.7– 4.6 m thick and extend ≈ 100 m east–west, transverse to flow. At most locations, the sandstone bodies do not show any vertical changes in grain size or sequences of sedimentary structures. The bulk of the sandstone has low-amplitude stratification (lithofacies Sh, Sl). The mudrock in this assemblage is bioturbated, root marked, and contains abundant plant material (lithofacies Fm, Table 2). Mudrock and bioturbated sandstone (lithofacies Sb) form sheets that extend over 250 m across outcrops. Sandy bedload rivers deposited the sheet-sandstone and mudrock–sandstone assemblages. The sheetsandstone assemblage accumulated in a multichannel system characterized by wide (≥ 75 m) and shallow channels. Channel formation and abandonment was rapid, producing the sheet geometry that characterizes this fluvial association. Avulsion and partial abandonment of channel tracts produced the fining-upward grain-size trends. The mudrock–sandstone assemblage reflects development of a fluvial network with a smaller channel belt, possibly a single-channel river, which had a vegetated flood basin. Suspended sediment was an
important component of the sediment load, but large width : thickness ratios of the sandstone bodies record channel geometry that was adjusted to transport sandy bed-material load. Channels were, however, deep enough for dunes to develop.
STRATIGRAPHY Sedimentary deposits in the East Fork area are characterized by alternating lacustrine and fluvial deposits. We have divided the section into five informal units based on lithofacies assemblages (Fig. 6, Table 5). Informal units are defined, in most cases, by distinct surfaces that are easily correlated between sections at major outcrop areas (Figs 3 & 4; Table 6). Fluvial to lacustrine surfaces Surfaces 1 and 3 are characterized by juxtaposition of fluvial deposits with offshore mudrock across a sharp surface (Figs 3, 4 & 5b; Table 6). The surfaces are not eroded, nor are shoreline deposits present at any of the 13 sections where surfaces 1 and 3 are exposed. Thus, these surfaces record relatively rapid drowning of the palaeoriver valley, causing a geologically instantaneous rise in base level. Lacustrine to fluvial surfaces Surfaces 2 and 7 are both large-scale erosional surfaces that can be correlated across the study area (Figs 3 & 4; Table 6). Different assemblages, however,
VSIC08 9/12/08 2:03 PM Page 191
191
Lacustrine–fluvial transitions
Fig. 6. Composite stratigraphy of the study area. We divide the sedimentary succession into informal units and show only parts of sections in the transition intervals that are the topic of this paper.
Table 5. Stratigraphy of the East Fork sedimentary succession Deposits
Challis volcanism
Thickness (m)
Lateral Extent
Fluvial3
Third phase: explosive eruptions at Van Horn Peak cauldron complex
18.3
Sand Hollow sections 6 – 8; > 4 km to Spar Canyon sections
Conglomerate assemblage
Lacustrine2
Third phase: explosive eruptions at Van Horn Peak cauldron complex
30
All Sand Hollow sections; > 4 km north to Spar Canyon and an additional 14 km to Malm Gulch area
Mudrock, wedge sandstone assemblages, turbidite sandstone and the tuffs of Ellis and Eightmile Creek
Fluvial2
Second phase: lava and caldera eruptions
15
> 4 km from Sand Hollow to Spar Canyon; 19 km north from Sand Hollow to Bradshaw Basin
Sheet-sandstone and mudrock–sandstone assemblages (Palmer, 1997)
Lacustrine1
First and second phase: lava and caldera-forming eruptions
c. 281
6.3 km from Road Creek to the mouth of Spar Canyon; 7.3 km across Spar Canyon
Mudrock assemblage, turbidites, lava (Palmer and Shawkey, 1997)
Fluvial1
First phase: lava eruptions
53+
1.2 km along Road Creek and possible equivalent units 20 km north at Malm Gulch
Diamictite assemblage and stacked debris-flow deposits (Lukert, 1996)
Informal unit
are present at the transitions. Surface 2 juxtaposes lake-fill deposits related to the tuffs of Ellis Creek and Eightmile Creek with the sheet sandstone (Fig. 4, Table 6). Surface 7 juxtaposes offshore mudrock with the fluvial conglomerate (conglomerate assemblage, Fig. 3, Table 6). The two surfaces record the
re-establishment of rivers in the Eocene study area. Neither of these transitions resembles deposits recording gradual lake fill by sediment. Instead, catchment-wide erosion at or near the end of each lacustrine period appears to have characterized the Eocene East Fork system.
Tuffs of Ellis and Eightmile Creek
Sandstone–mudrock Mudrock
Mudrock
Wedge sandstone
Mudrock
Mudrock
2
3
4
5
6
7
Sharp, non-erosional contact between assemblages; transition surface is exposed 1.7 km east–west (oblique to palaeoflow) at Road Creek and at mouth of Spar Canyon (6.3 km north–south along palaeoflow)
Characteristics of the surface
Conglomerate
Mudrock
Mudrock
Wedge sandstone
Rapid fluvial to lacustrine landscape change
Growth of a small sand spit at the lakeshore along mouth of a creek Gradual drowning of the sand spit and development of offshore conditions Little overall change in depositional system during a period of pyroclastic activity at cauldron complex
Lacustrine to fluvial landscape change preceded by fluvial downcutting
Gradational contact between assemblages; transition surface extends only 0.6 km between sections 1–5 in Sand Hollow Gradational contact between assemblages; transition surface extends only 0.6 km between sections 1–5 in Sand Hollow Sharp but non-erosional contact between mudrock assemblages marked by abrupt change in color due to variations in organic material; lower mudrock is brown and contains abundant twigs and shredded plant material; overlying mudrock is white with rare plant remains and common pyroclastic-fall deposits; outcrop exposure limits identifiable lateral extent of this transition surface to 1 km at Sand Hollow Erosional contact between assemblages; transition surface extends 0.7 km east–west (normal to palaeoflow) across limited outcrop exposure at Sand Hollow; relief on surface is 6 m
Lacustrine to fluvial landscape change preceded by fluvial downcutting
Rapid fluvial to lacustrine landscape change
Interpretation
192
Sharp, non-erosional contact between assemblages; transition surface exposed for 1.7 km east–west (normal to palaeoflow) across the Sand Hollow area; preliminary correlation to deposits at Bradshaw Basin suggests that surface extends at least 19 km north of Sand Hollow
Erosional contact between assemblages; transition surface exposed Sheet sandstone or sandstone–mudrock for c. 3 km east–west (normal to palaeoflow) across outcrops at Spar Canyon and 3.5 km east–west across outcrops at Sand Hollow; north-south (along paleoflow) extent at least 4 km; maximum relief on surface is 2 m at Spar Canyon and 20 m at Sand Hollow
Mudrock
Diamictite
1
Upper assemblage
Lower assemblage
Transition
Table 6. Description of transitions in the East Fork sedimentary succession
VSIC08 9/12/08 2:03 PM Page 192
B. A. Palmer and E. P. Shawkey
VSIC08 9/12/08 2:03 PM Page 193
193
Lacustrine–fluvial transitions Table 7. Sandstone composition in lacustrine and fluvial deposits associated with the transition intervals Assemblage Diamictite Conglomerate Sheet sandstone Mudrock–sandstone Wedge sandstone
Sample number
Crystals (%)
10 3 4 10 11 6
53 55 39 –77 68 9 48
Volcanic rock fragments (%)
Transitions within lacustrine deposits Transitions 4–6 are in lacustrine2 deposits (Fig. 3; Tables 5 & 6). Transitions 4 and 5 differ from all others in that they are gradational and are restricted to a small area in Sand Hollow. These transitions bound shoreline sandstone and offshore mudrock (wedgesandstone and mudrock assemblages) and represent progradation and gradual drowning of a small shoreline system. Transition 6 differs from all the others in that it juxtaposes deposits of the same assemblage (mudrock) across a sharp but non-erosional contact (Table 6). This transition is characterized by changes in sediment composition rather than lithofacies variation. Offshore mudrock below the transition surface is dark brown and contains a large proportion of plant debris when compared with the white offshore mudrock above the surface. Transition 6 represents a rather abrupt change in the nature of sediment delivered to offshore areas of the lake.
LANDSCAPE EVOLUTION The six lithofacies assemblages described above record an accumulation history with marked changes in sedimentation conditions. Lake development was rapid, as were changes in lake level. The East Fork fluvial system was also characterized by change. The four fluvial assemblages represent development of five fluvial systems that were different enough to produce deposits of contrasting lithofacies assemblages and deposit shape. Additionally, the diamictite and sheetsandstone assemblages are low-diversity assemblages that represent development of unusual sedimentation conditions in response to specific environmental factors. First fluvial period (fluvial1): stratovolcano eruptions The diamictite assemblage accumulated during the
43 40 7–24 13 57 36
Non-volcanic fragments (%)
Glass (%)
Crystal : lithic ratio
2 1 0 0 1 2
7 4 16 –53 19 33 13
1.4 0.6 –1.4 2.4 –11 6.3 0.2 1.2
first period of volcanism in the Challis volcanic field (fluvial1 deposits; Fig. 6, Table 5). Lava eruptions and growth of lava fields were well under way when sedimentation in this first river began (McIntyre et al., 1982; Ekren, 1985; Fisher et al., 1992). Crystal : lithic ratio and sandstone composition (Table 7) indicate that sediment was eroded from both pyroclastic deposits and lava-flow fields. Additionally, the diamictite assemblage records alternating periods of fluvial and laharic sedimentation, which are typical of stratovolcanoes (e.g. Vessel & Davies, 1981; Smith, 1987a,b; Palmer & Walton, 1990; Palmer & Neall, 1991; Smith, 1991; Pierson et al., 1992). The diamictite sheets were deposited by frequent lahars during and following eruptive periods. Influx of large amounts of sand-size pyroclastic material led to development of the multichannel, sandy-bedload river. Sedimentation during quiescent periods, in contrast, was mostly by fluvial processes. The lenticular conglomerate accumulated during quiescent periods when lahars were infrequent and the more normal gravel bedload river developed. Incision into the diamictite sheets was in response to declining sediment loads. Preliminary stratigraphic work suggests that the diamictite assemblage at Road Creek is the downstream equivalent of a thick accumulation of debrisflow deposits exposed 18 km north at Malm Gulch (Fig. 1; Table 5). Debris-flow deposits at Malm Gulch include a large proportion of matrix-supported diamictite with clasts in the metre size range. Similar lithofacies assemblages are found 20– 40 km from source at other locations (Palmer & Walton, 1990; Palmer & Neall, 1991). This estimate, however, is a minimum distance because it is derived from lithofacies relationships in unconfined settings. It is likely that lahars travelled farther from their source in the valley setting of the Eocene Challis area. If our interpretation is correct, then the diamictite assemblage at Road Creek accumulated at least 35–60 km downstream of the source volcano.
VSIC08 9/12/08 2:03 PM Page 194
194
B. A. Palmer and E. P. Shawkey
An important feature of the diamictite assemblage is the geometry of the deposits. The only record of deposition during quiescent intervals is the channel-fill conglomerate. Any flood-basin deposits that might have accumulated during this period were apparently removed by erosion. Preservation of only the channel part of the quiescent record is favoured in settings with relatively low subsidence rates (Smith, 1991). Thus, slow subsidence characterized the basin during accumulation of the fluvial1 deposits. First lacustrine period (lacustrine1): effusive and explosive volcanism The abrupt change from fluvial deposits of the diamictite assemblage to offshore mudrock across surface 1 (Figs 4 & 6) represents very rapid drowning of the palaeoriver valley. Growth of the lava field downstream of the study area may have blocked the drainage. Earthquakes and associated fault movement in the study area may also have caused rapid lake development. Local faulting could have created a low area in the study area, and disrupted both the groundwater and surface-water systems. Lakes respond rapidly to fault-related changes in basin physiography, and juxtaposition of lacustrine deposits and subaerial deposits is a common record of basin faulting (e.g. Blair, 1987; Blair & Bilodeau, 1988; Blair & McPherson, 1994). We believe that emplacement of volcanic materials was the primary cause of flooding. Evidence for low subsidence rates during accumulation of the diamictite assemblage does not support subsidence related to tectonism as a flooding mechanism. The lava field downstream of the study area was large enough to obstruct stream flow by blocking the East Fork Valley (Fig. 1). The lava dam was large enough, and subsequent eruptions frequent enough to permit development of a long-lived lake and accumulation of > 280 m of lacustrine deposits (lacustrine1 deposits, Table 4). The end of lacustrine sedimentation coincides with first caldera eruptions at the Van Horn Creek cauldron complex and opening of the second phase of Challis volcanism (Tables 1 & 5). The uppermost deposits in the lacustrine1 unit are part of a lake-filling event that began with growth of a turbidite ramp into the area following precursor Ellis Creek eruptions (Palmer & Shawkey, 1997). Turbidity currents and debris flows related to the Ellis Creek and Eightmile Creek eruptions completed the lake-fill process. Sediment deposited in the study area following these erup-
tions filled a lake that was ≈ 20 m deep in the Sand Hollow area, and pushed the shoreline southward. The erosive event that followed lake fill (surface 2, Figs 3 & 6) can be understood by considering the concept of critical stream power, which is a way of looking at the balance between stream power and the forces that resist erosion (Bull, 1990). Stream power is a function of discharge and slope (Leopold & Bull, 1979; Bull, 1990). Changes in the nature of sediment load can have a marked influence on the forces that resist erosion (Bull, 1990). At equilibrium, stream power and the resisting forces are balanced, and no net aggradation or degradation (erosion) occurs. Once stream power exceeds the resisting forces, net degradation occurs in a river. Possible causes of increased stream power include climatic fluctuations (increased discharge) and tectonism in the drainage basin (increased slope). Slope could also have been increased by local base-level fall downstream of the study area. Finally, a decrease in the sediment supply could have caused erosion by decreasing the resisting forces, pushing the river across the threshold of erosion (Leopold & Bull, 1979). Decrease in sediment supply was unlikely because the degradational event followed a major caldera eruption. Thick pyroclastic deposits in the palaeoriver valley and hillslopes would have provided a large supply of sandy bedload to the river. Degradation following the Eightmile Creek eruption records an unexpected delay in the fluvial aggradation that usually follows explosive eruptions. The geometry of transition surfaces provides evidence needed to identify the cause of the degradational event. Erosion on surface 2 was much greater at Sand Hollow (20 m) than at Spar Canyon (2 m). This difference indicates that the degradational episode began downstream, and that the Spar Canyon area was near the upstream limit of degradation (see Leopold & Bull, 1979; Bull, 1990). Downstream initiation of the degradational event records a downstream control. Variations in discharge and tectonic slope discussed above would have acted on the drainage basin, and produced an erosion event that began in the upper reaches of the catchment. Base-level change downstream of the study area is therefore the most likely cause of the degradational event. Degradation following deposition of thick pyroclastic deposits was a characteristic fluvial response to caldera-forming eruptions in the Eocene study area (Palmer, 1997). We believe that the degradation resulted from complex response to deposition of > 10 m of pyroclastic deposits in the study area (see Schumm, 1973; Womack & Schumm, 1977).
VSIC08 9/12/08 2:03 PM Page 195
Lacustrine–fluvial transitions Second fluvial period (fluvial2): caldera eruptions and rivers Fluvial2 deposits are the sedimentary record of caldera-forming eruptions (Palmer, 1997). Influx of voluminous sandy bedload caused development of a river characterized by rapid channel formation and abandonment, as well as avulsion of the active channel belt (Palmer, 1997). The mudrock–sandstone assemblage, in contrast, accumulated during a period when the landscape was recovering from the eruption. The transition to the mudrock–sandstone assemblage occurred when sediment yield began to decline and suspended sediment became an important part of the total sediment load of the river. The response was development of a more stable channel and flood-basin system. Lack of continuous channel shifting allowed vegetation to successfully colonize the flood basin. The transition from eruption- to quiescent-phase sedimentation was not characterized by degradation, as it was after earlier stratovolcano eruptions, and as has been documented at other stratovolcano settings (e.g. Smith, 1987a,b; Palmer & Neall, 1991). Instead, the sedimentary record records continuous aggradation through the period of recovery, marking a distinct increase in basin subsidence rate. This increase in subsidence rate coincides with graben development in the caldera source area (McIntyre et al., 1982). The aggradational fluvial2 deposits record the onset of tectonic activity in the study area. Valley flooding and the second lacustrine period (lacustrine2) Surface 3 resulted from the second lacustrine flooding episode in the area (Figs 3 & 6). It is possible that flooding was related to continued growth of the downstream lava field. At this time, however, effusive activity in the volcanic field was waning. As discussed above, underlying fluvial2 deposits are the records of accumulation during relatively rapid subsidence. It is likely that tectonism created a low area in the basin, causing the East Fork lake to flood northward. The wedge-sandstone assemblage and transitions 4–6 record variations in depositional conditions within the lake (Figs 3 & 6). Shoreline progradation recorded by the wedge-sandstone could have been caused by increased sediment load related to volcanism, or drought and loss of vegetation cover causing an increase in sediment yield (e.g. Bull, 1991). Alternatively, the creek mouth may have migrated laterally into the study area.
195
In the study area, the compositional signature of explosive eruptions is a large crystal : lithic ratio in basal deposits and a well-developed upward decrease in the ratio that is related to progressive removal of the pyroclastic material from the sediment source area and river valley (Palmer, 1997; Palmer & Shawkey, 1997). This compositional signature is not present in the wedge sandstone. Instead, the wedge sandstone displays little variation in crystal : lithic ratio with vertical position, and the ratios are all low (Fig. 3; Table 7). Progradation was therefore not related to eruptions. Coal in deposits underlying the transition surface records a wet, warm climate. Drought extensive enough to affect upland areas was a possible cause of progradation, but the simplest interpretation of the shoreline deposit was migration of the river mouth in and out of the area. Surface 6 (Figs 3 & 6) reflects increased eruption activity at the Van Horn Peak cauldron, but the impact on the East Fork lake was not dramatic. Sediment size does not change across this transition interval, but the amount of sediment delivered to the lake increased markedly. Increased sediment load effectively diluted the concentration of organic material in the sediment. The abundance of pyroclastic fall deposits in the white mudrock above the surface indicates that the increased sedimentation rate was related to a period of eruptions at the cauldron that blanketed the area in fine ash. During these periods, suspended loads from the stream and rivers draining into the lake, and ash from the atmosphere, increased the sedimentation rate in the lake. The result was intervals of structureless mudrock with very little plant debris. Laminated mudrock, bioturbated intervals, and intervals with abundant plant debris were deposited when sedimentation rates were slower. The eruptions, however, did not produce enough sandy material to change sedimentation patterns in the lake. Understanding the final history of the lake in the study area is a problem because erosion responsible for surface 7 removed the uppermost part of lacustrine2 deposits (Figs 3 & 6). Any evidence of events that led up to the final draining of the lake was destroyed in the erosional event. Lava eruptions ended long before this transition, and it is possible that the reservoir behind the lava dams finally filled. The lava field would have been a local knickpoint on the longitudinal profile of the river. Erosion related to the local base-level fall at the knickpoint would have migrated upstream, eventually triggering a degradational event in the study area that removed the uppermost deposits.
VSIC08 9/12/08 2:03 PM Page 196
196
B. A. Palmer and E. P. Shawkey
Third fluvial period (fluvial3 ): wetter climate and small eruptions The conglomerate assemblage overlying surface 7 records an unusual depositional interval in what is mostly a sandstone succession (Table 5). Development of a gravel bedload stream was an exceptional event in the Eocene study area. Possible explanations are development of a wetter climate, tectonic uplift in the drainage basin, and long-term quiescence at the caldera. As discussed above, wetter climate increases stream power by increasing discharge. Gravel bedload transport can occur if gravel-sized material is present. Janecke (1992) and Janecke & Snee (1993) reported tectonic activity in the Challis volcanic field at about this time. Finally, quiescent periods in mountainous settings are characterized by development of gravel bedload rivers because progressive erosion of sandy pyroclastic deposits and recovery of vegetation cause incision back into gravel bedload sources (e.g. Walton, 1986; Smith, 1991). Key features for interpreting the conglomerate assemblage are lithofacies relationships and crystal : lithic ratio in the fining-upward intervals (range 0.6 –11), size of the cross-beds (3.5 m), size of the channel forms, and the weathered sandstone near the base of the fluvial3 unit (Fig. 3; Tables 3 & 7). Sediment composition of fluvial3 sandstone shows that eruptions did occur during accumulation of this unit (Fig. 3). The sandstone cap in the lower fining-upward sequence has crystal : lithic ratios as large as 11 (range 2.4 –11; Fig. 3), recording a fresh pyroclastic source for this sandstone. Crystal : lithic ratios in the rest of the assemblage range from 0.6 to 1.1.4, recording a mixed pyroclastic–siliciclastic sediment source (Fig. 3). Variation in crystal : lithic ratios and the gradational nature of the change from conglomerate to sandstone within the sequences, however, indicate that avulsion and lateral channel migration are better explanations for the fining-upward sequences than volcanic eruptions. The lower fining-upward sequence probably formed when the river was close to threshold values for avulsion. A flood of sandy material from eruptions at this time would have choked the main channel, prompting its final abandonment. We believe that a combination of reduced volcanic activity and climatic variation best explains the features we describe in the conglomerate assemblage. The wetter climate increased discharge, resulting in a deeper, larger river than existed previously. Additionally, the rate of chemical weathering increased,
permitting deep alteration of sedimentary deposits in areas of slow sedimentation.
CONCLUSIONS Alluvial and lacustrine stratigraphy in the East Fork area is characterized by lithofacies changes that record the abrupt displacement of Eocene lacustrine and fluvial environments. Additionally, fluvial lithofacies are characterized by vertical variations. Five river systems are recorded by the deposits. 1 Multichannel, sandy bedload river with frequent lahars, wide and shallow channels. 2 Single-channel gravel bedload river, narrow and deep channels. 3 Multichannel, sandy bedload river characterized by rapid lateral migration and avulsion of the channel belt. 4 Single- or multichannel sandy mixed load river with well-developed, vegetated flood basin. 5 Multichannel gravel bedload river wide and deep channels. A ‘normal’ fluvial style did not exist in the Eocene study area. The East Fork depositional system was very sensitive to changes in extrinsic variables. These changes occurred frequently enough to keep the system close to threshold values for lake development and shoreline position, channel cross-sectional geometry and planform, avulsion and lateral migration, and gradient adjustment. Volcanic activity, which included lava eruptions and explosive eruptions from a stratovolcano and calderas, was the primary extrinsic variable. Volcanism influenced basin drainage patterns, sediment influx (size and rate), and sediment transport process (lahar versus streamflow). The most dramatic result of eruptions was formation of a lake in the palaeovalley and changes in lake level. Lava eruptions downstream of the study area blocked the palaeodrainage, producing a long-term change in basin palaeophysiography. Major shoreline migration occurred when caldera-forming pyroclastic eruptions emplaced thick (to 20 m total) deposits in the lake, causing the shoreline to retreat southward. Volcanism was also the major influence on development of fluvial style for all but the multichannel, gravel bedload river, and the only major episode of laharic activity occurred during a rare period of stratovolcano volcanism north of the study area. Eruptions too small to cause change in fluvial style or shoreline position left a signature
VSIC08 9/12/08 2:03 PM Page 197
Lacustrine–fluvial transitions either in sandstone composition or stratification style in offshore mudrock. Tectonism and climatic change also influenced accumulation history. Rapid basin subsidence set the stage for overall aggradational behaviour that preserved a remarkable amount of the sedimentary succession. Aggradation was enhanced during periods of explosive volcanism. Finally, a major change to a wetter climate in the latter part of the basin history increased discharge enough for the multichannel gravel bedload river to develop.
ACKNOWLEDGEMENTS This work was funded by the National Science Foundation (Grant EAR-9205190) and is the result of fieldwork by both authors. We wish to thank David Gaylord, Mario Mazzoni, and Nancy Riggs for their informative and helpful comments on the early version of this paper.
REFERENCES Alexander, J. & Leeder, M.R. (1987) Active tectonic control on alluvial architecture. In: Recent Developments in Fluvial Sedimentology (Eds Ethridge, F.G., Flores, R.M. & Harvey, M.D.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 39, 243 –252. Bestland, E.A. (1997) Alluvial terraces and palaeosols as indicators of early Oligocene climate change (John Day Formation, Oregon). J. sediment. Res., 67, 840 – 855. Bettis, E.A., III & Autin, J.W. (1997) Complex response of a midcontinent North American drainage system to late Wisconsinan sedimentation. J. sediment. Res., 67, 740–748. Blair, T.C. (1987) Tectonic and hydrologic controls on cyclic alluvial-fan, fluvial, and lacustrine rift-basin sedimentation, Jurassic–lowermost Cretaceous Todos Santos Formation, Chiapas, Mexico. J. sediment. Petrol., 57, 845 – 862. Blair, T.C. & Bilodeau, W.L. (1988) Development of tectonic cyclothems in rift, pull-apart, and foreland basins. Sedimentary response to episodic uplift. Geology, 16, 517–520. Blair, T.C. & McPherson, J.G. (1994) Historical adjustments by Walker River to lake-level fall over tectonically tilted half-graben floor, Walker Lake Basin, Nevada. Sediment. Geol., 92, 7–16. Bridge, J.S. & Leeder, M.R. (1979) A simulation model of alluvial stratigraphy. Sedimentology, 26, 617– 644. Bull, W.B. (1990) Stream-terrace genesis: implications for soil development. In: Soils and Landscape Evolution (Eds Kneupfer, P. & McFadden, L.), Geomorphology, 3, 351–367. Bull, W.B. (1991) Geomorphic Response to Climate Change. Oxford University Press, Oxford.
197
DeCelles, P.G., Gray, M.B., Ridgeway, K.D., et al. (1991) Controls on synorogenic alluvial-fan architecture, Beartooth Conglomerate (Palaeocene), Wyoming and Montana. Sedimentology, 38, 567–590. Ekren, B. (1985) Eocene cauldron-related volcanic events in the Challis quadrangle. In: Symposium on the Geology and Mineral Deposits of the Challis 1o × 2o Quadrangle, Idaho (Ed. McIntyre, D.H.), Bull. US geol. Surv., Denver, CO, 1658, 43–59. Fisher, F.S., McIntyre, D.H. & Johnson, K.M. (1992) Geologic map of the Challis 1° × 2° Quadrangle, Idaho. US Geol. Surv. Miscellaneous Invest. Ser., Map I-1819, 1 : 250 000 scale. Gordon, I. & Heller, P.L. (1993) Evaluating controls on basinal stratigraphy, Pine Valley, Nevada: implications for syntectonic deposition. Geol. Soc. Am. Bull., 105, 47–55. Janecke, S.U. (1992) Kinematics and timing of three superposed extensional systems, east central Idaho: evidence for an Eocene tectonic extension. Tectonics, 11, 1121–1138. Janecke, S.U. & Snee, L.W. (1993) Timing and episodicity of middle Eocene volcanism and onset of conglomerate deposition, Idaho. J. Geol., 101, 603– 621. Leopold, L.B. & Bull, W.B. (1979) Base level, aggradation, and grade. Proc. Am. Phil. Soc., 123, 168 –202. Lukert, G.M. (1996) Volcaniclastic sedimentation of Malm Gulch and Germer basin, in the Challis volcanic field. MS thesis, University of Idaho, Moscow. McIntyre, D.H., Ekren, E.B. & Hardyman, R.F. (1982) Stratigraphic and structural framework of the Challis volcanics in the eastern half of the Challis 1o × 2o Quadrangle, Idaho. In: Cenozoic Geology of Idaho (Eds Bonnichsen, B. & Breckenridge, R.M.), Bull. Idaho Bur. Mines Geol., 26, 3 –22. Middleton, L.T. & Kraus, M.J. (1987) Contrasting architecture of two alluvial suites in different structural settings. In: Recent Developments in Fluvial Sedimentology (Eds Ethridge, F.G., Flores, R.M. & Harvey, M.D.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 39, 253 –262. Moye, F.J., Hackett, W.R., Blakley, J.D. & Snider, L.G. (1988) Regional geologic setting and volcanic stratigraphy of the Challis volcanic field, central Idaho. In: Guidebook for the Geology of Central and Southern Idaho (Eds Link, P.K. & Hackett, W.R.), Bull. Idaho Geol. Surv., 27, 87–97. Palmer, B.A. (1997) Sedimentary record of caldera-forming eruptions, Eocene Challis volcanic field, Idaho. Geol. Soc. Am. Bull., 109, 242 –252. Palmer, B.A. & Neall, V.E. (1991) Contrasting lithofacies architecture in ring-plain deposits related to edifice construction and destruction, the Quaternary Stratford and Opunake formations, Egmont Volcano, New Zealand. In: Volcaniclastic Sedimentation (Eds Cas, R. & Busby-Spera, C.), Sediment. Geol., 74, 71– 88. Palmer, B.A. & Shawkey, E.P. (1997) Lacustrine sedimentation processes and patterns during effusive and explosive volcanism, Challis volcanic field, Idaho. J. sediment. Res., 67, 154 –167. Palmer, B.A. & Walton, A.W. (1990) Accumulation of volcaniclastic aprons in the Mount Dutton Formation (Oligocene–Miocene), Marysvale volcanic field, Utah. Geol. Soc. Am. Bull., 102, 734 –748. Pierson, T.C., Janda, R.J., Umbal, J.V. & Daag, A.S.
VSIC08 9/12/08 2:03 PM Page 198
198
B. A. Palmer and E. P. Shawkey
(1992) Immediate and long-term hazards from lahars and excess sedimentation in rivers draining Mt. Pinatubo, Philippines. US geol. Surv. Water Resour. Invest. Rep. 92-4039. Ross, C.P. (1937) Geology and Ore Deposits of the Bayhorse Region, Custer County, Idaho. Bull. US Geol. Surv., Denver, CO, 877. Ryang, W.H. & Chough, S.K. (1997) Sequential development of alluvial / lacustrine system: southeastern Eumsung basin (Cretaceous), Korea. J. sediment. Res., 67, 274 –285. Schumm, S.A. (1973) Geomorphic thresholds and complex response of drainage systems. In: Fluvial Geomorphology (Ed. Morisawa, M.). SUNY, Binghamton, NY. Siegenthaler, C. & Huggenberger, P. (1993) Pleistocene Rhine gravel: deposits of a braided river system with dominant pool preservations. In: Braided Rivers (Eds Best, J.L. & Bristow, C.S.), Spec. Publ. geol. Soc., London 75, 147–162. Smith, G.A. (1987a) Sedimentology of volcanism-induced aggradation in fluvial basins: examples from the Pacific Northwest, U.S.A. In: Recent Developments in Fluvial Sedimentology (Eds Ethridge, F.G., Flores, R.M. & Harvey, M.D.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 39, 217–228. Smith, G.A. (1987b) The influence of explosive volcanism on
fluvial sedimentation: the Deschutes Formation (Neogene) in central Oregon. J. sediment. Petrol., 57, 613 – 629. Smith, G.A. (1991) Facies sequences and geometries in continental volcaniclastic sediments. In: Sedimentation in Volcanic Settings (Eds Smith, G.A. & Fisher, R.V.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 45, 109 –121. Vessel, R.K. & Davies, D.K. (1981) Nonmarine sedimentation in an active forearc basin. In: Recent and Ancient Nonmarine Depositional Environments: Models for Exploration (Eds Ethridge, F.G. & Flores, R.M.), Spec. Publ. Soc. econ. Palaeont. Miner., Tulsa, 31, 31– 48. Walton, A.W. (1986) Effect of Oligocene volcanism on sedimentation in the Trans-Pecos volcanic field of Texas. Geol. Soc. Am. Bull., 97, 1192–1207. Wing, W.L. & Greenwood, D.R. (1993) Fossils and fossil climate: the case for equable continental interiors in the Eocene. Phil. Trans. R. Soc. London, Ser. B, 34, 243–252. Wing, W.L. & Wolfe, J.A. (1993) Stable isotope study of fluid inclusions in fluorite from Idaho: implications for continental climates during the Eocene, comment. Geology, 21, 1051. Womack, W.R. & Schumm, S.A. (1977) Terraces of Douglas Creek, northwestern Colorado: an example of episodic erosion. Geology, 5, 72–76.
VSIC09 9/12/08 2:02 PM Page 199
Spec. Publs int. Ass. Sediment. (2001) 30, 199 –222
Volcanic and hydrothermal influences on middle Eocene lacustrine sedimentary deposits, Republic Basin, northern Washington, USA D . R . G A Y L O R D * , S . M . P R I C E † and J . D . S U Y D A M ‡ *Department of Geology, Washington State University, Pullman, WA 99164-2812, USA; †Great Basin Gold, Reno, NV 89510, USA; ‡2617 Custer Avenue, Billings, MT 59102, USA
ABSTRACT Volcanogenic sedimentary deposits are common in regions of highly extended crust and often accumulate in extensional basins formed on top of uplifted metamorphic core complexes. An ≈ 1.1 km-thick succession of middle Eocene strata from the upper Sanpoil Volcanics and lower Klondike Mountain Formation in the Republic Basin, Okanogan Highlands, northern Washington, records the transition from active volcanism and hydrothermal activity to dominantly lacustrine sedimentation in a highly extended region. Rapid subsidence along basin-margin faults in upper Sanpoil Volcanics time led to development of an asymmetric graben in which a deep, thermally stratified lake formed. Strata from the upper Sanpoil Volcanics reflect mixed volcanic, hydrothermal, and sedimentary deposition as andesitic and dacitic volcanism waned. Klondike Mountain Formation strata are characterized by three facies associations that reflect the changing influence of volcanism on the region: (i) association A (hydrothermally influenced lacustrine basin margin) consists of complexly interstratified hydrothermal eruption breccia and volcanogenic sedimentary conglomerate deposited primarily from cohesive and cohesionless sediment gravity flows, and sandstone and mudstone deposited from turbidity currents and suspension; (ii) association B (lacustrine basin plain) is composed of interstratified mudstone and thin, fine-grained sandstone deposited by distal turbidity currents and suspension; strata include abundant plant, insect and fish fossils and occasional volcanic ash beds; (iii) association C (Gilbert-type delta) consists of progradational deltaic successions composed of interstratified, pebble–boulder conglomerate and sandstone deposited primarily from linked cohesionless sediment gravity flows and turbidity currents. Overall, facies associations B and C form a single coarsening-upward megasequence. Volcanogenic lake sedimentation appears to have been favoured in this highly extended region because of a combination of rapid basin subsidence, moist climatic conditions, an abundant supply of loose volcanic detritus, and the topographically elevated and isolated nature of the Okanogan Highlands.
INTRODUCTION basins that developed following late Mesozoic and early Cenozoic crustal overthickening (e.g. Coney & Harms, 1984; Wernicke et al., 1987; Burchfiel et al., 1992; Varsek & Cook, 1994). It should be noted that the terminology ‘volcanogenic sediments’ used in this paper follows the definition of McPhie et al. (1993, pp. 94–97). In sum, ‘volcanogenic sediments’ are those derived primarily from erosion and reworking of preexisting volcanic deposits. The bulk of the middle Eocene sediments preserved within the grabens and half-grabens of the Okanogan Highlands of Washington accumulated in the standing waters of lakes. These lakes and their sedimentary deposits were strongly
Volcanogenic sedimentary deposits are common in many parts of the North American Cordillera that underwent rapid Cenozoic crustal extension (e.g. Howard & John, 1987; Nielson & Beratan, 1990; Beratan, 1991; Dickinson, 1991; Fedo & Miller, 1992; Suydam & Gaylord, 1997). These deposits accumulated in a variety of continental depositional settings including alluvial fans, streams, and lakes that preserve a sensitive record of upper crustal responses to extensional tectonism and volcanism. In the Okanogan Highlands of northern Washington and southern British Columbia, volcanogenic sediments accumulated in middle Eocene extensional
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
199
VSIC09 9/12/08 2:02 PM Page 200
200
D. R. Gaylord et al. 119
118
Toro da C Gra reek ben
49
N
Republic
Okanogan Metamorphic Core Complex
Kettle Metamorphic Core Complex
Columbia River
Lake
Roo
seve
lt
Re
pu
blic
Gr
ab
en
Study area
48 0
10
20
30
km
WA Middle Ecocene Klondike Mountain Formation Eccene volcanic and sedimentary rocks (undifferentiated) Mesozoic–Cenozoic plutonic rocks Low-grade metamorphic rocks (Permian–Triassic protolith) High-grade metamorphic rocks (Palaeozoic–Cenozoic protolith)
Fig. 1. Generalized geological map of the Okanogan Highlands of northern Washington showing relation of the Republic Basin study area to major tectonic, lithological, and physiographic features. ( Note major detachment faults (solid rectangles appear on downthrown sides of faults) along western and eastern margins of the Okanogan and Kettle metamorphic core complexes, respectively.)
VSIC09 9/12/08 2:02 PM Page 201
Republic Basin, Washington, USA influenced by volcanism and hydrothermal activity that provided ready sources for easily disaggregated and transported volcanogenic detritus. Sedimentary deposits from these lakes provide evidence of the tectonic and palaeogeographical evolution of the Okanogan Highlands and highlight the complementary relationship that volcanism, hydrothermal activity, and extensional tectonism had with lacustrine deposition and preservation. This paper focuses on the depositional history of the sedimentary-dominated portions of the middle Eocene upper Sanpoil Volcanics and lower Klondike Mountain Formation in the Republic Basin of northern Washington. The Republic Basin is the southernmost of two dominantly lacustrine basins that developed along the western margin of the Republic graben (Fig. 1), a prominent structural depression that cuts the Okanogan Highlands, a highly extended region of high- and low-grade metamorphic rocks, and volcanic and sedimentary strata (Muessig, 1962, 1967; Fox et al., 1976; Pearson & Obradovich, 1977; Fox & Rinehart, 1988; Holder & Holder, 1988; Parrish et al., 1988; Holder et al., 1990; Price, 1991). Development of the Republic Basin (as well as the larger Republic graben) coincided with metamorphiccore-complex uplift that raised the basin (and graben) to elevations ≈ 2 km greater than present-day heights of 1–1.5 km above sea level (Wolfe et al., 1998). As elevations increased, the potential for erosion of sedimentary deposits in the basins likewise increased. Continued downfaulting of lake and adjacent sedimentary deposits and subsequent burial of these sediments beneath lava flows, however, promoted their longer-term preservation. The research presented here provides evidence for and examples of volcanic and hydrothermal influences on lake sedimentation; a revised sedimentary and stratigraphic context for an unusually diverse and well-preserved assemblage of plant macrofossils, insect, and fish fossils previously identified from the Klondike Mountain Formation lake strata; and insight into the palaeogeographical, tectonic, and volcanic evolution of a rapidly extending terrain.
GEOLOGICAL SETTING The Republic graben is the largest of a series of north-north-east-trending extensional basins in the Okanogan Highlands (Fig. 1). The graben is ≈ 20 km wide and 100 km long, is bounded by the Bacon Creek and Scatter Creek normal faults on the west and the
201
Sherman normal fault on the east, and bisects the Okanogan and Kettle metamorphic core complexes. The Bacon Creek, Scatter Creek, and Sherman faults have surface dips from 65° to nearly vertical (Parker & Calkins, 1964; Muessig, 1967; Moye, 1984). A single splay of the Sherman fault, however, dips at ≈ 40° (Parker & Calkins, 1964) and the dip of the Scatter Creek fault decreases systematically to ≈ 35° near the southern end of the graben (Staatz, 1964). The Okanogan Highlands are situated at the southern terminus of the Omineca crystalline belt, a regional tectonic zone that straddles the boundary between allochthonous terranes and cratonic North America. Uplift of metamorphic core complex rocks that surround the Republic graben was accompanied by significant and rapid extension along crustal-scale detachment faults that delimit the eastern and western margins of the Okanogan Highlands (Price, 1979; Price et al., 1981; Harms, 1982; Brown & Journeay, 1987; Orr & Cheney, 1987; Hansen & Goodge, 1988; Parrish et al., 1988; Harms & Price, 1992; Varsek & Cook, 1994). These metamorphic core complexes are structurally similar to, but c. 30 Myr older than, metamorphic core complexes in the southern Basin and Range (Coney, 1980; Armstrong, 1982). The Okanogan Highlands consist of four major lithological elements: 1 high-grade metamorphic rocks (lower-plate rocks of metamorphic core complexes); 2 Permo-Triassic low-grade, allochthonous metamorphic rocks (upper-plate rocks of metamorphic core complexes); 3 Mesozoic and Tertiary calc-alkaline plutonic rocks (primarily the Colville batholith of Holder & Holder, 1988); and 4 Tertiary volcanic and sedimentary rocks (including those of the Republic Basin) (Fig. 1). The sedimentary–tectonic evolution of the Republic graben has been the subject of some debate. Cheney (1994) and Cheney & Rasmussen (1996) regarded the Tertiary stratified deposits in the Republic graben as the segmented and translocated remains of a oncecontinuous, regional-scale basin. Such a basin may have existed at least in part during deposition of the O’Brien Creek Formation, the oldest stratified Tertiary unit in the graben (Matthews & Gaylord, 1994; Matthews, 1997). Sedimentary and stratigraphic evidence, however, does not support such an interpretation for upper Sanpoil Volcanics and Klondike Mountain strata deposited either in the Republic graben (this paper) or in the Toroda Creek half-graben to the west (Suydam & Gaylord, 1997). Instead, the evidence
VSIC09 9/12/08 2:02 PM Page 202
202
D. R. Gaylord et al.
indicates that the graben geometry and /or internal basin features largely controlled patterns of sedimentation, a conclusion also reached by Holder (1990) for the Sanpoil Volcanics. In her study, Holder (1990) noted that both the localization of hypabyssal feeders along the graben margin as well as thickening of lava flows into the graben demonstrated that a pre-existing, and then synchronously developing, Republic graben largely controlled the distribution of that unit. The original margins as well as the structural character of the Republic basin within the Republic graben are somewhat obscure, as a result of significant postdepositional erosion and faulting. In spite of this various lines of sedimentary and stratigraphic evidence, including facies trends and distributions, isopach maps, and (rare) palaeocurrent trends, indicate that much of the original basin is intact. As described in this paper, infilling of the Republic Basin proceeded in a manner that is roughly similar to half-graben models described by Blair (1987), Leeder & Gawthorpe (1987), Blair & Bilodeau (1988), and Mack & Seager (1990). However, in an important departure from a strictly halfgraben (trap-door) genesis, the Republic Basin has structural attributes that more closely resemble that of an asymmetric graben. Republic graben strata have been the focus of numerous geological investigations over the years because of exploration and mining for precious metals hosted in both Tertiary and upper Palaeozoic to early Mesozoic rocks (Umpleby, 1910; Lindgren & Bancroft, 1914; Muessig, 1967; Full & Grantham, 1968; Braun, 1989; Tschauder, 1989; Fifarek et al., 1996; Lasmanis, 1996). The Republic mining district (an area largely corresponding to the Republic Basin) has been a major gold- and silver-producing area with relatively continuous activity since 1896. Continuous sections of drill core acquired during mineral exploration provided sedimentary and stratigraphic details essential to the analyses presented here.
Basin (Matthews & Gaylord, 1994; Matthews, 1997). The stratigraphy of the upper Sanpoil Volcanics and Klondike Mountain Formation is summarized in Fig. 2. Sanpoil Volcanics strata consist of andesitic and dacitic flows with intercalated fine- to coarsegrained volcanogenic sedimentary rocks. Sanpoil Volcanics in the graben have yielded K–Ar ages of c. 48–53 Ma; however, most ages fall between c. 50 and 51 Ma (Pearson & Obradovich, 1977; Stoffel et al., 1991). The Klondike Mountain Formation consists of lower, sedimentary-dominated and upper, lava-flowdominated strata deposited between c. 48 and 49 Ma (Stoffel et al., 1991; Berger & Snee, 1992). Together, the Sanpoil Volcanics and Klondike Mountain Formation have a composite thickness that exceeds 3 km (Muessig, 1967; Full & Grantham, 1968). The contact between these units is generally comformable within the basin (Gaylord, 1986) although it rarely is exposed. Apart from Quaternary glacial deposits, strata of these two units dominate the surficial geology in the basin (Fig. 3).
SEDIMENTOLOGY AND STRATIGRAPHY OF THE REPUBLIC BASIN Background The middle Eocene volcanic and sedimentary rocks deposited in the Republic graben are (from oldest to youngest) the O’Brien Creek Formation, the Sanpoil Volcanics, and the Klondike Mountain Formation. The O’Brien Creek Formation lies beneath the deepest extent of drilling within the Republic Basin and probably pre-dates development of the Republic
Fig. 2. Composite stratigraphic section of uppermost Sanpoil Volcanics and Klondike Mountain Formation in the Republic Basin. Facies associations and inferred depositional settings are indicated to right of column. Data derived primarily from logging of > 12 000 m of drill core (from > 40 drill holes), 2000 m of exposed section, and reconnaissance mapping.
VSIC09 9/12/08 2:02 PM Page 203
203
Republic Basin, Washington, USA
Tsv
N
Klondike Mountain Formation
U 22
D
Tkb
U
Tkc
ree kF
Ba co
nC
Tkv
5
9 25
Middle Eocene
aul t
DH-1
A
DH-5 DH-6
B'
Facies association C
Tkb
Facies association B
Tka
Facies association A
Tsv
Sanpoil volcanics
Thc
Herron Creek Suite quartz monzonite
unt ke
Mo
4
20
U
U
D
Klo
Tkv
DH-4
Tsv
ndi
D
tem
ys
lt S
au
Tsv
aF
rek
ain
Eu
DH-3
Fa
35
B
D
Tkc
ult
DH-2
Tkb U
Upper volcanics (aphyric lava flows)
18
D
Thc
Tkv
A' D U
0 Republic
0
1 1
2 km 2 miles
Scale
Fig. 3. Generalized geological map of Republic Basin highlighting distributions of facies associations of the Klondike Mountain Formation (described in text), other major lithological units and structural features (modified from Muessig, 1967; Price, 1991); balls on downthrown sides of faults). Lines of N–S and E–W cross-sections (Figs 6 & 7, respectively) with DH (drill-hole) locations are superimposed on the map.
Sedimentary and volcanic facies Sanpoil Volcanics and Klondike Mountain Formation sedimentary and volcanic facies were categorized using a scheme modified from Miall (1978), Rust (1978), Johnson & Baldwin (1986), and Smith (1987) to facilitate interpretation of transport and depositional histories (summarized in Fig. 4). Limited outcrop and drill core from the upper Sanpoil strata prohibited a similar detailed treatment of facies and facies associations to that possible for the Klondike Mountain Formation strata. Sanpoil Volcanics The Sanpoil Volcanics (uppermost strata summarized in Fig. 5) consist primarily of high-K, calc-alkaline andesite and dacite lavas (Holder, 1990; Morris & Hooper, 1997). Most flow rocks are porphyritic with a
lithoidal to glassy groundmass of quartz and feldspar containing phenocrysts of plagioclase, hornblende, biotite and pyroxene (Moye, 1984; Holder, 1990). Wide distribution and thicknesses > 2200 m make the Sanpoil Volcanics the most voluminous lithostratigraphic unit in the Republic graben and the Republic Basin. Sanpoil Volcanics strata have been subdivided into informal lower and upper subunits on the basis of mineralogy and geochemistry (Holder et al., 1990). Porphyritic andesitic and dacitic Sanpoil Volcanics deposits in the Republic Basin have distinctive fragmental textures (primarily autobreccia and peperite). Autobreccia, composed of porphyritic lithoidal fragments set in a phenocryst-deficient, fine-grained lithoidal or glassy volcanic matrix is the most common textural variant. Autobreccia clasts have irregular, blocky shapes and commonly display jigsaw-puzzle fabrics. Autobreccias are best developed at the bases and tops of flow units, but can be pervasive throughout
VSIC09 9/12/08 2:02 PM Page 204
204
D. R. Gaylord et al.
Fig. 4. Simplified classification of upper Sanpoil Volcanics and Klondike Mountain Formation sedimentary and volcanic facies. Symbols apply to all stratigraphic sections and cross-sections used elsewhere in paper.
entire flows, and commonly grade into non-fragmental flow rocks. Sedimentary interbeds in the Sanpoil Volcanics are clearly subordinate to lava beds except in the upper 200 m of the formation, where an assortment of gravel-, sand-, and mud-dominated facies are concentrated. Finer-grained sand- and mud-rich deposits often contain organic-rich and coalified fragments as well as authigenic pyrite. These sedimentary beds are decimetres to metres thick, laminated to massive, ungraded to normally graded, and locally convolutely deformed, brecciated, and /or cut by small-scale faults. Beds of massive to laminated mudstone often separate ungraded and normally graded sandstone beds; flame and ball-and-pillow structures are common. Peperitic breccia, with clasts of porphyritic lava set in a matrix of mud, or mud and sand, is common in the uppermost, sedimentary-dominated portions of the Sanpoil Volcanics. Peperites are best developed along the margins of flow units where they are interstratified with fine- to coarse-grained lacustrine deposits. The
sedimentary matrix is black, siliceous, and pyritic within these breccia bodies. Hypabyssal feeders to the Sanpoil Volcanics, known as the Scatter Creek rhyodacite (Muessig, 1967), have been identified within and adjacent to the Republic Basin. Full & Grantham (1968) described bodies of Scatter Creek rhyodacite in underground mine workings along the Eureka fault (near the western basin margin). Mineral foliations and fracture sets in andesite and dacite coplanar with the Klondike Mountain fault on the eastern margin of the basin suggest that volcanic feeder systems were active along this margin too. West of the Bacon Creek fault (footwall of the Republic graben), plutons of both Devils Elbow granodiorite and Herron Creek quartz monzonite flank the Republic Basin; these plutons are deeper feeders to the Sanpoil Volcanics (Holder et al., 1990). Interpretation. Sedimentary deposits intercalated with upper Sanpoil lava beds are interpreted as primarily lacustrine in origin, based largely on the predomin-
VSIC09 9/12/08 2:02 PM Page 205
Republic Basin, Washington, USA
205
Strata across the contact between the Sanpoil Volcanics and Klondike Mountain Formation in the Republic Basin reflect a change from a prolonged interval of volcanism to one marked at first by limited hydrothermal activity and then by widespread sedimentation (followed by increasing, although limited, hydrothermal activity and then by increasing sedimentation). Peperitic breccias and textures in the upper Sanpoil flows are consistent with features generated by lavas that either intruded and/or flowed over soft, water-saturated sediment (Kokelaar, 1982). The concentration of peperite and hydrothermally altered volcanic and sedimentary strata at this contact and along the Bacon Creek and Eureka faults suggests a rough coincidence of volcanism, sedimentation, and basin subsidence. Klondike Mountain Formation
Fig. 5. Columnar stratigraphic section of typical sedimentary and volcanic facies from upper Sanpoil Volcanics in the Republic Basin. (Refer to Fig. 4 for explanation of symbols.)
ance of mud-rich, massive to laminated suspensioncurrent and normally graded turbidity-current deposits. The abundance of organic and coalified material and pyrite strongly suggests stratified, probably meromictic conditions within the lake. As the rate of subsidence increased along the Bacon Creek and Eureka faults (relative to that of the concurrently active Klondike Mountain fault) near the end of Sanpoil time, the western margin of the basin became a preferred locus for fine-grained lacustrine sedimentation in a similar manner to that depicted in half-graben infilling models (Blair, 1987; Leeder & Gawthorpe, 1987; Blair & Bilodeau, 1988; Mack & Seager, 1990).
Background. Klondike Mountain Formation strata were first formally described and named by Muessig (1962, 1967), who subdivided the unit into one formal and two informal members. The lower, fine-grained sedimentary rocks of the Klondike Mountain Formation were grouped into and formally named the Tom Thumb Tuff Member, essentially the ‘lake beds’ of Umpleby (1910) and Lindgren & Bancroft (1914). The rest of the formation was subdivided into the informally named middle and upper ‘members’ (Muessig, 1962, 1967). Subsequent workers (Gaylord et al., 1990; Price, 1991; Suydam, 1993; Gaylord et al., 1996; this paper) have adopted an informal two-tiered system for the Klondike Mountain Formation, dividing the unit into lower (sedimentary-dominated) and upper (volcanic-dominated) members (Fig. 2). The lower member includes fine-grained lake deposits of the Tom Thumb Tuff Member as well as coarsegrained alluvial–deltaic sandstone and conglomerate of Muessig’s (1962, 1967) informal middle member. Upper member volcanic-dominated deposits consist of dark, aphyric andesitic to rhyolitic flow rocks (Wagoner, 1992; Morris & Hooper, 1997) that unconformably overlie sedimentary strata of the lower member and correspond to Muessig’s upper member. Klondike Mountain Formation strata in the Republic graben are largely restricted to small basins near Republic (Fig. 1) and Curlew, ≈ 30 km to the north. The composite stratigraphic thickness of the Klondike Mountain Formation in the Republic Basin exceeds 0.9 km (Fig. 2) and consists almost entirely of detrital volcanogenic sedimentary mudstone, sandstone, breccia, and conglomerate derived from erosion
VSIC09 9/12/08 2:02 PM Page 206
206
D. R. Gaylord et al.
Fig. 6. N–S cross-section of the upper Sanpoil Volcanics and lower Klondike Mountain Formation showing stratigraphic relations between Sanpoil Volcanics facies and facies associations A, B, and C. (Refer to Fig. 3 for drill-hole locations and Fig. 4 for explanation of symbols.) Vertical exaggeration ≈ 1.9 ×.
Fig. 7. E–W cross-section of the upper Sanpoil Volcanics and lower Klondike Mountain Formation showing stratigraphic relations between Sanpoil Volcanics facies and facies associations A, B, and C. (Refer to Fig. 3 for drill-hole locations and Fig. 4 for explanation of symbols.) No vertical exaggeration.
of the Sanpoil Volcanics (Gaylord, 1986; Price, 1991). Klondike Mountain Formation detrital clasts also include chalcedonic and quartz vein fragments, hot springs sinter, silicified and /or argillized volcanic and sedimentary fragments, and rare crystalline plutonic grains. The spatial distribution of the Klondike Mountain Formation strata in both the Republic and Curlew basins suggests that much of this unit accumulated in restricted (< 200 km2) volcano-tectonic depressions (Gaylord et al., 1990, 1996; Price, 1991; Suydam & Gaylord, 1991; Suydam, 1993). Plant fossil assemblages indicate that the climate was humid and temperate throughout Klondike Mountain Formation accumulation (Wolfe & Wehr, 1987, 1991). Physiognomic evidence from fossil leaves collected from Klondike Mountain Formation strata in the Republic Basin
indicates that elevations were ≈ 2 km higher in middle Eocene times than at present (Wolfe et al., 1998). Sedimentary, volcanic, and hydrothermal facies and facies associations Klondike Mountain Formation sedimentary and volcanic facies were grouped into facies associations A, B, and C. Outcrop patterns of these three facies associations appear on the Republic Basin geological map (Fig. 3). Generalized cross-sections (Figs 6 & 7) reveal facies associations of variable thickness that generally progress stratigraphically from facies association A upwards through facies association C. Volcanism and related hydrothermal activity influenced the character of upper Sanpoil Volcanics and Klondike Mountain Formation sedimentary strata as
VSIC09 9/12/08 2:02 PM Page 207
Republic Basin, Washington, USA
207
evidenced by their compositions, textures, and stratigraphic relations. Primary volcanic rocks are most abundant in the lower portions of the section and in the ‘upper member’ rocks that cap the Klondike Mountain Formation. Facies association A: hydrothermally influenced lacustrine basin margin Facies association A consists of up to 200 m of mixed volcanogenic sedimentary and volcanic deposits. The sedimentary component of this facies association consists primarily of gravel-dominated beds with subordinate heterolithic mud- and sand-dominated strata (Fig. 8). Sedimentary deposits are interstratified with relatively thin (< 15 m) beds of autobrecciated, locally peperitic, andesitic and dacitic flow rocks. Deposits of facies association A vary in thickness along the N–S cross-section that roughly parallels the Bacon Creek and Eureka faults (Figs 3 & 6), thinning over a Sanpoil Volcanics topographic high and reaching a maximum thickness in a topographic low. Viewed along the E–W cross-section (Fig. 7), facies association A thickens to the west toward the Bacon Creek and Eureka faults and thins to the east. Gravel facies are compositionally and texturally immature and consist largely of matrix- to clastsupported pebble to boulder conglomerate and breccia. Detrital particles are composed primarily of argillized and, less commonly, silicified andesite and dacite. Other detritus includes quartz vein and chalcedony clasts, coal fragments, and silicified and/or argillized mudstone and sandstone clasts (Fig. 9a & b). Matrix-supported gravel beds are 0.5–12 m thick, contain angular to subrounded clasts, are massive (unstratified), and ungraded to distribution and coarse-tail normally graded with occasional inversely graded bases. Matrix consists of poorly sorted sand, silt, and clay. Maximum clast sizes range from a few centimetres to 2 m in diameter and tend to vary directly with bed thickness. A limited number of laterally continuous, matrix-supported breccia bodies thin radially away from (underlying) vertically orientated conduits of argillically altered to silicified pebble to cobble breccia. These breccia bodies are interstratified with other breccia bodies, thin coal lenses and beds, and sandstone and mudstone beds and lenses containing macerated and often coalified organic matter. Clast-supported gravel facies are generally thinner (0.3 –7.5 m) than the matrix-supported gravels, often have subrounded to rounded clasts, are massive, ungraded, and display distribution and coarse-tail normal
Fig. 8. Columnar stratigraphic section of typical facies association A deposits. (Refer to Fig. 4 for explanation of symbols.)
and occasionally inverse grading (Fig. 8). Inverse grading generally occurs in the lower halves of beds whereas the upper halves characteristically are massive or normally graded. Interstitial matrix is generally sand enriched and silt and clay depleted. Granule- and pebble-sized, clast-supported gravels commonly occupy the basal few centimetres or decimetres of normally graded sand beds whereas cobble-sized, clast-supported gravel facies tend to occur in thicker (> 3 m) beds that are interstratified with sand- and/or mud-rich facies. Heterolithic facies are characterized by thinly and thickly interlaminated to thinly interbedded massive and normally graded couplets of mud- and sand-dominated strata. Mud-dominated strata usually consist of laminated and, less commonly, massive, unstratified
VSIC09 9/12/08 2:02 PM Page 208
208
D. R. Gaylord et al.
(b)
(a) Fig. 9. Typical gravel facies from facies association A. (a) Matrix-supported, massive pebble–boulder breccia (of hydrothermal eruption–debris flow origin) with incorporated angular pebble-sized quartz vein clasts (light-coloured clasts immediately below and to left of tip of hammer (exposed length of head of hammer is 11 cm) ). Darker clasts are composed of variably altered and silicified andesite and dacite (derived from the Sanpoil Volcanics). (b) Matrix- and clast-supported pebble – cobble breccia (of hydrothermal eruption– debris flow origin) with silicified, laminated, mud-dominated heterolithic facies clasts. Other clasts include subrounded, white (altered) and darker (relatively unaltered) andesite and dacite. Matrix is relatively clay and silt depleted and sand enriched.
mudstone. Occasionally, thin (< 5 cm) beds of devitrified ash and white clay are intercalated with the muddominated strata. Sand-dominated heterolithic facies are commonly normally graded and occasionally massive, but also display horizontal lamination, inverse grading, and rare ripple cross-lamination. Sand particles typically are angular to subangular, very fine to very coarse grained, and poorly to moderately sorted. Cosets of sand- and mud-dominated heterolithic facies are decimetres to metres thick and commonly deformed. Load and deformation features include flame and ball-and-pillow structures, microfaults, microbrecciation, and, less commonly, convolute strata. Other facies association A deposits include lava flow, dark, siliceous, pyritic mudstone, and accretionary lapilli. Lava flow occurs in a few discontinuous, 3–15 mthick beds of autobrecciated porphyritic andesite and dacite with peperitic textures that overlie highly deformed and locally baked, fine-grained sedimentary deposits. Black, strongly siliceous and pyritic mudstone beds are intercalated with heterolithic (interstratified sandstone and mudstone) facies near two of the breccia conduits. Spherical to subspherical, 2–5 mm-diameter accretionary lapilli with thin white clay rims were observed in breccia and massive mudstone. Strata in facies association A are concentrated near the contact of the Sanpoil Volcanics and Klondike Mountain Formation and along the hanging walls of the Eureka and Bacon Creek faults (Fig. 3). These strata tend to thin away from these faults to the north
and east and are in gradational contact with underlying Sanpoil Volcanics flow rocks and the overlying strata of facies association B. Interpretation. The compositions and distributions of sedimentary and volcanic facies indicate that facies association A accumulated primarily along the margins of a lake, in an area subject to hydrothermal alteration and, periodically, eruptive hydrothermal activity. The ubiquity of lacustrine mudstones and turbidites throughout the association demonstrates that the bulk of deposition was subaqueous. Silicified tree trunks (Full & Grantham, 1968) and thin coal beds near the base of the association are evidence for vegetated shorelines and marshes. Fluvial deposits, originally if present, are either relatively thin or absent as a result of erosion. Sedimentary detritus was transported to the lake from the footwalls of the Bacon and Eureka faults and from hydrothermal vents and craters located along the shoreline or submerged within the lake. Sediment was introduced into the lake by grain-enriched debris flows and turbidity currents. Deposits from these sediment gravity flows became interstratified with fine-grained suspension and tephra fall accumulations. The humid temperate palaeoclimate (Wolfe & Wehr, 1987, 1991), abundance of preserved organic matter and pyrite, and absence of any evidence for surface exposure and weathering suggest that the lake was perennial, stratified, and probably meromictic.
VSIC09 9/12/08 2:02 PM Page 209
Republic Basin, Washington, USA The generally massive (unstratified), ungraded to distribution and coarse-tail normally graded, matrixsupported gravel beds are interpreted as cohesive debris flow deposits similar to those described from modern volcanic terrains (e.g. Scott, 1988; Scott et al., 1995; Vallance & Scott, 1997). The massive deposits could have resulted from either en masse freezing (Johnson, 1970; Middleton, 1970), or incremental accretion of internally massive to graded and ungraded flows (Scott, 1988; Vallance, 1994; Scott et al., 1995; Major, 1996; Vallance & Scott, 1997), or both. The inferred cohesive character of the flows is attributed to the relatively high concentrations of clay generated by the hydrothermal alteration of country rock including the Sanpoil Volcanics. Common interstratification of these debris flow deposits with laminated, fine-grained lacustrine sedimentary strata indicates that the flows maintained their cohesive character subaqueously, a transport phenomenon noted elsewhere (Lowe, 1982; Scott et al., 1995; Mohrig et al., 1998). Intercalation of coalified carbonaceous lenses and beds as well as the prominence of pyrite in the lake strata indicates that deposition often occurred within a chemically isolated anoxic hypolimnion. The relatively sand-enriched, mud-poor and massive (unstratified) to inversely and normally graded, clast-rich gravel beds in this facies association have textural and structural characteristics consistent with deposition from cohesionless, granular sediment gravity flows. Mechanisms that may have collectively played roles in particle support and segregation include hindered settling, grain–grain collisions, matrix buoyancy, dispersive pressures, turbulence (Lowe, 1982), and kinetic sieving (Middleton, 1970; Vallance, 1994). Deposition of these non-cohesive flows occurred both subaqueously (dominant mode of deposition) and subaerially (relatively minor). Sand-enriched, mud-depleted conglomerate with similar textural characteristics and internal grading has been ascribed elsewhere to en masse freezing and /or collapse of subaqueous gravelly ‘traction carpets’ that develop at the bases of high-density turbidity currents (Lowe, 1982; Postma et al., 1988; Falk & Dorsey, 1998). Deposition of such sediment need not have been entirely en masse, however. As described elsewhere by Scott & Vallance (1995) and Vallance & Scott (1997), it also is plausible that these conglomerate beds could have accumulated incrementally from longitudinally segregated flows. The normally graded sand and heterolithic sand beds in our study may represent deposition from overlying low-density turbidity currents.
209
The heterolithic mud-dominated facies are interpreted as the finer-grained and more distal portions of turbidity currents. Massive and laminated mudstone beds also may have had a genetic relation to turbidity currents, but they also may reflect seasonal suspension deposition in relatively quiet waters subject to occasional overturn (Sturm & Matter, 1978). The abundant flame and ball-and-pillow structures in the heterolithic and mud-dominated facies reflects the influence of rapid loading of the surface from various sources including turbidity currents. Slumping, sliding, and intraformational folding in these strata probably reflect failure of inclined depositional surfaces (Woodcock, 1979; Coleman & Prior, 1982). In the few cases where deformed mud-rich strata are juxtaposed with peperites and andestic flow and sill rocks, a volcanic mechanism for deformation is apparent. Thin beds of volcanic ash in some of these facies hint at active volcanism in the region, but no definitive geochemical linkages to source vents have been identified. The accretionary lapilli were probably produced during local hydrothermal eruptive events. A hydrothermal eruption or debris flow origin for the matrix-supported breccia bodies (Fig. 9a & b) that contain abundant silicified sedimentary and volcanic detritus is supported by the close spatial association of these bodies with deeper epithermal vein rocks, the radial thinning of these deposits away from breccia conduits (vents), and the common incorporation of accretionary lapilli in the deposits. Further, isopach mapping of overlying lake deposits revealed a collapsed and filled circular depression 700 m in diameter and 50 m deep that was centrally located atop a breccia conduit. That depression is interpreted as a former hydrothermal eruption crater. Such collapse features and brecciated silicified beds are common products of hydrothermal eruptions in modern hydrothermal fields (White, 1955; Lloyd, 1959; Cross, 1963; Muffler et al., 1971; Nairn et al., 1979; Nairn & Wiradiradja, 1980; Hedenquist & Henley, 1985). Similar features and deposits also have been associated with ancient epithermal gold–silver deposits (Berger & Eimon, 1983; Sillitoe et al., 1984; Jennings, 1990; Rytuba et al., 1990; Rytuba & Vander Meulen, 1991; Fifarek et al., 1996). In the Republic Basin, breccia bodies interstratified with massive to laminated, carbonaceous mudstone are interpreted as lake deposits. Distribution and coarse-tail normal grading in these breccia bodies probably reflects variable incorporation of lake water into the eruption-produced debris flows. Massive, ungraded breccia bodies are interpreted primarily as
VSIC09 9/12/08 2:02 PM Page 210
210
D. R. Gaylord et al.
subaerial hydrothermal eruption deposits, on the basis of their generally massive character and lack of either interstratification with or incorporation of finer-grained lake sediment. Stratiform, silica-rich ‘sinter’ deposits have also been described from Klondike Mountain Formation strata (Braun, 1989; Tschauder, 1989; Price, 1991; Fifarek et al., 1996). Classic, thinly banded fragments of chalcedonic sinter are common in some of the hydrothermal eruption breccia bodies (Fifarek et al., 1996). Very crudely stratified chalcedonic deposits also are present and commonly are overlain by coal lenses (Braun, 1989; Price, 1991). Banded sinter forms above the water table, whereas the crudely stratified chalcedony is consistent with deposition within sinter pools located slightly below the water table (Rytuba & Vander Meulen, 1991). Coalified carbonaceous matter overlying the stratiform silica probably is remnant plant matter that either grew in or was transported into sinter pools as they cooled or became inactive. The vent-proximal, black, siliceous, pyritic mudstone in facies association A may have been generated at interfaces between exhalative hydrothermal fluids (rich in silica, sulphur, and iron) and water-saturated, organic-rich lake sediment. The abundance of hydrothermally altered sand and gravel clasts within the facies association A detritus suggests a local provenance for these coarser-grained sediments, especially given the hydrothermally altered character of the Eureka and Bacon Creek fault zones. By contrast, the unaltered nature of clasts within intercalated mud-rich sedimentary strata points to their derivation from upland areas not affected by hydrothermal activity. Suspended, unaltered mud was widely distributed across the lake and periodically deposited, probably during overturn events.
Fig. 10. Columnar stratigraphic section of typical facies association B deposits. (Refer to Fig. 4 for explanation of symbols.)
Facies association B: lacustrine basin plain deposits Facies association B deposits are up to 250 m thick. Fossiliferous, interlaminated to interbedded mudstone, siltstone, and very fine sandstone facies make up over 75% of the association (Fig. 10). In general, strata in this facies association coarsen and become more sand enriched up-section. Laminated couplets of sandstone and mudstone preserve abundant plant macrofossils (Wolfe & Wehr, 1987, 1991) (Fig. 11a), fish fossils (Wilson, 1996), and insect fossils (Lewis, 1992) (Fig. 11b), as well as macerated and coalified organic debris. The best preservation of fossils occurs in thin mudstone and siltstone laminations that cap normally graded, thickly laminated to thinly bedded,
very fine- to medium-grained sandstone (heterolithic sand-dominated facies). Organic matter also is common in the normally graded sand of this facies, but usually occurs as macerated organic fragments. Other association B facies include normally graded couplets of sandstone and mudstone, and massive, ungraded sandstone. Sand grains are typically angular to subangular and unaltered. Deposits of this facies association display no distinct trends along the N–S cross-section (Fig. 6) but do thicken significantly towards the Bacon Creek and Eureka faults in the E–W cross-section (Fig. 7).
VSIC09 9/12/08 2:02 PM Page 211
Republic Basin, Washington, USA
211
(a)
(b) Fig. 11. Examples of preserved fossils in lake basin deposits. (a) Metasequoia sp. plant fossil preserved in a laminated siltstone that caps a normally graded, medium–fine-grained sandstone from facies association B; coin is 2.1 cm in diameter. (b) Insect fossil preserved on siltstone lamination capping normally graded, medium–fine-grained sandstone from facies association B.
Interlaminated mudstone and siltstone facies typically are organic rich (Fig. 12a), and brown to black except where they contain intercalated very thin to medium beds of light grey–white volcanic ash. Mudstone and siltstone strata are interlaminated on the scale of millimetres and contacts between laminae are generally sharp. Microfaulting is common and usually shows normal displacement. Flame and ball-andpillow structures and convolute strata also are common (Fig. 12b & c), especially where mud-dominated deposits are interstratified with sand-rich beds. Sandstone beds make up < 25% of this facies association, generally are < 20 cm thick, fine –medium grained, normally graded to occasionally ungraded, horizontally laminated, and rarely ripple cross-laminated. Organic detritus is usually aligned with bedding whether in sand- or in silt-rich strata. Four unusually thick (0.5– 5 m), normally graded sandstones form the bases of up to 15-m-thick fining-upward successions capped by silt- and mud-rich deposits. These normally graded successions thicken towards the western side of the basin. Interpretation. The character and distribution of facies within facies association B as well as relations with the other facies associations indicate that facies association B accumulated on a lacustrine basin plain, in a position relatively isolated from all except the largest clastic discharge events. The dominance of laminated mudstone–siltstone couplets is evidence for semiregular release of fine-grained sediment that was temporarily suspended in the thermocline (Sturm & Matter, 1978). Stratification and overturn in temperate climates generally follow seasonal patterns of temperature change
(Sturm & Matter, 1978). Associations of plant fossils collected from the Klondike Mountain Formation in and near the study area suggest mean annual temperatures of 10–11°C and mean annual ranges of 5°C (Wolfe & Wehr, 1991), conditions that probably were favourable for seasonal stratification and at least occasional overturn. Variable thicknesses of mudstone laminations indicate that stratification cycles were complex (Sturm & Matter, 1978), perhaps owing to local climate perturbations and the superposed effects of local tectonic and volcanic–hydrothermal activity. Abundant authigenic pyrite and coalified, allochthonous plant detritus reveal that lake water mixing often was incomplete, leaving uncirculated, anoxic bottom waters (Horne & Goldman, 1994). The generally unaltered, andesitic character of detritus within facies association B indicates that hydrothermally altered Sanpoil Volcanics source rocks were not significant contributors to Klondike Mountain Formation sediment as was the case for the sediments of facies association A. Strata containing intact plant, insect, and fish fossils were deposited in the lake basin by a combination of processes. Normally graded sand at the bases of the heterolithic sand-dominated couplets accumulated from the distal portions of turbidity currents. The thin, fossil-bearing mud and silt laminations reflect deposition from suspension. Given the wealth of whole fossils in these strata, scour by turbidity currents was not adequate to destroy intact plant, insect, and fish remains. The unusually thick, graded sandstone–mudstone beds are interpreted as the products of large turbidity currents that flowed towards the deepest parts of the
VSIC09 9/12/08 2:02 PM Page 212
212
D. R. Gaylord et al.
(a)
(b)
(c) Fig. 12. Representative examples of lake basin deposits from facies association B. (a) Thinly laminated, carbonaceous, lake basin mudstone. Laminations reflect distal turbidity current and suspension deposition in deep waters. (b) Flame and ball-and-pillow structures in interstratified fine sandstone and mudstone deposited by distal portions of turbidity currents. (c) Convolutely deformed interstratified fine – coarse-grained sandstone and mudstone deposited by proximal to distal turbidity currents along the coarser-grained margins of the lake basin. These deposits grade into the generally coarser-grained deposits of facies association C.
VSIC09 9/12/08 2:02 PM Page 213
Republic Basin, Washington, USA
213
Fig. 13. Columnar stratigraphic sections of typical facies association C deposits. (a) Gravel-dominated, more proximal portion of Gilberttype delta (topset and foreset beds). (b) Sand-dominated, more distal portion of Gilbert-type delta (lower foreset and bottomset beds). (Refer to Fig. 4 for explanation of symbols.)
basin. The relative rarity and great thicknesses of these units suggest they were the products of infrequent and catastrophic earthquake, flood, or volcanic events. Volcanic activity during this time is best represented by the thin ash laminations and beds that are traceable across the lake basin. Too little geochemical evidence exists to pinpoint the eruptive centres responsible for these ashes. The presence of accretionary lapilli in correlative lake deposits in the Curlew Basin (50 km to the north), however, suggests that active eruptive centres were relatively near. The coarsening-upward trend observed within the sedimentary deposits of facies association B reflects the gradation of this association with the coarser-
grained deposits of facies association C. The finergrained character of facies association B also reflects the generally distal nature of these deposits relative to those of facies association C. Facies association C: Gilbert-type delta Facies association C is volumetrically the most abundant of the three facies associations, locally exceeding thicknesses of 700 m. Strata within this facies association form the upper portions of a coarsening-upward megasequence that consists dominantly of sand- and gravel-dominated facies (Figs 7 & 13). Facies association C coarsens and thickens to the east and generally
VSIC09 9/12/08 2:02 PM Page 214
214
D. R. Gaylord et al.
(a)
(c)
(b) Fig. 14. Representative facies from Gilbert-delta type deposits, facies association C. (a) Crudely stratified pebble – cobble conglomerate foreset bed with sand-rich and mud-depleted matrix interstratified with thin massive to normally graded sandstone. Deposition occurred on upper portions of delta front where accumulation was primarily from cohesionless debris flows; sediment contributions from turbidity currents were secondary here. (b) Amalgamated, normally graded sandstone beds deposited along lower delta front; deposition was primarily from turbidity currents. It should be noted that each 10–15-cm-thick bed has granulitic particles (at base) overlain by plane-laminated, fine to very fine sand (in upper half ) and thin massive mud (at top). Hammer is 32 cm long. (c) Massive to crudely stratified, turbidity current-generated, coarse- to very coarse-grained, granular sand deposited on lower portion of delta front. Sandstone is encased between organic-rich, partially coalified, deformed, locally brecciated and microfaulted finer-grained strata.
grades into strata of facies association B down-section and to the west. Facies association C consists primarily of Sanpoil Volcanics clasts that largely were unaltered by hydrothermal activity. Gravel-dominated facies within facies association C consist primarily of 0.5 – 8-m-thick beds of mixed matrix- and clast-supported pebble–boulder conglomerate (Figs 13a & 14a) and mixed breccia– conglomerate that have a combined thickness > 500 m. Matrix consists of mud-deficient, moderately sorted, coarse to very coarse granular sand. Conglomerate beds are
massive to crudely stratified and tabular and commonly exhibit normal and inverse grading; basal contacts are generally planar. Gravel-dominated deposits are concentrated in horizontal to subhorizontal beds as well as cross-beds in the upper portions of the stratigraphic succession. The horizontal and subhorizontal beds unconformably truncate very large-scale crossstrata (foresets) that dip from 15 to 20° to the west (Fig. 15). Foreset dips decrease down-section and grade into finer-grained, subhorizontally and horizontally stratified deposits of facies association B. Gravel-rich
VSIC09 9/12/08 2:02 PM Page 215
Republic Basin, Washington, USA
215
Fig. 15. Typical outcrop of Klondike Mountain Formation strata showing Gilbert-type delta foresets (left and centre) and overlying topset bed (right side) from upper part of facies association C. View is to WNW across Republic Basin looking towards Okanogan metamorphic core complex on the far horizon. Angular discordance between the topset beds at right edge of photograph and foreset beds that dips ≈ 15° west should be noted. Topset beds are nearly flat-lying (in spite of the distorted perspective provided by the photograph).
foresets are interstratified with thin, centimetres- to decimetres-thick beds of ungraded, inversely graded, and, commonly, normally graded sand and locally, granule-rich, plane-stratified sandstone. The sand-dominated deposits of facies association C (Fig. 14b & c) consist dominantly of normally graded sandstone beds with intercalated beds of inversely graded, massive, and crudely stratified sandstone. The combined thickness of sand-dominated beds in this association locally exceeds 200 m. Sandstone beds also contain thin lenses of horizontally laminated silty sand, flattened carbonaceous debris, lenses of clast-supported, normally graded granular gravel, and rarely, ripple cross-laminated sand. Mudstone intraclasts are relatively common near the contacts with mud-rich heterolithic units but otherwise are rare. Soft-sediment deformation is common in the lower portions of the association. Heterolithic facies within facies association C are concentrated near the base of the association and consist primarily of thinly interbedded mudstone and sandstone intercalated with up to 1-m-thick, normally graded mudstone-capped sandstone beds. Unlike similar units in facies association A, these heterolithic facies contain no gravel-rich beds. Common smallscale deformation features include microfaults, and flame and ball-and-pillow structures. Interpretation. The gravel- and sand-rich sediments of facies association C are interpreted as the coarsergrained topset and foreset deposits of Gilbert-type deltas (Gilbert, 1885, 1890; Nemec, 1990) and possibly Gilbert-type fan deltas that prograded into the lake.
The scale of limited field exposures and comparisons with Gilbert-type deltas documented elsewhere (e.g. Colella, 1988; Kazanci, 1988; Dorsey et al., 1995) leads us to believe that facies association C consists of several stacked Gilbert-type deltas that were part of the overall coarsening-upward trend. The largest of these deltas were fed by streams that drained the eastern portions of the Republic graben and crossed the Klondike Mountain fault. Surface exposures reveal that at least one large Gilbert-type delta prograded into the lake; foreset geometries indicate that waters were at least 100 m deep (Fig. 15). Basins with steep topographic margins provide especially favourable conditions for Gilbert-type delta development (Colella, 1988; Leeder et al., 1988; Prior & Bornhold, 1988). Thus, the tectonically active Republic Basin was well suited for the generation of these deposits. The overall wedge-shaped geometry produced by the deltaic progradation is revealed in the E–W cross-section (Fig. 7). Activity on the Klondike Mountain fault ceased near the end of facies association C accumulation, as evidenced by the unfaulted gravel topset beds that overlie fault-displaced, gravelly delta deposits. Sedimentary and stratigraphic evidence indicates that the Gilbert-type delta deposits in this association were deposited primarily by cohesionless gravel-rich sediment gravity flows that transformed downslope into turbidity currents. Similar bipartite successions of strata composed of relatively thick ungraded and graded gravel and intervening normally graded and horizontally laminated sand have been explained as clast-rich debris flows that transformed into high- and
VSIC09 9/12/08 2:02 PM Page 216
216
D. R. Gaylord et al.
low-density turbidity currents (Postma et al., 1988; Falk & Dorsey, 1998). Deposition from the cohesionless debris flows probably occurred via two mechanisms: en masse freezing (Johnson, 1970; Middleton, 1970) and incremental accretion (Scott & Vallance, 1995; Major, 1996; Vallance & Scott, 1997). We did not attempt to differentiate between the relative importance of these mechanisms primarily because generally poor exposures did not permit us to identify or trace changes in texture and /or fabric, data so crucial to such analyses (Scott & Vallance, 1995; Major, 1996; Vallance & Scott, 1997). The genetic link between cohesionless, grainconcentrated gravel-rich deposits and generally thinner, and overlying, normally graded and thinly horizontally stratified sand beds (that grade distally into the finergrained basin plain deposits) is consistent with models developed for other Gilbert-type deltas (Colella et al., 1987; Colella, 1988; Nemec, 1990; Falk & Dorsey, 1998). Stratigraphic and sedimentary relations in the Republic Basin suggest that (i) gravel-dominated facies generally accumulated as topsets and upper to middle foresets, (ii) sand-dominated facies usually represent lower foresets, and (iii) heterolithic facies tended to accumulate in lower foreset and bottomset positions. Fine-grained sediments that accumulated beyond the delta in bottomsets and deeper lake-basin sediments probably were derived from suspended load stream sediment and fine-grained sediment winnowed from coarse-grained debris flows and turbidity currents. Map, facies, and stratigraphic relations suggest
that facies association C sediment was introduced into the lake via multiple inlets of which eastern sources were prominent. South-east-directed palaeocurrents, deduced from rare ripple cross-lamination in facies association C sediments exposed near the Bacon Creek fault, indicate that the lake also was fed from the western or north-western margins of the basin. Unfortunately, without additional sedimentary evidence, it is not possible to fully reconstruct the patterns of basin infilling.
BASIN EVOLUTIONcDISCUSSION Development of the Republic Basin coincided with extensional faulting, waning volcanism, hydrothermal activity, and increasing lacustrine sedimentation. Relations between sedimentary deposits, tectonic activity, and palaeogeographical evolution are summarized in a schematic block diagram (Fig. 16), which integrates map patterns, facies distributions, and cross-sectional data. During the earliest stages of Republic Basin evolution in late Sanpoil Volcanics time, vertical motion along the Bacon Creek and Eureka faults deepened the westernmost portion of an asymmetric graben basin, inducing early lacustrine sedimentation; downward motion also occurred along the Klondike Mountain fault, but at a slower rate. Evidence that these faults were active at this time is demonstrated by the focusing of Sanpoil Volcanics-equivalent dykes and intrusions along and within these faults and fault
Fig. 16. Schematic block diagram of tectono-sedimentary setting during late stages of Klondike Mountain Formation volcanogenic sedimentation within the Republic Basin, an asymmetric graben. Noteworthy features are concentration of breccia deposits (including hydrothermal eruption breccia bodies) along and near Eureka Creek fault, thickening of finer-grained lake deposits towards the western margin of the basin, and progradation of multiple, stacked Gilbert-type deltas.
VSIC09 9/12/08 2:02 PM Page 217
Republic Basin, Washington, USA zones. Subsequently, during latest Sanpoil Volcanics and earliest Klondike Mountain Formation time, the Eureka fault zone was the locus of considerable hydrothermal alteration and gold–silver mineralization (Full & Grantham, 1968; Tschauder, 1989; Fifarek et al., 1996). As the basin deepened, a perennial, stratified, meromictic lake developed that at first hosted the texturally diverse and stratigraphically complex suite of volcanogenic, hydrothermally influenced lacustrine basin margin (facies associations A) and lacustrine basin plain (facies association B) sediments. Continued hydrothermal activity including explosive hydrothermal eruptions coincided with these early stages of Klondike Mountain lake development and provided an abundant source of loose detritus, including hydrothermally altered volcanic and sedimentary rocks, minor hot-springs silica clasts, and hydrothermally produced clay. The abundance of clay contributed to the cohesive character of the numerous sediment gravity flows that were transported from hydrothermally altered sources into the lake. Continued faulting and local hydrothermal vent collapse maintained water depths in which turbidity-current- and suspensiongenerated mudstone and siltstone accumulated. At the same time, streams deposited coarse-grained gravel and sand along the basin margins. Thickening of facies associations A and B deposits towards both the Eureka and Bacon Creek faults (Fig. 7) indicates that movement along these structures both influenced sedimentation and was of a greater magnitude than the fault motion along the Klondike Mountain fault on the eastern side of the basin. Reduction in basin subsidence rates commonly is used to explain progradational infilling as deltas and fan-deltas expand into basins (Leeder & Gawthorpe, 1987; Blair & Bilodeau, 1988). Such a sedimentary– tectonic scenario helps explain the generally coarseningupward trend of facies association B and C sediments. Progradation of Gilbert-type deltas probably was tied to decreased downfaulting as well as expansion of source area drainage basins. Local preservation of unfaulted, upper Klondike Mountain Formation gravel deposits on top of the Klondike Mountain fault indicates that motion along this fault ceased near the end of basin filling. The gravelly topset sediments that cap facies association C and form a laterally extensive plain in the eastern part of the basin may record the final local base level in the Republic Basin, as basin subsidence ended. Subsequently, these gravel deposits were incised and later capped by lava flows of the upper Klondike Mountain Formation.
217
Volcanism was waning in the Republic Basin (the Republic graben) at the same time that rapid extension was driving both regional and tectonic changes. The expected relationship between extension and volcanism is controversial (e.g. Gans et al., 1989; Axen et al., 1993), but as Gans & Bohrson (1998) have recently documented, highly extended regions undergoing very rapid extension may experience corresponding episodes of diminished volcanism. Gans & Bohrson (1998) also noted that elevated hydrothermal activity (such as accompanied accumulation of the upper Sanpoil Volcanics and the lowermost Klondike Mountain Formation) might be expected at such times. Although the similarity of Gans & Bohrson’s (1998) observations is striking, more precise isotopic age constraints relating the timing of extension, volcanism, hydrothermal activity and sedimentation wil1 be needed, to more thoroughly evaluate the application of their research into the Republican Basin. The prominence of volcanogenic lacustrine sedimentary deposits in the Republic Basin, and elsewhere in the grabens and half-grabens of the Okanogan Highlands at this time, is noteworthy. Synextensional deposits from other highly extended regions in the southern US Cordillera are similarly rich in volcanogenic sedimentary accumulations; however, many of those accumulations also tend to include a much more voluminous record of alluvial deposition than found in the Republic Basin (Howard & John, 1987; Nielson & Beratan, 1990; Beratan, 1991; Dickinson, 1991; Fedo & Miller, 1992). The fact that volcanogenic lacustrine rather than alluvial deposits tended to accumulate and be preserved in the Republic Basin is thus curious. We believe the tendency for accumulation of lake sediment may be attributed to a combination of factors, including the relatively moist climate in the Okanogan Highlands (as opposed to the generally dry climates that influenced synextensional deposits in the southern US Cordillera) that promoted persistent lake development, and the relatively high elevations within the Okanogan Highlands (Wolfe et al., 1998), which could have helped isolate this area from surrounding lowlands. As a result, internal rather than through-flowing axial drainages developed within the grabens and half-grabens. The tendency for preservation of lake sediment at the expense of alluvial deposits also may be attributed to a combination of factors, including the resistant nature of the volcanic flow rocks that capped the relatively non-resistant Klondike Mountain Formation lake sediments (preferentially protecting those strata from erosion), and the probability that alluvial sediments accumulated
VSIC09 9/12/08 2:02 PM Page 218
218
D. R. Gaylord et al.
adjacent to, but not within, the downfaulted lacustrine depocentre.
SUMMARY AND CONCLUSIONS Middle Eocene strata preserved in the Republic Basin provide a useful example of how volcanism, hydrothermal activity, and extensional tectonism can influence the character of lacustrine sedimentation. Sedimentation in the Republic Basin changed systematically from upper Sanpoil Volcanics to lower Klondike Mountain Formation time. As volcanism waned, sedimentation increased and a pulse of hydrothermal activity coincided with that transition. The humid, temperate climate also was a factor, promoting plant growth, helping maintain a perennial lake, and probably accelerating local rates of weathering and erosion. Hydrothermal alteration of the country rock and occasional hydrothermal eruptions that spanned the transition from the Sanpoil Volcanics to the Klondike Mountain Formation produced abundant clay-sized particles and provided an abundant source of loose, readily reworked detritus. The high clay concentrations contributed to the cohesive nature of the debris flows that were often responsible for transporting volcanic and hydrothermal detritus into the (tectonically) deepening lake. These cohesive flows made up a significant portion of lake margin deposits that characterized facies association A. The remaining deposits of facies association A were generated by non-cohesive sediment gravity flows, turbidity currents, and suspension. Continued downfaulting along the margins of the developing asymmetric graben that defines the Republic Basin led to the most widespread deposition of the fine-grained, laminated, and organic-rich lake basin deposits of facies association B, including highly fossiliferous plant, insect, and fish-bearing strata. Rare volcanic eruptions in the region at this time produced thin ash deposits. Sedimentation in the deeper parts of this stratified, meromictic lake was accomplished via suspension settling as well as from distal turbidity current deposition. The unusually thick and widespread turbidity current deposits probably were triggered by catastrophic events, the most likely candidates of which were volcanic eruptions, earthquakes, or floods. Organic matter was preferentially preserved in the lake because of a semipermanent anoxic hypolimnion. Concurrent progradation and stacking of gravel- and sand-rich Gilbert-type delta deposits
(facies association C) into the lake contributed to the overall coarsening-upward megasequence that defines the bulk of the Klondike Mountain Formation sedimentary succession. Gilbert-type delta foresets characterized by bipartite gravel and sand cross-strata were generated by linked cohesionless debris flows and overriding turbidity currents. Surface exposures of the Gilbert delta foresets suggest that the lake was at least 100 m deep near the end of Klondike Mountain Formation sedimentation. Filling of the lake roughly corresponded to the cessation of Republic Basin subsidence. Subsequent tectonic tilting of the western side of the basin and later fluvial incision of the Klondike Mountain Formation sedimentary strata were followed by a final phase of volcanism. This volcanic activity produced the capping andesitic to rhyolitic flows of the upper Klondike Mountain Formation.
ACKNOWLEDGEMENTS The authors wish to thank Hecla Mining Company for their generous access to their Republic district property and to their drill core collection. K. A. Lindsey helped supervise and J. M. Matthews assisted in core logging during the research. Financial support was provided by grants from Washington State University and Hecla Mining Company. We appreciated the many useful discussions with K. L. Stoffel, R. J. Tschauder, G. McCauley-Holder, R. W. Holder, C. M. Knaack, K. A. Lindsey, J. M. Matthews, G. Morris, P. R. Hooper, R. L. Thiessen, J. W. Vallance, and A. J. Watkinson during our research in the Republic area. The authors particularly thank reviewers R. J. Dorsey, L. T. Middleton, and N. R. Riggs for their insightful comments; their diligent reviews significantly improved the manuscript.
REFERENCES Armstrong, R.L. (1982) Cordilleran metamorphic core complexesafrom Arizona to southern Canada. Ann. Rev. Earth planet. Sci., 10, 129 –154. Axen, G.J., Taylor, W.J. & Bartley, J.M. (1993) Space– time patterns and tectonic controls of Tertiary extension and magmatism in the Great Basin of the western United States. Geol. Soc. Am. Bull., 105, 56 –76. Beratan, K.K. (1991) Miocene synextension sedimentation patterns, Whipple Mountains, southeastern California: implications for the geometry of the Whipple detachment system. J. geophys. Res., 96, 12425 –12442. Berger, B.R. & Eimon, P.I. (1983) Conceptual models of
VSIC09 9/12/08 2:02 PM Page 219
Republic Basin, Washington, USA epithermal precious metal deposits. In: Cameron Volume on Unconventional Mineral Deposits (Ed. Shanks, W.C.), pp. 191–205. Society of Mining Engineers of the American Institute of Mining, Metallurgical and Petroleum Engineers, New York. Berger, B.R. & Snee, L.W. (1992) Thermochronological constraints on mylonite and detachment fault development, Kettle Highlands, northeastern Washington and southern British Columbia. Geol. Soc. Am. Abstr. Prog., 24, 65. Blair, T.C. (1987) Tectonic and hydrologic controls on cyclic alluvial fan, fluvial, and lacustrine rift basin sedimentation, Jurassic–lowermost Cretaceous Todos Santos Formation, Chiapas, Mexico. J. sediment. Petrol., 57, 845 – 862. Blair, T.C. & Bilodeau, W.L. (1988) Development of tectonic cyclothems in rift, pull-apart, and foreland basins: sedimentary response to episodic tectonism. Geology, 16, 517–520. Braun, E.R. (1989) Epithermal and hot-spring gold –silver deposits, Republic Mining District, Ferry County, Washington. PhD thesis, University of Montana, Missoula. Brown, R.L. & Journeay, J.M. (1987) Tectonic denudation of the Shuswap metamorphic terrane of southeastern British Columbia. Geology, 15, 142 –146. Burchfiel, B.C., Cowan, D.S. & Davis, G.A. (1992) Tectonic overview of the Cordillera orogen in the western United States. In: The Cordillera Orogen, Conterminous US (Eds Burchfiel, B.C., Lipman, P.W. & Zoback, M.L.), Geology of North America, G-3, pp. 407– 479. Geological Society of America, Boulder, CO. Cheney, E.S. (1994) Cenozoic unconformity-bounded sequences of central and eastern Washington. In: Regional Geology of Washington State (Conv. Lasmanis, R. & Cheney, E.S.), pp. 115 –139. Bull. Wash. Div Geol. Earth Resour., 80. Olympia, Washington. Cheney, E.S. & Rasmussen, M.G. (1996) Regional geology of the Republic area. Wash. Geol., 24, 3–7. Colella, A. (1988) Pliocene –Holocene fan deltas and braid deltas in the Crati Basin, southern Italy: a consequence of varying tectonic conditions. In: Sedimentology and Tectonic Settings (Eds Nemec, W. & Steel, R.J.), pp. 50 –74. Blackie, Glasgow. Colella, A., De Boer, P.L. & Nio, S.D. (1987) Sedimentology of a marine intermontane Pleistocene Gilbert-type fan-delta complex in the Crati Basin, Calabria, southern Italy. Sedimentology, 34, 721–736. Coleman, J.M. & Prior, D.B. (1982) Deltaic environments of deposition. In: Sandstone Depositional Environments (Eds Scholle, P.A. & Spearing, D.), Mem. Am. Ass. petrol. Geol., Tulsa, 31, 139–178. Coney, P.J. (1980) Cordilleran metamorphic core complexes. In: Cordilleran Metamorphic Core Complexes (Eds Crittenden, M.D., Jr, Coney, P.J. & Davis, G.H.), Geol. Soc. Am. Mem., Boulder, CO, 153, 7–31. Coney, P.J. & Harms, T.A. (1984) Cordilleran metamorphic core complexes: Cenozoic extensional relics of Mesozoic compression. Geology, 12, 550 –554. Cross, D. (1963) Soils and geology of some hydrothermal eruptions in the Waiotapu District. N.Z. J. Geol. Geophys., 6, 70–87. Dickinson, W.R. (1991) Tectonic Setting of Faulted Tertiary Strata Associated with the Catalina Core Complex in Southern Arizona. Geol. Soc. Am. Spec. Paper, Boulder, CO, 264.
219
Dorsey, R.J., Umhoefer, P.J. & Renne, P.R. (1995) Rapid subsidence and stacked Gilbert-type fan deltas, Pliocene Loreto basin, Baja California Sur, Mexico. Sediment. Geol., 98, 181–204. Falk, P.D. & Dorsey, R.J. (1998) Rapid development of gravelly high-density turbidity currents in marine Gilberttype fan deltas, Loreto Basin, Baja California Sur, Mexico. Sedimentology, 45, 331–349. Fedo, C.M. & Miller, J.M.G. (1992) Evolution of a Miocene half-graben basin, Colorado extensional corridor, southeastern Colorado. Geol. Soc. Am. Bull., 104, 481–493. Fifarek, R.H., Devlin, B.D. & Tschauder, R.J. (1996) Au–Ag mineralization at the Golden Promise Deposit, Republic District, Washington: relation to graben development and hot spring processes. In: Geology and Ore Deposits of the American Cordillera (Eds Coyner, A.R. & Fahey, P.L.), Geol. Soc. Nevada Symposium Proceedings, Reno –Sparks, 1063 –1088. Fox, K.F., Jr & Rinehart, C.D. (1988) Okanogan gneiss domeaa metamorphic core complex in north-central Washington. Wash. geol. Newslett., 16, 3–12. Fox, K.F., Jr, Rinehart, C.D., Engels, J.C. & Stern, T.W. (1976) Age of emplacement of the Okanogan gneiss dome, north-central Washington. Geol. Soc. Am. Bull., 87, 1217–1224. Full, R.P. & Grantham, R.M. (1968) Ore deposits of the Republic Mining District, Ferry County, Washington. In: Ore Deposits of the United States, 1933–1967, GratonSales Volume (Ed. Ridge, J.D.), Am. Inst. Mining Metall. petrol. Eng., 2, 1481–1494. New York. Gans, P.B. & Bohrson, W.A. (1998) Suppression of volcanism during rapid extension in the Basin and Range Province, United States. Science, 279, 66 – 68. Gans, P.B., Mahood, G. & Schermer, E. (1989) Synextensional Magmatism from the Basin and Range Province: a Case Study from the Eastern Great Basin. Geol. Soc. Am. Spec. Paper, Boulder, CO, 233. Gaylord, D.R. (1986) Cenozoic sedimentation in the Republic graben, north-central Washington: economic implications. Soc. econ. Paleont. Miner. Abstr. Prog., 3, 42. Gaylord, D.R., Price, S.P., Suydam, J.D. & Thiessen, R.L. (1990) Middle Eocene sedimentation, hydrothermal activity, tectonism and volcanism, Klondike Mountain Formation, central Republic graben, Washington. Geol. Assoc. Can. Abstr. Prog., 15, A44. Gaylord, D.R., Suydam, J.D., Price, S.M., Matthews, J. & Lindsey, K.A. (1996) Depositional history of the uppermost Sanpoil Volcanics and Klondike Mountain Formation in the Republic basin. Wash. Geol., 24, 15 –18. Gilbert, G.K. (1885) The topographic features of lake shores. Ann. Rep. US geol. Surv., 5, 69–123. Gilbert, G.K. (1890) Lake Bonneville. Monograph 1. US geol. Surv., Government Printing Office, Washington. Hansen, V.L. & Goodge, J.W. (1988) Metamorphism, structural petrology, and regional evolution of the Okanogan Complex, northeastern Washington. In: Metamorphism and Crustal Evolution of the Western United States, Rubey Vol. VII (Ed. Ernst, W.G.), pp. 233 –270. Prentice -Hall, Englewood Cliffs, NJ. Harms, T.A. (1982) The Newport fault: low-angle normal faulting and Eocene extension, northeast Washington and northwest Idaho. MS thesis, Queens University, Kingston, Ont.
VSIC09 9/12/08 2:02 PM Page 220
220
D. R. Gaylord et al.
Harms, T.A. & Price, R.A. (1992) The Newport fault: Eocene listric normal faulting, mylonitization, and crustal extension in northeast Washington and northwest Idaho. Geol. Soc. Am. Bull., 104, 745–761. Hedenquist, J.W. & Henley, R.W. (1985) Hydrothermal eruptions in the Waiotapu geothermal system, New Zealand: their origin, associated breccias, and the relation to precious metal mineralization. Econ. Geol., 80, 1640 –1668. Holder, G.M. (1990) Relationship of the Sanpoil Volcanics with intrusive rocks of the Colville batholith, Washington. PhD thesis, Washington State University, Pullman. Holder, R.W. & Holder, G.M. (1988) The Colville batholith: Tertiary plutonism in northeast Washington associated with graben and core-complex (gneiss dome) formation. Geol. Soc. Am. Bull., 100, 1971–1980. Holder, G.M., Holder, R.W. & Carlson, D.H. (1990) Middle Eocene dike swarms and their relation to contemporaneous plutonism, volcanism, core-complex mylonitization, and graben subsidence, Okanogan Highlands, Washington. Geology, 18, 1082–1085. Horne, A.J. & Goldman, C.R. (1994) Limnology. McGrawHill, New York. Howard, K.A. & John, B.E. (1987) Crustal extension along a rooted system of low-angle normal faults, Colorado River extensional corridor, Calfornia and Arizona. In: Continental Extensional Tectonics (Eds Coward, M.D., Dewey, J.F. & Hancock, P.L.), Spec. Publ. geol. Soc., London, No. 28, pp. 299–311. Geol. Soc. London, Bath. Jennings, S. (1990) Geology of the west sinter deposit at the Buckhorn Mine, Eureka County, Nevada. In: Geology and Ore Deposits of the Great Basin (Eds Raines, G., Lisle, R.E., Schafer, R.W. & Wilkinson, W.H.), pp. 947–956. Geological Society of Nevada. Johnson, A.M. (1970) Physical Processes in Geology. Freeman, Cooper, San Francisco, CA. Johnson, H.D. & Baldwin, C.T. (1986) Shallow siliciclastic seas. In: Sedimentary Environments and Facies, 2nd edn (Ed. Reading, H.G.), pp. 229–282. Blackwell Scientific Publications, Oxford. Kazanci, N. (1988) Repetitive deposition of alluvial-fan and fan-delta wedges at a fault-controlled margin of the Pleistocene–Holocene Burdur Lake graben, southwestern Anatolia, Turkey. In: Fan Deltas: Sedimentology and Tectonic Settings (Eds Nemec, W. & Steel, R.J.), pp. 186–196. Blackie, Glasgow. Kokelaar, B.P. (1982) Fluidization of wet sediments during the emplacement and cooling of various igneous bodies. J. geol. Soc. London, 139, 21–33. Lasmanis, R. (1996) A historical perspective on ore formation concepts, Republic Mining District, Ferry County, Washington. Wash. Geol., 24, 8–11. Leeder, M.R. & Gawthorpe, R.L. (1987) Sedimentary models for extensional tilt-block / half-graben basins. In: Continental Extensional Tectonics (Eds Coward, M.P., Dewey, J.F. & Hancock, P.L.), Spec. Publ. geol. Soc. London, No. 28, pp. 139–152. Geol. Soc. London, Bath. Leeder, M.R., Ord, D.M. & Collier, R. (1988) Development of alluvial fans and fan deltas in neotectonic extensional settings: implications for the interpretation of basin-fills. In: Fan DeltasaSedimentology and Tectonic Settings (Eds Nemec, W. & Steel, R.J.), pp. 173 –185. Blackie, Glasgow.
Lewis, S.E. (1992) Insects of the Klondike Mountain Formation, Republic, Washington. Wash. Geol., 20, 15 – 19. Lindgren, W. & Bancroft, H. (1914) Republic Mining District. In: The Ore Deposits of Northeastern Washington (Ed. Bancroft, H.), Bull. US geol. Surv., Denver, CO, 550, 133 –166. Lloyd, E.F. (1959) The hot springs and hydrothermal eruption Waiotapu. N.Z. J. Geol. Geophys., 2, 141–176. Lowe, D.R. (1982) Sediment gravity flows: II. Depositional models with special reference to the deposits of high-density turbidity currents. J. sediment. Petrol., 52, 279–297. Mack, G.H. & Seager, W.R. (1990) Tectonic control on facies distribution of the Camp Rice and Palomas Formations (Pliocene–Pleistocene) in the southern Rio Grande rift. Geol. Soc. Am. Bull., 102, 45–53. Major, J.J. (1996) Experimental studies of deposition by debris flows: process, characteristics of deposits, and effects of pore-fluid pressure. PhD thesis, University of Washington, Seattle. Matthews, J.M. (1997) The stratigraphy and sedimentology of the Eocene O’Brien Creek Formation, Okanogan Highlands, Northeast Washington. PhD thesis, Washington State University, Pullman. Matthews, J.M. & Gaylord, D.R. (1994) Sedimentology, stratigraphy, and volcanic history of the Eocene O’Brien Creek Formation, north-central Washington. Geol. Soc. Am. Abstr. Prog., 26, A242. McPhie, J., Doyle, M. & Allen, R. (1993) Volcanic Textures. University of Tasmania, Centre for Ore Deposit and Exploration Studies (CODES), Hobart. Miall, A.D. (1978) Lithofacies types and vertical profile models in braided river deposits: a summary. In: Fluvial Sedimentology (Ed. Miall, A.D.), Mem. Can. Soc. petrol. Geol., Calgary, 5, 597– 604. Middleton, G.V. (1970) Experimental studies related to flysch sedimentation. In: Flysch Sedimentology in North America (Ed. Lajoie, J.), Spec. Pap. geol. Assoc. Can., Toronto, Ont., 7, 253 –272. Mohrig, D., Whipple, K.X., Hondzo, M., Ellis, C. & Parker, G. (1998) Hydroplaning of subaqueous debris flows. Geol. Soc. Am. Bull., 110, 387–394. Morris, G.A. & Hooper, P.R. (1997) Petrogenesis of the Colville Igneous Complex, northeast Washington: implications for Eocene tectonics in the northern U.S. Cordillera. Geology, 25, 831– 834. Moye, F.J. (1984) Geology and petrochemistry of Tertiary igneous rocks in the western half of the Seventeenmile Mountain Quadrangle, Ferry County, Washington. PhD thesis, University of Idaho, Moscow. Muessig, S.F. (1962) Tertiary Volcanic and Related Rocks of the Republic Area, Ferry County, Washington. US geol. Surv. Prof. Pap., Denver, CO, 450-D, D56 –D58. Muessig, S.F. (1967) Geology of the Republic Quadrangle and a Part of the Aeneas Quadrangle, Ferry County, Washington. Bull. US geol. Surv., Denver, CO, 1216. Muffler, L.J.P., White, D.E. & Truesdale, A.H. (1971) Hydrothermal explosion craters in the Yellowstone National park. Geol. Soc. Am. Bull., 82, 723 –740. Nairn, I.A. & Wiradiradja, S. (1980) Late Quaternary hydrothermal explosion breccias at Kawerau geothermal field, New Zealand. Bull. Volcanol., 43, 1–13.
VSIC09 9/12/08 2:02 PM Page 221
Republic Basin, Washington, USA Nairn, I.A., Wood, C.P. & Hewson, C.A.Y. (1979) Phreatic eruptions of Ruapehu: April, 1975. N.Z. J. Geol. Geophys., 22, 155–173. Nemec, W. (1990) Aspects of sediment movement on steep delta slopes. In: Coarse-Grained Deltas (Eds Colella, A. & Prior, D.B.), Spec. Publs int. Ass. Sediment., No. 10, pp. 29–73. Blackwell Scientific Publications, Oxford. Nielson, J.E. & Beratan, K.K. (1990) Tertiary basin development and tectonic implications, Whipple detachment system, Colorado River extensional corridor, California and Arizona. J. geophys. Res., 95, 571–580. Orr, K.E. & Cheney, E.S. (1987) Kettle and Okanogan domes, northeastern Washington and southern British Columbia. In: Selected Papers on the Geology of Washington (Ed. Schuster, J.E.), Wash. Div. Geol. Earth Resour. Bull., 77, 55 –71. Parker, R.L. & Calkins, J.A. (1964) Geology of the Curlew Quadrangle, Ferry County, Washington. Bull. US geol. Surv., Denver, CO, 1169. Parrish, R.R., Carr, S.D. & Parkinson, D.L. (1988) Eocene extensional tectonics and geochronology of the southern Omineca belt, British Columbia and Washington. Tectonics, 7, 182 –212. Pearson, R.C. & Obradovich, J.D. (1977) Eocene Rocks in Northeast WashingtonaRadiometric Ages and Correlation. Bull. US geol. Surv., Denver, CO, 1433. Postma, G., Nemec, W. & Kleinspehn, K.L. (1988) Large floating clasts in turbidites: a mechanism for their replacement. Sediment. Geol., 58, 47– 61. Price, R.A. (1979) Intracontinental ductile crustal spreading linking the Fraser River and Northern Rocky Mountain Trench transform fault zones, south-central British Columbia and northeast Washington. Geol. Soc. Am. Abstr. Prog., 11, 499. Price, S.M. (1991) Geology of the Klondike Mountain Formation and upper Sanpoil Volcanics in the Republic Mining District, Ferry County, Washington. MS thesis, Washington State University, Pullman. Price, R.A., Archibald, D. & Farrar, E. (1981) Eocene stretching and necking of the crust and tectonic unroofing of the Cordilleran metamorphic infrastructure, southeastern British Columbia and adjacent Washington and Idaho. Geol. Assoc. Can. Abstr. Prog., 6, A47. Prior, D.G. & Bornhold, B.D. (1988) Submarine morphology and processes of fjord fan deltas and related high gradient systems: modern examples from British Columbia. In: Fan DeltasaSedimentology and Tectonic Settings (Eds Nemec, W. & Steel, R.J.), pp. 125–143. Blackie, Glasgow. Rust, B.R. (1978) Depositional models for braided alluvium. In: Fluvial Sedimentology (Ed. Miall, A.D.), Mem. Can. Soc. petrol. Geol., Calgary, 5, 605– 625. Rytuba, J.J. & Vander Meulen, D.B. (1991) Hot-spring precious metal systems in the lake Owyhee volcanic field, Oregon–Idaho. In: Geology and Ore Deposits of the Great Basin (Eds Raines, G., Lisle, R.E., Schafer, R.W. & Wilkinson, W.H.), pp. 1085–1096. Geol. Soc. Nevada. Reno, Nevada. Rytuba, J.J., Vander Meulen, D.B. & Barlock, V.E. (1990) Epithermal precious-metal systems in the Lake Owyhee volcanic field, Oregon–Idaho. In: Geology and Ore Deposits of the Great Basin (Chairperson Cuffney, B.), p. 83. Geol. Soc. Nevada Abstr. Prog. Reno, Nevada.
221
Scott, K.M. (1988) Origins, behavior, and sedimentology of catastrophic lahars at Mount St. Helens, Washington. In: Sedimentologic Consequences of Convulsive Geologic Events (Ed. Clifton, H.E.), Geol. Soc. Am. Spec. Pap., Boulder, CO, 229, 23–36. Scott, K.M., Vallance, J.W. & Pringle, P.T. (1995) Sedimentology, Behavior, and Hazards of Debris Flows at Mount Rainier, Washington. US geol. Surv. Prof. Pap., Denver, CO, 1547. Sillitoe, R.H., Baker, E.M. & Brook, W.A. (1984) Gold deposits and hydrothermal eruption breccias associated with a maar volcano, Papua New Guinea. Econ. Geol., 79, 638– 655. Smith, G.A. (1987) The influence of explosive volcanism on fluvial sedimentation: the Deschutes Formation (Neogene) in central Oregon. Geol. Soc. Am. Bull., 97, 1–10. Staatz, M.H. (1964) Geology of the Bald Knob Quadrangle, Ferry and Okanogan Counties, Washington. Bull. US geol. Surv., Denver, CO, 1161-F. Stoffel, K.L., Joseph, N.L., Waggoner, S.Z., Gulick, C.W., Korosec, M.A. & Bunning, B.B. (1991) Geologic map of WashingtonaNortheast quadrant. Geol. Map GM39. Wash. Div. Geol. Earth Resour. Olympia, Washington. Sturm, M. & Matter, A. (1978) Turbidites and varves in Lake Brienz (Switzerland): deposition of clastic detritus by density currents. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker, M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 147–168. Blackwell Scientific Publications, Oxford. Suydam, J.D. (1993) Stratigraphy and sedimentology of the Klondike Mountain Formation, with implications for the Eocene paleogeography and tectonic development of the Okanogan Highlands, northeast Washington. PhD thesis, Washington State University, Pullman. Suydam, J.D. & Gaylord, D.R. (1991) Middle Eocene paleogeography of the Okanogan Highlands, Washington: evidence from the sedimentary and volcanic record. Geol. Soc. Am. Abstr. Prog., 23, 68. Suydam, J.D. & Gaylord, D.R. (1997) Toroda Creek halfgraben, northeast Washington: late-stage sedimentary infilling of a synextensional basin. Geol. Soc. Am. Bull., 109, 1333–1348. Tschauder, R.J. (1989) Gold deposits in northern Ferry County, Washington. Inform. Circ. Wash. Div. Geol. Earth Resour, 86, 241–253. Olympia, Washington. Umpleby, J.B. (1910) Geology and Ore Deposits of Republic Mining District. Bull. geol. Surv. Wash., 1. Olympia, Washington. Vallance, J.W. (1994) Experimental and field studies related to the behavior of granular mass flows and the characteristics of their deposits. PhD thesis, Michigan Technological University, Houghton. Vallance, J.W. & Scott, K.M. (1997) The Osceola mudflow from Mount Rainier: sedimentology and hazard implications of a huge clay-rich debris flow. Geol. Soc. Am. Bull., 109, 143 –163. Varsek, J.L. & Cook, F.A. (1994) Three-dimensional crustal structure of the Eastern Cordillera, southwestern Canada and northwestern United States. Geol. Soc. Am. Bull., 106, 803–823. Wagoner, L.C. (1992) Geochemistry of the Eocene Klondike Mountain Formation lavas of the Republic graben, northeast
VSIC09 9/12/08 2:02 PM Page 222
222
D. R. Gaylord et al.
Washington. MS thesis, Washington State University, Pullman. Wernicke, B.P., Christiansen, R.L., England, P.C. & Sonder, L.J. (1987) Tectonomagmatic evolution of Cenozoic extension in the North American cordillera. In: Continental Extensional Tectonics (Eds Coward, M.D., Dewey, J.F. & Hancock, P.L.), Spec. Publ. geol. Soc. London, No. 28, pp. 203–221. Geol. Soc. London, Bath. White, D.E. (1955) Violent mud-volcano eruption of Lake City hot springs, northeastern California. Geol. Soc. Am. Bull., 66, 1109–1130. Wilson, M.V.H. (1996) The Eocene fishes of Republic, Washington. Wash. Geol., 24, 30 –31.
Wolfe, J.A. & Wehr, W.C. (1987) Middle Eocene dicotyledenous plants from Republic, northeastern Washington. Bull. US Geol. Surv., Denver, CO, 1597. Wolfe, J.A. & Wehr, W.C. (1991) Significance of the Eocene fossil plants at Republic, Washington. Wash. Geol., 19, 18–24. Wolfe, J.A., Forest, C.E. & Molnar, P. (1998) Paleobotanical evidence of Eocene and Oligocene paleoaltitudes in midlatitude western North America. Geol. Soc. Am. Bull., 110, 664 – 678. Woodcock, N.H. (1979) The use of slump structures as paleoslope orientation indicators. Sedimentology, 26, 83 – 99.
VSIC10 9/12/08 1:00 PM Page 223
Lakes as sensitive recorders of eruptions and the response of distal landscapes
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
VSIC10 9/12/08 1:00 PM Page 225
Spec. Publs int. Ass. Sediment. (2001) 30, 225 –246
Tephra layers in a sediment core from Lake Hestvatn, southern Iceland: implications for evaluating sedimentation processes and environmental impacts on a lacustrine system caused by tephra fall deposits in the surrounding watershed J . H A R D A R D Ó T T I R 1 * , Á . G E I R S D Ó T T I R 2 and T . T H Ó R D A R S O N 3 1Institute of Arctic and Alpine Research and Department of Geological Sciences, University of Colorado,
CB 450, Boulder, CO 80309-0450, USA; 2Department of Geosciences, University of Iceland, 101 Reykjavík, Iceland; 3CSIRO, Magmatic Ore Deposit Group, Division of Exploration and Mining, Private Bag, PO Wembley, W.A. 6014, Australia
ABSTRACT A 570 cm sediment core (94-HV01) from Lake Hestvatn in southern Iceland records continuous lacustrine deposition over the last 6200 yr, including numerous layers of primary fallout tephra. Significant downcore changes are observed in sediment magnetic, chemical and sedimentological characteristics of the deposit, and four zones are identified based primarily on changes in sediment magnetic properties that are supported by chemical analyses. Zone I (3830 – 6200 14C yr bp) is characterized by low sediment flux, low concentrations of magnetic and chemical substances, and high amounts of biogenic silica, and reflects a period of high productivity in the lake and minimal erosion in the watershed. However, an increase in magnetic and chemical concentrations in the uppermost part of the zone suggests increased erosion within the watershed that is possibly caused by a slight change in vegetation composition towards the end of this period. Zone II (2180 –3830 14C yr bp) is identified by an abrupt increase in sediment flux into the lake and greatly increased concentrations of magnetic and chemical substances. This change coincides with a shift in vegetation composition within the watershed that is possibly caused by lower temperatures and /or increased precipitation, as well as in conjunction with the deposition of the H4 tephra and subsequent less voluminous tephra layers. Zone III (1180 –2180 14C yr bp) is marked by low concentrations of magnetic and chemical materials and relatively low sedimentation rates as a result of less frequent deposition of fallout tephra into the watershed and less erosion, possibly caused by a slight change in composition of vegetation cover. Zone IV (present– 1180 14C yr bp) is characterized by a sudden increase in sedimentation rates and high concentrations of magnetic and chemical materials, and reflects an increase in surface erosion within the watershed following the settlement of Iceland. Seventy-three tephra layers are identified in the Lake Hestvatn sediment core (94-HV01). The Katla and Hekla volcanoes are the source of most of the tephra layers in the core, although tephra layers representing the Vestmannaeyjar, Veidivötn and Grímsvötn volcanoes are also present. These tephra layers represent fallout ash deposited directly into the lake along with subsequent contributions from reworking of tephra within the lacustrine and terrestrial environment. On the basis of major and trace element concentration profiles across the Hekla 4 (H4) and Katla N (KN) tephra layers and adjacent sediment, we show that the influx of tephra particles decreased to pre-eruption values, 66 and 90 yr after the initial tephra deposition. This study indicates that in regions such as southern Iceland, where eruptions that produce great amounts of tephra are frequent, the effects of tephra deposition on lacustrine systems have to be evaluated before other environmental proxies (e.g. pollen and diatoms) in lake sediments can be interpreted accurately.
*Present address: National Energy Authority, Grensasvegur 9, 108 Reykjavik, Iceland. Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
225
VSIC10 9/12/08 1:00 PM Page 226
226
J. Hardardóttir et al.
INTRODUCTION Superposition of a mantle plume and mid-ocean ridge magmatism makes Iceland one of the most productive volcanic regions in the world, with four volcanic zones and at least 31 active volcanic systems (Fig. 1a). On average one eruption occurs every 5 yr in Iceland (Thórarinsson, 1981), and the high frequency of explosive eruptions makes it well suited for tephrochronological studies. Rapid accumulation of the Icelandic loess enhances preservation of tephra layers and results in a high-resolution record because even tephra layers from eruptions closely separated in time are clearly isolated in soil profiles. Individual tephra layers can be traced to their source because each volcanic system erupts magma with distinct chemical composition and produces tephra with distinct physical properties (Larsen, 1981). Furthermore, the eruption history of Iceland during the last 1100 yr is well documented in numerous historical accounts, resulting in an excellent tephrochronological record with a resolution of roughly one to two decades. A framework for the Holocene tephra stratigraphy has been established in Iceland through regional mapping and dating of key marker layers (Larsen & Thórarinsson, 1977; Thórarinsson, 1981) and through studies of selected soil profiles containing many tephra layers (e.g. Thórarinsson, 1958, 1967, 1968; Larsen, 1982, 1984). The use of tephrochronology in Iceland has mainly been restricted to dating geological formations or
(a)
archaeological structures intercalated with soils containing tephra layers of known age (Thórarinsson, 1961; Larsen, 1982, 1984), although glacial ice (Steinthórsson, 1982; Larsen et al., 1996) and lake sediments (e.g. Thompson et al., 1986; Björck et al., 1992; Haflidason et al., 1992; Geirsdóttir et al., 1995; Hardardóttir et al., 1996) have been dated recently with the aid of tephrochronology. Most of these studies have used tephra layers as a chronological and correlational tool, although Thompson et al. (1986) evaluated the relative importance of tephra accumulation from suspension, reworking and bioturbation in Icelandic lake sediments. Tephra deposition in a watershed can have a significant impact on biological and physical processes that operate within a lake basin. Studies have shown that density and/or composition of vegetation cover may change in an area covered by tephra deposit as a result of: 1 suffocation of the existing vegetation by thick tephra; 2 the abrasive character of the sharp pumice particles; 3 deposition of toxic volcanigenic fluoride (e.g. Fridriksson, 1981; Lotter et al., 1995; Wilmshurst & McGlone, 1996); 4 climatic cooling caused by high concentration of atmospheric aerosols (LaMarche & Hirschboeck, 1984; Baillie & Munro, 1988), although several
(b) N 0
2 km
Hestvatn 94-HV01 60 300
Hestvatn
275
Lake Hestvatn
250 10 20
Veidivötn
ítá
Hekla
Hv
Grímsvötn
Hestfjall
30 40 50
Katla
225
Feldarvík
200
0
Vestmannaeyjar North Atlantic Ocean
50
100 km 75
100 175 125 150
Fig. 1. (a) Volcanic zones of Iceland (in grey) and the main volcanic systems active or dormant since deglaciation () and related fissure swarms. The locations of Lake Hestvatn (), Hekla, Katla, Veidivötn, Vestmannaeyjar, and Grímsvötn () are shown. (b) Simplified bathymetry of Lake Hestvatn and location of core 94-HV01.
VSIC10 9/12/08 1:00 PM Page 227
Tephra fall deposits in the watershed workers (e.g. Birks, 1994; Hall et al., 1994; Dwyer & Mitchell, 1997; Caseldine et al., 1998) have questioned the significance of such effects from volcanic activity on remote areas. A few studies indicate an increase in diatom productivity in lakes affected by tephra fall, and this increase is probably caused by elevated silica content of the lake water (Haflidason & Einarsson, 1989; Lotter et al., 1995). Reduction in vegetation cover caused by deposition of airborne tephra and /or transport and redeposition of newly fallen tephra is likely to increase erosion within the affected watersheds and consequently alter depositional processes in the lake basins. Consequently, changes in vegetation cover, lake productivity, and erosional processes induced by tephra deposition must be accounted for before sedimentological or biotic proxies (e.g. pollen, diatoms) from lake sediments and /or soil sections are used to interpret past environmental changes. In this paper we present sedimentological, chemical, and magnetic analyses from a sediment core (94HV01) obtained from Lake Hestvatn, Iceland (Fig. 1). The lake lies close to the main volcanic zone in southern Iceland, and its watershed has repeatedly been affected by tephra fall, as is demonstrated by 73 tephra layers in the ≈ 570 cm core that spans the last 6200 14C yr. We introduce the tephrochronological record of the core; high-resolution chemical analyses through two major tephra layers to evaluate the processes responsible for transport and deposition of these tephras in Lake Hestvatn; and the chemical and physical variations observed in the sediment core. We show that four main processes are responsible for the downcore changes in magnetic, chemical, and sedimentological variables in core 94-HV01: 1 Middle to Late Holocene climatic variability (lower temperature and/or increased precipitation) that caused gradual decline in vegetation cover and increased erosion rates; 2 deposition of thick tephra layers in the watershed that caused an abrupt escalation in the flux of volcanogenic material to the lake, disturbed vegetation cover, and increased erosion rates; 3 frequent deposition of thin tephra layers, which tended to maintain high erosion rates within the watershed by inhibiting vegetation growth; 4 the Norse settlement of Iceland in late ninth century ad that initiated major vegetation disturbance and generated higher sediment erosion rates than had been seen in the watershed over the previous 5000 yr or more. The complicated relationships between the interacting processes demonstrate that extreme caution is
227
advised when sediment records from regions subjected to tephra deposition are interpreted, because of the significant impact that tephra layers can have on lacustrine systems.
GEOLOGICAL AND GEOGRAPHICAL SETTING Lake Hestvatn is located in the southern lowlands of Iceland close to the South Iceland fracture zone (Fig. 1a). It rests on a 0.7–2.5 Ma basalt crustal wedge between the western rift zone and the southern transgressive volcanic zone. Nearby are two of the three most active volcanoes in IcelandaHekla and the glacially covered Katla volcanoes. Located further to the east, the Veidivötn volcanic centre and the Grímsvötn volcano (Fig. 1a) have also been very productive during the Holocene. Easterly winds are very common in South Iceland (Einarsson, 1976), and consequently the region of Lake Hestvatn has been covered repeatedly during Holocene time by tephra deposits that originate from these volcanoes, especially from the Katla volcano because of its position relative to Lake Hestvatn (Fig. 1a). The thickest Holocene tephra layers, Hekla 4 (H4) and Katla lower (KN), are studied in detail in this paper. The H4 tephra layer originated from a large Plinian eruption in the Hekla volcano at 3830 ± 30 14C yr bp (Dugmore et al., 1995), and on-land studies indicate that the tephra layer covered more than ≈ 78 000 km2 (Larsen & Thórarinsson, 1977). The silica composition of H4 ranges from 74% at the base to 57% at the top of the tephra layer, and the change in composition is accompanied by a colour change from white to brownish black at the top (Larsen & Thórarinsson, 1977). The KN tephra layer is one of the large prehistoric Katla layers that covered a large region in southern Iceland, and it has been dated to c. 3300 ± 100 14C yr bp by accumulation rate calculations from numerous soil sections (Róbertsdóttir, 1992b). Like other Katla layers, it is of transitional alkali basalt composition and contains high concentrations of TiO2 and FeO. Lake Hestvatn lies within a glacially eroded bedrock basin that has been affected by tectonic processes. It has a surface area of 6.8 km2 and a maximum water depth of 61.5 m (Fig. 1b). The lake is surrounded by rolling hills, except towards the east where Hestfjall (317 m above sea-level (a.s.l.) ) rises above the lake. The glacially fed river Hvítá curves along the eastern and southern foothills of Hestfjall. The
VSIC10 9/12/08 1:00 PM Page 228
228
J. Hardardóttir et al.
bedrock strata consist of predominantly basaltic lava flows intercalated with minor hyaloclastites. Heathland vegetation dominates the higher terrain around the lake whereas mires and wetlands are found at lower elevations. The area is cultivated and has been used for centuries as pasturage for grazing animals. In extremely cold years, ice-dams may form in the Hvítá river and cause floods into Lake Hestvatn if they break. Such floods may suddenly raise the lake’s water level by a few metres and cause a temporary but substantial increase in the flux of suspended sediment into the lake.
METHODS Core collction, core description and sediment sampling Several sediment cores were collected through lake ice from Lake Hestvatn (cores 94-HV01, 94-HV02 and 94-HV03) in 1994 with a Nesje gravity coring system (Nesje et al., 1987; Nesje, 1992). This paper introduces results from core 94-HV01 (568 cm long) collected from 60 m water depth in the northern basin of the lake (Fig. 1b). After cutting the core into two Dshaped halves we described the lithology and colour of the sediment; took colour photographs and Xradiographs of the halves for further description and tephra-layer counts; collected sediment samples for chemical, magnetic and sedimentological analyses; obtained macrofossils and bulk sediment samples for radiocarbon dating; and sampled a representative suite of tephra layers (> 1 mm thick) for chemical analysis, including sampling at 1 cm interval across the Hekla 4 (H4) layer and a Katla (KN) tephra layer (Fig. 2). Identification of tephra layers and chronology Light-colour intermediate to silicic tephra layers and many of the thicker, dark-colour basaltic tephra layers were initially identified visually using colour, grain size and relative stratigraphic position, whereas thin basaltic layers (≤ 2 mm) were identified by the aid of X-radiographs. Visual identification of marker tephra layers was verified by major element analysis of the groundmass glass in tephra clasts using a Cameca SX 50 energy dispersive microprobe and instrumental settings of 15 kV for the accelerating voltage, 15 nA beam-current, and a 10 µm focused beam. Estimated analytical precision is ≈ 1%. An additional 27 tephra layers were linked to their source volcano by microprobe analysis. Finally, we documented the number
Fig. 2. A simplified log of core 94-HV01 indicating main tephra layers, radiocarbon ages (see Table 1), and the corresponding whole-core magnetic susceptibility. K-1500, KE, and KN mark Katla tephra layers; Vö is the Settlement layer; and HA, H3, H4, and H5 indicate tephra layers that originate from Hekla. The T-tephra layer may originate from Hekla and /or Vestmannaeyjar.
and thickness of tephra layers to evaluate whether their frequency and/or thickness affected other sedimentological characteristics. Tephrochronology was used as the main chronological tool in the Nesje cores because of the abundance of identifiable tephra layers. The light-colour Hekla layers (H3, HA, H4, and H5), the Settlement layer (Vö), and the thicker Katla layers (KE and KN) provide the basis of the tephrochronology for the core (Fig. 2). The radiocarbon ages of these tephra layers as they are determined in land-based soil sections are listed in Table 1. In addition, seven bulk acid-insoluble sediment samples and one macrofossil sample were analysed for radiocarbon ages: three at the AMS Laboratory at the University of Aarhus, Denmark (Sveinbjörnsdóttir et al., 1998; Table 1), and five at the AMS Laboratory at Lawrence Livermore, USA (introduced in Table 1). Radiocarbon ages were calibrated into calendar ages using the CALIB calibration program version 3.0.3c (Stuiver & Reimer, 1993a).
VSIC10 9/12/08 1:00 PM Page 229
229
Tephra fall deposits in the watershed
Table 1. Chronology of the Nesje core based on known tephra layers and AMS radiocarbon dates. The age–depth model is based on the ages represented in bold type. Radiocarbon ages have been converted to calendar age using CALIB 3.0.3c 14C calibration program (Stuiver & Reimer, 1993a). All calendar ages are reported as years bp, with present being the top of core (year 1994). H layers, tephra layers erupted from Hekla; K, tephra layers originating from Katla; Vö, Settlement layer Tephra layer or depth (cm) K-1500 Vö (188) 14 C (231) HA (276) 14 C (278) 14 C (278.3) KE (299) H3 (308) KN (340) H4 (401) 14 C (404) 14 C (425) 14 C (499) T-tephra 14 C (499) 14 C (543) H5 (566)
C age (yr bp)
14
360 1180 ± 5* 2060 ± 70 2500 ± 100† 3260 ± 50 2875 ± 45 2850 ± 10† 2880 ± 30 3300 ± 100† 3830 ± 10 4230 ± 70 5060 ± 50 5765 ± 55 6170 ± 50 6750 ± 50 6200
Calendar age [yr bp (1994)] 494 1120 2595 ± 195 2975 ± 60 3055 ± 60 3555 ± 120 4275 ± 10 6615 ± 90 7115 ± 130
Reference (lab no.) Thórarinsson, 1975 Grönvold et al., 1995 this study (CAMS-19977) Róbertsdóttir (1992a) this study (CAMS-19973) Sveinbjörnsdóttir et al., 1998 (AAR-3474) Róbertsdóttir, 1992b Dugmore et al., 1995 Róbertsdóttir, 1992b Dugmore et al., 1995 Sveinbjörnsdóttir et al., 1998 (AAR-3475) this study (CAMS-19974) Sveinbjörnsdóttir et al., 1998 (AAR-3476) this study (CAMS-19975) this study (CAMS-19976) Larsen & Thórarinsson, 1977
*Calculated 14C ages from calendar ages based on Stuiver & Reimer (1993b). †Ages calculated based on accumulation thicknesses in soil sections.
Other sedimentological, chemical and magnetic analyses Particle size of the inorganic sediment fraction (< 2000 µm) was measured with low-angle laser light scattering using a Malvern Long Bed Mastersizer. These data are presented as volume percentages of the analysed fraction. To assess the impact of tephra deposition on chemical and physical characteristics of the clastic sediment, we divided the collected sediment samples into two groups: Group 1 contains no tephra layers and Group 2 contains samples that possibly include tephra (either very thin fallout tephra layers or tephra that had been washed into the lake basin after the primary tephra deposition). The classification is based on the vertical distance of the sample to an identified tephra, and further on the basis of chemical analyses and studies of the core material and X-radiographs. We classified samples as Group 1 (includes 20 samples, whereas Group 2 includes 19 samples) if they were 1 cm or more above the minor tephra layers (1–2 mm thick), but this minimum distance was up to 10 cm for the samples above the thicker tephra layers. Principal component analysis on 35 sedimentological, magnetic and chemical variables measured in the core was used to evaluate which variables to use for discriminant statistical analysis (we continued further statistical
analyses on only variables that had component scores above 0.5 and below – 0.5 on the first principal component factor). Additionally, we used only prehuman-settlement samples for this study because results discussed further below show that postsettlement samples were subjected to very different environmental forcing. We then used discriminant analysis to study whether the samples of Group 1 and Group 2 could be distinguished based on the chosen sedimentological and chemical variables. Samples used for total carbon analysis were milled to < 500 µm and run on a CM5120 furnace device (combusted to 950°C), and then measured on a CO2 coulometer with a detection limit of 0.01 wt %. Biogenic silica was measured on selected subsamples from the core by the procedure described by DeMaster (1981).Values of biogenic silica may to some extent be used to evaluate productivity within lakes, especially diatom production. Duplicates of several samples were run for both total carbon and biogenic silica, and those showed a mean reproducibility of ± 0.5% and ± 4%, respectively. In addition, to evaluate productivity variations in the core, we use these analyses to evaluate possible dilutional effects on magnetic variables. Major and trace element analyses of Ti, Fe, Ni, Cu, Rb, Sr, Zn, Y, Zr, Mn, and Nb were conducted on selected bulk sediment samples and on the sample
VSIC10 9/12/08 1:00 PM Page 230
230
J. Hardardóttir et al.
suite across the H4 and KN tephra layers by energy dispersive X-ray fluorescence analysis (XRF). Analytical error for these elements was calculated as the standard deviation of three replicate standard runs and is ± 0.04 wt%, ± 0.4 wt%, and ± 12 p.p.m., for Ti, Fe, and Zr, respectively. We completed these major and trace element analyses to estimate possible changes in sediment source (including tephra input) to the lake; to evaluate downcore alterations of iron oxides, which are usually noted by loss of Fe relative to Ti and Zr concentrations (Rosenbaum et al., 1995, 1996) because Fe is a mobile element whereas Ti and Zr are immobile under most environmental conditions; and to document the bulk sediment chemical changes across the H4 and KN tephra layers, which will vary significantly with variable influx of tephra particles into the lake. Thus, we use the chemical analyses to assess the subsequent tephra accumulation in the lake after the initial tephra deposition and evaluate the depositional processes affecting this sustained tephra deposition. Additionally, we used sediment magnetic measurements to record the concentration, grain size, and type of magnetic minerals in the sediment. Changes in these components can imply changes in depositional and transport mechanisms and source changes within the lake basin, as well as post-depositional alterations of magnetic minerals in the sediment. For clarity, we use the term ‘magnetite’ for low-coercivity ferrimagnetic minerals, and ‘hematite’ for high-coercivity antiferromagnetic minerals because results from petrographic examination introduced later suggest that these iron oxides are the main magnetic minerals in the core. Whole-core magnetic susceptibility was measured on the unsplit core sections with Bartington MS.2 equipment and a corresponding loop sensor. A suite of sediment magnetic analyses was then performed on 5 cm3 cubes sampled at 5 cm intervals. Magnetite concentration is inferred from magnetic susceptibility (χ) measured on Bartington MS.2 equipment in an alternating magnetic field of ≈ 1 mT at low (0.47 kHz) (χlf ) and high (4.7 kHz) (χhf ) frequencies. Isothermal remanent magnetization (IRM) was introduced to the samples in an Impulse Magnetizer with a 1.2 Tesla (T) magnetic field in forward direction and – 0.3 T and – 0.1 T fields in the opposite direction. The magnetization of the samples was then measured on a Molspin spinner magnetometer. Because magnetite saturates below 0.3 T whereas hematite has much higher coercivity (saturates at fields > 1.2 T), the ratio of –IRM– 0.3 to IRM1.2 (the S-parameter) represents the relative amounts of magnetite and hematite (S-
parameter = 1 indicates that magnetite type minerals are the predominant magnetic minerals). Similarly, the HIRM parameter ((IRM1.2 + IRM– 0.3)/2) can be used to evaluate the amount of hematite in a sample (Thompson & Oldfield, 1981; King & Channel, 1991). Anhysteretic remanent magnetization, here represented as susceptibility of ARM (χ(arm)), was induced in the samples using a decreasing alternating field demagnetization with a peak induction of 100 mT and a biasing direct field of 0.1 mT. When normalized for magnetite content (χ), the ratio χ(arm)/χ can be used to evaluate the domain size of magnetic minerals [often referred to as magnetic grain size (King et al., 1982)] within the sediment. Higher value of χ(arm)/χ suggests smaller magnetic grain size; however, other factors, such as mineral intergrowth, composition, and shape, can complicate interpretations. The magnetic grain sizes introduced in this paper represent only relative changes in magnetic domain size because direct measurement of domain size was not performed. Frequency-dependent magnetic susceptibility (FDMS = (χlf – χhf )/(χlf ) ) determines the presence of very small (superparamagnetic ≈ < 0.3 µm) magnetic grains in the sample; higher values represent increased influence of superparamagnetic grain sizes. Finally, we performed petrographic analysis on polished grain mounts from magnetic separates of representative sediment samples to identify alteration of iron oxides and to verify the type of magnetic minerals inferred from the sediment magnetic measurements.
RESULTS Sedimentological, magnetic and chemical analyses The sediment in the core is olive–grey organic material intercalated with numerous tephra layers (Fig. 2). The grain size within the core is rather monotonous. Silt content varies between 70 and 80% and is highest below 400 cm in the core. Clay content ranges from 10 to 20% and reaches its highest values in the uppermost 200 cm of sediment (Fig. 3). Thin laminations (1– 3 mm thick) are present below ≈ 200 cm in the core, both on the fresh core surface and on X-radiographs. Laminations typically have a sharp upper and lower boundary, and indicate that the sediment has not been affected by bioturbation. Sections of the upper part of the core show some soft sediment deformation that Geirsdóttir et al. (1994) interpreted as sediment liquefaction caused by earthquakes during accumulation of the sediments.
VSIC10 9/12/08 1:00 PM Page 231
Tephra fall deposits in the watershed
231
Fig. 3. The four magnetic zones and selected magnetic, chemical, and physical variables from core 94-HV01. The locations of the main tephra layers and radiocarbon ages (from Table 1) are shown.
Discriminant analysis (Davis, 1986) indicates that pre-settlement samples within the core can be classified into Group 1 (without tephra) or Group 2 (possibly with tephra) with over 85% accuracy based on their total carbon, Ti, Sr, Y, Zr, Fe/Ti, χlf, IRM(1.2 T), HIRM, S-parameter, and χ(arm)/χ values (Table 2). This result indicates that there is significant statistical difference between clastic samples taken close to tephra layers and those taken farther away from them. Thicker tephra layers show up as distinct peaks in the whole-core susceptibility and reflect a high concentration of magnetic minerals within these deposits (Fig. 2). The susceptibility peaks have proved useful for correlating tephra layers among cores from Lake Hestvatn and among cores from other lakes in the area (Geirsdóttir et al., 1995; Hardardóttir et al., 1996). To evaluate the magnetic record of the sediment, we use the measurements made on the discrete sediment samples. On the basis of specific magnetic susceptibility (χ), we divide the core into four zones (I–IV, Table 3, Fig. 3). Throughout the paper we discuss the downcore changes in relation to these magnetic zones. Zone I, which extends in the range c. 3830 – 6200 14C yr bp (depth interval 404 –569 cm),
is characterized by relatively low magnetite and hematite concentrations and by relatively fine magnetic grain size compared with the other zones. Despite high variability, an increase in magnetic grain size and concentration of magnetite and hematite is recorded between 2180 and 3830 14C yr bp, defining zone II (depth interval 255–395 cm). Zone III (1180– 2180 14C yr bp, depth interval 195–250 cm) is marked by a decrease in magnetic grain size and relatively low concentrations of magnetite and hematite. Zone IV extends from the present to 1180 14C yr bp (depth interval 0–190 cm) and is characterized by relatively coarse magnetic grain size and high concentration of both magnetite and hematite. The S-parameter is high throughout the sediment record (> 0.88) and indicates that magnetite is the dominant magnetic phase in the sediment. Slightly higher S-parameter values are seen in zone I. Petrographic analysis of the magnetic separates confirms that magnetite in conjunction with titanomagnetites and hematite are the dominant magnetic minerals within the core. The petrographic analysis also indicates that the magnetic minerals within the sediment have not been subjected to major alteration.
VSIC10 9/12/08 1:00 PM Page 232
232
J. Hardardóttir et al.
Table 2. Results from discriminant analysis between pre-settlement samples of Group 1 (without tephra) and Group 2 (possibly with tephra) in core 94-HV01. The variables were chosen based on results from principal component analysis and include only variables with principal component scores above 0.5 and below – 0.5. Bold text indicates the variables that are statistically different between Group 1 and Group 2 at the 95% significance level. SS, sum of squares; d.f., degrees of freedom; MS, mean squares; F, F-test; P, probability Variable Sand (%) error Silt (%) error % Total carbon error Ti (wt %) error Zn (p.p.m.) error Rb (p.p.m.) error Sr (p.p.m.) error Y (p.p.m.) error Zr (p.p.m.) error Nb (p.p.m.) error Fe/Ti (wt %) error Fe/Zr error Zr/Ti (p.p.m.) error χlf (µm3 kg –1) error IRM1.2T (mAm2 kg –1) error HIRM (Am2 kg –1) error S-parameter error χ(arm)/χ error
SS
d.f.
MS
F
P
Significance
27.397 646.904 28.273 637.657 3.454 15.647 0.467 2.061 441.295 7182.113 11.697 521.546 7486.618 36093.78 580.936 2217.773 12786.73 83850.06 135.965 2939.057 34.451 143.852 0.003 0.029 398.240 34943.30 0.988 5.728
1 37 1 37 11 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37 1 37
27.397 17.484 28.273 17.234 3.454 0.423 0.467 0.056 441.295 194.111 11.697 14.096 11.697 975.508 580.936 59.940 12786.73 2266.218 135.965 79.434 34.451 3.888 0.003 0.001 398.240 944.414 0.988 0.155 415.462 68.336 0.000 0.000 0.003 0.001 23.544 4.194
1.567
0.219
Not significant
4.641
0.208
Not significant
8.167
0.007
Significant
8.387
0.006
Significant
2.273
0.140
Not significant
0.830
0.368
Not significant
7.675
0.009
Significant
9.692
0.004
Significant
5.642
0.023
Significant
1.712
0.199
Not significant
8.861
0.005
Significant
3.354
0.075
Not significant
0.422
0.520
Not significant
6.382
0.016
Significant
6.080
0.018
Significant
6.271
0.017
Significant
4.748
0.036
Significant
5.614
0.023
Significant
2528.439 0.000 0.000 0.003 0.023 23.544 155.180
Table 3. The ages of the four magnetic zones (I–IV) in core 94-HV01. The radiocarbon ages are based on dated tephra layers (Vö, H4, H5; Table 1) and calculated sediment accumulation rates (SAR) (between Vö and HA). Calendar ages are calculated as in Table 1 Magnetic zones IV III II I
Calendar yr bp 0 –1120 1120 –2150 2150 – 4275 4275 –7120
C yr bp
14
0 –1180 118 –2180 2180 –3830 3830 – 6200
Many major and trace element concentrations of selected bulk sediment samples show similar downcore trends, although to some degree they vary independently of each other (Fig. 4). Ti, Cu, Mn, Y, and Zr show good correlation with χ, and the calculated correlation coefficients (r) are 0.82, 0.74, 0.72, 0.71, and 0.68, respectively (number of samples was 63). Most element concentrations show relatively low but increasing concentrations in zone I, higher concentrations in zone II, then a slight decrease in zone III and, finally, high concentrations in zone IV. In the lowermost part of the core (zone I), the total carbon content increases gradually from 2.5 to 4 wt%
VSIC10 9/12/08 1:00 PM Page 233
Tephra fall deposits in the watershed
233
Fig. 4. Results from XRF analyses for core 94-HV01. Concentrations of selected trace and major elements and the fourfold magnetic zonation of the core are shown.
and drops to values between 2 and 3 wt% in zone II. The carbon content increases abruptly in the uppermost part of zone II, and from there the total carbon content remains high to the top of the core. These changes in the total carbon content (Fig. 3), however, do not always correlate with the variations in sediment magnetic parameters and major and trace element values. Biogenic silica values vary substantially throughout the core, with the lowest values around 20 wt% in zone IV, reaching a peak of 99 wt% in zone I, at 475 cm depth (Fig. 3). This peak is represented by only a single analysis and should be regarded with caution until it has been confirmed by further analyses. Below the 99 wt% peak the biogenic silica values are fairly high (between 40 and 60 wt%). Above the peak the biogenic silica content decreases to 20 – 40 wt% and remains at that level through zones II and III. Tephra layers Tephra characteristics and concentration Most of the 73 tephra layers in the 94-HV01 core are < 1 cm thick, and many of them are 1–2 mm thick. However, the important chronostratigraphic tephra layers are usually 1.5 – 4.5 cm thick. Although primary tephra layers are abundant, they represent only ≈ 5% of the total thickness in the core. This percentage is a minimum estimate, and it does not include the tephra that is admixed into the sediment as the result of reworking
of freshly fallen tephra in the surrounding watershed. Such upward mixing is observed above most of the thicker tephra layers, but is not seen beneath them. Most of the tephra layers consist of fine ash, although some of the thicker layers include medium to coarse ash. The tephra layers were counted and their accumulated thicknesses calculated for each magnetic zone (I–IV). Only slight differences are observed in the number of tephra layers between the zones (Table 4). Zone IV has the lowest number of tephra layers (0.8 tephra layers per 100 yr) and the other zones contain between 0.9 and 1.3 tephra layers per 100 yr. The tephra thicknesses vary more between zones. Greatest average thickness of 1.28 mm of tephra per centimetre (8.9 mm per 100 yr) is observed in zone II, whereas zone IV has the lowest average thickness of 0.25 mm cm–1 (Table 4). The trend of total thicknesses calculated on the basis of time differs substantially from the trend of thicknesses per centimetre among the four magnetic zones (Table 4) because sediment accumulation rates vary significantly downcore (Table 5). The difference is even greater when this same variable is calculated in radiocarbon ages, and this demonstrates the need to use calendar ages for all rate calculations. Microprobe analysis of tephra particles Major-element composition of 36 tephra layers was determined by microprobe analyses of the groundmass glass in clasts. The analyses confirm our visual
VSIC10 9/12/08 1:00 PM Page 234
234
J. Hardardóttir et al.
Magnetic zones Number of and sample tephra layers depths (cm) per 100 yr IV (0–190) III (195–250) II (255–395) I (404–569)
0.8 1.0 1.3 0.9
Total thickness (mm) of tephra layers per 100 yr
Number of tephra layers per cm
Total thickness (mm) of tephra layers per cm
4.3 2.7 8.9 3.1
0.05 0.17 0.18 0.16
0.25 0.85 1.28 0.55
Table 4. Ash distribution in core 94HV01 for each magnetic zone (I–IV)
Table 5. Sediment accumulation rates (SAR) in core 94-HV01 based on radiocarbon ages of tephra layers (calibration to calendar ages as in Table 1) Core interval Top of core–K1500 K1500–Vö Vö–HA HA–KE KE–H3 H3–KN KN–H4 H4–T-tephra T-tephra–H5
SAR including tephra (cm per 100 yr)
SAR excluding tephra (cm per 100 yr)
Difference (in %) between the two calculations
15.2 18.2 6.0 6.2 9.4 6.8 8.4 3.8 13.6
14.8 17.7 5.6 5.2 9.0 5.7 7.5 3.6 13.0
2.6 2.7 6.7 16.1 4.3 16.2 10.7 5.3 4.4
identification of all marker layers. Major-element compositions show that the core contains tephra layers derived from at least five volcanic systems; Katla, Hekla, Veidivötn, Vestmannaeyjar, and Grímsvötn volcanoes (Fig. 1a). On the basis of these compositions, 25 tephra layers show affinities typical for the Katla volcano (69% of the analysed layers). Seven of the tephra layers were erupted by the Hekla volcano, and one tephra originates from each of the Grímsvötn, Veidivötn, and Vestmannaeyjar centres (Table 6). The T-tephra originated either from Hekla (as microprobe results of the intermediate tephra grains suggest) or from Vestmannaeyjar (as microprobe analysis of the basaltic tephra grains implies) (Table 6). Alternatively, the T-tephra may represent a deposit that originates from nearly contemporaneous eruptions at both volcanoes. Physical and chemical characteristics across the H4 and KN tephra layers The H4 layer originated from a large Plinian eruption in Hekla at 3830 ± 30 yr bp (Larsen & Thórarinsson, 1977; Dugmore et al., 1995). The fallout H4 tephra layer in the core is made of two components: a 0.5-cm layer of light grey, very fine ash of silicic composition overlain by 1 cm of blackish brown medium ash of intermediate composition. A 2.5-cm layer of grey,
laminated, very fine ash (mixture of the silicic and intermediate components) is found immediately above the fallout tephra layer. Above the laminated layer a slight colour change is seen in the clastic sediment for another 2–3 cm. The increase in Zr and Y concentrations and the decrease in Fe and Cu concentrations across the H4 tephra layer (Fig. 5b) are consistent with an admixture of tephra of dacitic and rhyolitic composition into the clastic sediment. The XRF analyses also represent the compositional difference between the silicic and the intermediate components of the tephra and show that the concentrations of Ti, Fe, Sr, and Zr are higher in the intermediate ash than in the silicic ash component (Fig. 5b). The chemical composition of the laminated layer above the fallout tephra corresponds to the chemical composition of the silicic ash, although it appears to be mixed with the intermediate ash to some extent. The colouring of the clastic sediment noted above the H4 and KN tephra layers represents the influx of tephra into the lake following the initial fallout of the tephras. The KN layer is 4.5 cm thick and is one of two thick prehistoric Katla layers that cover a large region in southern Iceland. Like other Katla layers it is of transitional alkali basalt composition and contains high concentrations of TiO2 and FeO. The high Ti value in Katla deposits is well documented by a change in Ti
Katla (AD 1500) Katla-Eldgjá (AD 934) ″ Veidivötn-Vatnaöldur (Vö) ″ ″ ″ Katla ″ Katla Katla ″ Katla Katla Katla Katla Katla Katla Hekla (HA) Grímsvötn Hekla Katla (KE) Hekla (H3) ″ ″ ″ Katla Katla (KN) ″ Katla Katla
Volcanic system 1 2 ″ 3 ″ ″ ″ 4 ″ 5 6 ″ 7 8 9 10 11 12 13 14 15 16 17 ″ ″ ″ 18 19 ″ 20 21
Tephra no. 8 5 2 7 10 3 3 2 3 8 6 2 8 8 8 10 8 16 11 8 6 16 3 4 3 1 9 8 11 11 9
n 47.04 (0.28) 47.00 (0.54) 49.86 (0.42) 49.92 (0.32) 49.75 (0.20) 47.60 (1.49) 72.19 (0.28) 47.71 (0.34) 72.23 (0.05) 47.73 (0.30) 47.76 (0.20) 50.07 (0.20) 47.60 (0.53) 49.14 (0.15) 47.63 (2.14) 48.33 (0.28) 48.32 (0.32) 47.90 (0.26) 59.98 (5.27) 49.39 (0.26) 54.41 (0.39) 47.30 (0.35) 48.23 (0.74) 58.26 (0.97) 64.77 (2.51) 72.56 47.15 (0.27) 47.02 (0.43) 47.16 (0.18) 46.95 (0.22) 49.42 (0.24)
SiO2 4.74 (0.28) 5.00 (0.50) 3.63 (0.12) 1.99 (0.48) 1.87 (0.09) 3.84 (1.85) 0.28 (0.05) 4.59 (0.06) 0.33 (0.04) 4.56 (0.15) 4.57 (0.06) 2.15 (0.01) 4.60 (0.21) 4.22 (0.15) 5.11 (1.94) 4.45 (0.11) 4.45 (0.17) 4.37 (0.13) 1.49 (0.67) 2.54 (0.06) 2.19 (0.18) 4.47 (0.14) 1.56 (0.29) 1.70 (0.11) 0.95 (0.30) 0.26 4.57 (0.09) 4.47 (0.09) 4.43 (0.09) 4.39 (0.11) 3.99 (0.13)
TiO2 12.97 (0.07) 12.26 (0.87) 16.56 (0.06) 13.85 (0.23) 13.83 (0.15) 13.22 (0.73) 15.05 (0.05) 13.02 (0.14) 13.94 (0.05) 13.07 (0.24) 12.82 (0.11) 13.98 (0.11) 12.60 (0.61) 13.14 (0.11) 13.37 (0.73) 13.08 (0.07) 13.19 (0.50) 13.23 (0.19) 15.81 (1.02) 13.83 (0.15) 13.80 (0.95) 13.12 (0.37) 15.02 (0.48) 14.75 (0.89) 15.11 (0.48) 13.95 13.04 (0.11) 13.04 (0.19) 12.95 (0.09) 13.09 (0.10) 13.12 (0.14)
Al2O3 15.38 (0.25) 16.36 (0.74) 11.43 (0.71) 13.00 (0.44) 13.22 (0.37) 14.89 (1.39) 2.34 (0.03) 15.27 (0.06) 3.73 (0.10) 15.05 (0.26) 15.14 (0.21) 11.79 (0.00) 15.20 (0.48) 14.37 (0.26) 14.60 (0.54) 14.71 (0.29) 14.62 (0.46) 14.77 (0.33) 8.77 (2.49) 13.02 (0.22) 13.00 (0.79) 15.14 (0.54) 11.31 (0.42) 10.98 (1.30) 7.40 (1.69) 4.27 15.25 (0.28) 15.53 (1.33) 15.30 (0.21) 15.24 (0.25) 14.33 (0.29)
FeO 0.26 (0.07) 0.32 (0.12) 0.14 (0.19) 0.26 (0.04) 0.23 (0.10) 0.28 (0.10) 0.14 (0.03) 0.19 (0.06) 0.10 (0.14) 0.22 (0.10) 0.21 (0.11) 0.20 (0.08) 0.26 (0.07) 0.26 (0.07) 0.22 (0.06) 0.24 (0.06) 0.26 (0.07) 0.25 (0.04) 0.25 (0.05) 0.25 (0.04) 0.33 (0.05) 0.27 (0.07) 0.30 (0.08) 0.27 (0.11) 0.28 (0.03) 0.00 0.22 (0.05) 0.25 (0.06) 0.23 (0.06) 0.23 (0.12) 0.27 (0.07)
MnO
Table 6. Chemical composition of 36 tephra layers in the 94-HV01 core as determined by microprobe analyses
5.18 (0.12) 4.86 (0.51) 3.26 (0.35) 6.48 (0.46) 6.59 (0.15) 5.68 (1.00) 0.24 (0.02) 4.89 (0.10) 0.17 (0.00) 4.96 (0.18) 5.05 (0.06) 6.78 (0.05) 5.20 (0.50) 4.54 (0.07) 4.81 (0.15) 4.87 (0.07) 4.84 (0.25) 5.08 (0.10) 2.01 (0.84) 6.48 (0.21) 3.55 (0.99) 5.17 (0.22) 8.18 (0.13) 1.96 (0.16) 0.98 (0.11) 0.15 5.20 (0.09) 5.27 (0.48) 5.36 (0.08) 5.45 (0.06) 4.38 (0.10)
MgO 10.04 (0.22) 9.79 (0.60) 10.02 (0.94) 11.57 (0.65) 11.63 (0.16) 10.65 (0.90) 0.88 (0.02) 9.84 (0.13) 0.94 (0.07) 9.94 (0.16) 9.96 (0.12) 11.80 (0.06) 10.35 (0.48) 9.30 (0.16) 9.82 (0.46) 9.74 (0.17) 9.74 (0.13) 10.02 (0.29) 5.83 (1.39) 11.29 (0.14) 6.71 (0.24) 10.21 (0.18) 12.77 (0.69) 5.89 (0.25) 4.00 (0.84) 1.32 10.22 (0.08) 10.34 (0.66) 10.31 (0.15) 10.47 (0.15) 9.05 (0.18)
CaO 3.04 (0.09) 2.90 (0.22) 3.84 (0.03) 2.46 (0.16) 2.45 (0.09) 2.77 (0.27) 4.17 (0.10) 3.12 (0.08) 4.49 (0.17) 3.05 (0.14) 3.02 (0.15) 2.61 (0.03) 2.99 (0.16) 3.36 (0.11) 3.04 (0.22) 3.12 (0.06) 3.16 (0.22) 3.04 (0.10) 3.65 (0.34) 2.60 (0.10) 3.42 (0.30) 3.03 (0.14) 2.19 (0.27) 3.71 (0.33) 4.12 (0.52) 3.76 3.05 (0.09) 2.86 (0.15) 3.04 (0.07) 2.98 (0.10) 3.55 (0.05)
Na2O
0.63 (0.18) 0.63 (0.10) 0.63 (0.07) 0.22 (0.12) 0.17 (0.10) 0.49 (0.30) 0.00 (0.00) 0.54 (0.11) 0.08 (0.12) 0.60 (0.09) 0.65 (0.07) 0.29 (0.01) 0.55 (0.09) 0.71 (0.08) 0.54 (0.15) 0.54 (0.09) 0.55 (0.07) 0.51 (0.09) 0.72 (0.51) 0.27 (0.05) 1.26 (0.15) 0.51 (0.11) 0.23 (0.07) 0.87 (0.11) 0.30 (0.05) 0.00 0.54 (0.07) 0.51 (0.09) 0.51 (0.08) 0.51 (0.09) 0.94 (0.11)
P2O5
continued on p. 236
0.74 (0.08) 0.87 (0.12) 0.64 (0.09) 0.26 (0.10) 0.25 (0.02) 0.58 (0.34) 4.71 (0.09) 0.84 (0.08) 3.98 (0.04) 0.82 (0.05) 0.84 (0.02) 0.32 (0.06) 0.76 (0.08) 0.96 (0.06) 0.87 (0.21) 0.92 (0.06) 0.88 (0.05) 0.84 (0.05) 1.49 (0.34) 0.34 (0.04) 1.35 (0.17) 0.78 (0.06) 0.20 (0.08) 1.61 (0.25) 2.09 (0.66) 3.74 0.75 (0.03) 0.71 (0.04) 0.70 (0.03) 0.69 (0.04) 0.95 (0.04)
K 2O
VSIC10 9/12/08 1:00 PM Page 235
Tephra fall deposits in the watershed 235
8 10 3 4 14 8 10 4 9 2 4 10 7 2 10 10 12 9 2 2 9 8 11 8 3 11 15
n 57.38 (0.95) 49.41 (0.61) 47.25 (0.13) 53.89 (0.98) 57.93 (1.82) 47.19 (0.57) 58.37 (0.43) 47.02 (0.26) 58.11 (0.63) 76.11 (0.02) 47.37 (0.50) 48.25 (0.19) 46.83 (0.19) 46.69 (0.31) 47.65 (0.27) 47.86 (0.28) 47.60 (0.29) 46.29 (0.18) 58.08 (4.26) 60.45 (0.65) 46.42 (0.29) 48.15 (0.48) 48.15 (0.21) 47.86 (0.28) 49.62 (0.85) 48.73 (0.54) 65.64 (0.91)
SiO2 1.70 (0.45) 4.11 (0.24) 4.11 (0.28) 2.11 (0.74) 1.63 (0.26) 3.79 (0.65) 1.53 (0.15) 3.86 (0.26) 1.55 (0.44) 0.10 (0.14) 4.35 (0.10) 4.35 (0.24) 2.55 (0.10) 4.12 (0.45) 4.43 (0.08) 4.38 (0.19) 4.39 (0.09) 2.66 (0.11) 1.49 (0.34) 1.30 (0.10) 2.60 (0.08) 4.24 (0.16) 4.20 (0.10) 4.37 (0.09) 3.34 (0.34) 4.20 (0.14) 1.36 (0.10)
TiO2 15.66 (2.30) 13.32 (0.58) 13.29 (0.23) 16.38 (4.33) 14.71 (0.83) 13.76 (1.18) 15.32 (1.04) 13.61 (0.34) 15.07 (1.92) 13.46 (0.16) 13.18 (0.13) 12.95 (0.59) 15.69 (0.16) 13.64 (0.48) 13.06 (0.12) 13.23 (0.70) 13.11 (0.15) 15.56 (0.24) 17.05 (3.48) 14.13 (0.56) 15.63 (0.11) 13.08 (0.43) 13.17 (0.14) 13.07 (0.15) 16.30 (0.25) 13.42 (0.53) 14.50 (0.11)
Al2O3 10.27 (2.19) 14.17 (0.53) 15.08 (0.51) 10.58 (3.96) 10.63 (0.91) 14.57 (0.90) 10.12 (0.92) 14.39 (0.61) 10.27 (1.75) 2.06 (0.10) 14.89 (0.22) 15.21 (0.49) 13.16 (0.36) 16.17 (0.06) 15.12 (0.31) 14.91 (0.64) 15.00 (0.23) 13.71 (0.24) 9.06 (1.59) 11.11 (0.47) 13.72 (0.18) 15.27 (0.65) 15.03 (0.31) 15.38 (0.16) 12.33 (0.61) 14.57 (0.56) 6.44 (0.59)
FeO 0.25 (0.13) 0.25 (0.06) 0.26 (0.05) 0.25 (0.14) 0.28 (0.09) 0.19 (0.08) 0.30 (0.08) 0.21 (0.05) 0.28 (0.09) 0.15 (0.07) 0.18 (0.06) 0.25 (0.07) 0.24 (0.06) 0.30 (0.03) 0.20 (0.09) 0.24 (0.05) 0.24 (0.07) 0.25 (0.04) 0.27 (0.01) 0.35 (0.04) 0.19 (0.07) 0.27 (0.05) 0.22 (0.07) 0.25 (0.05) 0.15 (0.00) 0.28 (0.06) 0.17 (0.07)
MnO 2.05 (0.69) 4.51 (0.27) 5.34 (0.34) 2.92 (1.18) 2.87 (1.39) 5.87 (0.73) 2.18 (0.31) 5.74 (0.35) 2.52 (0.95) 0.05 (0.00) 5.29 (0.15) 4.78 (0.18) 7.23 (0.25) 5.17 (0.56) 5.20 (0.07) 4.88 (0.32) 5.18 (0.11) 7.28 (0.56) 1.36 (0.22) 1.54 (0.11) 7.20 (0.13) 4.62 (0.40) 4.84 (0.17) 4.78 (0.14) 3.55 (0.12) 4.71 (0.26) 1.36 (0.14)
MgO 6.51 (0.47) 9.12 (0.17) 10.51 (0.44) 7.68 (0.85) 6.40 (1.50) 10.73 (0.46) 5.99 (0.36) 11.06 (0.23) 6.14 (0.38) 1.38 (0.02) 10.44 (0.40) 9.69 (0.25) 10.86 (0.20) 9.72 (0.20) 9.98 (0.11) 9.99 (0.13) 10.04 (0.12) 10.88 (0.22) 6.32 (2.08) 4.81 (0.20) 10.88 (0.14) 9.44 (0.16) 9.71 (0.28) 9.75 (0.11) 9.73 (0.42) 9.40 (0.25) 3.71 (0.28)
CaO 3.92 (0.56) 3.45 (0.11) 3.21 (0.16) 3.96 (0.83) 3.53 (0.53) 2.78 (0.39) 4.08 (0.19) 2.99 (0.05) 3.90 (0.53) 3.79 (0.21) 3.14 (0.07) 3.16 (0.09) 2.73 (0.04) 3.03 (0.09) 3.10 (0.09) 3.18 (0.11) 3.14 (0.09) 2.66 (0.15) 4.14 (0.36) 3.90 (0.31) 2.68 (0.10) 3.21 (0.10) 3.26 (0.05) 3.17 (0.04) 3.62 (0.15) 3.27 (0.15) 3.84 (0.17)
Na2O 1.32 (0.27) 0.91 (0.11) 0.68 (0.09) 0.95 (0.35) 1.28 (0.31) 0.64 (0.22) 1.35 (0.12) 0.59 (0.02) 1.31 (0.21) 2.86 (0.00) 0.69 (0.12) 0.84 (0.06) 0.44 (0.05) 0.64 (0.08) 0.75 (0.04) 0.77 (0.03) 0.77 (0.05) 0.44 (0.06) 1.45 (0.65) 1.73 (0.05) 0.41 (0.03) 0.92 (0.08) 0.83 (0.06) 0.83 (0.04) 0.82 (0.08) 0.88 (0.08) 2.65 (0.12)
K2O
0.93 (0.28) 0.76 (0.09) 0.27 (0.24) 1.29 (0.48) 0.78 (0.21) 0.46 (0.12) 0.76 (0.12) 0.53 (0.05) 0.84 (0.28) 0.06 (0.08) 0.48 (0.09) 0.52 (0.07) 0.29 (0.08) 0.53 (0.13) 0.51 (0.05) 0.55 (0.08) 0.53 (0.07) 0.27 (0.04) 0.78 (0.02) 0.67 (0.11) 0.28 (0.09) 0.55 (0.05) 0.60 (0.10) 0.54 (0.10) 0.54 (0.14) 0.54 (0.08) 0.34 (0.06)
P2O5
Numbers in parentheses are one standard deviation of the mean; n, number of analyses. *The basaltic tephra component has alkali basalt composition similar to that of the Vestmannaeyjar volcanic system, whereas the intermediate tephra has composition similar to that of the Hekla volcanic system.
22 23 24 ″ 25 ″ ″ ″ ″ ″ ″ 26 27 ″ 28 29 30 31 ″ ″ ″ 32 33 34 ″ 35 36
Tephra no.
236
Hekla? Katla Hekla? ″ Hekla (H4) ″ ″ ″ ″ ″ ″ Katla Vestmannaeyjar? ″ Katla Katla Katla Hekla? /Vestmannaeyjar? (T)* ″ ″ ″ Katla Katla Katla ″ Katla Hekla (H5)
Volcanic system
Table 6. (continued )
VSIC10 9/12/08 1:00 PM Page 236
J. Hardardóttir et al.
VSIC10 9/12/08 1:00 PM Page 237
237
Tephra fall deposits in the watershed (a) Katla (KN) Total carbon
Ti
Sr
%wt. 1 1.5 2 2.5 3 3.5
325
Y
Zr
ppm 35 40 45 50 55
Fe
Cu
%wt. 10 11 12 13 14 15
Zr/Ti ppm 0.005 0.01 0.015 0.02
Depth (cm)
330
335
340
345
350 0
1
2 3 %wt.
4
(b) Hekla (H4) Total carbon
0
Ti
200 400 600 ppm
Sr
%wt. 0 0.5 1 1.5 2
385
150 200 250 300 ppm
Y ppm 50 100
Zr
60
Fe
80 100 120 ppm
Cu
%wt. 4 6 8 10 1214 16
Zr/Ti 0
ppm 0.06
0.12
Depth (cm)
390
395
400
405
410
0
1
2 3 %wt.
4
100 200 300 400 ppm
100 200 300 400 500 ppm
0
40 80 ppm
120
Fig. 5. Concentrations of total carbon and trace and major elements through two tephra layers in core 94-HV01. (a) Katla N (KN); the tephra is indicated in grey. (b) Hekla 4 (H4); the silicic, intermediate, and the mixed laminated parts of the tephra are indicated by successively darker grey colour implying more basaltic composition.
content in the sediment below the tephra, through the tephra, and above it (Fig. 5a). Directly underneath the tephra, the Ti concentration is uniform at ≈ 1.2%, but rises abruptly to 3% within the tephra layer and gradually decreases to a value of ≈ 1.5% at 7 cm above the
tephra layer. Concentrations of Sr and Zr show very similar trends across the KN tephra layer, but other trace and major elements are variable. Carbon contents vary substantially across the studied tephra layers (Fig. 5). The total carbon value is
VSIC10 9/12/08 1:00 PM Page 238
238
J. Hardardóttir et al.
≈ 3% before deposition of both tephra layers, but decreases dramatically to < 0.3% in H4 and is almost undetectable within KN. Above both tephra layers, the total carbon value increases abruptly, but in neither case does it reach the pre-tephra concentrations. Chronology Radiocarbon dating has proved difficult in core 94HV01 because the radiocarbon dates of acid-insoluble bulk sediment samples give consistently older ages than radiocarbon dates from time-equivalent horizons in soil sections (Table 1, CAMS-19973, CAMS-19974, CAMS-19975, CAMS-19976, and CAMS-19977). Three other radiocarbon dates were obtained on a low-density fraction of bulk sediments that formed after the samples had been soaked in HCl (Table 1, samples AAR-3474, AAR-3475, and AAR-3476, Sveinbjörnsdóttir et al., 1998). Although the age discrepancy was less for these samples, they were up to 300 14C yr older than the correlative tephra layers and 14C ages from a core obtained from Lake Vestra Gíslholtsvatn, 10 km east of Hestvatn (Geirsdóttir et al., 1995; Sveinbjörnsdóttir et al., 1998). Sveinbjörnsdóttir et al. (1998) discussed this dating problem in Lake Hestvatn, which probably arises from older plant remains being washed into the lake and incorporated into the sediment. Similar problems have been reported elsewhere, e.g. from Baffin Island, Canada, where Abbott & Stafford (1997) found the sediment– water interface in present-day lakes to be as old as
1000 14C yr bp. To avoid this problem, our chronology is predominantly based on correlation between identified tephra layers in the core and equivalent tephra layers that have been radiocarbon dated in soil sections. Identification of marker tephra layers was performed both visually and by chemical analysis, which confirms the existence of the marker tephra layers shown in Table 1. In Iceland, these tephra layers are used as the basis for tephrochronology because consistent dates have been obtained for most of them (Larsen & Thórarinsson, 1977; Dugmore et al., 1995; Grönvold et al., 1995). An additional radiocarbon age of 5765 ± 55 14C yr bp from 499 cm depth (Table 1, sample AAR 3476) has been used to evaluate the maximum age of the T-tephra. We are aware of the possibility that this date may be up to 200 14C yr too old. Sediment accumulation rates Sediment accumulation rates in the core (both including and excluding tephra layers, Table 5, Fig. 6) have been calculated between the marker tephra layers (Fig. 2) to evaluate trends in material influx into the lake. The difference between the two calculations gives a first-order evaluation of tephra deposition into the lake. Sediment accumulation rates (excluding tephra layers) in zones I, II, and III fluctuate between 3.6 and 13 cm per 100 yr and are highest in the zone between the T-tephra and H5 (Table 5, Fig. 6). This high value may be an overestimate of as much as 5 cm per 100 yr if the 14C age of the T-tephra proves too high. A major
Approximate 14C years BP
7000
6000
Main tephra layers
5000
4000
ZONE I
3000 ZONE II
2000
1000
ZONE III
0
ZONE IV 0
14.8 cm/100 yrs — 2.6%
K-1500 100 17.7 cm/100 yrs — 2.7%
200 5.6 cm/100 yrs — 6.7%
HA H3
KE KN
5.2 cm/100 yrs — 16.1% 9.0 cm/100 yrs — 4.3% 5.7 cm/100 yrs — 16.2%
300
7.5 cm/100 yrs — 10.7%
H4 400 3.8 cm/100 yrs — 5.3%
T
13.0 cm/ 100 yrs 4.4%
500
H5 8000
7000
6000
5000
4000
3000
Calendar years BP
2000
1000
0
Depth (cm) excluding tephra
Vö
Fig. 6. Sediment accumulation rates (cm per 100 yr) in core 94-HV01 excluding tephra layers. The four magnetic zones and the main tephra layers are shown. The mean rate is represented by the thick line, the shaded areas represent the error on the calendar ages that are used for calculations. The percentage shown for each section depicts the difference in rate calculations including and excluding tephra layers.
VSIC10 9/12/08 1:00 PM Page 239
Tephra fall deposits in the watershed change is observed at the Vö tephra, which represents the time of settlement of Iceland. Sediment accumulation rates increase by a factor of three from 5.6 cm per 100 yr below the Vö to 17.7 cm per 100 yr above the tephra (Fig. 6). This value decreases to 14.8 cm per 100 yr at the K-1500 tephra. Incorporation of tephra layers into sediment accumulation-rate calculations changed the results for some sections substantially, e.g. the rate increases for the intervals between H3 and KN and between HA and KE were 16.2 and 16.1%, respectively (Table 5). For most other zones the change was < 10%.
INTERPRETATION OF RESULTS
239
1 both volcanoes are close to the lake (at 105 and 55 km, respectively); 2 these volcanoes have erupted frequently during the last millennia, and eruptions are typically explosive and produce large amounts of tephra (Hekla during an initial Plinian eruption with a very high (> 20 km) eruption column and the subglacial Katla by its phreatomagmatic eruptions); 3 because only tephra layers that are thicker than 1 mm were analysed by microprobe, the tephra frequency may be biased towards the thicker tephra layers from neighbouring volcanoes rather than thin layers from more distant volcanic systems. Of these, however, there is little doubt that Hekla and Katla volcanoes have had the greatest influence on the sedimentological history in the lake.
Depositional processes and impact of tephra Deposition of tephra into the lake systemaa case study Frequency and origin of tephra With the exception of magnetic zone II (1180–2180 14C yr bp, Fig. 3), the tephra frequency is fairly uniform between 0.8 and 1.0 tephra layers per 100 yr (Table 4). The low number of tephra layers in the core compared with modern eruption frequencies (one eruption every 5 yr) can readily be explained by the frequent occurrence of fissure eruptions in Iceland, which are normally associated with low tephra production, and because the tephra in core 94-HV01 originates predominantly from four volcanoes of the 31 active volcanic centres. Also, strong unidirectional winds cause volcanic plumes to be narrow and confine the tephra dispersal to a restricted area downwind from the volcanoes. Because studies indicate that eruption rates in southern Iceland have been stable during Holocene time (Jakobsson, 1979), higher tephra frequencies (1.3 tephra layers per 100 yr) in zone II are puzzling (Table 4). This zone includes the thickest and the greatest number of tephrochronostratigraphic markers (e.g. HA, KE, H3, and KN). Consequently, the average tephra thickness (7 mm) within zone II is the highest in the core. The reason for the increased tephra abundances and thicknesses is unknown, although larger and more explosive eruptions and changes in the prevailing wind direction could cause such an increase without a corresponding change in the long-term average eruption rates. The tephra chemistry shows that at least five active volcanic systems contribute to the tephra deposition into Lake Hestvatn and that most of the tephra layers are from the Katla and Hekla volcanoes (Fig. 1a). Several factors may be the cause of the high frequency of Katla and Hekla tephra layers in the core:
We closely examined two of the tephra layers, H4 from Hekla and KN from Katla, to gain a better knowledge of the tephra influx into the lake after a major explosive eruption and of how the influx affected the lacustrine environment. Both layers have a very sharp lower boundary, as do all tephra layers in the core, which indicates that bioturbation does not affect the deposits on the millimetre scale. Both tephra layers are distinct and massive. H4 is made of two components, one of silicic and the other of intermediate composition (Fig. 5b). The two components of H4 represent the fallout deposit produced by initial phases of the eruption, which is estimated to have lasted for 24–30 h (Larsen & Thórarinsson, 1977). The occurrence of fine laminated ash above the primary fallout tephra may represent either tephra that was resuspended or reworked in the lake, or tephra that was washed into the lake shortly after deposition of the primary deposit. The lowermost part of the laminated layer has total carbon values as low as those within the fallout tephra (Fig. 5b). This low carbon content indicates that little mixing with organic catchment material has occurred and that the lowest part of the layer probably accumulated from resuspended material within the lake. Furthermore, the increase in carbon content in the upper part of the laminated layer may indicate subsequent input of reworked tephra from the catchment area. Concentrations of major and trace elements (Ti, Sr, Zr, Y, Fe, Cu) through the tephra layers show abrupt changes that can be related to the tephra input (Fig. 5). For example, within 1 cm of the base of the tephra layers, the concentration of Ti increases from 1.1% to
VSIC10 9/12/08 1:00 PM Page 240
240
J. Hardardóttir et al.
> 3% in KN, and decreases from ≈ 1.4% to 0.3% in H4. The major and trace element concentrations either decrease or increase exponentially to ≈ 5 cm above the KN and H4 tephra layers, where the concentrations approach background values (Fig. 5). The time over which the tephra input affected the lake basin is evaluated by using the calculated sediment accumulation rates excluding tephra layers. The results show that tephra from H4 was washed into the lake until 66 yr after the H4 eruption, whereas abundant KN tephra particles remained in the system for almost 90 yr. One explanation for the 30% longer influx time for the KN tephra compared with the H4 tephra layer may be that the KN tephra layer is thicker within the lake catchment than the H4 tephra layer, and greater amounts of tephra within the catchment may have supplied the lake with tephra particles for a longer time period. Because of high input of H4 and KN tephra from the catchment to the lake, it is probable that the sediment accumulation rates were higher in the years after the eruption than average sedimentation rates used in our calculation. Thus, we may overestimate the duration of high tephra influx from the catchment. Yet it is likely that traces of tephra were washed into the lake for a longer time, as is suggested by the difference in concentrations of the trace and major elements before and after the tephra deposition. Likewise, concentrations of trace elements within secondary H4 tephra in lake sediments in northern Iceland decreased by half in 36 yr (Thompson et al., 1986), comparable with the time it took the concentration of trace and major elements in Lake Hestvatn to reach equilibrium again. The study of H4 and KN demonstrates that several processes are responsible for the tephra deposition in the lake. The initial impact is caused by direct deposition of fallout tephra to the lake, which is followed by remobilization of tephra in the lake and in the catchment immediately afterwards, as suggested by the laminated mixed layer above the primary H4 tephra layer. For decades after both H4 and KN eruptions, tephra was being washed from the catchment area and redeposited with other material within the lake. Downcore changes Most studies of environmental variability in southern Iceland have focused on the great vegetation and environmental changes coinciding with the settlement of Iceland, although studies of the Holocene environment have increased steadily during the last decade. Available records of Holocene environmental changes
in southern Iceland are primarily based on glacier oscillations (see Gudmundsson, 1997) and pollen studies from mires and lakes (Einarsson, 1961, 1963; Vasari, 1972; Hallsdóttir, 1987, 1995; Vasari & Vasari, 1990; Geirsdóttir et al., 1995; Hallsdóttir et al., 1996). Although the precise timing of the environmental changes established by these studies may be debated, their results indicate that c. 7000 14C yr ago the southern lowlands of Iceland were covered with birch woodland. This forest started to retreat from the area c. 1000–2000 14C yr later, and mires and heath expanded until the birch woodland increased again at c. 2500 14C yr bp. Potential effects of tephra fall on the Icelandic pollen records and the inferred climate record, however, have not been examined rigorously. Our fourfold zonation of core 94-HV01 was based originally on magnetic variables, although the zonation is also observed in many of the sedimentological and chemical variables. Variability within the measured factors was apparently caused by basinwide shifts in the source material and/or changes in depositional processes. Because tephra layers are very common (> 70) in the core, and two of the zonal boundaries approximately coincide in time with the deposition of two significant tephra layers (H4 between zones I and II and Vö between zones III and IV, Fig. 3), the effect of tephra on the lacustrine system has to be evaluated before the sedimentological, magnetic, and chemical downcore changes can be correctly interpreted. These changes are examined in further detail in the following sections. Magnetic zone I The magnetic mineral and element concentrations are low within zone I (3830–6200 14C yr bp; depth interval 404–569 cm), and the χ(arm)/χ values suggest a relatively fine-grained magnetic assemblage (Figs 3 & 4). The low magnetite and hematite concentrations within zone I may to some extent be explained by the overall high amount of biogenic silica within the same zone (Fig. 3). High biogenic silica would dilute the magnetic signal but, because fluctuations of magnetic and biogenic silica concentrations are not in phase within zone I, other factors must also contribute to the changes in magnetic concentrations. Significant variations are observed in the other variables within zone I. Below 500 cm depth in the core, the total carbon content is ≈ 2.5%, but it increases to ≈ 4% above it (Fig. 3). Simultaneously, an increase is observed in Fe concentration whereas concentrations of Nb, Cu, and Sr decrease. Other elements, such as
VSIC10 9/12/08 1:00 PM Page 241
Tephra fall deposits in the watershed Ti and Zr, show less distinct change, although their concentrations tend to decrease slightly (Fig. 4). Sediment accumulation rates in zone I also differ substantially between the lower and upper part of the zone. Using the 6615 ± 90 cal yr bp age for the T-tephra (Table 1), the sediment accumulation rate decreased by a factor of 3.6 across the T-tephra. Because the age of the T-tephra may be as much as 200 14C yr too old, however, this value is a maximum for the decrease in sediment accumulation rate. The total carbon increase after the deposition of the T-tephra may have been caused by either increased lake productivity or a reduction in the clastic sediment influx. A maximum in biogenic silica may indicate that productivity did indeed increase (Fig. 3); however, the biogenic silica peak at 475 cm depth is represented by only one measurement and should be used with caution until it is confirmed. Alternatively, the high sediment accumulation rates below the T-tephra may indicate that a greater amount of inorganic sediment was transported into the lake. Normalizing the total carbon data with calculated sediment accumulation rates to evaluate the total carbon flux, however, shows that variations in sediment accumulation rates do not fully account for the carbon differences. Frequency of tephra layers is somewhat higher before the deposition of the T-tephra than in the section above, but most of the layers are very thin (mostly 1–2 mm thick). Consequently, the post-tephra-fall influx of tephra particles is small and cannot explain the difference in sediment accumulation rates. Although the number of tephra layers is lower in the upper part than in the lower part of zone I, the layers are thicker in the upper part. Thicker tephra layers are likely to have greater impact than thinner layers on the physical and chemical characteristics of the overlying sediment because particles from these tephra layers are washed into the lake for a longer time period. Such an effect from thicker tephra layers is not detected within zone I because this effect would increase the values of IRM and χ, which is opposite to what is observed in the upper part of zone I. Thus, we conclude that tephra deposition had a minor effect on the variations seen within zone I, with the exception of few samples immediately above the T-tephra layer, which show elevated values of IRM and χ. Our data do not allow us to assess to what extent the increase in the carbon value at 500 cm depth in the core reflects decreases in the influx of inorganic material or factors controlling biological productivity, although it is likely that both processes worked simultaneously to increase the total carbon percentages at
241
this time. The high biogenic silica values, elevated carbon concentrations, and the pollen data (Hallsdóttir, 1995) in zone I indicate that this period was characterized by the most favourable climatic conditions for the growth of birch patches within the catchment and for high productivity in the lake. It is, however, unclear whether this climate condition involved higher temperatures, decreased precipitation, and/or changes in seasonality. Magnetic zone II A major change in many of the measured variables is observed between zone I and zone II (2180–3830 14C yr bp; depth interval 255–395 cm). Sediment magnetic variables suggest a high content of both magnetite and hematite (Fig. 3), and a marked increase is observed in concentrations of Rb, Sr, Y and Zr, and in the Zr/Ti ratio (Fig. 4). Concentrations of Ti and Fe are also high, although this change to elevated concentrations occurred in the uppermost part of zone I. These changes represent increased influx of inorganic material compared with organic material, a relationship that is also demonstrated by a lowering of total carbon and biogenic silica values. Although the magnetic mineral and the elemental concentrations fluctuate somewhat, they tend to stay high during the 1650 14C yr period spanned by zone II. Marked changes are also seen in the pollen record from the same core at the boundary between zone I and II (Hallsdóttir, 1995; Hallsdóttir et al., 1995). The birch woodland decreases and grasses, heath, and mires become the dominant vegetation, although woodland patches still exist. Pollen concentration (pollen grains cm–3) is also less than in most samples within zone I. The change between zone I and zone II is marked by the deposition of the H4 tephra layer. As one of the largest postglacial tephra layers to have been deposited in Iceland (Larsen & Thórarinsson, 1977), H4 is likely to have had large environmental effects. Therefore, one of the principal questions in this study is whether the physical and chemical changes that are delineated above and lasted for over 1650 14C yr in zone II are solely due to this tephra fall. The timing of the change can offer substantial clues to this question. Some of the geochemical and magnetic variables such as Ti, Sr, Zr, Nb, total carbon, and χ(arm)/χ begin to change towards values observed in zone II before the deposition of the H4 tephra (Figs 3 & 4). Furthermore, Hallsdóttir et al. (1995) showed that diminished pollen concentrations and a change toward declining woodland are evident in the sediment samples from core 94-HV01 before H4 was deposited. The
VSIC10 9/12/08 1:00 PM Page 242
242
J. Hardardóttir et al.
initiation of chemical and vegetational changes before the H4 tephra fall suggests that environmental factors other than the tephra deposition probably initiated these changes, although the tephra fall may have speeded them up or acted as a final blow to a changing system. As described above, the deposition of the H4 layer is characterized by major changes in chemical elements (Fig. 5b). However, the trace and major element concentrations reverted to approximately pre-eruption values < 100 yr after the deposition of the tephra layer. The return to pre-eruption values suggests that, although the tephra deposition may have caused the major change observed in the measured variables, the subsequent input of the H4 tephra particles cannot have been solely responsible for the large changes in chemical concentrations sustained for 1650 14C yr after the tephra deposition. The same conclusion is suggested by some of the changes in the trace and major elements themselves, such as the concentration of Cu, which decreases within the H4 tephra layer from about 80 to 20 p.p.m., but increases to values of 90–100 p.p.m. in the samples above the tephra layer (Figs 4 & 5b). The tephra deposition may have caused other significant changes in the watershed that may not have been retrogressive. Vegetation in the catchment area appears to have undergone a shift from birch woodland to increased heath, grass, and sedge vegetation simultaneous with the tephra deposition (Hallsdóttir, 1995; Hallsdóttir et al., 1995). In most cases, it takes only a few centuries or less to restore the original vegetation after a tephra fall (Thórarinsson, 1961; Lotter et al., 1995; Wilmshurst & McGlone, 1996). Rapid recovery of vegetation is especially the case if the tephra layer is < 5 cm thick (Fridriksson, 1981), as the thickness of H4 is in soil sections around Lake Hestvatn. Other effects of volcanic eruptions, however, have to be evaluated. Fluorine in the tephra, especially in the fine ash, can damage and even kill vegetation over large areas (Fridriksson, 1981). Furthermore, several studies have suggested correlations between atmospheric loading from volcanic eruptions and atmospheric temperature changes (e.g. Self et al., 1981; LaMarche & Hirschboeck, 1984; Rampino & Self, 1984; Sigurdsson, 1990). Significant cooling of surface air temperatures in the Northern Hemisphere followed the large Laki eruption in Iceland (ad 1783 –1784) (Angell & Korshover, 1985; Wood, 1992; Thórdarson et al., 1993, 1996; Fiacco et al., 1994) and the Icelandic Eldgjá eruption in ad 938 ± 4 (Stothers, 1998). No studies are available that estimate the effect of the H4 eruption on Icelandic climate, but atmospheric cooling caused by the H4 eruption has been
proposed to be a major factor in the decline in pine pollen in Scotland during Middle Holocene times (Blackford et al., 1992). Similar pollen–tephra studies from the northern part of Ireland, however, indicate no such temporal link between the pollen decline and the H4 eruption (Hall et al., 1994; Dwyer & Mitchell, 1997; Caseldine et al., 1998). Aerosol loading from the H4 eruption may have caused temporary air-temperature cooling, which could be one of the many factors affecting the vegetation shift and the resulting erosional changes observed in the Lake Hestvatn catchment. The increased concentrations of trace and major elements, magnetite, and hematite suggest that erosion within the catchment area increased substantially after the deposition of H4. The increased erosion is reflected in higher sediment accumulation rates, although part of the incoming material must have been reworked H4 tephra. Unless other factors are affecting the environment, a high rate of erosion in response to tephra deposition is not a major long-term factor because the catchment would rehabilitate after a few centuries. Changes in climate, either increased precipitation or slight cooling (or both), are likely to have been one of the underlying environmental factors that caused the initial long-term changes seen in the core, and are probably responsible for the permanent change in vegetation. Another possible cause for the sustained environmental changes initiated around the H4 deposition is continued volcanigenic input from Katla, Hekla, and other volcanoes into the lake sediments. Statistical analysis of clastic samples of Group 1 (without tephra) and Group 2 (possibly with tephra) (Table 2) shows that magnetic susceptibility, HIRM, Ti, Sr, Zr, and Y are significantly higher for samples that have tephra than for samples without tephra, whereas the Fe : Ti ratio is significantly lower for samples that have tephra. Major and trace element concentrations, however, differ substantially between tephras of basaltic, intermediate, and silicic concentrations. The elevated values of Ti and Sr within zone II may thus partly reflect increased input of tephra of basaltic and intermediate composition into the lake, whereas high concentrations of Zr, Y, and Zr/Ti suggest higher input of relatively silicic tephra particles than in zones I and III. In fact, the greatest frequency and thicknesses of tephra layers are observed within zone II in core 94-HV01 (Table 4), and this zone includes the greatest number of andesite to rhyolite tephra layers (most of the H tephra layers). Consequently, interaction of several environmental impacts caused the sustained changes observed in zone II:
VSIC10 9/12/08 1:00 PM Page 243
Tephra fall deposits in the watershed 1 change in vegetation composition (Hallsdóttir, 1995; Hallsdóttir et al., 1995) caused by either lowered temperatures and/or increased precipitation, and related increase in erosion within the watershed; 2 the combined effects of the H4 tephra deposition, including high influx of tephra particles into the lake, increased erosion within the watershed caused by stress on the vegetation cover, and possible temporary lowering of air temperatures; 3 frequent deposition of basaltic, intermediate, and silicic tephra layers within zone II. Magnetic zone III An abrupt change in the chemical and magnetic variables is observed at the boundary between zone II and zone III (1180 –2180 14C yr bp; depth interval 195– 250 cm) (Figs 3 & 4). Total carbon content abruptly increases and the amounts of magnetite and hematite decrease to values similar to those seen in the upper part of zone I. Magnetic grain size and concentrations of many trace and major elements (Ti, Sr, Y, Zr, and Nb) decrease. However, a slight trend towards increased values in these variables is observed in the upper part of this zone before the main environmental changes associated with the settlement of the region started. The decline in chemical and magnetic variables suggest that, between c. 1180 and 2180 14C yr bp, the input of inorganic material into Lake Hestvatn and erosion in the catchment was generally lower than in zone II, although it increased again in the decades or centuries before the settlement. Our results do not differentiate among possible causes for the decreased erosion, although part of the decline is attributed to decreased tephra frequency and thickness compared with zone II (Table 4). A modest increase in birch pollen and a small decrease in grass concentrations are observed in core 94-HV01 at depths corresponding to zone III (Hallsdóttir, 1995; Hallsdóttir et al., 1995). This floral change may be caused by minor rise in temperatures and/ or decreased precipitation, and suggests slight stabilization of the vegetation cover in the catchment consistent with the decreased erosion we observe in our record. We therefore suggest that a combination of (i) decreased tephra input and (ii) slight temperature rise and /or precipitation decrease was responsible for the decrease in chemical and magnetic concentrations observed in zone III. Magnetic zone IV A significant increase is observed in magnetic zone IV
243
(present–1180 14C years bp; depth interval 0–190 cm) in most of the measured variables including: (i) type, concentration, and grain size of magnetic minerals (Fig. 3); (ii) most trace and major element concentrations (Fig. 4); and (iii) sediment accumulation rates (Table 5, Fig. 6), whereas biogenic silica concentration and particle size decrease (Fig. 3). These changes coincide with the Norse settlement of Iceland and the deposition of the Vö tephra layer at 1180 14C yr bp. Contemporaneous changes are recorded in various other studies in Iceland. Pollen studies from South Iceland reveal that grass heath and mires expanded during this time and the scattered birch woodland disappeared (Hallsdóttir, 1987). A three- to six-fold increase in sediment accumulation rates in peat sections shows that soil erosion increased substantially at the same time (e.g. Thórarinsson, 1961; Gudbergsson, 1975). Similarly, a greater than three-fold increase in sediment accumulation rates is observed in the Lake Hestvatn core between zones III and IV, but the sediment accumulation rates decreased again by 20% around ad 1500 (360 14C yr bp, Fig. 6). Studies by Dugmore & Erskine (1994) in South Iceland indicate that soil erosion increased around ad 1500 and that the effects of local rather than regional factors became more pronounced for soil development. Thus, the decrease in sediment accumulation rates in Lake Hestvatn at ad 1500 may be spurious, and it suggests that we did not recover the sediment–water interface during coring, which would have the apparent effect of lowering the sediment accumulation rate above the K 1500 tephra layer. Although factors such as volcanic eruptions and climate variations may to some extent have contributed to the increase in sediment accumulation rates in Lake Hestvatn and in soil sections at 1180 14 C yr bp, the main cause for the regional soil and vegetation changes at this time was primarily the environmental modifications initiated by the habitation of Iceland (Bjarnason, 1942; Thórarinsson, 1961; Hallsdóttir, 1987; Arnalds, 1988). Animal grazing and deforestation by the early settlers weakened the vegetation cover and made the land more susceptible to erosion. The elevated concentrations of both magnetic minerals and trace and major elements at this time in the Hestvatn core correlate well with elevated sediment accumulation rates that resulted from increased erosion. Because the period from 1180 14C yr bp to the present is greatly affected by farming and other land use, it is not possible to distinguish the effects of volcanic eruptions or climate change from other environmental disturbances within zone IV.
VSIC10 9/12/08 1:00 PM Page 244
244
J. Hardardóttir et al.
CONCLUSIONS Sedimentological, sediment magnetic, and chemical analyses on lacustrine sediment samples from core 94-HV01 obtained from Lake Hestvatn, southern Iceland, show that although individual variables are difficult to interpret, together they complement each other and offer significant evidence of the Holocene depositional processes in Lake Hestvatn. The study shows that the core can be divided into four zones (I–IV) that are primarily identified by changes in sediment magnetic variables, and to some extent by general chemical distinctions. Zone I (3830 – 6200 14C yr bp) represents a low influx of inorganic material, as is characterized by relatively fine magnetic grain size, low concentrations of magnetic minerals and trace and minor elements, and high concentrations of biogenic silica and total carbon. Variations within this zone probably reflect both changes in the input of inorganic material and changes in productivity within the lake. Except during the deposition of the T-tephra layer, tephra deposition had only a minor effect on the sediment accumulation within zone I. This zone represents the best conditions (possibly higher temperatures and /or less precipitation) for birch vegetation and high productivity in the catchment that are observed within core 94-HV01, although these optimal growth conditions started to deteriorate towards the upper part of the zone. The shift from zone I to zone II (2180 –3830 14C yr bp) is marked by the deposition of the H4 tephra layer. The tephra was deposited during an initial fallout phase and during subsequent reworking of the fallout tephra within the lacustrine sediments and the watershed. A large increase in magnetic mineral concentration, magnetic grain size, and concentrations of many chemical substances is observed at this time. These variations indicate that dilution from biogenic silica was less in zone II than in zone I, and that erosion within the watershed increased significantly after the deposition of the H4 tephra layer, although erosion had already started to increase within zone I. The cause of the sustained erosional change in zone I is complex. Vegetational change towards decreased birch patches and increased grasses and mire vegetation may have increased the erosion within the catchment, and this change was accelerated after the deposition of the H4 tephra. The H4 deposition may have caused additional stress on the vegetation by high fluorine concentration, the abrasive character of the tephra particles, and possible brief lowering of air tempera-
tures. Additionally, the high frequency of basaltic, intermediate, and silicic tephra deposition after the H4 deposition greatly affected the chemical and magnetic mineral concentration within zone II, and helped maintain the high erosion rates within the watershed during this time period. Our study across the H4 and KN tephra layers indicates that it took the Hestvatn catchment 66 and 90 yr, respectively, to regain a steady state in terms of trace and major element influx to the lake after the deposition of these tephra layers. Zone III (1180–2180 14C yr bp) has relatively finer magnetic grain size and decreased concentrations of magnetic minerals and of many trace and minor elements compared with zone II. Erosion within the catchment declined during this time, reflecting both decreased tephra input and stabilization of the vegetation cover. This period is represented by a slight shift in vegetation towards increased birch density (Hallsdóttir, 1995). This zone may represent a favourable climate for tree expansion (higher temperatures and/ or decreased precipitation), which is also represented in higher biogenic silica concentrations within zone III. Increased concentrations of magnetic minerals and chemical substances in the samples just below the deposition of the Vö tephra layer (1180 14C yr bp) suggest increased erosion at that time. Zone IV (present–1180 14C yr bp) depicts the postsettlement sediment accumulation, which has been greatly affected by extensive land use within the area. Erosion increased significantly during this time and is characterized by both increased magnetic grain size and greater concentrations of magnetic minerals and trace and major elements. However, it is not possible to infer a climatic signal for this period, because of human disturbance within the region. Consequently, the variability in magnetic, chemical, and sedimentological characteristics observed in the sediments of core 94-HV01 are caused by an interaction of four main processes. 1 Climatic variability with either declining temperatures and/or increased precipitation that could decrease vegetation cover, and thus increase erosion within the catchment and lower the productivity in the lake. 2 Deposition of thick tephra deposits, such as the H4 tephra layer, within the catchment. 3 Frequent deposition of thin tephra layers, which helped sustain high erosion rates. 4 The settlement of Iceland c. 1100 yr ago, which caused major vegetation modifications and a subsequent increase in erosion rates as trees were cut down and the land was used as pasturage for animals.
VSIC10 9/12/08 1:00 PM Page 245
Tephra fall deposits in the watershed
ACKNOWLEDGEMENTS This work has been funded by the Icelandic Research Council (Grants 95-N-043, 961830096, and 961830097) and the National Science Foundation (Grants ATM9224554 and ATM-9531397). We would like to thank everyone who helped with fieldwork at Lake Hestvatn, as well as G. Larsen, who helped with visual tephra identification, and Drs M. Hallsdóttir and H. Norddahl for assistance with core opening and sampling. R. Kihl and Dr F. Luiszer helped with analytical procedures. The use of the USGS Magnetic Laboratory in Lakewood, Colorado, supported by USGS Global Change and Climate History Program, is appreciated. Comments from the co-editor Dr N. Riggs, the reviewers Drs P. Pringle and S. Königer, and from Drs J. T. Andrews and R. Reynolds greatly helped to improve the manuscript. This is PALE (Palaeoclimate of Arctic Lakes and Estuaries) Contribution 103.
REFERENCES Abbott, M.B. & Stafford, T.W. (1997) Radiocarbon geochemistry of modern and ancient Arctic lake systems, Baffin Island, Canada. Quat. Res., 45, 300 –311. Angell, J.K. & Korshover, J. (1985) Surface temperature changes following the six major volcanic episodes between 1780 and 1980. J. Clim. appl. Meteorol., 24, 937–951. Arnalds, A. (1988) Landgaedi á Islandi fyrr og nú. Graedum Island, I, 13 –31. Baillie, M.G.L. & Munro, M.A.R. (1988) Irish tree rings, Santorini and volcanic dust veils. Nature, 322, 344 –346. Birks, H.J.B. (1994) Did Icelandic volcanic eruptions influence the postglacial vegetational history of the British Isles. Trends Ecol. Evol., 9, 312–314. Bjarnason, H. (1942) Ábúd og örtröd. Ársrit Skógraektarfélags Íslands, 1942, 8 – 40. Björck, S., Ingólfsson, Ó., Haflidason, H., Hallsdóttir, M. & Anderson, N.J. (1992) Lake Torfadalsvatn: a high resolution record of the North Atlantic ash zone I and the last glacial–interglacial environmental changes in Iceland. Boreas, 21, 15 –22. Blackford, J.J., Edwards, K.J., Dugmore, A.J., Cook, G.T. & Buckland, P.C. (1992) Icelandic volcanic ash and the mid-Holocene Scots pine (Pinus sylvestris) pollen decline in northern Scotland. Holocene, 2, 260 –265. Caseldine, C., Hatton, J., Huber, U., Chiverrell, R. & Wooley, N. (1998) Assessing the impact of volcanic activity on mid-Holocene climate in Ireland: the need for replicate data. Holocene, 8, 105 –111. Davis, J.C. (1986) Statistics and Data Analysis in Geology, 2nd edn. Wiley, New York. DeMaster, D.J. (1981) The supply and accumulation of silica in the marine environment. Geochim. Cosmochim. Acta, 45, 1715 –1732. Dugmore, A.J. & Erskine, C.C. (1994) Local and regional patterns of soil erosion in southern Iceland. In: Environ-
245
mental Change in Iceland (Eds Stötter, J. & Wilhelm, F.), Münchener Geographische Abhandlungen, B12, 63 –78. Dugmore, A.J., Cook, G.T., Shore, J.S., Newton, A.J., Edwards, K.J. & Larsen, G. (1995) Radiocarbon dating tephra layers in Britain and Iceland. Radiocarbon, 37, 379 –388. Dwyer, R.B. & Mitchell, F.J.G. (1997) Investigation of the environmental impact of remote volcanic activity on north Mayo, Ireland, during the mid-Holocene. Holocene, 7, 113 –118. Einarsson, M. (1976) Vedurfar Á Íslandi. Idunn, Reykjavík. Einarsson, Th. (1961) Pollenanalytische Untersuchungen zur Spät- und postglazialen Klimageschichte Islands. Sonderveröff. geol. Inst. Univ. Köln, 6, 1–52. Einarsson, Th. (1963) Pollen-analytical studies on the vegetation and climate history of Iceland in late and postglacial times. In: North Atlantic Biota and Their History (Eds Löve, Á. & Löve, D.), pp. 355 –365. Pergamon Press, Oxford. Fiacco, R.J., Thórdarson, Th., Germani, M.S., et al. (1994) Atmospheric aerosol loading and transport due to the 1783 –1784 Laki eruption in Iceland, interpreted from ash particles and acidity in the GISP2 ice core. Quat. Res., 42, 231–240. Fridriksson, S. (1981) Áhrif gjósku á gródur. Freyr, 77, 668 – 671. Geirsdóttir, Á., Hallsdóttir, M., Larsen, G., et al. (1994) Setmyndun í sunnlenskum stöduvötnum. In: Geoscience Society of Iceland Spring Meeting, Conference Proceedings, Reykjavík, 7. Geirsdóttir, Á., Hallsdóttir, M. & Hardardóttir, J. (1995) Environmental change in southern Iceland during the Holocene. In: Second Annual PALE Research Meeting, Conference Proceedings, Eatonville, WA, 51–56. Grönvold, K., Óskarsson, N., Johnsen, S.J., et al. (1995) Ash layer from Iceland in the Greenland GRIP core correlated with oceanic and land sediments. Earth planet. Sci. Lett., 135, 149 –155. Gudbergsson, G. (1975) Myndun móajardvegs í Skagafirdi. J. agric. Res. Iceland, 7, 20 – 45. Gudmundsson, H. (1997) A review of the Holocene environmental history of Iceland. Quat. Sci. Rev., 16, 81–92. Haflidason, H. & Einarsson, Á. (1989) Repeated tephra falls and their effect on the nutrient loading in a volcanogenic eutrophic lake, north Iceland. Terra Abstr., 1, 227. Haflidason, H., Larsen, G. & Ólafsson, G. (1992) The recent sedimentation history of Thingvallavatn, Iceland. Oikos, 64, 80–95. Hall, V.A., Pilcher, J.R. & McCormac, F.G. (1994) Icelandic volcanic ash and the mid Holocene Scots pine (Pinus sylvestris) decline in the north of Ireland: no correlation. Holocene, 4, 79–83. Hallsdóttir, M. (1987) Pollen analytical studies of human influence on vegetation in relation to the Landnám tephra layer in southwest Iceland. LUNDQUA Thesis, University of Lund. Hallsdóttir, M. (1995) On the pre-settlement history of Icelandic vegetation. Icelandic Agric. Res. Sci., 9, 17–29. Hallsdóttir, M., Geirsdóttir, Á., Hardardóttir, J., et al. (1995) Umhverfisbreytingar á Sudurlandi á nútíma í ljósi setmyndunar í stöduvötnum. In: Geoscience Society of Iceland Spring Meeting, Conference Proceedings, Reykjavík, 53 –55. Hallsdóttir, M., Geirsdóttir, Á., Róbertsdóttir, B.G., et al. (1996) Gródurfar á Sudurlandi á sídari hluta
VSIC10 9/12/08 1:00 PM Page 246
246
J. Hardardóttir et al.
nútímaaí ljósi frjógreiningar á seti Vestra Gíslholtsvatns í Holtum. In: Geoscience Society of Iceland, Spring Meeting, Conference Proceedings, Reykjavík, 48 –50. Hardardóttir, J., Geirsdóttir, Á. & Sveinbjörnsdóttir, A.E. (1996) New findings regarding the deglaciation of southern Iceland; evidence from lake sediments. Geol. Soc. Am. Abstr. Prog., 28, 57. Jakobsson, S. (1979) Petrology of Recent Basalts of the Eastern Volcanic Zone, Iceland. Acta Nat. Islandica, 26, p. 103. King, J., Banerjee, S.K., Marvin, J. & Özdemir, Ö. (1982) A comparison of different magnetic methods for determining the relative grain size of magnetite in natural materials: some results from lake sediments. Earth planet. Sci. Lett., 59, 404–419. King, J.W. & Channel, J.E.T. (1991) Sedimentary magnetism, environmental magnetism, and magnetostratigraphy. Rev. Geophys. Suppl., 29, 358 –370. LaMarche, V.C. & Hirschboeck, K.K. (1984) Frost rings in trees as records of major volcanic eruptions. Nature, London, 307, 121–126. Larsen, G. (1981) Tephrochronology by microprobe glass analysis. In: Tephra Studies (Eds Self, S. & Sparks, S.J.), pp. 95–102. D. Reidel, Dordrecht. Larsen, G. (1982) Gjóskutímatal Jökuldals og nágrennis. In: Eldur Er Í Nordri (Eds Thórarinsdóttir, H., Óskarsson, Ó.H., Steinthórsson, S. & Einarsson, Th.), pp. 51– 65. Sögufélagid, Reykjavík. Larsen, G. (1984) Gjóskurannsóknir vegna fornleifauppgraftrar í Herjólfsdal, Vestmannaeyjum. Report No. 8204, Nordic Volcanological Institute, Reykjavík. Larsen, G. & Thórarinsson, S. (1977) H4 and other acid Hekla tephra layers. Jökull, 27, 28– 46. Larsen, G., Gudmundsson, M.T. & Björnsson, H. (1996) Gjóskulög í Tungnaárjökli: gossaga, aldur íss og dvalartími í jökli. In: Geoscience Society of Iceland Spring Meeting, Conference Proceedings, Reykjavík, 16. Lotter, A.D., Birks, H.J.B. & Zolitschka, B. (1995) Lateglacial pollen and diatom changes in response to two different environmental perturbations; volcanic eruption and Younger Dryas cooling. J. Paleolimnol., 14, 23 – 47. Nesje, A. (1992) A piston corer for lacustrine and marine sediments. Arctic Alpine Res., 24, 257–259. Nesje, A.K.S., Elgersma, A. & Dahl, S.O. (1987) A piston corer for lake sediments. Norsk geogr. Tidskr., 41, 123–125. Rampino, M.R. & Self, S. (1984) Sulphur-rich volcanic eruptions and stratospheric aerosols. Nature, 310, 677–679. Róbertsdóttir, B.G. (1992a) Thrjú forsöguleg gjóskulög frá Heklu, HA, HB og HC. In: Geoscience Society of Iceland Spring Meeting, Conference Proceedings, Reykjavík, 6 –7. Róbertsdóttir, B.G. (1992b) Forsöguleg gjóskulög frá Kötlu, ádur nefnd ‘Katla 5000’. In: Geoscience Society of Iceland Spring Meeting, Conference Proceedings, Reykjavík, 8–9. Rosenbaum, J.G., Reynolds, R.L., Drexler, J., Best, P.J. & Rodbell, D. (1995) Geochemical proxies for heavymineral content in sedimentsaconstraints on interpretations in environmental magnetism. In: IUGG, XXI General Assembly, Boulder, CO, A95. Rosenbaum, J.G., Reynolds, R.L., Adam, D.P., Drexler, J., Sarna-Wojcicki, A.M. & Whitney, G.C. (1996) Record of middle Pleistocene climate change from Buck Lake, Cascade Range, southern Oregona evidence from sediment magnetism, trace-element geochemistry, and pollen. Geol. Soc. Am. Bull., 108, 1328–1341.
Self, S., Rampino, M.R. & Barbera, J.J. (1981) The possible effects of large 19th and 20th century volcanic eruptions on zonal and hemispheric surface temperatures. J. Volcanol geothermal Res., 11, 41– 60. Sigurdsson, H. (1990) Evidence of volcanic loading of the atmosphere and climate response. Palaeo, 89, 277–289. Steinthórsson, S. (1982) Gjóskulög í jökulkjarna frá Bárdarbungu. In: Eldur Er I Nordri (Eds Thórarinsdóttir, H., Oskarsson, O.H., Steinthórsson, S. & Einarsson, T.), pp. 361–368. Sögufélagid, Reykjavík. Stothers, R.B. (1998) Far reach of the tenth century Eldgjá eruption, Iceland. Climate Change, 715 –723. Stuiver, M. & Reimer, P.J. (1993a) Extended 14C data base and revised CALIB 3.0 14C age calibration program. Radiocarbon, 35, 215 –230. Stuiver, M. & Reimer, P.J. (1993b) High precision bidecadal calibration of the radiocarbon time scale, ad 1950–500 bc and 2500 – 6000 bc. Radiocarbon, 35, 1–33. Sveinbjörnsdóttir, Á.E., Heinemeier, J., Kristensen, P., Rud, N., Geirsdóttir, Á. & Hardardóttir, J. (1998) 14C AMS datings of Icelandic lake sediments. Radiocarbon, 40, 865 – 872. Thompson, R. & Oldfield, F. (1981) Environmental Magnetism. Allen & Unwin, London. Thompson, R., Bradshaw, H.W. & Whitley, J.E. (1986) The distribution of ash in Icelandic lake sediments and the relative importance of mixing and erosion processes. J. Quat. Sci., 1, 3 –11. Thórarinsson, S. (1958) The Öraefajökull Eruption of 1362. Acta Nat. Islandica II, 2. Thórarinsson, S. (1961) Uppblástur á Íslandi í ljósi öskulagarannsókna. Ársrit Skógraektarfélags Íslands, 1961, 17–55. Thórarinsson, S. (1967) The eruption of Hekla in historical times. A tephrochronological study. The eruption of Hekla 1947– 48. Social Sci. Islandica, 1, 1–170. Thórarinsson, S. (1968) Heklueldar. Sögufélagid, Reykjavík. Thórarinsson, S. (1975) Katla og annáll Kötlugosa. Árbók Ferdafélags Íslands, 1975, 129 –149. Thórarinsson, S. (1981) The application of tephrochronology in Iceland. In: Tephra Studies (Eds Self, S. & Sparks, S.), pp. 109 –134. D. Reidel, Dordrecht. Thórdarson, Th., Self, S. & Steinthórsson, S. (1993) Aerosol loading of the Laki fissure eruption and its impact on climate. EOS (Trans. Am. geophys. Union), 74, 106. Thórdarson, Th., Self, S., Óskarsson, N. & Hulsebosch, T. (1996) Sulphur, chlorine and fluorine degassing and atmospheric loading by the 1783–84 ad Laki (Skaftár Fires) eruption in Iceland. Bull. Volcanol., 58, 205 –225. Vasari, Y. (1972) The history of the vegetation of Iceland during the Holocene. In: Climatic Changes in Arctic Areas During the Last Ten Thousand Years (Eds Vasari, Y., Hyvärinen, H. & Hicks, S.), Acta Universitatis Ouluensis, A3, G-1, 239 –251. Vasari, Y. & Vasari, A. (1990) L’histoire Holocene des lacs Islandais. In: Extrait de Pour Jean Malaurie (Ed. Devers, S.), pp. 277–293. Plon, Paris. Wilmshurst, J.M. & McGlone, M.S. (1996) Forest disturbance in the central North Island, New Zealand, following the 1850 BP Taupo eruption. Holocene, 6, 399 – 412. Wood, C.A. (1992) Climatic effects of the 1783 Laki eruption. In: The Year Without a Summer? (Ed. Harrington, C.R.), pp. 58–77. Canadian Museum of Nature, Ottawa, Ont.
VSIC11 9/12/08 1:00 PM Page 247
Spec. Publs int. Ass. Sediment. (2001) 30, 247–261
Late Pleistocene–Holocene volcanic stratigraphy and palaeoenvironments of the upper Lerma basin, Mexico M. CABALLERO*, J. L. MACÍAS*, S. LOZANO-GARCÍA*, J. URRUTIAF U C U G A U C H I * and R . C A S T A Ñ E D A - B E R N A L * *Instituto de Geofísica, UNAM, Coyoacán 04510, México D. F., Mexico
ABSTRACT Detailed analysis of a series of subaerial and lacustrine volcaniclastic sequences in the upper Lerma basin provides evidence of the eruptive history of Nevado de Toluca volcano. These data, together with magnetic susceptibility and microfossil analyses from the studied sequences, are used to document the environmental evolution of this area during the last 40 kyr. The activity of Nevado de Toluca includes three welldocumented dome destruction events that occurred at c. 37 000, c. 28 000, and c. 15 000 yr bp, and two Plinian eruptions at c. 24 000 yr bp (Lower Toluca Pumice) and c. 11 600 yr bp (Upper Toluca Pumice), the latest having the most extensive effect in the upper Lerma and neighbouring basins. Monogenetic volcanism at the Tres Cruces volcano by c. 8500 yr bp is also recorded in the upper Lerma basin. Volcanic activity had a great influence on lacustrine sedimentation, as thick packets of volcanic material were repeatedly deposited in the lake basin. Pollen data from a highland pond indicate extensive grasslands and cold, dry conditions during the last interstadial (> 30 000 yr bp). The diatom record from the lowlands indicates that a shallow, freshwater lake covered the area during the last 15 kyr. An episode of lower lake level followed by a trend towards slightly higher water level is recorded during late Pleistocene time (c. 14 000–11 600 yr bp). Relatively dry early Holocene conditions, followed by a recovery of the lake level after c. 5000 yr bp are recorded in the upper Lerma basin; this lake-level pattern is also recorded in the neighbouring Lake Chalco basin. Throughout these periods, emplacement of volcanic material resulted in changes to lake geometry and, in part, to lake levels.
INTRODUCTION is, however, not yet clear given the few records that span the last 10+ kyr. Central Mexico was characterized by intense Quaternary volcanism. Recent research on the volcanic activity of the central part of the Trans-Mexican Volcanic Belt and on the palaeolimnological and palaeoenvironmental evolution of the Basin of Mexico (lakes Texcoco and Chalco) during Quaternary time (Siebe et al., 1995; Caballero & Ortega, 1998; LozanoGarcía & Ortega-Guerrero, 1997) indicate that volcanism has been one of the major forces driving environmental change in central Mexico, and that this volcanism has obscured the record of the climatic evolution in the area. The upper Lerma basin, located immediately west of the Basin of Mexico, was selected
Central Mexico is an important region for Quaternary palaeoenvironmental research, because it is a highaltitude, tropical area in which relatively few studies covering the entire last glacial–interglacial cycle have been undertaken. Palaeoenvironmental studies in several of the lacustrine basins in the Trans-Mexican Volcanic Belt (Fig. 1A) (Deevey, 1944; Watts & Bradbury, 1982; Straka & Ohngemach, 1989; Metcalfe et al., 1991; Metcalfe, 1992), particularly in the Basin of Mexico (Clisby & Sears, 1955; Bradbury, 1989; Lozano-García et al., 1993; Urrutia et al., 1994, 1995; Caballero et al., 1996), have documented the occurrence of important environmental changes in the region during Quaternary time. The pattern of climatic changes in this region since the last glacial maximum
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
247
VSIC11 9/12/08 1:00 PM Page 248
248
M. Caballero et al.
Fig. 1. (A) General location of the upper Lerma basin in central Mexico. NT, Nevado de Toluca; BM, Basin of Mexico; TMVB, Trans-Mexican Volcanic Belt. (B) Simplified geological map of the study area, including major morphological features surrounding the upper Lerma basin. Zacango, Metepec, La Isla I and II represent studied stratigraphic sites. Murillo & Carbajal (1993) and Murillo (1994) recovered important Pleistocene faunal remains in Metepec.
for palaeoenvironmental research because it is a smaller basin with a volcanic record dominated by the activity of Nevado de Toluca stratovolcano. Furthermore, this is the highest basin in Mexico and it is considered to be more sensitive to climatic variability than others in the area. The record of climatic changes is preserved in the lacustrine sediments that accumulated during periods of volcanic quiescence. Previous studies in the upper Lerma basin have documented
environmental changes during the last c. 10 kyr (Metcalfe et al., 1991; Sugiura et al., 1994). There is, however, a lack of palaeoenvironmental information in the region for earlier periods, although remains of Pleistocene fauna (mammoths and other vertebrates) have been recovered near the towns of Metepec (Murillo & Carbajal, 1993; Murillo, 1994) and San Mateo Atenco (O. Carranza, personal communication). In this paper we present the first results of a
VSIC11 9/12/08 1:00 PM Page 249
249
Upper Lerma Basin, Mexico series of projects on the eruptive history of Nevado de Toluca and the environmental evolution of the upper Lerma basin over the last 40 kyr. These include detailed volcanic stratigraphy, pollen, diatom and magnetic susceptibility studies coupled with radiocarbon dating of key stratigraphic layers in subaerial and lacustrine volcaniclastic sequences.
DESCRIPTION OF THE AREA The upper Lerma basin lies at an altitude of 2570 m above sea-level (a.s.l.) and is bounded to the east by the late Tertiary Las Cruces volcanic range, to the south by Nevado de Toluca, scattered cinder cones and fissural lava flows of Late Pleistocene–Holocene age, and to the south-west by the San Antonio volcano of probable Late Miocene age. Lake Chignahuapan is the highest of a series of three water bodies that are connected by the Lerma river, which flows towards the north-west into Lake Chapala, located in western Mexico (Fig. 1a). At present, these lakes have been considerably reduced as a result of water extraction for Mexico City, and are particularly shallow during the dry winter season. Among the volcanic features in the basin, Nevado de Toluca (4680 m a.s.l.) has greatly contributed to the input of primary volcanic materials and secondary volcaniclastic sediments into the upper Lerma basin during the last 40 kyr. Previous work on Nevado de Toluca has documented three major eruptions during this 40 kyr time span (Bloomfield & Valastro, 1974, 1977; Bloomfield et al., 1977). These are a vulcanian-type explosion that deposited ‘blue–grey lahars’ (c. 28 000 yr bp), a Plinian-type eruption that deposited the Lower Toluca Pumice (LTP) fallout (24 000 yr bp), and a large Plinian event that dispersed the Upper Toluca Pumice (UTP) layer (c. 11 600 yr bp). Recent stratigraphic studies of Nevado de Toluca have found that the early workers misidentified the 28 000 yr bp eruption. Several exposures around the volcano show at least two deposits produced by two large domedestruction events dated at c. 37 000 and c. 28 000 yr bp (Macías et al., 1997). Monogenetic activity has also been reported for the area. The Tres Cruces volcano, a cinder cone located south of Lake Chiconahuapan (Fig. 1B), produced an ash-fall deposit and a series of basaltic andesitic lava flows at c. 8500 yr bp (Bloomfield, 1975; Metcalfe et al., 1991). During the last 10 kyr Lake Chiconahuapan experienced a series of lake-level fluctuations as well as the impact of volcanic activity (Metcalfe et al., 1991;
Sugiura et al., 1994). Metcalfe et al. (1991) suggested that relatively high lake level stands were established by c. 8200 yr bp, after the fall of the UTP, but that this transgression was interrupted by the eruption of the Tres Cruces volcano. Probable low levels were present after this eruption, between c. 8200 and 6000 yr bp, but higher water stands were then re-established. Low lake levels are suggested for c. 4600 yr bp, followed by higher water stands recorded after c. 3600 yr bp, with maximum levels by c. 1600 yr bp. Another period of lower lake level is proposed for the c. 1400–900 yr bp interval, which correlates with the development of Teotenango archaeological site, followed by a slight rise in the water level in more recent times.
METHODS We report analyses for two sites (Zacango and Metepec) that contain the most complete sequences of volcanic and volcaniclastic deposits from Nevado de Toluca (Fig. 1B). In the Zacango quarry, samples from selected horizons were taken for diatom, pollen and tephra analyses. At Metepec, samples were taken at 0.1-m intervals for loss on ignition (LOI), diatom, pollen, magnetic susceptibility and tephra analyses. In addition, two cores were drilled in the northern part of Lake Chignahuapan, La Isla I (5.5 m deep) and La Isla II (5.7 m deep), from which we describe the volcanic stratigraphy and diatom contents (Fig. 1B). The cores were extracted using percussion, non-rotary drilling equipment. La Isla II was selected as the master core for palaeolimnological studies because it contains the best sediment record. Samples for LOI, diatom analyses, and magnetic susceptibility were taken at 0.05 m intervals, except at tephra levels where only samples for analyses of tephra components and magnetic susceptibility were taken. Samples for radiocarbon dating were collected at selected horizons from both cores and from subaerial sequences. Pollen and diatom samples were prepared and counted following standard techniques (Lozano et al., 1993; Caballero & Ortega, 1998; Lozano & Ortega, 1998). For palaeolimnological interpretation, diatoms were grouped into three assemblages according to their limnological preferences (Hustedt, 1930, 1959, 1960–1966; Gasse, 1980, 1986; Kramer & LangueBertalot, 1986, 1988, 1991; Caballero, 1995). 1 Acid pond assemblage. This assemblage is characterized by high abundance of Eunotia spp. (> 10%) in association with lower percentages of Pinnularia spp. and other acidophilous taxa such as Cymbella
VSIC11 9/12/08 1:00 PM Page 250
250
M. Caballero et al.
perpusilla. This association indicates the presence of shallow, low-conductivity and slightly acid waters (pH ≈ 6). 2 Freshwater pond assemblage. This assemblage is characterized by high numbers of Cocconies placentula (> 10%) in association with other periphytic (occurring on a substrate), circumneutral (pH ≈ 7) taxa such as Cymbella spp., Epithemia spp. and Nitzschia paleacea. Occasionally, some aerophilous taxa such as Hantzchia amphioxys or Pinnularia spp., or some alkaliphilous species such as Nitzschia amphibia, Amphora ovalis var. affinis and Cyclotella meneghiniana can be present. This assemblage indicates a shallow, circumneutral (pH ≈ 7), freshwater pond with abundant aquatic vegetation. Aerophilous and alkaliphilous varieties suggest that the system experienced important, probably periodic, water-level fluctuations during which pH reached slightly higher values (pH ≈ 8). 3 Shallow freshwater lake assemblage. This assemblage is dominated by the presence of tychoplanktonic (occurring on a substrate, occasionally entering the plankton) Fragilaria spp. (> 20%) in association with some periphytic taxa such as Cocconeis placentula and occasionally some planktonic species such as Aulacoseria ambigua. This association indicates the presence of a shallow, circumneutral (pH ≈ 7), freshwater lake with aquatic vegetation. Samples for magnetic susceptibility were collected in 2 cm acrylic cubes. Low-field magnetic susceptibility (k) was measured using a Bartington MS2 susceptibility system equipped with a dual-frequency laboratory sensor. Results are reported as 10 –5 SI dimensionless units. For LOI determinations, samples were ignited at 550°C for 2 h (Bengtsson & Enell, 1986). For the analyses of tephra components samples were sieved using 1-µm screens. Representative fractions (0φ and 1φ) were then cleaned using dilute HCl and distilled water, and ultrasound. Tephra particles were then observed under the binocular microscope.
VOLCANIC STRATIGRAPHY We describe here the composite stratigraphic sequence observed in the subaerial exposures of Zacango and Metepec on the northern flanks of Nevado de Toluca and in the two cores from Lake Chiconahuapan (Figs 1B and 2). This composite stratigraphy provides a record of the eruptive history of volcanism in the area and its effects on the lacustrine evolution of the upper Lerma basin.
Block-and-ash-flow deposit (BAF c. 37 000 yr BP) The basal unit is a grey, massive (disorganized) block-and-ash-flow deposit that is distributed around Nevado de Toluca. Where best exposed (i.e. at Zacango) this deposit is up to 20 m thick and is composed in some places of at least four flow units that commonly show gas escape pipes (Zacango section, Fig. 2). These units consist of grey, gravel- to bouldersize, dense dacite clasts with minor amounts of poorly vesiculated pumice, glassy lithic clasts and red altered accidental dacite clasts from the volcanic edifice set in a coarse sand-size matrix. The dense dacite clasts have millimetre-sized phenocrysts of plagioclase, hornblende, augite, and minor hypersthene, quartz, and biotite embedded in an aphanitic groundmass of the same constituents. The block-and-ash-flow deposit is covered at Zacango by a sequence of finely stratified beds composed of silt- to clay-size particles interbedded with organic-rich horizons (Fig. 2). Elsewhere, for example at Metepec, the block-and-ash-flow deposit correlates by its stratigraphic position with a grey, massive sand–silt, ash-flow deposit rich in crystals. This deposit is overlain by 15 cm of massive organicrich sediments. Charcoal found within the block-and-ash flow deposit on the south-eastern slopes of Nevado de Toluca gave an accelerator mass spectrometry (AMS) date of 37 000 ± 1125 yr bp (Macías et al., 1997). The basal block-and-ash-flow deposit has a minimum age of 35 160 ± 960 yr bp (Fig. 2, Table 1), and organic material in the sediments that overlie the deposit at Metepec was dated at 35 160 ± 960 yr bp (Table 1, Fig. 2). This correlates well with the date of 35 600 + 2600/ – 1800 yr bp for the ‘gray lahar’ reported by Heine (1978) and an underlying palaeosol dated at 38 000 yr bp by Cantagrel et al. (1981). Block-and-ash-flow deposit (BAF c. 28 000 yr BP) The younger block-and-ash-flow deposit is a massive blue–grey pyroclastic unit that is widely distributed across the northern flanks of Nevado de Toluca, reaching up to 20 m thick in some outcrops (Fig. 1b). It is at least 5 m thick at Zacango, where it is composed of three flow units (Fig. 2). The base of the deposit is exposed at Zacango, where it overlies the fine stratified beds described above. This deposit is lithologically very similar to the 37 000-yr pyroclastic flow deposit but is barren of pumice clasts and is supported by a finer-grained matrix.
VSIC11 9/12/08 1:00 PM Page 251
251
Upper Lerma Basin, Mexico
Fig. 2. Stratigraphic correlation of studied sites in exposures (Zacango and Metepec) and cores (La Isla I and II). The location of sections is shown in Fig. 1B. Scale for the Metepec section also applies to La Isla I and II cores, and to layers without a specified thickness at the Zacango sequence. Table 1. Diatom assemblages from the Isla II core, upper Lerma basin, Mexico Assemblage (1) Acid pond (2) Freshwater pond
(3) Shallow freshwater lake
Distinctive diatoms
Environmental interpretation
Eunotia spp. (> 10%), Pinnularia spp. and Cymbella perpusilla Cocconies placentula (> 10%), Cymbella spp. and Epithemia spp. with aerophilous (Hantzchia amphioxys, Pinnularia spp.) and alkaliphilous (Nitzschia amphibia, Amphora ovalis var. affinis, Cyclotella meneghiniana) spp. Fragilaria spp. (> 20%) and Cocconeis placentula with Aulacoseria ambigua
Shallow, low-conductivity, slightly acid waters (pH ≈ 6) Shallow, circumneutral (pH ≈ 7), freshwater pond with abundant aquatic vegetation; aerophilous varieties suggest important (periodic?) level fluctuations with slightly higher pH values (pH ≈ 8)
At Metepec the stratigraphically equivalent deposit is a grey–greenish, massive layer entirely composed of silty sand rich in crystals. This deposit is overlain by a series of fluvial deposits from which the in situ remains
Shallow, circumneutral (pH ≈ 7), freshwater lake with aquatic vegetation
of two mammoths associated with other Pleistocene fauna (equid, camelid and bison) were excavated (Murillo & Carbajal, 1993; Murillo, 1994; Castañeda, 1998) (Table 2).
VSIC11 9/12/08 1:00 PM Page 252
252 Site Metepec Metepec Metepec Metepec La Isla I La Isla I La Isla II La Isla II
M. Caballero et al. Laboratory code*
Conventional age (yr bp)
δ13CPDB (‰)†
Beta-70712 Beta-70712 Beta-70709 Beta-70708 A-9317 A-9318 A-9778 A-9705
35 160 ± 960 27 180 ± 410 4820 ± 90 105.2 ± 0.8% 12 320 + 475/ – 450 11 890 + 220/ – 215 10 820 + 360/ – 345 13 870 + 445/ – 420
– 25 – 25 – 25 – 25 – 21.9 – 28.8 – 23.3 – 27.5
Table 2. Radiocarbon determinations on samples from three sequences in the upper Lerma basin: Zacango, Metepec and La Isla cores
*Beta, Beta Analytic; A, Geochronology Laboratory, University of Arizona. †PDB, Peedee Belemnite standard correction of isotopic fractionation.
Carbonized wood within the younger block-andash-flow deposit has been dated in two outcrops around Nevado de Toluca, yielding dates of 28 140 + 865/ – 780 and 28 925 + 625/ – 580 yr bp (Macías et al., 1997). These figures represent the age of the volcanic event and correlate with the upper age limit of this deposit proposed by Bloomfield & Valastro (1977), who reported a date of 27 580 ± 650 yr bp for a layer of fluvial gravel overlying it. This age is supported by the stratigraphic position of the deposit at Metepec. Here, the age of the deposit is constrained by a 35 160 ± 960 yr bp date on the clay-rich layer that underlies the deposit and a date of 27 180 ± 410 yr bp (Table 1, Fig. 2) on the overlying fluvial sediments (Urrutia-Fucugauchi et al., 1996). Lower Toluca Pumice (24 000 yr BP) The Lower Toluca Pumice (LTP) consists of an ochre, massive, clast-supported fall layer (average thickness 55 cm). The LTP is rich in ochre pumice with lesser amounts of grey dense juvenile dacite, hydrothermally altered lithic clasts, and schist fragments from the local basement. At Zacango, the LTP is represented from the base upwards by at least three fall layers separated by thin cross-bedded pumice-rich layers, and a massive silty bed on top. Here, the LTP overlies a thick palaeosol dated at 24 260 ± 670 yr bp (Bloomfield & Valastro, 1977). According to Bloomfield et al. (1977), the LTP has a dispersal axis trending north-eastwards from Nevado de Toluca, covers an area of 400 km2, and has an approximate total volume of 0.33 km3. The LTP at Metepec is a 15-cm-thick deposit composed of gravel- to sand-sized pumice and minor lithic clasts (Metepec section, Fig. 2). The LTP unconformably overlies the fluvial deposits described above, and its upper contact is undulating and eroded, and is covered by a sequence of two massive heterolithic matrix-supported lahars and fluviatile deposits.
> 14 000 yr BP) Ash-flow deposit (> The bottom of both La Isla cores comprises a grey, massive, sand–silt ash-flow deposit (AFD) rich in angular crystals and glass. This deposit was not fully penetrated in any of the cores but it shows a minimum thickness of 80 cm in La Isla I and 85 cm in La Isla II. In both cores it underlies a dark brown, massive, silty sequence rich in organic material. A radiocarbon age on bulk sediment 50 cm above the AFD in La Isla II core gave an age of 13 870 ± 445 yr bp. In several exposures on the north-north-eastern flanks of the volcano the AFD is a grey ash-flow deposit and thin, grey, cross-stratified deposits that lie between the LTP and UTP layers. The radiocarbon date and the field relationships suggest that this AFD has an minimum age of c. 4 000 yr bp. Upper Toluca Pumice (c. 11 000 yr BP) The Upper Toluca Pumice (UTP) is the most widespread tephra produced during the volcanic history of Nevado de Toluca, and has been used as a stratigraphic marker in the upper Lerma and Mexico basins (e.g. Mooser, 1967; Metcalfe et al., 1991; LozanoGarcía et al., 1993). Bloomfield & Valastro (1974, 1977) described the UTP deposit as being composed of two main fall members separated by fine sandy ash and covered in some places by ‘pink lahars’. The fallout members of the UTP present a main dispersal axis oriented N65°E towards Mexico City. They cover an approximate area of 2000 km2 and have an approximate total volume of 3.5 km3 (Bloomfield et al., 1977). The UTP is found in all of our stratigraphic sections and consists of up to four layers (Fig. 2). It is made of white, angular, vesiculated dacite pumice, grey juvenile lithic clasts and crystals. At Zacango, where it is 2.5 m thick, the UTP overlies a thick palaeosol. The fall deposits at Metepec are 0.47 m thick, partly
VSIC11 9/12/08 1:00 PM Page 253
Upper Lerma Basin, Mexico
253
reworked and cover two heterolithological units composed of gravel-size lithic fragments in a sandsized matrix. At La Isla I and II (0.94 and 1.33 m thick, respectively) the UTP covers a black massive sequence of silty lacustrine sediments rich in organic matter. Two organic-rich horizons that underlie the UTP were dated, one at La Isla I core (3.65 and 2.65 m), giving a maximum age of 11 890 + 220/ – 215 yr bp, and one at La Isla II core (3.19 m), yielding a maximum age for this eruption of 10 820 + 365/ – 345 yr bp (Table 1; La Isla I and II sections, Fig. 2). These dates are in good agreement with other maximum dates reported for Lake Chalco, Basin of Mexico, of 12 800 ± 90 and 12 520 ± 135 yr bp (Lozano-García et al., 1993) as well as with the average date of 11 600 yr bp proposed by Bloomfield & Valastro (1974, 1977).
samples but diatoms were not preserved in any of the eight levels (Fig. 3). The oldest of the pollen-positive samples has remarkably high values of grasses (up to 75%). Other herbs such as Asteraceae, Chenopodiaceae– Amaranthaceae, Artemisia and Cirsium are also present. The arboreal pollen assemblage shows low values (Quercus, 10%; Alnus, 4%), particularly Pinus, which reaches only 2% of the counts. Other elements recorded at this level are subaquatic taxa such as Cyperaceae and some zygospores of Zygnema. The second sample has a different pollen composition: there is a reduction in the grass pollen (45%) and an increase in the arboreal taxa, mainly Quercus (23%), Pinus (15%), Alnus (11%) and Liquidambar (1%). Pollen of Cyperaceae is also present, as are remains of two algae: Botryococcus and Zygnema.
Tres Cruces Tephra (c. 8500 yr BP)
Metepec
The Tres Cruces Tephra (TCT) has a restricted distribution compared with the UTP and LTP deposits. It was recorded only at La Isla II sequence and is present in the two pits described by Metcalfe et al. (1986, 1991). The TCT is a black, massive, fine sand layer (31 cm thick at La Isla II), partly indurated towards the top, and mainly composed of dark grey vesiculated scoria and yellowish angular glass (La Isla II section, Fig. 2). This deposit and a series of basaltic andesitic lava flows are the products of the Tres Cruces volcano (Fig. 1). An exposure of palaeosol below these deposits yielded a maximum age for the event of 8440 ± 70 yr bp (Bloomfield, 1975). Metcalfe et al. (1986, 1991) dated an organic-rich sediment under the TCT in their Pit 2, which gave another maximum age of 8160 ± 100 yr bp.
The sequence at Metepec (7.1 m, Fig. 2) is dominated by fluvial and volcaniclastic deposits and only in the lower section (< 4 m) are fine-grained lacustrine horizons present (Fig. 4). Diatom analyses were negative for all the samples from this site. The LOI values were extremely low, indicating low organic accumulation. Pollen content was also relatively poor in terms of diversity and abundance of taxa (Castañeda, 1998). According to the volcanic stratigraphy, this record is divided into the following four units (Fig. 4). Unit I (7.1–5.0 m) is bounded by the two BAF deposits and dates from 37 000 to 28 000 yr bp; it therefore correlates in time with the samples from Zacango. These sediments show intermediate values of magnetic susceptibility (≤ 100 SI), except for the BAFs, which have higher values (300–100 SI). Pollen is absent in the 37 000 yr bp BAF, but present at medium to low concentrations ((15–17) × 102 grains cm–3) in the overlying silts. At the base of the 28 000 yr bp BAF, pollen is present at low concentrations ((3–7) × 102 grains cm–3) but disappears at higher levels (5.7– 5 m). The dominant pollen type (up to 98% of counts) is the aquatic fern Isöetes mexicana, which grows in very shallow (50 cm), cool and clear waters. It occurs in association with other non-arboreal taxa (Asteraceae, Poaceae, Cyperaceae) and with extremely low values of Pinus pollen (1–2%). This assemblage suggests that Metepec was a littoral environment at this time. In the 27 000 yr bp BAF an apparent increase in Pinus pollen (78%) is recorded. Other arboreal elements, such as Alnus, are present at low percentages (1–3%) whereas Isöetes mexicana is less abundant. This pollen
MAGNETIC AND MICROFOSSIL ANALYSES Zacango In the Zacango sequence, a series of dark, sandy silt and clay horizons are present between the two BAF deposits (Fig. 2). These beds represent an exposure of 1.8 m that date between 37 000 and 28 000 yr bp. Although these horizons suggest subaerial or transitional environments, and therefore low probabilities for microfossil preservation, pollen and diatom analyses were carried out at eight levels. Only two of these, however, had enough pollen grains to allow standard counting. Chrysophyte cysts were also present in these
VSIC11 9/12/08 1:00 PM Page 254
254
M. Caballero et al.
Fig. 3. Pollen percentage diagram of Zacango, upper Lerma basin. The pollen sum includes arboreal and non-arboreal pollen taxa. Aquatic taxa and algae are expressed in number of grains. Chronology is indicated to the left.
assemblage could, however, have been transported within the BAF and therefore not represent a change in the local vegetation. Unit II (5.0 –3.0 m) is bracketed by the c. 28 000 yr bp BAF and the top of the LTP (24 000 yr bp). Above the BAF, fine silty sediments are overlain by fluvial deposits (Fig. 2). The Pleistocene faunal remains reported for this sequence (Murillo & Carbajal, 1993; Murillo, 1994) were preserved at the transitions between these two layers. Magnetic susceptibility in the silty sediments is low (< 50 SI) but increases in the fluvial deposits (100 –250 SI). The pollen record shows a similar pattern, with low values at the base of the silts ((1– 4) × 102 grains cm–3), increasing towards the top and reaching the highest values in the fluvial deposit (up to 60 × 102 grains cm–3). In the silty sediment Isöetes mexicana dominates the assemblage (up to 100%). In contrast, at the base of the fluvial deposits Pinus is present (up to 60%), in association with some non-arboreal elements (Asteraceae, Poaceae) and high values (23%) of Cyperaceae. Pollen is not preserved in the upper layers of the fluvial deposit. The LTP has medium magnetic susceptibility values (100 SI) and pollen is not preserved, except for the top 20 cm of this
tephra, where Isöetes mexicana reappears, again dominating the assemblage. Unit III (3.0–0.8) is confined by the LTP (24 000 yr bp) and the top of the UTP (11 600 yr bp). A thick series of lahar and fluvial deposits is present between these two tephras. These sediments have medium to high magnetic susceptibilities (50–330 SI) and a discontinuous pollen record (Fig. 4). The few samples in which pollen was preserved had low concentration values and were dominated by Isöetes mexicana. In the UTP, magnetic susceptibility is low (≤ 50 SI) and pollen is absent. Unit IV (0.8–0.0 m) lies above the UTP. This tephra is overlain by recent sediments whose base was dated at 4820 ± 90 yr bp (Table 1). This suggests the presence of a sedimentation hiatus between latest Pleistocene (c. 11 600 yr bp) and middle Holocene time. Magnetic susceptibility values in these sediments are intermediate (50–100 SI) and there is no pollen record. Lake Chignahuapan The sequence in La Isla II core is dominated by
VSIC11 9/12/08 1:00 PM Page 255
Upper Lerma Basin, Mexico
255
Fig. 4. Pollen percentage diagram from Metepec pit, upper Lerma basin, showing stratigraphy and magnetic susceptibility (10–5 SI). The pollen sum includes arboreal, non-arboreal and aquatic pollen taxa. Pollen concentration is expressed in grains × 102 cm–3. Radiocarbon dates and tephra dates are indicated to the left of the diagram. Pollen units were established by the presence of tephra layers.
organic-rich silts that are interrupted by the presence of the > 14 000 yr bp AFD, the UTP and the TCT. Alternate layers of dark brown and brown silt characterize the sediments at the bottom part of the core (between the AFD and the UTP); the rest of the core is formed by massive, organic-rich brown silts. In total, La Isla II core represents predominantly quiet-water lacustrine sedimentation. Magnetic susceptibility of the sediments is low whereas organic content (LOI) is high, suggesting that sediment influx to the lake during the last c. 14 000 yr bp was generally low and that organic sedimentation was predominant in the system. Stratigraphy as well as the results of the diatom and magnetic susceptibility analyses from this core are summarized in Fig. 5a & b. This record can be divided by the presence of the tephra layers into three lacustrine periods that corre-
late with Metepec units III and IV; unit IV, however, is divided by the presence of the TCT into IVa and IVb (before and after the deposition of this tephra). Unit III (4.35–3 m) is defined by the presence of the > 14 000 yr bp AFD and the UTP (c. 11 600 yr bp). The bottom of this unit is formed by a layer of weathered ash with some root remains (4.39–4.23 m), which is overlain by a series of alternating layers of dark brown and brown organic-rich (LOI 15–40%) silts (4.23–3.00 m). Macroscopic charcoal particles were present in the silts directly under the UTP (3.30– 3.00 m). The magnetic susceptibility of the altered ash layer is of the order of 10 SI whereas the values of the organic silts are extremely low, around 1 SI. The low magnetic susceptibility together with the high LOI values indicate low sediment influx and high organic accumulation, with a sedimentation rate of
VSIC11 9/12/08 1:00 PM Page 256
256
(a)
(b)
M. Caballero et al.
VSIC11 9/12/08 1:00 PM Page 257
Upper Lerma Basin, Mexico ≈ 0.35 mm yr –1. According to its diatom content, this unit can be divided in two phases, below and above the 3.78 m level (c. 13 000 yr bp). The lower interval (4.35–3.78 m) is characterized by lower LOI values (< 30%) and by low to medium diatom abundance ((5–500) × 106 valves (v) g–1 of dry sediment (gds), Fig. 5a). The first sample above the > 14 000 yr bp AFD has a shallow freshwater lake assemblage (Fig. 5b), but this changes in the next sample to an acid pond association. In the rest of the samples a freshwater pond assemblage is present. Sponge spicules and some chrysophyte statocysts are also preserved in these sediments (Fig. 5a). The diatom record suggests that during this interval (> 14 000 to c. 13 000 yr bp) Lake Chignahuapan changed from a slightly higher to a lower lake level, passing through a slightly acidic period. In the upper interval (3.78–3.00 m) LOI has higher values (> 30%) and the total diatom abundance reaches the highest values in the record (> 1000 × 106 v gds–1). The diatom assemblage indicates a shallow freshwater lake. Chrysophyte cysts are present at higher values (500 × 103 cysts gds–1). The diatom record indicates a transition towards a slightly higher lake level during this interval (c. 13 000 –11 600 yr bp). Unit IVa (1.12– 0.64 m) lies between the UTP (c. 11 600 yr bp) and the TCT (c. 8500 yr bp). Brown, massive, organic silts (LOI 15– 40%) rich in small root remains represent this period. The sedimentation rate of this unit was slightly lower than in the previous unit (≈ 0.2 mm yr–1). Diatom abundance is medium ((225– 725) × 106 v gds–1), except for one sample (at 0.65 m) with low values (8 × 106 v gds–1). Magnetic susceptibility is very low but reaches higher values towards the top of the unit (10 SI, Fig. 5a). According to its diatom content the unit can be divided into two parts: one between 1.12 and 0.80 m and the other formed by the sample collected at 0.65 m. The diatom assemblages of the first part represent a freshwater pond, followed by a shallow freshwater lake. Some alkaliphilous taxa are also present, as are sponge spicules. This suggests that the lake level was initially low but that it later reached higher levels, allowing a shallow, freshwater lake with probable periodical fluctuations to be established. The sample at 0.65 m has, in contrast, a lower diatom abundance and a shallow pond assemblage, in association with alkaliphilous taxa, sponge spicules and chrysophyte
257
cysts. This suggests a significant reduction in the lake level before the deposition of the TCT. The presence of an indurated surface on top of the TCT at La Isla II suggests that this tephra was exposed to subaerial conditions and that at least the area around this core was seasonally dry. Unit IVb (0.28–0.0 m) is bounded below by the TCT and above by the top of the sequence, and is therefore younger than 8500 yr bp. This period of lacustrine sedimentation is represented by dark brown organic silts (LOI 5–20%) with root remains and some altered pumice. Magnetic susceptibility in these sediments is low, but represents the highest values associated with lake deposits in the sequence (10 SI). This higher magnetic susceptibility in recent lake sediments is common to Holocene sequences studied in the Basin of Mexico (Urrutia et al., 1994, 1995; Lozano-García & Ortega, 1997) and could be related to the full establishment of human populations in the area (e.g. agricultural activities). Diatom abundance is, however, very low (< 20 × 106 v gds–1) and, assuming that the top of the sequence dates to the present, sedimentation rates seem to be particularly low (≈ 0.03 mm yr–1), suggesting the presence of hiati. The diatom assemblage of the bottom sample (0.30 m) has a freshwater pond assemblage in association with planktonic (Aulacoseira ambigua) and alkaliphilous taxa. The rest of the samples in this unit have a freshwater pond assemblage in association with higher tychoplanktonic Fragilaria spp. These assemblages suggest that the lake was initially a shallow freshwater pond subject to major depth fluctuations that allowed the planktonic and alkaliphilous taxa to develop, but afterwards, it became a permanent, freshwater to slightly alkaline system.
PALAEOENVIRONMENTAL HISTORY Approximately 37 000 yr ago Nevado de Toluca produced the largest eruption known in its Late Pleistocene history. This event destroyed a large dacitic dome (≈ 0.5 km3) that blocked the central crater and produced pyroclastic flows that filled major gullies and mantled topographic highs to a distance of 15–20 km from the source. This event blocked the stream valleys around Nevado de Toluca, forming closed lakes, small ponds, and mud floods. Although charcoal has not
Fig. 5. (opposite) Stratigraphic column from La Isla II core, upper Lerma basin. (a) Magnetic susceptibility (10–5 SI); loss on ignition (LOI); total diatom abundance (number of valves per gram of dry sediment (v gds–1) ) and total abundance of sponge spicules and chrysophyte cysts (spicules or cysts per gram of dry sediment (gds–1) ). (b) Diatom percentage diagram (taxa (10%)). Units were established by the presence of tephra layers.
VSIC11 9/12/08 1:00 PM Page 258
258
M. Caballero et al.
been found within the eruptive deposit, overlying sediments in the closed ponds preserve pollen. In Zacango, the record suggests that after the 37 000 yr bp eruption of Nevado de Toluca, a small pond was established in this high-altitude area. Although pollen preservation in this environment was poor, it gives an insight into the composition of the plant communities at this altitude. Palynological results indicate the presence of a water body that was surrounded by extensive grasslands. These could represent a well-established plant community subjected to periodic fires as suggested by abundant charcoal particles in the pollen slides, or it could be a successional stage before the development of forests after volcanic disruption. In both cases a dry environment can be inferred. The presence of Cyperaceae, Zygnema and chrysophyte cysts indicate that the pond was a shallow, stagnant (Van Geel & Van der Hammen, 1978), freshwater body. The low coverage of arboreal vegetation in the area is confirmed by the Metepec sequence, where these taxa are present in low values in the silts of unit I (Fig. 4). Some time after the 37 000 yr bp eruption, a new dacite dome developed within the crater of Nevado de Toluca (Macías et al., 1997). At c. 28 000 yr bp the volcano reached a mature magmatic stage that resulted in the destruction of the dome. The event produced a series of pyroclastic flows that moved along the volcano flanks, burning the forest and assimilating pollen particles, as found in the Metepec record. These flows (BAF) reached the edge of the lake, where they suffered a sudden decrease in velocity and deposited large amounts of material. At Metepec, the flow deposited a fine-grained ash layer. On the basis of the granulometric and component (crystal-rich deposit) characteristics of this flow (BAF) deposit and the pollen data, it is likely that Metepec was located at the edge of former Lake Chignahuapan at the time of the 28 000 yr bp eruption. This event should have had an important influence on the lake characteristics, pushing it eastwards and reducing its overall size and depth; however, the nature of the lake itself cannot be determined at this time. Shortly after deposition of the 28 000 yr bp BAF, the stratigraphic sequence at Metepec indicates the end of lacustrine sedimentation and a transition to a fluvial environment. At lower elevations, however, a lacustrine environment could have persisted. This event correlates with the mammoths and other Pleistocene faunal remains found in situ within the fluviatile deposits dated at 27 180 ± 410 yr bp (Urrutia-Fucugauchi et al., 1996; Table 2). Nevado de Toluca erupted again at c. 24 000 yr bp depositing a thick pumice deposit (LTP) close to the
volcano, and at least 10–20 cm of fine gravel near the lake, as observed at Metepec and at other subaerial exposures. This event was probably followed by remobilization of this material by secondary debris flows that deposited massive layers on the northern flanks of the volcano and in transitional environments such as Metepec. Debris flow emplacement disrupted the surface of the LTP to produce an undulated contact, which suggests that the emplacement took place shortly after the LTP was deposited. Some time before c. 14 000 yr bp another explosive eruption of Nevado de Toluca produced a series of dilute pyroclastic flows and surges that travelled along the volcano flanks. As indicated by La Isla cores, only the ash flow reached the western shores of Lake Chignahuapan and continued to move across the shallow water body as a dilute ash cloud depositing a uniform grey, silt-rich layer. After the emplacement of the > 14 000 yr bp AFD the diatom record indicates that the lake experienced a series of changes from a freshwater, shallow lake to a freshwater pond, passing through a slightly acidic phase. After c. 13 000 yr bp a trend towards a higher lake level was established. Stratigraphic evidence suggests that this trend correlates in time with debris flow deposits overlying the LTP in Metepec. In general terms both events give evidence of an increased surface runoff. At c. 11 600 yr bp Nevado de Toluca emitted the largest Plinian tephra known in its stratigraphic record. This eruption produced a Plinian column that was dispersed to the north-east by the dominant winds, in the direction of the cities of Toluca and Mexico. The event deposited a pumice layer (UTP) up to 7–8 m thick close to the volcano, 15 cm at Tlapacoya, and ≈ 10 cm in cores drilled in the Chalco basin (Lorenzo & Mirambell, 1986; Lozano-García et al., 1993; Urrutia-Fucugauchi et al., 1995). This volcanic event must have drastically changed the bathymetry and environmental conditions of Lake Chignahuapan, as La Isla sequences show that at least 1.5 m of gravel-size pumice was deposited in the lake. After this event the diatom record indicates that Lake Chignahuapan was shallower and had a slightly higher alkalinity (unit IVa), suggesting that the water flow though the system was generally reduced. There was, however, a recovery of lacustrine levels but this trend was interrupted by a major change in environmental conditions that favoured a significant reduction in lake depth. Shallow lake levels by c. 9000–8500 yr bp are suggested by the diatom assemblages and by the presence of an indurated surface on top of the TCT at La Isla II. In one of the pits studied by Metcalfe
VSIC11 9/12/08 1:00 PM Page 259
Upper Lerma Basin, Mexico et al. (1991) the TCT showed signs of weathering, giving further evidence of low water stands before and after the emplacement of the TCT. Metcalfe et al. (1991) considered that this dry event occurred immediately after the deposition of the TCT. However, the record from La Isla II indicates that this dry phase started just before the deposit of the TCT and continued afterwards. This early Holocene dry phase correlates with the sediment hiatus in the Metepec sequence. The TCT (c. 8500 yr bp) was the product of monogenetic volcanic activity at the Tres Cruces volcano that was associated with a fissure emission of the Tenango andesite (Bloomfield, 1974, 1975). The volcanic activity at Tres Cruces generated a series of basaltic lava flows followed by a black tephra fall layer that was mainly dispersed towards the north-west (Lake Chignahuapan). This layer has a thickness of 60 cm in Pit 2 reported by Metcalfe et al. (1991), at the south-western edge of the lake. At La Isla II this tephra has a thickness of 31 cm that correlates with the wind-dispersal axis of the deposit. Thus, this tephra is a good stratigraphic marker for the Holocene record of Lake Chignahuapan. As discussed above, the Tres Cruces eruption occurred during a period of shallow lake conditions during which some areas of the lake might have dried out, at least seasonally. Lake level, however, recovered afterwards and the diatom record indicates the presence of an initially shallow pond with important fluctuations that later became a permanent, freshwater to slightly alkaline pond. This is consistent with the Metepec record, where sediment accumulation continued after c. 5000 yr bp. This trend to a lake level recovery was also reported by Metcalfe et al. (1991) after c. 6000 yr bp. Those workers reported a series of shorter-period lake-level fluctuations during late Holocene time that were apparently not recorded at La Isla, where the stratigraphic evidence suggests a sedimentation hiatus during this interval. The record from Pit 1 reported by Metcalfe et al. (1991), however, provides a better record of late Holocene sedimentation than do La Isla cores and therefore gives more detailed information for this later period.
CONCLUSIONS Volcanic activity of Nevado de Toluca during Late Pleistocene times was characterized by major cataclysmic eruptions that produced large amounts of volcanic material. Dome-destruction events generated hot pyroclastic flows some 37 000, 28 000, and 14 000
259
yr ago. These flows affected the arboreal vegetation around the volcano and markedly changed the morphology of the surrounding areas including the upper Lerma basin. Two other cataclysmic Plinian eruptions occurred at Nevado de Toluca some 24 000 and 11 600 yr ago. The later event deposited more than 1 m of gravel-size pumice into Lake Chignahuapan and at least 10 cm of fine pumice in the neighbouring basins of Texcoco and Chalco. The microfossil record studied indicates that volcanic activity had a great impact on the environmental evolution of the area and it also suggests the presence of some climate-related changes. Athough the records from Zacango and Metepec show poor microfossil preservation, they provided useful palaeoenvironmental information from contrasting altitudes, which, together with the good quality of the volcanic record, and with the diatom assemblages from La Isla II, give an insight into the environmental evolution of this area during the last 40 000 yr. These are the first palaeoenvironmental data for Late Pleistocene time in this basin and this is the first attempt to reconstruct the palaeoenvironment of a central Mexican site where Pleistocene megafaunal remains have been found. Of particular interest are the pollen results from Zacango, which gave an unexpected grass-dominated assemblage at an altitude where pine forests are usually dominant (Rzedowski, 1994). This suggests the presence of a cold, dry environment by the end of the last interstadial (> 30 000 yr bp). The open landscape dominated by grasslands was favourable for the presence of Pleistocene megafauna, as indicated by the remains preserved in the Metepec sequence. This sequence also suggests that the quick and massive accumulation of volcanic deposits at the western part of the basin favoured an eastward displacement of the lacustrine area as indicated by the transition from a littoral to a fluvial environment at this site. The results from the core sediments indicate that an extensive, relatively shallow lake covered the lowlands of the upper Lerma basin during at least the last c. 14 000 yr and that, unlike the lakes of the Basin of Mexico, this lake never experienced particularly saline episodes. This suggests that the discharge through the Lerma River was more or less continuous during this period. The generally low levels in the basin during this time seem to be related to the repeated rapid influx of volcanic and sedimentary materials into the lake. Fluctuations in lacustrine levels are, however, detected. Some of these fluctuations could be a response to climatic changes because there is a good correlation
VSIC11 9/12/08 1:00 PM Page 260
260
M. Caballero et al.
with the trends found in nearby Lake Chalco (LozanoGarcía et al., 1993; Urrutia et al., 1995; Caballero & Ortega, 1998). In both basins, available palaeoenvironmental data (diatoms and magnetic susceptiblity in Lerma; diatoms, pollen, and magnetic susceptibility in Chalco) indicate slightly higher water levels during Late Pleistocene time (between c. 14 000 and 11 600 yr bp). This trend suggests an increase in surface runoff that was probably related to glacial retreat during the Late Pleistocene from Nevado de Toluca in the upper Lerma basin, and from Iztaccíhuatl and Popocatépetl in Lake Chalco. Likewise, low lake-level conditions during early to middle Holocene times followed by a recovery after c. 5000 yr bp are recorded in both basins. This early to middle Holocene low lake-level period correlates in time with the maximum in summer insolation for the Northern Hemisphere (Berger, 1978) and could therefore be related to a rise in summer temperature and evaporation rather than to a reduction in precipitation.
ACKNOWLEDGEMENTS This study was financed by CONACyT (0179P-T9506 and 1824-T9211) and DGAPA, UNAM (IN102297, IN303394, IN107196 and INI04797). The collaboration of Linda Manzanilla and Silvia Murillo with the Metepec radiocarbon dating and palaeontological studies, and the technical support of Susana Sosa, Martín Espinoza, José Luis Arce and Lucia Capra was very much appreciated. We also thank Sarah Metcalfe and Beth Palmer for their valuable comments, as well as William Bandy, Greg Mackintosh and Gustavo Tolson for their comments on the English manuscript.
REFERENCES Bangtsson, L. & Enell, M. (1986) Chemical analyses. In: Handbook of Holocene Palaeoecology and Palaeohydrology (Ed. Berglund, B.E.), pp. 423–453. Wiley, New York. Bloomfield, K. (1974) The age and significance of the Tenango Basalt, Central Mexico. Bull. Volcanol., 37, 585–595. Bloomfield, K. (1975) A Late-Quaternary monogenetic volcano field in Central Mexico. Geol. Rundsch., 64, 476 – 497. Bloomfield, K. & Valastro, S. (1974) Late Pleistocene eruptive history of Nevado de Toluca volcano, Central Mexico. Bull. Volcanol., 85, 901–906. Bloomfield, K. & Valastro, S. (1977) Late Quaternary tephrachronology of Nevado de Toluca volcano, Central Mexico. Overseas Geol. Miner. Resour., 46, 1–15.
Bloomfield, K.S., Sánchez-Rubio, G. & Wilson, L. (1977) Plinian eruptions of Nevado de Toluca volcano, Mexico. Geol. Rundsch., 66, 120 –146. Bradbury, J. (1989) Late Quaternary lacustrine palaeoenvironments in the Cuenca de Mexico. Quat. Sci. Rev., 8, 75–100. Caballero, M. & Ortega, B. (1998) Lake levels since about 40,000 years ago at Lake Chalco, near Mexico City. Quat. Res., 50, 69–79. Caballero-Miranda, M. (1995) Late Quaternary palaeolimnology of Lake Chalco, the Basin of Mexico. PhD thesis, Hull University, UK. Caballero-Miranda, M., Lozano-García, S., OrtegaGuerrero, B. & Urrutia-Fucugauchi, J. (1996) Historia ambiental del sistema lacustre del sureste de la Cuenca de México. Memorias del Segundo Seminario Internacional Sobre Xochimilco, Parque Ecológico de Xochimilco, Ciudad de México, 12–26. Cantagrel, J.M., Robin, C. & Vincent, P. (1981) Les grandes étapes d’évolution d’un volcan andesitique composite: exemple du Nevado de Toluca. Bull. Vulcanol., 44, 177–188. Castañeda, B.R. (1998) Estudio palinológico de una secuencia volcaniclastica del Pleistoceno tardío, asociado a una osamenta de Mammuthus sp. en el Municipio de Metepec, Estado de México. BSc thesis, UNAM, Mexico City. Clisby, K.H. & Sears, P.B. (1955) Palynology in southern North America. Part III: Microfossil profiles under Mexico City correlated with the sedimentary profiles. Geol. Soc. Am. Bull., 66, 511–520. Deevey, E.S. (1944) Pollen analysis and Mexican archaeology, an attempt to apply the method. Am. Antiquity, 10, 135–149. Gasse, F. (1980) Flore des diatomées lacustres PlioPléistocenes du Gadeb (Éthiopie). Revue Algologique, Mémoire hors-serie no. 3, Muséum national d’histoire naturelle, Paris. Gasse, F. (1986) East African Diatoms: Taxonomy and Ecological Distribution. J. Cramer, Berlin. Hustedt, F. (1930, 1959, 1960 –1966) Die Kieselalgen Deutschlands, Öesterreichs und der Schweiz, Volumes I–III, 1991 re-edition. Koeltz, Champaign, IL. Kramer, K. & Langue-Bertalot, H. (1986, 1988, 1991) Süsswasserflora von Mitteleuropa 2. Bacillariophyceae 2/1–2/3. Gustav Fischer, New York. Lorenzo, J.L. & Mirambell, L. (Eds) (1986) Tlapacoya, 35,000 años de historia del lago de Chalco. INAH, Mexico City. Lozano-García, S. & Ortega-Guerrero, B. (1998) Late Quaternary environmental changes of the central part of the Basin of Mexico; correlation between Texcoco and Chalco basins. Rev. Palaeobot. Palynol., 99, 77–93. Lozano-García, S., Ortega-Guerrero, B., CaballeroMiranda, M. & Urrutia-Fucugauchi, J. (1993) Late Pleistocene and Holocene palaeoenvironments of Chalco Lake, Central México. Quat. Res., 40, 332–342. Macías, J.L., Arce, J.L., García, P.A., et al. (1997) Late Pleistocene–Holocene cataclysmic eruptions at Nevado de Toluca and Jocotitlán volcanoes, Central Mexico. In: Guidebook of Geological Excursions for the 1997 Annual Meeting of the Geological Society of America, Salt Lake City, Book 1 (Eds Link, K.P. & Kowallis, B.J.), pp. 1–28. Brigham Young University Press, Provo, Utah.
VSIC11 9/12/08 1:00 PM Page 261
Upper Lerma Basin, Mexico Metcalfe, S.E. (1992) Changing environments of the Zacapu Basin, Central Mexico: a diatom-based history spanning the last 30,000 years. Research Paper No. 48, School of Geography, University of Oxford, Oxford, 1–38. Metcalfe, S.E., Street-Perrott, F.A., Perrott, R.A. & Harkness, D.D. (1986) Environmental changes during the late Quaternary in the upper Lerma basin, estado de México, México. In: Proceedings of the Eighth International Diatom Symposium, Paris, 1984 (Ed. Ricard, M.), pp. 417– 482. Koeltz, Koenigstein. Metcalfe, S.E., Street-Perrott, F.A., Perrott, R.A. & Harkness, D.D. (1991) Palaeolimnology of the upper Lerma basin, Central Mexico: a record of climatic change and anthropogenic disturbance since 11,600 yr BP. J. Paleolimnol., 5, 197–218. Mooser, F. (1967) Tefracronología de la Cuenca de México para los últimos treinta mil años. Bol. Inst. Nac. Antropol. Hist. Mexico, 30, 12–15. Murillo, R.S. (1994) Hallazgo de Fauna Pelistocénica en Metepec. In: Homenaje Al Dr. Roman Piòa Chan, pp. 1–17. INAH, Mexico City. Murillo, R.S. & Carbajal, C.M.C. (1993) El Mamut de Metepec. Arqueol. Mexicana, 1, 78–79. Rzedowski, J. (1994) Vegetación de México. Limusa, Mexico City. Siebe, C.G., Macías, J.L., Abrams, M., Elizarraras, R.S., Castro, R. & Delgado, H. (1995) Quaternary explosive volcanism and pyroclastic deposits in East– Central Mexico: implications for future hazards. In: Guidebook of Geological Excursions for the 1995 Annual Meeting of the Geological Society of America, New Orleans, Book 1 (Eds
261
Chacko, J.J. & Whitney, J.A.), pp. 1– 48. Basin Research Institute, Baton Rouge, Louisiana. Straka, H. & Ohngemach, D. (1989) Late Quaternary vegetation history of the Mexican highland. Plant Syst. Evol., 162, 115–132. Sugiura, Y., Flores, A., Ludlow, B., Valadez, F., Gold, M. & Maillol, J.-M. (1994) El agua, la tierra y el hombre en el Alto Lerma: un estudio multidisciplinario. Resultados preliminares. Arqueología, 11–12, 29– 45. Urrutia-Fucugauchi, J., Lozano-García, S., Ortega Guerrero, B., et al. (1994) Palaeomagnetic and palaeoenvironmental studies in the southern basin of Mexico aI. Volcanosedimentary sequence and basin structure of Chalco lake. Geofís. Int., 33, 421– 430. Urrutia-Fucugauchi, J., Lozano-García, S., Ortega Guerrero, B. & Caballero-Miranda, M. (1995) Palaeomagnetic and palaeoenvironmental studies in the southern basin of MexicoaII. Late Pleistocene–Holocene Chalco lacustrine record. Geofís. Int., 34, 33–53. Urrutia-Fucugauchi, J., Murillo, S., Lozano-García, S., et al. (1996) Paleomagnetismo and fechamientos por radiocarbono de una secuencia del Pleistoceno TardioHoloceno en Metepec, Centro de México. XXIV Sociedad Mexicana de Antropología, Tepic, Nayarit, México, 129. Van Geel, B. & Van der Hammen, T. (1978) Zygmemataceae in Quaternary Colombian sediments. Rev. Palaeobot. Palynol., 25, 377–392. Watts, W.A. & Bradbury, J.P. (1982) Paleoecological studies at Lake Patzcuaro on the West–Central Mexican Plateau and at Chalco in the Basin of Mexico. Quat. Res., 17, 56 –70.
VSIC12 9/12/08 12:59 PM Page 263
Spec. Publs int. Ass. Sediment. (2001) 30, 263 –284
Environmental and tectonic controls on preservation potential of distal fallout ashes in fluvio-lacustrine settings: the Carboniferous–Permian Saar–Nahe Basin, south-west Germany S . K Ö N I G E R and H . S T O L L H O F E N Institut für Geologie, Universität Würzburg, Pleicherwall 1, D-97070 Würzburg, Germany
ABSTRACT Thin but widespread fallout tuff layers form important stratigraphic markers in purely continental settings such as the Carboniferous–Permian Saar–Nahe Basin. They provide tools for the correlation of laterally variable lacustrine –deltaic sections and for the recognition of tectonically induced facies and thickness changes across faults. Not all the ash layers, however, were preserved as primary deposits; many were affected by reworking to varying degrees to form mixed pyroclastic–siliciclastic beds. The preservation of volcanic ash in the Saar–Nahe Basin was strongly influenced by the depositional environment. Specifically, volcanic ash falls deposited during the transgressive phases of sedimentary cycles had a high preservation potential. In addition, the preservation potential was strongly controlled by contemporaneous fault displacements. Ash beds deposited in footwall positions were highly affected by reworking and erosion and usually preserve only reduced thicknesses. In contrast, hanging-wall blocks provide an enhanced preservation potential for both primary and reworked ashes. Such sections record enhanced thicknesses, and the intercalation of siliciclastic interbeds cause complex lithological build-ups. The Late Variscan, intermontane Saar–Nahe Basin in south-west Germany developed as a north-east– south-west-trending 120 km × 40 km half-graben, which is subdivided by sets of orthogonal transfer faults. Its basin fill is characterized by purely continental, fluvio-lacustrine sediments revealing complex thickness and facies patterns as a result of contemporaneous tectonic faulting. During Stephanian (Kasimovian– Gzelian) and earliest Permian (Asselian) time, extrabasinally derived acidic volcanic ashes were deposited as distal pyroclastic fallout in lacustrine–deltaic dominated settings and subsequently altered to various clay mineral assemblages. Essentially three depositional settings with contrasting preservation potential can be distinguished: tuffs interbedded with offshore-lacustrine (black) shales have the highest preservation potential, and are usually preserved as primary deposits showing sharp and planar contacts, planar lamination (multiple) graded bedding, and laterally constant thickness; ashes reworked by turbidity currents form erosionally based, climbing-ripple and planar-bedded graded tuff horizons with load and flute casts at their bases; and tuffs interbedded with flood plain and crevasse splay sediments are often crossbedded and display both current and tool marks. Reworking is common and associated with abundant admixture of siliciclastic detritus or erosion of the entire tuff bed.
INTRODUCTION eruptive topography of the depositional surface and thus provide excellent chronostratigraphic marker horizons, which allow both the correlation of isolated sedimentary sections and the unravelling of complex sedimentary facies architectures. After primary deposition, however, the unconsolidated ash can be easily affected by erosion, reworking, and admixture of epiclastic material as a result of wind activity, water currents, or gravitational forces. These processes not
Explosive volcanic eruptions can produce huge amounts of fine-grained pyroclastic material, which may spread laterally by wind drift over large areas. Characteristic features of the resulting fallout deposits are their lateral continuity, relatively constant thickness, comparatively high sedimentation rates and a primary distribution covering various landscapes independent of the environmental setting. Such fallderived pyroclastic layers tend to mantle the pre-
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
263
VSIC12 9/12/08 12:59 PM Page 264
264
S. Königer and H. Stollhofen
only modify the original sedimentary fabrics, they also affect the compositional properties of the deposits and favour the formation of tuffaceous sediments transitional to siliciclastic composition. The preservation of primary deposits of unconsolidated tephra is thus highly dependent on various factors such as the environment of deposition in general, subsidence and sediment delivery rates, and the length of inter-eruptive intervals. Potential processes that rework unconsolidated fallout ash deposits include ephemeral river floods (e.g. Nakayama & Yoshikawa, 1997), aeolian deflation, wave action (e.g. Houghton & Landis, 1989), turbidity currents (e.g. Busby-Spera, 1988; Palmer & Shawkey, 1997), subaqueous slumping and slope failure (e.g. White & Busby-Spera, 1987; BusbySpera, 1988; Houghton & Landis, 1989), and bioturbation (e.g. Houghton & Landis, 1989). Palmer & Shawkey (1997) described the contrasting preservation potential of tephra deposits in lacustrine and fluvial settings. Although much fallout ash may be initially stored on flood plains, post-eruptive tuffaceous loads in rivers can be sufficiently increased through tephra reworking to influence sedimentation in lakes, even in areas distal to the source (Nakayama & Yoshikawa, 1997). In this paper, we document depositional processes and patterns of reworking not of modern but of ancient Carboniferous–Permian pyroclastic material in the Saar–Nahe Basin in south-west Germany. Because of intense alteration, many of the tuff horizons in the Saar–Nahe Basin were formerly described as ‘tonsteins’ (kaolinite-rich mudstones) in the literature (e.g. Heim, 1960, 1961, 1970) even if their volcanic origin was already established. It is our aim to discuss the effects of the depositional environment and synsedimentary tectonism on the preservation potential of the fallout ashes in the Saar–Nahe Basin and the architecture and composition of deposits formed in the ancient fluvio-lacustrine setting. We feel that the influence of synsedimentary tectonism on preservation potential has been largely neglected in the sedimentological literature to date.
GEOLOGICAL SETTING AND STRUCTURAL FRAMEWORK The Saar–Nahe Basin is situated in SW Germany and extends NE–SW from ≈ 40 km west of Frankfurt to the French–German border near Saarbrücken (Fig. 1). Considering its exposed dimensions of 120 km × 40 km, it is one of the largest of about 70 intermontane basins
that developed during late- and post-orogenic (Late Carboniferous–Permian) extension of the Variscan Mountain Belt. Towards the south, east and west of the basin, sediments of Triassic and Tertiary age largely cover the Carboniferous–Permian basin fill, and the true basin dimensions are known only from a few boreholes and seismic lines. The basin fill comprises exclusively continental sediments with a preserved thickness of ≈ 6500 m. Deposition started at the Namurian–Westphalian boundary (late Bashkirian) and continued until late Early Permian times, recording a shift from humid towards semiarid climates over > 20 Myr (see Lippolt et al., 1984; Lippolt & Hess, 1989). From the extensive development of coal seams and from palaeomagnetic investigations (Witzke, 1990) it is inferred that the basin occupied an equatorial position during Westphalian times (late Bashkirian–Moscovian) but thereafter continuously drifted northward (Ziegler, 1990). Consequently, long-term drying started when Europe moved out of the tropical rainy zone into latitudes characterized by pronounced seasonality. From Stephanian time (Kasimovian–Gzelian) onwards, these periods of seasonal drying are recorded by the occurrence of caliche nodules (Möhring & Schäfer, 1990), whereas ‘wet’ periods supported restricted peat formation.
BASIN STRUCTURE Seismic sections across the Saar–Nahe Basin (DEKORP 1C and 9N; see Henk, 1993) and the asymmetric distribution of sediment thicknesses and facies reveal the half-graben structure of the basin. The main basin-bounding fault is a south-east-dipping detachment that coincides with the surface trace of the Hunsrück Boundary Fault (Fig. 1) at the northern basin margin (Henk, 1993). Initial basin formation started at the Namurian–Westphalian boundary in a transpressional regime during late orogenic convergence. However, from Stephanian times onwards extensional plate boundary forces caused an overall E–W- to ENE–WSW-trending extensional regime within the late Palaeozoic dextral megashear zone between the Appalachians and the Urals (Arthaud & Matte, 1977). Palaeocurrent patterns, combined with facies and thickness variations, indicate a continuous north-eastward shift of the depocentre during the postWestphalian basin evolution (Schäfer, 1989). This and kinematic considerations argue for left-lateral oblique slip movements at the master fault, with different
VSIC12 9/12/08 12:59 PM Page 265
Preservation potential of fallout ashes
265
Fig. 1. Geological maps showing (A) the distribution of the Upper Carboniferous and Lower Permian units in the Saar–Nahe Basin and the location of the study area; and (B) the distribution of the Lower Permian Lebach Group containing the studied section R2–Humberg Black Shale and locations of measured logs. Major structural elements include sets of NW–SE-trending transfer faults and the NE–SW-trending synclines and anticlines.
amounts of extension along the basin axis being accommodated by pronounced sets of NW–SEtrending dextral strike-slip (transfer) faults running perpendicular to the NE–SW basin margins. Such transfer faults dominate the intrabasinal structural framework and create sub-basinal compartments characterized by differing structural style, thickness, and facies evolution (Stollhofen, 1998). In addition, syn- and antithetic normal faults and several major synclines and anticlines have developed, which run parallel to the NE–SW basin margins. The synsedimentary formation of the anticlines is indicated by reduced sediment thicknesses and coalification. In contrast, enhanced sediment thicknesses and coalification are associated with the synclines (Teichmüller et al., 1983). Most probably, the formation of the synclines and anticlines is related to a ramp and flat geometry of the basin-bounding detachment causing deformation of the hanging wall (Henk, 1993).
SYN-RIFT EVOLUTION OF THE SAAR–NAHE BASIN As a whole, the Saar–Nahe Basin fill registered four main tectono-sedimentary phases: 1 an early transpressive proto-rift phase (Namurian– Westphalian); 2 a pre-volcanic syn-rift phase (Stephanian); 3 a volcanic syn-rift phase (earliest Permian ‘Rotliegend’); 4 the final post-rift phase. Only the syn-rift development will be considered in this paper. The pre-volcanic syn-rift phase comprises a 3800– 4700 m sequence of Stephanian deposits, comprising the Göttelborn to Thallichtenberg Formations (Fig. 2). These record a time interval of 14 Myr (304–290 Ma) according to radiometric dating by Lippolt et al. (1984) and Lippolt & Hess (1989). The erosional
VSIC12 9/12/08 12:59 PM Page 266
266
S. Königer and H. Stollhofen
PERMIAN
'Rotliegend'
Group
Nahe
290Ma
Tholey Lebach
Formation
Donnersberg
Thallichtenberg Oberkirchen Disibodenberg
Odernheim-Unit
Meisenheim Lauterecken
Jeckenbach-Unit
Quirnbach
Stephanian
CARBONIFEROUS
STUDIED SECTION
Kusel
Wahnwegen Altenglan Remigiusberg
300Ma
Ottweiler
Breitenbach Heusweiler Dilsburg Göttelborn
Fig. 2. Generalized Stephanian and ‘Rotliegend’ stratigraphy of the Saar–Nahe Basin illustrating the stratigraphic position of the studied section. Radiometric dates from Hess (1985) and Lippolt & Hess (1989).
unconformity between Westphalian and Stephanian strata clearly represents a major megasequence boundary coinciding with a complete reorganization of intrabasinal fault kinematics, as well as the drainage patterns within the basin. In contrast to the earlier north-westward sediment derivation in Westphalian time, from Stephanian time onwards, sediment supply was mainly from source rocks located to the southeast, such as the granites, gneisses, and mica schists of the Northern Vosges, the Odenwald, and the Black Forest areas (Schäfer, 1989). Lithology is dominated by a complex intercalation of grey lacustrine mudstones and fine-grained sandstones, as well as some thicker sand-rich fluvio-deltaic wedges with minor but significant units of fluvial conglomerates, limestones, and coal (Schäfer, 1986; Stapf, 1990b). Thin (< 0.5 m) but laterally extensive ash tuff beds in the Stephanian and lowermost Permian sequences of the Saar–Nahe Basin are related to extrabasinal sources (Stollhofen, 1994; Königer et al., 1996; Königer, 1999). Planar bed contacts, planar bedding, good sorting, and grain-size characteristics of typical primary preserved layers imply a deposition by fallout, distinctly distal from the source. All of the tuff beds at this stratigraphic level reveal an increasing number and thickness, associated with decreasing grain sizes of pyroclastic particles, towards the southern basin margin. Considering the overall southerly palaeowind directions during Late Carboniferous time (e.g. Parrish, 1982), together with rock chemistry
and radiometric datings, Stollhofen (1994) suggested that granites within the Black ForestaNorthern Vosges region represent the subvolcanic part of volcanic centres, which provided a potential source for the pyroclastic deposits. Most commonly, lithologies are arranged in superimposed upward-coarsening and -shallowing cycles recording gradual transitions from offshore-lacustrine mudstones through delta-front sandstones to pebbly cross-stratified sandstones of the delta plain (Schäfer & Sneh, 1983). The latter have been successively numbered as horizons R1–R7 by Boy et al. (1990) and are important marker beds together with packages of offshore-lacustrine black shales (Schäfer, 1986). Intrabasinal facies patterns suggest repeated progradation of such fluvial deltas and a predominant sediment input from the southern, tectonically less active margin of the half-graben. There, a progressive southward enlargement of the depositional area is indicated by onlapping geometries in seismic sections such as the DEKORP 9N line (Henk, 1993). In contrast, at the northern, tectonically active boundary of the halfgraben, local alluvial fans developed (Stapf, 1990b) and maximum thicknesses of deep-water lake facies occur. Basin evolution was associated with a general north-eastward shift of the depocentre, which is also the principal direction of the palaeodrainage system (Schäfer, 1986). During the volcanic syn-rift phase, extensional tectonics and the north-eastward shift of the depocentre continued but were associated with widespread intrabasinal magmatic activity (Stollhofen & Stanistreet, 1994). Basaltic–andesitic lava flows (Schwab, 1981) and rhyolitic–dacitic pyroclastic rocks (Stollhofen, 1994) form the dominant volcanic extrusive rock types. Shallow intrusions occur predominantly as rhyolitic–dacitic domes and basaltic–andesitic sills and dykes. The resulting lithological record comprises reddish, mainly fluvial sandstones deposited by meandering rivers and several laterally extensive tuffaceous units and lavas. The sedimentary and associated volcanic rocks belong to the Donnersberg Formation of the ‘Rotliegend’ Nahe Group. They form a 1100 m succession (Stapf, 1990a), which was deposited during a time interval of ≈ 4 Myr according to Rb/Sr ages of 290–286 Ma (Lippolt et al., 1989; J. C. Hess, personal communication). Considering the Carboniferous– Permian boundary of 298 Ma as proposed by ClaouéLong et al. (1995), underlying strata of the Remigiusberg to Thallichtenberg Formations, which are traditionally referred to by the lithostratigraphic term ‘Lower Rotliegend’ in the literature (e.g. Schäfer,
VSIC12 9/12/08 12:59 PM Page 267
Preservation potential of fallout ashes 1986), partly relate to the higher parts of the Stephanian succession on the basis of available numerical age data (see Menning, 1995). For any regional comparison of strata it is important to keep this discrepancy between lithostratigraphic nomenclature and radiometric ages in mind. All other lithostratigraphic terms used in this paper are taken from Boy (1989) and Stapf (1990a).
UPPER STEPHANIAN DEPOSITIONAL ENVIRONMENTS AND VOLCANISM Tephrostratigraphy The present study deals with an interval of the lower-
Fig. 3. Overview log of the studied section illustrating positions of important tephrostratigraphic and lithostratigraphic marker horizons.
267
most Permian succession (Fig. 2) that is particularly well exposed in the field and has been sampled from several drill cores. Figure 3 gives a lithostratigraphic compilation of the entire examined section and introduces the stratigraphic terminology used in the text. The section includes parts of the Jeckenbach and Odernheim Units of the Lebach Group with the mappable stratigraphic markers R2 at the base and the Humberg Black Shale at the top of the succession. The total thickness varies regionally between 260 m in the area of the Pfalz Anticline (borehole Münsterappel 1) and 370 m in the adjacent Nahe and Pfalz Syncline structures (Fig. 1). More than 20 different tuff horizons can be distinguished within the studied section but only seven provide important tephrostratigraphic markers: the
VSIC12 9/12/08 12:59 PM Page 268
268
S. Königer and H. Stollhofen
Humberg, Gaugrehweiler, Kappeln, Hesselberg, Raumberg, St Alban, and Pappelberg Tuffs. These horizons are up to 70 cm thick and occur widely within the basin over at least 60 km lateral extent. On the basis of their fabrics and compositional characteristics they are easy to distinguish from the enclosing siliciclastic ‘background’ sediments. Two tuff horizons (Hesselberg and Raumberg Tuff ) split laterally into two or more layers, which are separated by finegrained siliciclastic sediments. Other tuff horizons, such as the Odernheim, Kuhtrift, Jeckenbach, Hoferhof, and Windhof Tuffs have a more restricted lateral distribution within the study area and thicknesses of only a few centimetres. As a result of sedimentary reworking, some of the latter tuffs show significant admixture of siliciclastic detritus and reveal sedimentary structures indicating subaqueous transport. Petrography All of the examined tuff beds show a uniform assemblage of juvenile magmatic components. These comprise completely recrystallized platy and cuspate relict glass shards up to 200 µm in diameter, which are embedded in an equigranular, crypto- to microcrystalline quartz–kaolinite matrix. Pumice fragments are rarely preserved but may reach 500 µm in length. Crystal components are dominated by thorn-shaped, inclusion-free volcanic quartz splinters of up to 650 µm maximum grain size. Crystals of euhedral sanidine and marginally corroded plagioclase are up to 500 µm in diameter and are usually extensively altered to kaolinite and sericite, respectively. Biotite shows euhedral crystal outlines, may be up to 1.5 mm in diameter and contains a few apatite microlites besides abundant dark brown pleochroic haloes around inclusions of zircon and monazite. The heavy-mineral assemblage comprises mainly zircon, apatite, and monazite besides minor amounts of sphene, hornblende, tourmaline and rutile. Altogether, the juvenile components of the tuffs, together with geochemical analyses, indicate a rhyolitic– dacitic composition of the parent magma (Königer, 1999). No pyroclastic lithic components were found in primary deposits. Pyroclastic material may be mixed with detrital material derived from the siliciclastic ‘background’ sedimentation to varying degrees. The adjacent siliciclastic sediments include pebbles of quartzite, granite, chert, and a few gneiss particles. Inclusion-rich milky quartz, ‘plutonic’ quartz (characterized by inclusion trains), orthoclase, microcline, plagioclase, and muscovite form the majority of the components found in the sand fraction. Heavy minerals
comprise abundant rounded zircon, tourmaline, and apatite as well as less common rutile, epidote, and staurolite. According to the classification of McBride (1963) the Stephanian sandstones are sublitharenites and lithic arkoses. Lithofacies Figure 4 shows a detailed section through the lithofacies assemblages of two of the lacustrine–deltaic cycles between the marker beds R4 and R6. This section is from the Münsterappel 1 borehole core, drilled into a condensed sequence developed on top of the Pfalz Anticline. The section shows the development of three major facies assemblages, which are arranged in combined, upward-fining and upward-coarsening cycles: 1 an offshore-lacustrine facies association; 2 a prodelta to delta front facies association; 3 a delta plain facies association. Table 1 gives a summary of the characteristics of each lithofacies. The offshore-lacustrine facies association This association consists of four lithofacies: 1 laminated black shales; 2 micritic limestones; 3 normally graded, fine-grained sandstones and tuffites; 4 normally or reverse-graded tuffs. Laminated black shales are 0.2–2.0 m thick and can be traced over at least 10 km with several shales (e.g. Jeckenbach, Odernheim, Kappeln, and Humberg Black Shale) extending over > 80 km in a north-east– sout-west direction (see Stapf, 1990b). The black shales are composed of ferruginous, carbonaceous, biogenic, or clastic ‘non-glacial varves’ (Stapf, 1990b) with an average lamina-couplet thickness of 0.38 mm (Bangert, 1994). Occasionally they are rich in wellpreserved fossil remains such as amphibians, fishes (Paramblypterides, rarely Xenacanthus), ostracods and conchostracans (see Boy, 1989, for compilation). They are further characterized by significant heavy metal (Cu, V, U, Pb, Zn, Mo) concentrations (Herzberg, 1966) in association with the occurrence of siderite and pyrite as layers, nodules, and finely dispersed particles. It is inferred that these deposits originated from suspension fallout within temporarily closed lakes developing euxinic lake-floor conditions and a stable chemical and/or thermal density stratification. Only rarely was bedding disturbed by wave activity in the shallow (up to 10 m depth) lakes (Schäfer, 1986). This observation may be explained by the protective influence of algal mats floating at the water surface,
VSIC12 9/12/08 12:59 PM Page 269
Preservation potential of fallout ashes
269
Fig. 4. Detailed log of the R4 –R6 cycles including sedimentological interpretation.
thus suppressing wave activity (Boy & Hartkopf, 1983), rather than by a general absence of storm events. Micritic limestones are up to 6 cm thick and show a wavy lamination. They formed either as precipitates of planktonic algae caused by seasonal mass extinctions (Rast & Schäfer, 1978) or as inorganic precipitates of Mg-calcite, produced during periods of closed lake conditions and enhanced evaporation (Willems & Wuttke, 1987). Grey, fine-grained sandstones and tuffites, only a few millimetres thick, are normally graded with sharp
planar bases and occasionally developed small load structures. These are interpreted as microturbidites with increasing thicknesses towards the lake margins. Examples of more locally developed tuffaceous turbidite deposits in an offshore-lacustrine setting are associated with the Humberg tephrastratigraphic marker horizon. Tuff horizons such as the Humberg, Odernheim, and Jeckenbach Tuffs are only a few millimetres to 8 cm thick but laterally show fairly constant thicknesses. They are characterized by sharp planar contacts,
Tabular Tabular
Tabular
Sheet Lenticular to tabular Lenticular Tabular Tabular to lenticular
Normal grading, plane lamination–ripple cross- or plane bedding, scour marks, load, dewatering structures, rip-up clasts Horizontal to irregular wavy lamination Ripple cross-bedded or massive, slide, slump, ball-and-pillow, flame structures Massive or horizontal lamination, normal or reverse grading Trough cross-bedded, scoured bases Climbing-ripple cross-bedded, massive, normally graded Plane bedding, plant-rich Plane lamination, desiccation cracks Normal or reverse grading, thin-, thick-, cross-bedded
Fine- to mediumgrained sandstones and ash tuffites
Biogenic limestones
Sandstones
Ash tuffs
Pebbly sandstones
Sandstones and ash tuffites
Fine-grained sandstones
Mudstones
Ash tuffs
Tabular to lobate
Tabular
Tabular
Tabular
Massive, hummocky crosslamination, wave-rippled top contacts, normal grading
Multiple normal reverse grading, plane lamination
Ash tuffs
Non-erosive, sharp planar
Non-erosive, sharp planar
Non-erosive, sharp planar
Non-erosive, sharp planar
Erosive, sharp
Non-erosive, sharp planar
Non-erosive, irregular
Non-erosive, sharp planar
Erosive, irregular
Erosive, sharp planar
Non-erosive, sharp planar
Non-erosive, sharp planar
Non-erosive, sharp planar
Sheet to lobate
Mudstones and fine-grained sandstones
Normal grading, small basal load structures
Fine-grained sandstones and tuffites
Non-erosive, sharp planar
Tabular
Plane bedding, wavy rippled
Wavy lamination
Micritic limestones
Non-erosive, sharp planar
Basal contact
Tabular
Geometry
Mudstones and fine-grained sandstones
Horizontal lamination
Sedimentary structures
Black shales
Lithofacies
None
Rootlets
Plant remains, plant debris
Plant debris
None
None
Plant debris
Ostracods, conchostracans
None
None
Plant debris
None
None
Planktonic algae
Amphibians, fishes, ostracods, conchostracans, plant debris
Fossils
Primary [phreato] Plinian volcanic ash fall deposits including reworked ash
Poorly to well-drained flood plain deposits
Interdistributary bay area deposit under permanent water cover
Distal crevasse splay deposits
Minor distributary channel deposits
Primary [phreato] Plinian volcanic ash fall deposits including turbiditic deposits
Laterally coalescing mouth bar forming progradational sheet sand deposits
Shallow-water stromatolite and oncolite beds
Proximal to distal turbidites
Storm-deposited beds
Suspension fallout
Primary [phreato] Plinian volcanic ash fall deposits
Distal microturbidites
Algae blooms, low sediment supply
Suspension fallout
Facies interpretation
270
Delta plain
Prodelta to delta front
Offshorelacustrine
Facies association
Table 1 Summary of siliciclastic lithofacies characteristics of various depositional settings in the studied fluvio-lacustrine sequence of the Saar–Nahe Basin
VSIC12 9/12/08 12:59 PM Page 270
S. Königer and H. Stollhofen
VSIC12 9/12/08 12:59 PM Page 271
Preservation potential of fallout ashes planar lamination, and multiple normal grading, with individual layers not thicker than 1 mm. Usually the tuffs are well sorted and consist mainly of fine ash grains that have been completely recrystallized. Rarely, pumice fragments, up to 500 µm in size, appear concentrated at the top of individual layers and reflect density grading. The pyroclastic deposits are interpreted to represent Plinian or phreatoplinian ash falls derived from source areas located ≈ 150 –250 km to the south of the Saar–Nahe Basin in the Black ForestaNorthern Vosges area (Stollhofen, 1994; Königer, 1999). Gravity sorting within the winddriven eruption plume led to aeolian fractionation and a lateral grain-size fining of individual components (e.g. heavy minerals) from south to north across the field area, proportional to the transport distance. The prodelta to delta front facies association This association dominates the section and involves six lithofacies: 1 planar-bedded or wave-rippled mudstones and finegrained sandstones; 2 massive or hummocky cross-laminated mudstones and fine-grained sandstones; 3 normally graded, fine- to medium-grained sandstones and ash tuffites; 4 biogenic limestones; 5 ripple cross-bedded or massive sheet-like sandstones; and 6 normally or reverse-graded tuffs. Planar-bedded or wave-rippled mudstones and fine-grained sandstones show a greyish to yellowish grey coloration. They are characterized by accumulations of plant debris and muscovite flakes on bedding planes and are interpreted to result from suspension fallout in a setting that was only intermittently wave influenced. Massive or small-scale hummocky cross-laminated mudstones and fine-grained sandstones typically developed erosional base and wave-rippled top contacts. Massive layers are up to 6 cm thick with tabular geometries. They drape the erosional surfaces of underlying beds and occasionally show normal grainsize grading. Hummocky cross-laminated units, up to 0.3 m thick and with a wavelength of up to 0.5 m, are restricted to sandy substrates. Both subfacies are interpreted to result from storm-depositional events, when enhanced wave energy led to local reworking of the lake floor. Under relatively shallow-water conditions where waves affected the lake floor, this resulted in hummocky cross-bedded units, whereas massive
271
units formed in deeper-water settings where the stormreworked detritus was resedimented from suspension. The normally graded, fine- to medium-grained sandstone and ash tuffite beds are up to 1.4 m thick and show tabular geometries. They are characterized by abundant scour marks, load (load casts, ball-andpillow) and dewatering (flame, dish-and-pillar) structures, and the occurrence of rip-up clasts. The upward succession of bedding features (planar lamination– ripple cross-bedding–planar bedding) broadly implies upward decreasing flow velocities. We agree with the interpretation of Negendank (1972) and Schäfer & Sneh (1983) that these sediments represent turbidites deposited in distal to proximal prodelta settings of well-oxygenated lakes. Biogenic limestones are up to 0.2 m thick and exhibit tabular geometries. In accordance with Stapf (1990b) they are interpreted to represent thin shallowwater stromatolite and oncolite beds containing ostracods and conchostracans, which together reflect periods of reduced sediment input. Ripple cross-bedded or massive, sheet-like sandstones may be up to 2 m in thickness and are rich in plant debris. Commonly, these beds are associated with a wide spectrum of soft-sediment deformation features such as slide and slump structures, ball-and-pillow structures and flame structures. Lithological characteristics and sediment body architectures are considered to indicate laterally coalescing mouth bar deposits forming progradational delta-front sheet sands. The normally or reverse-graded tuff facies is similar to that already defined in the previous facies association. Field examples comprise the Raumberg and Hesselberg Tuffs, which also occur in offshorelacustrine environments. Besides these, more massive or poorly bedded tuff horizons such as the Gaugrehweiler, Kappeln, and St Alban Tuffs are widespread, with enhanced thicknesses up to 60 cm and an admixture of siliciclastic detritus reflecting deposits of tuffaceous turbidites after primary fallout. The delta plain facies association This facies association usually forms the top part of the upward-coarsening cycles. Five lithofacies can be identified: 1 trough cross-bedded, pebbly sandstones; 2 climbing-ripple cross-bedded or massive sandstones and ash tuffites; 3 plant-rich planar-bedded fine-grained sandstones; 4 planar-laminated mudstones; 5 normally or reverse-graded tuffs.
VSIC12 9/12/08 12:59 PM Page 272
272
S. Königer and H. Stollhofen
Trough cross-bedded, pebbly sandstones are 0.5– 5.0 m thick and show distinctly scoured bases, which downcut as much as 1.5 m into the underlying succession. Palaeocurrent directions deduced from the dip of the tangential foresets are highly variable but reflect dominantly north-east-directed transport. Pebbles are strictly confined to the basal parts of the sandstone beds. They are well rounded with a maximum grain size of 1.5 cm and comprise quartzite, milky quartz, chert and, rarely, granite compositions. This facies is interpreted to represent minor distributary channel deposits of the delta plain. The climbing-ripple cross-bedded or massive sandstones and ash tuffites are sharp-based lenticular to tabular bodies. Beds may appear massive, and typically exhibit normal grain-size grading. Such facies form isolated sand bodies interbedded with the planarlaminated mudstone facies and reflect crevasse splay deposits relatively distal to the feeder channel. Plant-rich, planar-bedded fine-grained sandstones are dark grey because of the abundance of iron and manganese oxides and organic material, in the form of both finely dispersed fragments and aligned larger plant remains of a hygro- or mesophile floral assemblage (Kerp & Fichter, 1985). No rootlet horizons were observed. Deposits have lenticular geometries and vary between 10 and 30 cm in thickness. Such accumulations of ‘allochthonous’ plant material point to deposits of interdistributary bay areas with permanent water cover. Planar-laminated mudstones are characterized by tabular bed architectures and have a grey or reddish green colour. The grey mudstone facies locally preserves rootlets whereas the reddish green mudstones may contain desiccation cracks and pedogenic carbonate nodules up to 1 cm in diameter. These facies reflect contrasting deposits of both poorly and welldrained flood plains. The normally (rarely reverse-) graded tuff facies occurring in this setting is the same as that in the two previous facies associations. The tuff horizons are interbedded with both crevasse splay and flood plain sediments, and their commonly normally graded and occasionally ripple cross-bedded structure indicates subaqueous ash deposition at high rates of suspension fallout. Reverse grain-size grading developed only where pumice particles were available. An example of a tuff formed in this setting is the Pappelberg Tuff.
(average 45 m) thick. The average cycle duration has been estimated to be of the order of 245 kyr on the basis of the span of time represented by the entire prevolcanic syn-rift unit (14 Myr) subdivided into a total of 57 depositional sequences. Cycles are characterized by a pronounced asymmetry consisting of relatively thin transgressive but thick regressive successions. Rapid transgressive developments are reflected by either black shales or interbedded mudstones and fine-grained sandstones, of the offshore-lacustrine and prodelta facies associations, respectively, directly overlying distributary channel and flood plain deposits of the delta plain facies association. Less pronounced transgressions are recorded by a more sawtooth-like grain-size pattern reflecting repeated periods of flooding followed by infilling. Nevertheless, even in these cases, an overall upward-fining and upward-thinning trend of the bedsets is evident in the sections. Laminated black shales and limestones are interpreted to represent condensed intervals within the succession, which record maximum and more continuous flooding of the Permian lakes and a drastically reduced clastic input. Higher in the succession, grain-size trends reflect a regression. Minor black shale units are interbedded with sandy turbidite, tempestite, and mouth bar deposits that increase upward in frequency and thickness. These progradational delta front sheet sands are associated with a wide spectrum of soft-sediment deformation features such as slide and slump structures, which may be related to a falling lake level causing failure of up-dip fans and delta slopes. Mouth bar deposits and the delta plain facies association constitute the uppermost part of the overall upwardcoarsening succession. Occasionally, flood plain deposits in the uppermost parts of the cycles are oxidized and contain pedogenic features. During subsequent lake-level lowstands, these were sporadically incised by distributary channels of the delta plain. The whole cycle thus reflects a rapid deepening of the lake succeeded by a long-lasting shallowing trend from distal prodelta–offshore to subaerial delta plain environments. This development suggests an overall decrease in rates of accommodation space formation and consequently declining potential for vertical accretion and preservation of sediment.
PRESERVATION OF THE VOLCANIC ASH TRANSGRESSIVE–REGRESSIVE CYCLES Facies associations are organized into cycles 18–80 m
The relationship between preservation potential of fallout ash and the depositional environment has been considered in various continental and marine settings.
VSIC12 9/12/08 12:59 PM Page 273
Preservation potential of fallout ashes In general, fallout ash beds have a higher preservation potential in subaqueous rather than subaerial settings (Huff et al., 1996). Palmer & Shawkey (1997) discussed the high preservation potential of tephra deposits in the lacustrine record in contrast to the low preservation potential in fluvial settings. On alluvial plains, primary pyroclastic material can be easily remobilized by fluvial systems (Nakayama & Yoshikawa, 1997). In our study we evaluated the preservation potential of fallout ash in a variety of fluvio-lacustrine subenvironments. Figures 3 & 4 illustrate the positions of tuff beds within the framework of fluvio-lacustrine cycles. Generally, the intercalation of pyroclastic beds is not restricted, but appears to be distinctly concentrated within the transgressive parts of depositional cycles. This is the general case throughout the Stephanian sequence. Of the 20 tuffs identified within the studied section, 16 are contained within the transgressive parts of the cycles. Huff et al. (1996) made similar observations in marine-influenced depositional environments and recognized that the highest preservation potential of Palaeozoic ash beds is in transgressive deposits formed during times of widespread epicontinental flooding. In our study, tuff layers of the offshore-lacustrine facies association, deposited during transgressions, are suggested to have the highest preservation potential for several reasons: 1 the availability of accommodation space during lake-level rise; 2 the virtual absence of wave reworking and fluvially induced currents in offshore-lacustrine settings; 3 the negligible sedimentation rates of non-pyroclastic material; 4 the absence of bioturbation, as a result of anaerobic lake-floor conditions. Such conditions favour successions of individual ash layers amalgamating into thicker fallout units. This is illustrated by Fig. 5, which shows a vertically stacked succession of graded, 0.1– 0.2-cm-thick fallout layers without any reworking or intercalation of siliciclastic interbeds. In contrast, Fig. 6 shows individual fallout tuff layers preserved with primary fabrics and compositions but separated by coaly layers consisting of accumulated plant debris. As both tuff horizons were deposited in a similar depositional environment the latter could reflect longer-lasting inter-eruption intervals or reduced rates of ash fallout. Only minor amounts of reworked ash material were observed in offshore-lacustrine settings. Such horizons are exclusively formed by microturbidites of a few millimetres to 2 cm thickness (Fig. 7), which are preferentially
273
Fig. 5. Thin-section photograph of the Humberg Tuff showing a vertically stacked succession of normally graded, 0.1– 0.2-cm-thick fallout layers without any reworking or siliciclastic interbeds. Scale is 0.5 cm.
Fig. 6. Thin-section photograph of the Odernheim Tuff showing the individual fallout tuff layers preserved with primary fabric and composition but separated by coaly layers consisting of accumulated plant debris. Scale is 0.5 cm.
interbedded with fallout tuff layers or with black shales overlying fallout tuffs. As documented by the Humberg marker horizon, zones of interbedded ash– tuffaceous turbidites and black shales may reach cumulative thicknesses of up to 0.4 m comprising an average of 25 turbidite layers per 10 cm.
VSIC12 9/12/08 12:59 PM Page 274
274
S. Königer and H. Stollhofen Borehole core [cm] St.Alban-Tuff 60
Siliciclastic suspension fallout
Fallout tuff bed Plant debris rich suspension fallout
55
50
45
Fig. 7. Hand specimen of the Humberg Tuff showing yellowish tuffaceous layers intercalated with grey mudstone layers of an offshore-lacustrine setting. The normally graded tuffite layers have load structures at base contacts and are interpreted as microturbidites. Scale is 1 cm long.
40
Tuffaceous turbidite bed
35
30
Prodelta and delta front settings are characterized by high sediment delivery rates, particularly during regression, and ash layers are rare and highly diluted by siliciclastic material in delta front environments. Maximum thicknesses in prodelta settings are related not only to the preservation of primary deposited pyroclastic fallout but also to considerable accumulations of reworked pyroclastic material deposited from turbidity currents. Figure 8 illustrates an example from the St Alban marker bed, consisting of only four minor fallout layers, up to 7 cm thick, but two thicker, graded tuffaceous turbidite beds, which make up 41 cm of the 60 cm core interval. Reworking of the ash is reflected not only in the admixture of siliciclastic detritus, but also by current-induced sedimentary structures such as ripple cross-bedding shown in Fig. 9. As a result of high sedimentation rates, primary preserved fallout tuff horizons developed abundant hydroplastic deformation structures, particularly in the prodelta setting. Slide, slump, and water escape structures are widespread and modify the original planar lamination of the tuff layers as revealed by the Kappeln and Raumberg Tuff horizons (Fig. 10). In addition, in well-oxygenated parts of the lakes, bioturbation may cause considerable disturbance of the original fabrics. Tuff horizons deposited within a delta plain environment have only a moderate preservation potential as they may be easily affected by lateral channel switching causing rapid erosion of the ash. Fallout tuffs of this setting therefore tend to have lenticular bed geometries, and tephrostratigraphic markers such as the Kuhtrift Tuff are only of local importance.
25
20
Fallout tuff bed
15
Tuffaceous turbidite bed
10
Fallout tuff bed 5
Fallout tuff bed 0
Normal grain size grading No bedding, massive Lamination or plane-bedding Accumulation of plant debris Cross-bedding Dewatering structure Fig. 8. Detailed section of a borehole core comprising the St Alban tephrostratigraphic marker. Graded, tuffaceous turbidite beds dominate the succession and preserved primary fallout tuff layers form only minor interbeds.
VSIC12 9/12/08 12:59 PM Page 275
Preservation potential of fallout ashes
Fig. 9. Photograph of the St Alban tephrostratigraphic marker, illustrating ripple cross-bedding in a tuffaceous turbidite horizon.
Fig. 10. Photograph of the Raumberg Tuff. Diapiric dewatering structures fold and bend the original plane lamination of the fallout tuff layers. Scale is 1 cm long.
However, considerable thicknesses of marker horizons may result from intercalation of tuffaceous crevasse splay deposits, which show admixture of siliciclastic material to varying degrees. An example is provided by the Pappelberg Tuff. This marker consists of up to three stacked layers consisting of normally graded tuffaceous crevasse splay deposits with a total thickness of up to 20 cm. Non-pyroclastic flood plain and crevasse splay deposits of several centimetres thickness form local interbeds, which at their bases occasionally show minor features of erosion and reworking of the underlying tuffite beds. Tectonic control on ash preservation Not only the depositional environment but also the tectonic environment had considerable influence on the preservation potential and thickness distribution of volcanic ash. Figures 11–13 illustrate the lateral distribution of the tephrastratigraphic markers and associated lithofacies in the study area in the north-eastern
275
part of the Saar–Nahe Basin. On a regional scale, the entire studied interval thins towards the Pfalz Anticline in NW–SE cross-sections (Fig. 11). The lithological succession in Logs 13 and 30 within the Nahe Syncline is dominated by the prodelta to delta front facies association, whereas Log 28 along strike of the Pfalz Anticline shows thinner prodelta but relatively thick delta plain successions. We interpret these differences to reflect reduced subsidence in the vicinity of the syndepositionally evolving anticline structure. In particular, marker horizons dominated by reworked pyroclastic material of tuffaceous turbidites and crevasse splays such as the Hesselberg, Pappelberg, and Hoferhof Tuffs pinch out towards the anticline. South-west–north-east-trending cross-sections reveal the influence of syndepositionally active transfer faults on thickness and facies development of sedimentary units. Figures 12 & 13 demonstrate distinct fault-related thickness changes, interpreted to reflect hanging-wall and footwall positions, respectively, of the depositional area. Examples of footwall blocks are represented by the basin segments containing the measured Logs 14, 16 (Fig. 12) and 9 (Fig. 13). Only a few tuff horizons are preserved within basin segments occupying a footwall position during deposition. This is particularly valid for the succession R3–R6, which only locally preserved tephrostratigraphic markers south-west of the Lauter Fault. In contrast, hangingwall positions favour both the preservation of primary fallout tuffs and the accumulation of reworked pyroclastic material. This is demonstrated by Log 17, situated north-east of the Alsenz Fault. The hangingwall block preserved almost completely the succession of tephrostratigraphic markers and, because of the enhanced subsidence, the Raumberg marker splits south-westwards into two separate horizons interleaved with up to 6-m-thick deposits of the prodelta facies association. Moreover, different horizons offset by the same fault show variable magnitude and sense of separation. This suggests that the faults were repeatedly rejuvenated and that activity included alternating polarity of fault movement during basin history. An example is shown by Log 4 in Fig. 13. Considering relative thickness development, the depositional area initially occupied a footwall position, but after deposition of the Hesselberg Tuff, a hanging-wall position developed. The adjacent Log 5 implies the opposite succession, with an early hanging-wall development that later changed to a footwall position. It is suggested that fault movements caused the formation of topographic barriers, thus controlling
276
Fig. 11. NW–SE cross-section illustrating contrasting facies and thickness development of synsedimentary evolving anticline and syncline structures. (See Fig. 1 for locations of measured logs and Fig. 12A for legend.) Tephrastratigraphic markers do not pinch out towards the north-west but were not considered in available core descriptions of borehole Monzingen 1.
VSIC12 9/12/08 12:59 PM Page 276
S. Königer and H. Stollhofen
Preservation potential of fallout ashes
Fig. 12. SW–NE cross-section documenting the complex architecture of the Odernheim and Jeckenbach Units along the north-west flank of the Pfalz Anticline. (See Fig. 1 for locations of measured logs.) Rapid thickness changes occur across synsedimentary faults. Hanging-wall blocks such as the area containing Log 17 show a rather complete preservation of tephrastratigraphic markers, whereas footwall blocks (Logs 14, 16, 23, 24, and 25) tend to preserve an incomplete record.
VSIC12 9/12/08 12:59 PM Page 277
277
VSIC12 9/12/08 12:59 PM Page 278
S. Königer and H. Stollhofen
Fig. 12. (continued )
278
Fig. 13. SW–NE cross-section of the Odernheim and Jeckenbach Units along the south-east flank of the Pfalz Anticline, illustrating the changing magnitude and polarity of displacement at the Feilskopf Fault. (See text for further explanation and Fig. 1 for locations of measured logs.) The legend is given in Fig. 12A.
VSIC12 9/12/08 12:59 PM Page 279
Preservation potential of fallout ashes 279
VSIC12 9/12/08 12:59 PM Page 280
280
S. Königer and H. Stollhofen
intrabasinal sediment dispersal and formation of accommodation space. However, in offshore settings, fault development and tectonic activity are generally not as well recorded in the sedimentary succession for two reasons: the suspension-dominated sedimentation is not as sensitive to relatively small fault displacements; and because the lacustrine sediments accumulated below base level, the rate of accumulation is predominantly controlled by the rate of sediment supply and does not necessarily reflect fluctuations in the rate of accommodation development. Repeated fault movements are thought to cause a variety of deformation structures, such as clastic dykes, convolute lamination, intraformational breccia, and microfaults, which are restricted to the vicinity of fault zones and concentrated within the transgressive parts of depositional cycles. Most probably they are genetically related to seismic shocks associated with fault movements (see Plaziat et al., 1990). The whole spectrum of deformation structures has been described and a more
complete discussion of their origin given by Stollhofen (1998). The asymmetry of the depositional cycles, the evidence of rapid rather than gradual flooding, the concentration of seismogenic deformation features, and the most pronounced facies and thickness changes across synsedimentary faults within transgressive units all imply that subsidence events, probably coupled with earthquakes, played a major role in providing accommodation space. Blair & Bilodeau (1988) discussed the genetic link between lake-level rise and tectonic activity. A series of subsidence events rapidly increase accommodation space, and, assuming constant rates of sediment supply, the volume of sediment available per time unit is no longer able to fill the entire basin. This causes a relative lake-level rise and the restriction of alluvial fans to basin margins. In contrast, fluvial conditions and the progradation of marginal alluvial fans are favoured during periods of low subsidence when supplied sediment volumes exceed the available
Fig. 14. Schematic block diagram illustrating the preservation potential of fallout ash in various fluvio-lacustrine subenvironments.
VSIC12 9/12/08 12:59 PM Page 281
281
Preservation potential of fallout ashes accommodation space. In the Saar–Nahe Basin, maximum preservation potential for volcanic ash layers was reached when subsidence events occurred contemporaneously with wet climatic intervals, which allowed the tectonically created accommodation space to fill with water. A minimum preservation potential was reached when low rates of tectonically generated accommodation space coincided with dry climatic intervals.
CONCLUSIONS Both primary fallout tuffs and reworked tuffs of the Odernheim and Jeckenbach Units form important tephrastratigraphic marker horizons in the purely continental fluvio-lacustrine succession of the Saar– Nahe Basin. The broad-scale depositional subenvironments are shown in a schematic block diagram (Fig. 14). After rapid deposition of large amounts of unconsolidated tephra, even in topographically elevated areas (mantle bedding), subaerially exposed deposits were highly sensitive to reworking. This may have been triggered by gravitational forces or rainfall, and earthquakes may have mobilized even cohesive, partly consolidated tephra. The majority of
A
reworked pyroclastic material, mixed with siliciclastic detritus to varying degrees, was then redeposited as crevasse splay deposits and turbidites in prodelta settings. Fallout ashes were preferentially preserved in offshore-lacustrine and prodelta settings during transgressions and on extensive interdistributary areas of the delta plain. Table 2 gives an overview of characteristic thicknesses, textures, and the varying preservation potential associated with these depositional settings. However, the preservation potential and the anatomy of the entire tephrostratigraphic succession are related not only to the depositional environment but also to the tectonic setting (Table 2). Fault-generated topography modified the depositional surface and strongly influenced the available accommodation space. Figure 15A–E illustrates a hypothetical sequential development. The initial depositional surface (Fig. 15A) is flat and favours preservation of the original tabular geometry of a fallout ash layer (I). After onset of faulting (Fig. 15B), a succeeding fallout unit (II) is preserved in both footwall and hanging-wall blocks but the thickness of the resulting deposit is enhanced in the hanging-wall position as a result of accumulation of reworked ash. Other fallout ash layers (III and IV) are either partially (Fig. 15C) or completely
D
Depositional surface not modified by faulting
I
I
Ash completely eroded
IV
III II I
III II I
B
Hanging wall accumulation of footwall derived ash
II I
II I
Fig. 15. Tectonic model for fluviolacustrine settings showing a hypothetical succession of fault displacements favouring erosion of fallout ash in footwall positions versus enhanced preservation and accumulation of reworked ash in hanging-wall positions. (Note the contrasting tephrostratigraphic record in footwall and hanging-wall blocks (E).)
E Lateral splitting of tuff horizon due to hanging wall confined siliciclastic deposition and amalgamation of ash layers in footwall positions
V IV III
C III II I
Ash locally eroded in footwall position
IV+V III II I
II I III II I
I-V
Tephrostratigraphic marker
Hanging-wall and footwall blocks
Distal crevasse splay ash tuffites
Fallout ash tuffs
Klausweiler Tuff Gaugrehweiler Tuff Kappeln Tuff Hesselberg Tuff (locally) Raumberg Tuff (Fig. 10) Windhof Tuff
High
Normal or reverse grading; massive, plane lamination to bedding, wellsorted, occasional detrital contamination (especially at delta front), hydroplastic deformation
Pappelberg Tuff
Low Massive, normal grading, plane- or climbing-ripple cross-bedded, high siliciclastic contamination or thin interbeds
Up to 20 cm, thinning towards footwall blocks
Pappelberg Tuff Kuhtrift Tuff
Low to moderate Same as on hanging-wall blocks but often reworked or entirely eroded, higher detrital contamination
0–15 cm
Footwall blocks
Hanging-wall and footwall blocks
Pappelberg Tuff Kuhtrift Tuff
Low to moderate
Massive, commonly normally graded, planar thin- to thick-bedded, occasionally detrital contamination, sometimes siliciclastic interlayers
Up to 15 cm
Hanging-wall blocks
Moderate to high Kappeln Tuff St Alban Tuff (Fig. 8)
Massive, laminated or ripple crossbedded, normal grading, moderate to high detrital contamination, minor plane debris accumulations
Up to 36 cm, thinning towards footwall blocks
Gaugrehweiler Tuff Kappeln Tuff St Alban Tuff (Fig. 9) Low to moderate
Massive, planar thin- to thickbedding, ripple cross-bedding, high amounts of admixed siliciclastic material, local siliciclastic interlayers
Low to moderate Same as in hanging-wall blocks, but locally reworked or completely eroded, higher detrital contamination
Same as in hanging-wall blocks
Humberg Tuff (Fig. 7)
Very high
Multiple normal or reverse grading, plane lamination, well-sorted, sharp planar contacts, rarely minor detrital contamination, thin coaly intereruptive interlayers Normal grading, basal load structures, non-erosive, moderate detrital contamination
Examples Humberg Tuff (Fig. 5) Odernheim Tuff (Fig. 6) Hesselberg Tuff (locally) Jeckenbach Tuff Hoferhof Tuff
Preservation potential Very high
Textures
Up to 60 cm
Few mm to 7 cm
Footwall blocks
Proximal to distal Hanging-wall and tuffaceous turbidites footwall blocks
Reworked and redeposited ash tuffites
Few mm to 12 cm
Few mm to 2 cm (single flows), cumulative up to 0.4 m
Few mm to 8 cm
Bed thickness
Hanging-wall blocks
Only in hangingwall blocks
Distal tuffaceous turbidites
Primary fallout ash tuffs
Hanging-wall and footwall blocks
Tectonic setting
Primary fallout ash tuffs
Facies
282
Delta plain
Prodelta to delta front
Offshorelacustrine
Depositional environment
Table 2 Descriptive summary of pyroclastic lithofacies including interpreted interrelationships between depositional environment, tectonic setting, and preservation potential
VSIC12 9/12/08 12:59 PM Page 282
S. Königer and H. Stollhofen
VSIC12 9/12/08 12:59 PM Page 283
Preservation potential of fallout ashes eroded (Fig. 15D) in the footwall block but not redeposited in the hanging wall. If siliciclastic deposition is confined to the hanging wall, the following ash layer (V ) may preserve its original tabular geometry. However, as layer V then amalgamated with remains of the underlying ash layer IV, the resulting bed geometry produces a lateral splitting of the marker horizon in cross-section (Fig. 15E).
ACKNOWLEDGEMENTS We wish to thank all those who have contributed through their discussion to this paper, in particular Volker Lorenz, Ian G. Stanistreet, and Jost Haneke. The manuscript benefited from the reviews by Peter Ballance and Christoph Breitkreuz, and the editorial comments by James White. Thanks are extended to Dougal Jerram for reviewing the initial draft, and to Frank Holzförster for helpful assistance in completion of the manuscript. Research was funded by the German Research Foundation (DFG ref. no. Lo 171/22-1, 2), which is gratefully acknowledged. REFERENCES Arthaud, F. & Matte, P. (1977) Late Palaeozoic strike-slip faulting in southern Europe and northern Africa: result of a right-lateral shear zone between the Appalachians and the Urals. Geol. Soc. Am. Bull., 88, 1305–1320. Bangert, B. (1994) Analyse hochfrequenter Sedimentationszyklen in laminierten Tonsteinen des Permokarbon am Beispiel der Forschungsbohrung ‘Münsterappel 1’ (Saar– Nahe-Becken, SW-Deutschland). Diploma thesis, Institut für Geologie, Universität Würzburg. Blair, T.C. & Bilodeau, W.L. (1988) Development of tectonic cyclothems in rift, pull-apart, and foreland basins: sedimentary response to episodic tectonism. Geology, 16, 517–520. Boy, J.A. (1989) Zur Lithostratigraphie des tiefsten Rotliegend (?Ober-Karbon–?Unter-Perm) im Saar–NaheBecken (SW-Deutschland). Mainzer geowiss. Mitt., 18, 9–42. Boy, J.A. & Hartkopf, C. (1983) Paläontologie des saarpfälzischen Rotliegenden. Field Guide Excursion C, 53rd Meeting of the German Palaeontological Society, Mainz, 1983. Boy, J.A., Meckert, D. & Schindler, T. (1990) Probleme der lithostratigraphischen Gliederung im unteren Rotliegend des Saar–Nahe-Beckens (Oberkarbon–Unterperm; SW-Deutschland). Mainzer geowiss. Mitt., 19, 99–118. Busby-Spera, C.J. (1988) Evolution of a middle Jurassic back-arc basin, Cedro Island, Baja California: evidence from a marine volcaniclastic apron. Geol. Soc. Am. Bull., 100, 218–233. Claoué-Long, J.C., Jones, P.J., Foster, C., Roberts, J. & Mackinlay, S. (1995) Calibration of Late Paleozoic time.
283
13th International Congress on Carboniferous–Permian, Krakow, Abstracts, 24. Heim, D. (1960) Über die Petrographie und Genese der Tonsteine aus dem Rotliegend des Saar–Nahe-Beckens. Beitr. Miner. Petrol., 7, 281–317. Heim, D. (1961) Über die Tonsteintypen aus dem Rotliegenden des Saar–Nahe-Gebietes und ihre stratigraphischregionale Verteilung. Notizbl. Hess. Landesamtes Bodenforsch., 89, 377–399. Heim, D. (1970) Die Tonsteine im Unterrotliegend des Saar–Nahe-Gebietes. Z. dt. Geol. Ges., 120, 297–307. Henk, A. (1993) Subsidenz und Tektonik des Saar–NaheBeckens (SW-Deutschland). Geol. Rundsch., 82, 3 –19. Herzberg, W. (1966) Spurenelemente in den Unterrotliegend-Sedimenten der Saar–Nahe-Senke. Geol. Rundsch., 55, 48–59. Hess, J.C. (1985) Die Stellung des Permokarbons im Saar–Nahe-Gebiet aufgrund neuerer isotopischer Daten. Abstracts, Meeting for Carboniferous Stratigraphy, Nohfelden (Saarland),Germany, 1985. Houghton, B.F. & Landis, C.A. (1989) Sedimentation and volcanism in a Permian arc-related basin, southern New Zealand. Bull. Volcanol., 51, 433 – 450. Huff, W.D., Kolata, D.R., Bergström, S.M. & Zhang, Y.-S. (1996) Large-magnitude Middle Ordovician volcanic ash falls in North America and Europe: dimensions, emplacement and post-emplacement characteristics. J. Volcanol. geothermal Res., 73, 285–301. Kerp, H. & Fichter, J. (1985) Die Makrofloren des saarpfälzischen Rotliegenden (?Ober-Karbon–UnterPerm; SW-Deutschland). Mainzer geowiss. Mitt., 14, 159–286. Königer, S. (1999) Distal ash tuffs in the lowermost Permian of the Saar–Nahe Basin (SW Germany): distribution, sedimentology, volcanology, petrography, geochemistry, and zircon ages. PhD thesis, Universität Würzburg. Königer, S., Stollhofen, H. & Lorenz, V. (1996) Tephrostratigraphie in kontinentalen Ablagerungsräumen: Beispiele aus dem permokarbonen Saar–NaheBecken (SW-Deutschland). Kurzfassungen der Vorträge und Poster (Abstracts), Sediment ’96 Meeting, Wien, 1996, 82. Lippolt, H.J. & Hess, J.C. (1989) Isotopic evidence for the stratigraphic position of the Saar–Nahe Rotliegend volcanism. III. Synthesis of results and geological implications. Neues Jahrb. Geol. Paläontol. Monatsh., 9, 553–559. Lippolt, H.J., Hess, J.C. & Burger, K. (1984) Isotopische Alter von pyroklastischen Sanidinen aus Kaolin– Kohlentonsteinen als Korrelationsmarken für das mitteleuropäische Oberkarbon. Fortschr. Geol. Rheinl. Westfalen, 32, 119–150. Lippolt, H.J., Hess, J.C., Raczek, I. & Venzlaff, V. (1989) Isotopic evidence for the stratigraphic position of the Saar–Nahe Rotliegende volcanism. II. Rb –Sr investigations. Neues Jahrb. Geol. Paläontol. Monatsh., 9, 539–552. McBride, E.F. (1963) A classification of common sandstones. J. sediment. Petrol., 33, 664 – 669. Menning, M. (1995) A numerical time scale for the Permian and Triassic periods: an integrated time analysis. In: The Permian of Northern Pangea, Vol. 1. Paleogeography, Paleoclimates, Stratigraphy (Eds Scholle, P.A., Peryt, T.M. & Ulmer-Scholle, D.S.), pp. 77–97. Springer, Berlin.
VSIC12 9/12/08 12:59 PM Page 284
284
S. Königer and H. Stollhofen
Möhring, G. & Schäfer, A. (1990) Caliche im Stefan des Saar–Nahe-Beckens. Mainzer geowiss. Mitt., 19, 63–80. Nakayama, K. & Yoshikawa, S. (1997) Depositional processes of primary to reworked volcaniclastics on an alluvial plain; an example from the Lower Pliocene Ohta tephra bed of the Tokai Group, central Japan. Sediment. Geol., 107, 211–229. Negendank, J.F.W. (1972) Turbidite aus dem Unterrotliegenden des Saar–Nahe-Gebietes. Ein Beitrag zur Sedimentologie limnischer Ablagerungen. Neues Jahrb. Geol. Paläontol. Monatsh., 9, 561–572. Palmer, B.A. & Shawkey, E.P. (1997) Lacustrine sedimentation processes and patterns during effusive and explosive volcanism, Challis volcanic field, Idaho. J. sediment. Res., 67, 154–167. Parrish, J.T. (1982) Upwelling and petroleum source beds, with reference to the Paleozoic. Bull. Am. Assoc. petrol. Geol., 66, 750–774. Plaziat, J.C., Purser, B.H. & Philobbos, E. (1990) Seismic deformation structures (seismites) in the syn-rift sediments of the NW Red Sea (Egypt). Bull. Soc. géol. France, 8, 419–434. Rast, U. & Schäfer, A. (1978) Delta-Schüttungen in Seen des höheren Unterrotliegenden im Saar–Nahe-Becken. Mainzer geowiss. Mitt., 6, 121–159. Schäfer, A. (1986) Die Sedimente des Oberkarbons und Unterrotliegenden im Saar–Nahe-Becken. Mainzer geowiss. Mitt., 15, 239–365. Schäfer, A. (1989) Variscan molasse in the Saar–Nahe Basin (W-Germany), Upper Carboniferous and Lower Permian. Geol. Rundsch., 78(2), 499–524. Schäfer, A. & Sneh, A. (1983) Lower Rotliegend fluviolacustrine sequences in the Saar–Nahe-Basin. Geol. Rundsch., 72(3), 1135–1146. Schwab, K. (1981) Differentiation trends in Lower Permian effusive igneous rocks from the southeastern part of the Saar–Nahe-Basin / FRG. Proceedings of the International Symposium on Central European Permian, Jablonna, 27–29 April 1978, 180 –200.
Stapf, K.R.G. (1990a) Einführung lithostratigraphischer Formationsnamen im Rotliegend des Saar–Nahe-Beckens (SW-Deutschland). Mitt. Pollichia, 77, 111–124. Stapf, K.R.G. (1990b) Fazies und Verbreitung lakustriner Systeme im Rotliegend des Saar–Nahe-Beckens (SWDeutschland). Mainzer geowiss. Mitt., 19, 213–234. Stollhofen, H. (1994) Vulkaniklastika und Siliziklastika des basalen Oberrotliegend im Saar–Nahe-Becken (SWDeutschland): Terminologie und Ablagerungsprozesse. Mainzer geowiss. Mitt., 23, 95 –138. Stollhofen, H. (1998) Facies architecture variations and seismogenic structures in the Carboniferous–Permian Saar–Nahe Basin (SW Germany): evidence for extensionrelated transfer fault activity. Sediment. Geol., 119(1–2), 47–83. Stollhofen, H. & Stanistreet, I.G. (1994) Interaction between bimodal volcanism, fluvial sedimentation and basin development in the Permo-Carboniferous Saar– Nahe Basin (SW-Germany). Basin Res., 6, 245–267. Teichmüller, M., Teichmüller, R. & Lorenz, V. (1983) Inkohlung und Inkohlungsgradienten im Permokarbon der Saar–Nahe-Senke. Z. dt. Geol. Ges., 134, 153–210. White, J.D.L. & Busby-Spera, C.J. (1987) Deep-marine arc apron deposits and syndepositional magmatism in the Alistos Group at Puna Cono, Baja California, Mexico. Sedimentology, 34, 911–927. Willems, H. & Wuttke, M. (1987) Lithogenese lakustriner Dolomite und mikrobiell induzierte ‘Weichteil-Erhaltung’ bei Tetrapoden des Unter-Rotliegenden (Perm, Saar– Nahe-Becken, SW-Deutschland). Neues Jahrb. Geol. Paläontol. Abh., 174, 213–238. Witzke, B.J. (1990) Palaeoclimatic constraints for Palaeozoic palaeolatitudes of Laurentia and Euramerica. In: Palaeozoic Palaeogeography and Biogeography (Eds McKerrow, W.S. & Scotese, C.R.), Mem. geol. Soc. London, No. 12, pp. 57–73. Geol. Soc. London, Bath. Ziegler, P.A. (1990) Geological Atlas of Western and Central Europe. Shell International, The Hague.
VSIC13 9/12/08 12:59 PM Page 285
Spec. Publs int. Ass. Sediment. (2001) 30, 285 –298
Deposition of Mount Mazama tephra in a landslide-dammed lake on the upper Skagit River, Washington, USA J . L . R I E D E L * , P . T . P R I N G L E † and R . L . S C H U S T E R ‡ *North Cascades National Park, 7280 Ranger Station Road, Marblemount, WA 98267, USA; †Washington DNR, Geology and Earth Resources Division, PO Box 47007, Olympia, WA 98504-7007, USA; ‡US Geological Survey, Box 25046, MS 966, Denver, CO 80225, USA
ABSTRACT The cataclysmic eruption of Mount Mazama, Oregon, at c. 6730 14C yr bp, deposited tephra over 1.0 × 106 km2 of north-western North America. Primary tephra fall accumulated to a thickness of 2 cm in the upper Skagit River watershed, Washington. Mazama tephra eroded from this watershed was deposited in Lake Ksnea, of 14 km length and 40 m depth. This lake was created when a landslide blocked the Skagit River at 7040 14C yr bp. Horizontally bedded, dark grey silt and clay were deposited slowly by suspension settling in Lake Ksnea before the eruption of Mount Mazama. The 2-cm-thick primary Mazama tephra layer abruptly caps 7 m of pre-eruption sediments, and is overlain by as much as 17 m of Mazama tephra deposited relatively rapidly on a delta at the mouth of Damnation Creek. Most of a 13-m-thick section is composed of lacustrine tephra containing rhythmic stratified beds deposited by suspension settling. Turbidity currents deposited centimetrescale, cross-bedded silt and tephra at the top of some rhythmite beds. Lower in this section, tephra containing abundant fine-grained terrestrial sediments and other sedimentary structures interrupts the rhythmite beds. These structures include faulted and warped beds, flame structures and pendants created by softsediment deformation. Tephra deposits are overlain conformably with cross-bedded sands throughout most of a 200-m-long section. Coarse alluvial gravels and landslide deposits unconformably overlie the tephra and sand at several locations. The deposits described are interpreted as an inversely graded, prograding delta sequence composed almost entirely of Mount Mazama tephra. Despite a lack of age control on the rate of tephra deposition, the sedimentology of this section indicates that the tephra delta was deposited within 1 yr or less.
INTRODUCTION slope of Washington (Royse, 1967). In the upper Skagit River valley, secondary subaqueous deposition of Mazama tephra in a landslide-dammed lake resulted in an accumulation 17 m thick. These deposits were first described by Carithers (1946), who identified their lacustrine origin, and later by Misch (1977). The tephra was first identified as a product of Mount Mazama by Riedel (1990). Recent examination of these deposits at two exposures provided an opportunity to determine the rate and processes of Mazama tephra deposition in lacustrine, fluvial, and deltaic environments. Secondary objectives of this study were to date the emplacement of the landslide and to reconstruct the physical
The cataclysmic eruption of Mount Mazama, Oregon, at c. 6730 14C yr bp, deposited tephra across 1 × 106 km2 of north-western North America (Fig. 1; Williams, 1942; Mullineaux, 1974; Bacon, 1983; Hallet et al., 1997). In north-western Washington state (Fig. 1), primary subaerial deposition of the tephra resulted in a layer ≈ 2 cm thick. Subaerially deposited Mazama tephra that was subsequently eroded from watersheds was deposited in lakes and ponds throughout the fallout area (Foit & Mehringer, 1993). For example, Davis (1978) reported alluvial fills of Mazama tephra 10 m thick in northwestern Nevada. Fluvial deposits of Mount Mazama tephra have also been identified on the continental
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
285
VSIC13 9/12/08 12:59 PM Page 286
286
J. L. Riedel et al.
Fig. 1. Location of the upper Skagit River watershed and regional distribution of Mazama tephra fall (Mullineaux, 1974). Inset map shows detail of the upper Skagit river areas.
VSIC13 9/12/08 12:59 PM Page 287
287
Deposition of Mount Mazama tephra characteristics the landslide-dammed lake at the time Mount Mazama erupted at 6730 14C yr bp.
(1988). The Damnation Creek landslide dam impounded a lake that is herein called Lake Ksnea, which is a local Skagit Indian place name (Figs 3 & 4).
GEOLOGICAL SETTING METHODS Bedrock in the upper Skagit River watershed is part of a Cenozoic accreted-terrane margin. Between Newhalem and Marblemount (Figs 1 & 2), the Skagit River valley is superposed across the Chelan Mountains terrane (Tabor et al., 1989). The eastern edge of this terrane is the Ross Lake fault zone, whereas on the west it has a metamorphosed contact with the Nason terrane. The study area lies almost entirely within the Chelan Mountains terrane, which is composed primarily of metamorphic rocks, including Cascade River Schist and talc, but includes late Cretaceous tonalitic intrusions (Misch, 1966; Cater & Crowder, 1967). The upper Skagit River watershed currently receives ≈ 2 m of precipitation annually. Most precipitation falls between November and March, and floods during this period are large and frequent. Spring floods are more frequent, of longer duration, and have smaller peak flows than late autumn and early winter rain-on-snow floods. Abundant precipitation, together with the tectonic uplift of this area that began during the mid-Tertiary Period (Tabor et al., 1989), resulted in deep valley incision by Pleistocene glaciers and streams. Local relief near Damnation Creek (Fig. 2) is > 2000 m. Valley walls range in slope from 60 to 80% (30 – 40°). Relatively weak schist bedrock, heavy precipitation, steep valley walls, steep foliation planes, and seismic activity contributed to the occurrence of several large landslides between Bacon and Sky Creeks (Figs 1 & 2). The largest of the landslides originated at an elevation of 700 m on the north valley wall, and is referred to as the Damnation Creek landslide (Figs 2 & 3). The landslide deposits are poorly sorted and are dominated by blocks of angular schist as large as several metres in diameter. A smaller landslide from the south valley wall is superposed on the Damnation Creek landslide deposits (Fig. 2). The toe of this smaller landslide deposit does not reach across the valley, but may have added to the height of the dam created by the older Damnation Creek landslide. A large volume of material from the Damnation Creek landslide travelled across the floor of the Skagit River valley and created a natural dam that blocked the Skagit River. The dam extended for ≈ 1 km across the floor of the Skagit River valley, making it a type II dam in the classification scheme of Costa & Schuster
Dimensions of the Damnation Creek landslide deposits were measured from 1 : 24 000 scale topographic maps and electrical-transmission-line topographic surveys that crossed the landslide deposits. Lake Ksnea’s size and depth were reconstructed from well logs in the town of Newhalem and by field surveys. The surface area of the lake was approximated by tracing the 160 m contour at the top of the landslide dam up-valley into the Skagit River Gorge. Deposits of Lake Ksnea occur at several locations between the landslide dam and the Skagit River Gorge. Two sections from the lower end of the lake near the dam were exposed in a river-cut bank (Fig. 3; site 1) and in a highway cut (Fig. 3; site 2). These sections provided most of the information used to reconstruct the history of tephra deposition in the lake, and are described in detail below. Access to the exposed section at site 1 was by raft, whereas the section at site 2 is at mile 114.6 on Washington State Highway 20. The bottom of the section at site 1 is beneath the surface of the Skagit River, which limited the depth of excavation and sampling. Elevation control was brought to these sites from a benchmark located along Highway 20 between Sky and Damnation Creeks (Fig. 2). Material for radiocarbon dating was obtained from site 1, at the down-valley end of the lake. Samples consisted primarily of small pieces of wood and twigs that were not contaminated by roots or other modern organic material. Therefore, radiocarbon ages obtained have relatively low standard deviations (Table 1). Mount Mazama tephra is generally distinguished by its relatively low silica, and its high iron and titanium content. Results from analyses of two tephra samples were compared with the Fort Rock Valley (Oregon) standard for source identification (Table 2). Textural analysis of the sediments was conducted using the hydrometer method for the silt and clay fraction, whereas sieves were used to analyse the sand fraction of each sample. Statistical parameters of sample textures were determined using a program developed by Spears Engineering and Technical Services (1996). Nomenclature follows the North American Stratigraphic Code. Stratigraphic units were distinguished based on stratigraphic position, texture, structure, and
VSIC13 9/12/08 12:59 PM Page 288
288
+
+
+
+
+
+
+
+ +
+
+
+
+
+
+
1250
+
+
+
+
+
+
+
+
+
+
+
+
+
+
1451 m
+
+
+
+
Damnation Creek
+
+
Sky Creek
+
+
+
+
+
+
+
Todg
+
+
+
+
+
+
+
+
+
+
+
+
J. L. Riedel et al.
Qal
BM 132m +
+
+
+
Qaf
50
Site 1
Qlav
+
7
+
Site 2 Qlav
it
Ri
ve
r
Qlav
+
Damnation Creek Landslide
50
0
Sk
ag
Qls
Landslide Dam
Tkns
00
Qls
10
Qls
+
+ Qls +
+
+
+
Qal
+ kmd
+
Tephra exposed
Kmd
Qlav
State Highway 20
TKns
Qls
+
TKao
Qaf
+
Qls
Todg
Qal
+
+ +
Tkao +
+
Qaf
+
+
+
+ + +
+
0
+
+
150
+
+
+ +
+
N
Fault Bedrock contact
0
1
2 km
Contour intervals 250 metres
Fig. 2. Geological map of area near Damnation Creek landslide. Map units include: Kmd, Marblemount Metaquartz Diorite; Tkns, Cascade River Schist; Tkao, Alma Creek Orthogneiss; Todg, Chilliwack Granodiorite; Qal, Skagit River alluvium; Qaf, alluvial fan; Qls, landslide; Qlav, lacustrine tephra. Sources: Tabor et al. (1989), Fugro Northwest Inc. (1979) and this study.
Fig. 4. (opposite) Longitudinal profile of the upper Skagit River valley and the landslide-dammed Lake Ksnea. Elevation of modern Skagit River channel from 7.5 minute topographic maps. Qla, lacustrine deposits. See Fig. 2 for other stratigraphic units. See Fig. 3 for site and well locations. (Vertical exaggeration: 88 times.)
VSIC13 9/12/08 12:59 PM Page 289
Deposition of Mount Mazama tephra
Fig. 3. Extent of landslide-dammed lake in the Upper Skagit River valley. (Lake level approx. 160 m a.s.l.)
289
VSIC13 9/12/08 12:59 PM Page 290
290
J. L. Riedel et al.
Table 1. Radiocarbon dating analyses of wood samples taken from site 1, unit 1 (radiocarbon dating performed by Beta Analytic, Maimi, FL) Measured 14 C age ( yr bp)
Sample data Beta-96391 Sample 96072606 (1.5 m below contact with ash) Beta-96392 Sample 96072608 Beta-96393 Sample 96072609 (just above surface of river) Cataclysmic eruption of Mount Mazama
13
C/12C ratio*
Calibrated age (yr bp)† 2σ
6350 ± 70‡
– 23.4 0/00
(7380, 7220, 7040)
6790 ± 90‡
– 27.1 0/00 – 25.6 0/00
(7740, 7580, 7440)
7040 ± 60‡ 6845 ± 50§
(7932, 7880, 7680) 7966–7323 7316–7286
*Relative to PDB-1. †Calibrated ages determined by the Stuiver & Reimer (1983) program CALIB 3. No laboratory error multiplier was used in calibration of sample data. ‡No laboratory error multiplier was used for standard deviations reported in column 2. §Cataclysmic eruption of Mount Mazama date from Bacon (1983). No 13C/12C ratio adjustment was performed on this sample. That adjustment would yield an age about 20 yr younger for this sample.
Oxide
Reference Mazama glass composition standard*
Unit 3 primary deposit NOCA 607 2605
Unit 4 secondary deposit NOCA 2-89
SiO2 TiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O Cl
73.26 (0.19)† 0.42 (0.03) 14.34 (0.11) 2.26 (0.10) 0.43 (0.03) 1.59 (0.05) 4.80 (0.10) 2.74 (0.05) 0.16 (0.04)
73.12 (0.15) 0.42 (0.02) 14.46 (0.13) 2.07 (0.07) 0.45 (0.02) 1.60 (0.08) 4.98 (0.16) 2.73 (0.05) 0.17 (0.02)
73.37 (0.27) 0.42 (0.03) 14.34 (0.14) 2.08 (0.10) 0.45 (0.03) 1.59 (0.06) 4.86 (0.14) 2.69 (0.07) 0.18 (0.05)
Table 2. Results of microprobe glass chemistry analyses of Mount Mazama ash from several western North America sites. Values reported are normalized to 100% on a volatilefree basis. Microprobe analysis conducted at Washington State University by Dr N. Foit. (See Fig. 2 for exposure locations and Fig. 5 for stratigraphic locations.)
*Mazama standard from Fort Rock Valley, Oregon. †Standard deviations in parentheses.
lithology. Thus, four lithostratigraphic units are described that represent lacustrine, fluvial, and deltaic environments, and mass-wasting. Correlation between units exposed at sections and those described from a well log is based primarily on elevation and texture.
RESULTS Large-scale maps of the Damnation Creek landslide indicate that the top of the landslide dam was ≈ 160 m in elevation. A dam this tall would have impounded the Skagit River and created a lake of 14 km length
and 1 km width. Trees in growth position eroding from lacustrine deposits indicate that the current elevation of the Skagit River where it crosses the landslide dam is within a few metres of the elevation of the former floodplain. Therefore, the original depth of Lake Ksnea at the landslide dam was ≈ 30–40 m. Lacustrine, fluvial, deltaic, and mass-wasting deposits are exposed at two sites near the mouth of Damnation Creek (Figs 3 & 4). In general, the exposures at sites 1 and 2, 25 and 12 m tall, respectively, reveal only the upper one-half to two-thirds of the estimated total thickness of deposits at the downstream end of Lake Ksnea (Figs 4 & 5). The stratigraphy
VSIC13 9/12/08 12:59 PM Page 291
291
Deposition of Mount Mazama tephra
Fig. 5. Stratigraphy of tephra and lacustrine deposits near Damnation Creek landslide. Location of calibrated radiocarbon dates shown next to organic horizons. See Figs 2 and 4 for site locations and stratigraphic units.
of sediments examined at these sites does not include a contact with pre-lacustrine deposits. On the basis of a well that penetrated presumed Lake Ksnea lacustrine deposits in the town of Newhalem, the bottom of the exposed section at site 1 is probably within a few metres of the contact with pre-lacustrine deposits. In total, seven sedimentary units were identified at sites 1 and 2.
Table 3. Texture of sediments from Lake Ksnea (see Fig. 5 for stratigraphic locations) Unit 1 2 4A 4B 6
% Sand
% Silt
% Clay
12.1 4.3 8.1 0.8 83.7
77.6 82.5 86.2 87.2 16.3
10.3 13.2 5.7 12.0 0
Unit 1 The base of the section at site 1 is composed of 8.5 m of dense, grey mud that is very well sorted and has a texture dominated by silt (Table 3; Fig. 5). Unit 1 is also characterized by laminated beds. Faint horizontal
laminations occur throughout unit 1, which range in thickness from 2 to 5 mm, and consist of fine (light)– coarse (dark) couplets. The boundaries between laminated beds are sharp, and appear conformable.
VSIC13 9/12/08 12:59 PM Page 292
292
J. L. Riedel et al.
Thicker strata are more distinct than the laminations, and also occur throughout unit 1. We identified 15 of these thicker strata, which have an average thickness of 11 cm and a range of thickness from 6 to 28 cm. Thicker strata are conspicuous because of concentrations of organic material that contain abundant wood fragments. Four particularly prominent organic horizons in unit 1 were the sources of wood for radiocarbon dating. Radiocarbon dates were obtained from three of these layers, and the layers ranged in age from 7040 to 6350 yr bp (Table 1; Fig. 5). These thickest strata are spaced somewhat evenly throughout the lower section at site 1, and range in thickness from 0.5 to 2.15 m.
Unit 3 Unit 3 at site 1 is a 2 cm-thick layer of orange-to tan-coloured, ash-sized (< 2 mm) tephra (Figs 5 & 6). The ash consists of very fine sand- to silt-sized particles (primarily glass shards) along with common augite crystals (see Fig. 8, below). The composition of the glass in this tephra is an excellent match (similarity coefficient 0.98, Borchardt et al., 1972) to that in the Mazama tephra standard from Fort Rock Valley, Oregon. Post-depositional weathering of unit 3 has resulted in a concentration of iron-oxide minerals at its base, giving it an abrupt boundary with unit 2. Unit 4
Unit 2 The contact between units 1 and 2 is wavy and abrupt. Unit 2 is 25 cm thick and has a dark brown colour that contrasts with the dark grey colour of unit 1 (Figs 5 & 6). Furthermore, unit 2 contains less sand, and more silt and clay, than unit 1 (Table 3). No stratification or other sedimentary structures are visible in unit 2.
The primary Mazama tephra layer at site 1 has a diffuse, indistinct boundary with unit 4, which is a 17-m-thick deposit of reworked tephra. Sedimentary structures of unit 4 at site 1 are obscured by vegetation and highly disturbed by slope instability. A more detailed description of this unit was obtained at a better-exposed section at site 2.
Fig. 6. Field photograph of sedimentary Units 1, 2, 3 and 4 at site 1. Tape is approximately 70 cm long. Note change in colour of lacustrine units beneath Unit 3, which is the 2 cm thick primary Mazama tephra.
VSIC13 9/12/08 12:59 PM Page 293
293
Deposition of Mount Mazama tephra Three sedimentary units are described at site 2 along a 200-m-long road-cut near State Highway 20 (Figs 3–5). Beds are traceable through all of the exposures, dip gently to the west-south-west, and become thicker in that direction. Unit 4 at site 2 is chemically, texturally and lithologically equivalent to unit 4 at site 1. At site 2, however, Unit 4 exhibits sedimentary structures and bedding not observed at site 1. The chemistry of the glass in unit 4 is nearly identical to that of glass in the primary tephra layer (unit 3; Table 2). Carithers (1946) had previously examined a sample of tephra and noted that it consisted of 95% pumicite and 5% impurities, primarily quartz grains. Two distinct types of deposits were discovered in unit 4 at site 2 (Fig. 5). Unit 4A is composed primarily of tephra, and is at least 7 m thick. Unit 4B is 75 cm thick, and was deposited within unit 4A. A 25-cmthick bed from within unit 4B contains almost no sand and twice the amount of mud as unit 4A. Unit 4B also exhibits flames, pendants, and faults not observed in unit 4A (Table 3; Fig. 7). At the bottom of several sections at site 2, unit 4A exhibits weak horizontal stratification. Rhythmite
beds ranging in thickness from 13 to 19 cm are typically capped by faint, light grey and/or brown laminae. These beds are the most frequently observed at site 2, accounting for 7 m of the 13-m-thick measured section. Locally, lying between the massive and laminated components of the rhythmite beds are concaveupward cross-beds 5 cm tall and 16–21 cm wide. Some of the cross-beds were deposited in small troughs eroded into unit 4A (Fig. 7). Laminations become thicker and more prominent, and the frequency of cross-bedded deposits increases upward in unit 4A, before culminating in the deposition of silt-rich unit 4B. In upper unit 4A, above unit 4B, 10–20-cm-thick rhythmite bedding resumes. Near the 135 m elevation at site 2, beds of tephra alternate with cross-bedded sand deposits (unit 6). Unit 4B is traceable through three separate highway cuts spaced > 200 m apart (Figs 5 & 8). Unit 4B consists of numerous beds of brown silty ash that vary in thickness from a few millimetres to 25 cm. Thin, laminated beds occur throughout unit 4B, and are distinguished by concentrations of dark brown silt at the top of each bed. The silt beds of unit 4B are interbedded 80 cm Warped bedding
64 cm Micro fault
50 cm
Prominent silt bed
45 cm Cross bedding
Cavities 17 cm
0 cm
Fig. 7. Field photograph of Unit 4B at site 2. Unit 4B is distinguished on the basis of its high silt content and the common occurrence of sedimentary structures such as warped bedding, cross-bedding and micro-faulting. Cavities representing the remains of organic material are also commonly observed in this unit.
VSIC13 9/12/08 12:59 PM Page 294
294
J. L. Riedel et al. deposits (Fig. 5; unit 7). The alluvial gravels range in texture from pebble gravels to boulder cobbles, and are very similar in texture to the modern alluvial-fan deposits at the mouth of Damnation Creek. The finer gravels are interbedded with sands of unit 6, whereas the coarser gravels are deposited in channels scoured 1–1.5 m into units 4A and 6.
DISCUSSION
Fig. 8. Scanning electron microscope image of the Mazama Ash.
with cross-bedded deposits of 5 cm thickness and 15– 20 cm width. Warped and faulted beds occur within unit 4B and upper unit 4A (Fig. 7). Faults cutting across the horizontal beds record displacements on a scale of several millimetres, whereas warped bedding associated with the faulted areas records displacements of several centimetres. Flame and pendant structures also occur and are typically adjacent to the cross-bedded and faulted zones. Fossil casts are more common in unit 4B than in unit 4A, and some silt beds in unit 4B are deformed around fossil casts. Unit 5 Unit 5 is composed of coarse, angular clasts of schist at the top of the sections at sites 1 and 2 (Fig. 5). These mass-movement deposits have unconformable contacts with unit 4. At all locations, unit 5 was deposited in troughs, scoured out of Mazama tephra, that are as deep as several metres. Unit 6 Unit 6 caps the tephra deposit unit 4A (Fig. 5). It is composed of cross-bedded medium sand that includes climbing ripples (Jopling & Walker, 1968). Unit 6 is predominantly sand with no clay (Table 3). Unit 7 At the top of the exposed section at site 2, unit 6 is unconformably overlain by coarse-grained alluvial-fan
Several physical features of Lake Ksnea and its watershed influenced the deposition of Mazama tephra. First, because of the limited size and steep bedrock shorelines of the lake, fluvial processes are believed to have dominated over lacustrine depositional processes such as waves and wind-driven currents. Further, intense seasonal flooding occurs in the Skagit River watershed in spring and in late autumn. This pattern may have resulted in fluvial processes being out of phase with winter- and summer-dominated lacustrine depositional processes. Second, maximum lake depth at the time of Mount Mazama’s eruption was probably lower than when the landslide first dammed the river, as a result of 300 yr of outlet erosion. Without direct evidence of the lake level, we estimate maximum depth at the time of the eruption at between 10 and 30 m. Finally, Lake Ksnea drained a fairly large watershed of 3300 km2. Assuming a watershed-wide fallout of 2 cm, ≈ 66 × 106 m3 of tephra was potentially available for deposition in the lake. Mazama tephra was carried into Lake Ksnea by the Skagit River and several other large tributaries (Figs 3 & 4). Exposure of sediments at the head of the lake is poor, and no tephra has been identified to date. At the mouth of the Skagit River Gorge subsurface geology, recorded in the log of the 42-m-deep water supply well for the town of Newhalem, indicates that a 10-m thickness of sand and gravel overlies 15 m of mud (Fig. 4; Rittenhouse Zeeman & Associates, 1977). The clayey deposits overlie 19 m of sand and gravel with clasts up to 25 cm in diameter. The nature of the contact between the gravels and the mud is unknown. An elevation of 146 m at the top of the mud deposits in the well log is within a few metres of the top of lacustrine deposits from the lower end of Lake Ksnea (Fig. 4). Therefore, the mud described in the well log is tentatively assigned a Lake Ksnea origin. No well logs from this area mention the presence of tephra. At the deeper, downstream end of the lake, Damnation Creek is the only large tributary, entering the lake just above the landslide dam (Figs 2 & 3).
VSIC13 9/12/08 12:59 PM Page 295
Deposition of Mount Mazama tephra Radiocarbon dating of wood fragments from the base of the section at site 1 provided a means to determine the age of the Damnation Creek landslide and Lake Ksnea. A radiocarbon date obtained from unit 1 indicates that the landslide has a minimum age of 7040 14 C yr bp (Table 1; Fig. 5). This date from lacustrine deposits means that Lake Ksnea existed for at least several centuries before the cataclysmic eruption of Mount Mazama. Lake level was probably stable as a result of the 1 km width of the landslide dam and outlet armouring (Costa & Schuster, 1988). Large blocks of schist form a lag deposit that armours today’s narrow Skagit River channel at the site of the former lake outlet. The 6350 14C yr bp date from the top of unit 1 is problematic; it is younger than most estimates of the age of the cataclysmic eruption of Mount Mazama, yet lies beneath the primary tephra layer (unit 3; Fig. 5). Previous estimates of the age of the eruption are 6845 14C yr bp (Bacon, 1983), or, more recently, c. 6730 14C yr bp (Hallet et al., 1997). Stratigraphically, the 6350 14C yr bp date conforms with older dates collected beneath it. All of these dates were obtained from wood fragments, and had a low 1σ (Table 1). We offer no explanation for this anomalously young date, but note that it falls within a range of previous radiocarbon measurements of the maximum age of the Mazama tephra (Hallet et al., 1997). Several other features of unit 1 at site 1 also provide important information on the nature of the lacustrine depositional environment at the time that Mazama tephra was introduced. Chemical reduction of sediments observed in unit 1 indicates that the water of Lake Ksnea was thermally stratified during early Holocene time. Seasonal thermal stratification resulted in excellent preservation of organic material in clayey lake-bed deposits (unit 1), and probably influenced processes of lacustrine deposition as discussed below (Ashley, 1975; Sturm & Matter, 1978). Unit 1 also provided evidence on the processes and rates of lake sedimentation at the time of mount Mazama’s cataclysmic eruption. Fine texture, excellent sorting, and the presence of many light (fine)– dark (coarse) laminae couplets in unit 1 indicate that deposition was by suspension. Radiocarbon dating indicates that deposition of unit 1 averaged ≈ 7 mm yr –1. The laminated and stratified beds in unit 1 also indicate that sediment deposition in Lake Ksnea was cyclic. Cyclic beds are common in the stratigraphic records of ancient and modern lakes (Sturm & Matter, 1978; Einsele, 1992). The thinner laminations in unit 1 are interpreted as seasonal rhythmites because of their
295
bimodal texture and cyclic distribution throughout the section (Einsele, 1992; p. 92). The darker component of the laminae observed in unit 1 probably represents early spring and summer layers (Sturm & Matter, 1978). The formation of seasonal rhythmites is controlled by density stratification of lake water and seasonal changes in sediment yield and water discharge (O’Sullivan, 1983). Density currents are particularly important in the development of the summer layer, when the thermal contrast between lake water and cold runoff from melting snowpack is greatest (Ashley, 1975). Density currents, as well as wind and lake currents, can carry coarser summer sediments far from source areas into deeper water (Reading, 1978). The rhythmic bedding and reduced sediments in unit 1 indicate that lacustrine sedimentary processes at the time of tephra deposition were controlled by seasonal cycles of floods and thermal stratification of lake water. Contrasts in colour, texture and bedding characteristics, and an abrupt boundary between units 1 and 2, indicate that a significant change in the lacustrine environment of Lake Ksnea occurred immediately before the cataclysmic eruption of Mount Mazama (Fig. 6). We are not aware of any evidence linking the eruption to this change, but suggest that it may have been caused by landslides into the lake. An increase in lake surface elevation caused by a higher dam and/or an influx of sediment associated with a landslide could explain the change in the lacustrine environment represented by unit 2. Unit 3 is interpreted as the primary fall deposit of Mazama tephra into Lake Ksnea (Figs 5 & 6). Our conclusion is based on the chemistry of the tephra, the unmodified structure of individual glass shards, and the absence of contamination by terrestrial sediments or organic material. Further, the 2-cm thickness is typical for the primary fall deposits of Mazama tephra in this region. Physical differences between the primary watersettled Mazama tephra (unit 3) and the tephra fluvially transported from watershed to lake (unit 4) were examined chemically and physically. Individual glass shards from the fluvially transported tephra retain a jagged, fresh appearance like those of the primary tephra layer (Fig. 8). Glass shards from unit 4 contain slightly less Na2O and K2O than those from unit 3. Although these differences are within 1σ standard deviation, they may reflect hydrolytic leaching of the tephra during fluvial transport and/or post-depositional alteration (Bockheim et al., 1969). At the lower end of Lake Ksnea, Damnation Creek deposited a complex delta composed of fine tephra
VSIC13 9/12/08 12:59 PM Page 296
296
J. L. Riedel et al.
and coarse, locally derived clastic deposits. In sum, the sections exposed at site 2 represent a coarseningupward sequence interpreted as deposits of a prograding delta. Transition from deep-water deposition of unit 4 to shallow-water deposition of units 6 and 7 was probably caused by some combination of delta growth and a lowering of the level of Lake Ksnea by erosion at its outlet. Alternating sequences of silt (unit 4) and sand (unit 6) record seasonal fluctuations in stream discharge, sediment delivery, and lake level. Unit 4 grades conformably into unit 6, and together these deposits at site 2 form a coarsening-upward sequence. Unit 6 is primarily cross-bedded sand left by relatively shallowwater currents on the surface of the delta. Scour and fill deposits of coarse gravel (unit 7) represent fluvial deposition on an alluvial fan, erosively overlying deltaic deposits. Extensive erosion of tephra deposits followed drainage of the lake. Unit 4 at sites 1 and 2 records voluminous deposition of tephra on and near the Damnation Creek delta. Deltaic deposition of tephra at site 2 was primarily by suspension settling in a prodelta setting, which resulted in horizontally stratified beds 13 –19 cm thick in lower unit 4A. All three of the beds observed had similar thickness and structure, and are interpreted as rhythmite beds. The light grey, fine-grained laminae capping these beds are interpreted as the low deltadischarge components, whereas the underlying massive beds represent the high delta-discharge peaks in sedimentation rate. Several factors indicate that the cross-bedded tephra was deposited by turbidity (underflow) currents on the Damnation Creek delta. First, turbidity currents are common in mountain lakes where streams provide cold water and abundant fine-grained sediment (Sturm & Matter, 1978). Second, cross-bedding is a commonly observed feature of turbidity current deposits. Although no complete Bouma sequences were observed, the common light grey and brown silt caps on top of the cross-bedded deposits may represent the fine-grained deposits from the tails of turbidity currents (Sturm & Matter, 1978). Third, the cross-beds observed were scoured into underlying beds on reactivation surfaces, which is a common feature of turbidity-current deposits. Fourth, as the cross-bedded deposits in units 4A and 4B are isolated between horizontally stratified beds interpreted as deeper-water deposits, it is unlikely that they were created by shoreline processes. Finally, cross-bedded deposits in unit 4 occur in association with softsediment deformation structures, indicating that they
were created at a time of high sediment deposition on the delta. Low-density turbidity currents in lakes may be seasonally controlled, and occur at times of high river discharge and thermal stratification of lake water (Rupke, 1978). In the upper Skagit River watershed, high river discharge and strong thermal stratification coincide in late spring to early summer. Increasing frequency of cross-bedded deposits upward in unit 4 may therefore represent a progressive increase in sedimentation rate and development of thermal stratification in the first spring after the eruption of Mount Mazama. This interpretation is consistent with the observation that the nature of sedimentation on the delta temporarily changed during deposition of unit 4B. Unit 4B contrasts with unit 4A because of its finer texture, greater amount of terrestrial sediments, more prominent stratification, and the greater abundance of cross-bedded sediments and soft-sediment deformation structures (Table 3; Fig. 7). We suggest that these physical changes were caused by a peak in seasonal floods that washed tephra and a relatively larger amount of terrestrial sediments into the lake. This interpretation is also consistent with the strong seasonal control of lake sedimentation before the eruption of Mount Mazama indicated by unit 1. Wood deteriorates rapidly in well-drained deposits in this region, and we could not locate subfossil wood or other detrital organic materials in tephra deposits. Large cavities, including a possible tree mould, observed in unit 4, and abundant, isolated, smaller iron-oxide stains may represent the remains of organic material deposited with the tephra. In the absence of datable organic material, it is impossible to determine precisely how long it took the tephra at sites 1 and 2 to accumulate. The conformable unit 4A–4B–4A–6 transitions indicate that deposition was not interrupted for a significant period after tephra began washing into the lake. The purity of the tephra in unit 4A may indicate that it washed into the lake with a snow-melt event, rather than with a large event when the watershed was free of snow. This interpretation is generally consistent with that of Mehringer et al. (1977) and of Hallet et al. (1997), who suggested that the cataclysmic eruption of Mount Mazama occurred in the autumn. Considered together, this evidence indicates that most tephra washed into Lake Ksnea within a single year, and perhaps during a spring runoff of several months in length. This interpretation is consistent with the observations of Waldron (1967) and Collins & Dunne (1986),
VSIC13 9/12/08 12:59 PM Page 297
297
Deposition of Mount Mazama tephra who observed rapid and dramatic increases in tephra sedimentation after large volcanic eruptions. Further, these workers found that erosion and transport of tephra through watersheds decreased rapidly within a few years after eruption (Waldron, 1967; Collins & Dunne, 1986). One exception to rapid sedimentary response to tephra fallout occurred at Mount Tarawera in New Zealand (White et al., 1997). Storage of tephra on the floodplain and in a lake basin contributed to a delayed movement of tephra down the Tarawera river. As Damnation Creek is a high-gradient, third-order stream, and is prone to floods twice a year, it is unlikely that a large amount of tephra was stored in its watershed above Lake Ksnea for a significant length of time after the eruption of Mount Mazama.
CONCLUSIONS 1 A minimum age for the Damnation Creek landslide is 7040 14C yr bp (Table 1). This landslide dammed the Skagit River and created Lake Ksnea on the Skagit River floodplain. Lake Ksnea extended 14 km upstream from the dam, had a maximum original depth of 40 m, and existed for at least several centuries. 2 Physical changes in Unit 2 represent some physical change in the Lake Ksnea environment immediately preceding the cataclysmic eruption of Mount Mazama. 3 Mount Mazama tephra deposits as thick as 17 m dominate the sedimentology of lake-bed deposits at the deeper end of Lake Ksnea. Post-eruption accumulation of terrestrial sediments (unit 1) did not occur, presumably because tephra deposits substantially reduced lake depth, precluding suspension settling. 4 Deposition of Mount Mazama tephra on the Damnation Creek delta was by suspension settling and turbidity currents. Both processes were probably controlled by seasonal variations in sediment yield, runoff and thermal stratification of the lake. 5 Lake Ksnea lacustrine tephra deposits were prone to post-depositional soft-sediment deformation. Minor faulting, flame and pendant structures, and warped beds reveal deformation on the scale of millimetres to centimetres. Turbidity currents along the front of the Damnation Creek delta redistributed tephra deposits. 6 No significant chemical or physical differences exist between fallout (primary) and fluvial–lacustrine (secondary) Mazama tephra clasts deposited in Lake Ksnea. 7 Mazama tephra was deposited within a year or less after falling on the Lake Ksnea watershed. Although age control is lacking in the tephra deposits, several
factors suggest rapid deposition. These include the purity of the tephra, observations of the sedimentary response of streams during 20th-century volcanic eruptions, the high gradient of Damnation Creek, and the sedimentology of site 2. We concur with previous workers who inferred that the cataclysmic eruption of Mount Mazama occurred when snow covered a large part of the Skagit watershed, and that most tephra was deposited after Lake Ksnea became thermally stratified (in late spring). 8 Organic material deposited with tephra in the lacustrine environment was not preserved where those sediments were later exposed to extensive subaerial weathering. However, iron-oxide and organic stains and cavities may indicate that detrital organic material was deposited with the tephra. 9 Minor tephra fallout of a few centimetres from distant volcanic eruptions can have large impacts on lacustrine environments. Lakes that are most vulnerable are those that drain large, steep watersheds where tephra storage is unlikely, and those lakes with hydrological regimes dominated by large seasonal runoff events. When large amounts of tephra are carried rapidly into lakes by peak flow events, anticipated physical effects on the lacustrine environment may include raised lake level, higher turbidity, altered acidity, reduced light penetration and rapid sedimentation.
ACKNOWLEDGEMENTS This study was supported by the US National Park Service, the US Geological Survey, and the Washington State Department of Natural Resources. We thank these agencies and their staffs for support. In particular, Joanie Lawrence of the NPS was an invaluable field assistant, cartographer and editor. S. Benham of Pacific Lutheran University provided scanning electron microscope images of the tephra. We also acknowledge the thoughtful assistance of reviewers N. Foit, H. Mills, D. Gaylord and N. Riggs.
REFERENCES Ashley, G.M. (1975) Rhythmic sedimentation in glacial Lake Hitchcock, Massachusetts – Connecticut. In: Glaciofluvial and Glaciolacustrine Sedimentation (Eds Jopling, A.V. & McDonald, B.C.), Spec. Publ. Soc. econ. Paleont. Miner., Tulsa, 23, 304 –320. Bacon, C.R. (1983) Eruptive history of Mount Mazama and Crater Lake Caldera, Cascade Range, U.S.A. J. Volcanol. geothermal Res., 18, 57–115.
VSIC13 9/12/08 12:59 PM Page 298
298
J. L. Riedel et al.
Bockheim, J.G., Schlicte, A.K., Crecelius, E.A., et al. (1969) Composition variations of Mazama ash as related to variations in the weathering environment. Northwest Sci., 43, 162 –172. Borchardt, G.A., Aruscavage, P.J. & Millard, H.T., Jr (1972) Correlation of the Bishop ash, a Pleistocene marker bed, using neutron activation analysis. J. sediment. Petrol., 42, 301–306. Carithers, W. (1946) Pumice and pumicite occurrences of Washington. Report 15, Wash. State Dept. Conserv. Dev., pp. 60–61. Olympia, Washington. Cater, F.W. & Crowder, D.F. (1967) Geologic map of the Holden Quadrangle, Snohomish and Chelan counties, Washington. US geol. Surv. Quad. Map GQ-646, scale 1 : 62 500. Collins, B.D. & Dunne, T. (1986) Erosion of tephra from the 1980 eruptions of Mount St. Helens. Geol. Soc. Am. Bull., 97, 896 –905. Costa, J.E. & Schuster, R.L. (1988) The formation and failure of natural dams. Geol. Soc. Am. Bull., 100, 1054 –1068. Davis, J.O. (1978) Quaternary tephrochronology of the Lahontan Lake area, Nevada and California. Nevada archeol. Surv. Res. Pap., 7. Einsele, G. (1992) Sedimentary Basins. Springer, New York. Foit, F.F. & Mehringer, P.J. (1993) Age, distribution and stratigraphy of Glacier Peak tephra in eastern Washington and western Montana. Can. J. Earth Sci., 30, 535 –552. Fugro Northwest, Inc. (1979) Report on additional geologic studies for proposed Copper Creek Dam. Unpubl. rep. to Seattle City Light. Hallet, D.J., Hills, L.V. & Clague, J.J. (1997) New accelerator mass spectrometry radiocarbon ages for the Mazama tephra layer from Kootenay National Park, British Columbia, Canada. Can. J. Earth Sci., 34, 1202 –1209. Jopling, A.V. & Walker, R.G. (1968) Morphology and origin of ripple-drift cross-lamination, with examples from the Pleistocene of Massachusetts. J. sediment. Petrol., 38, 971–984. Mehringer, P.J., Jr, Blinman, E. & Peterson, K.L. (1977) Pollen influx and volcanic ash. Science, 198, 257– 261. Misch, P. (1966) Tectonic evolution of the North Cascades of Washington State. In: Symposium on Tectonic History and Mineral Deposits of the Western Cordillera in British Columbia and Neighboring Parts of the United States. Can. Inst. Min. Metal. Spec. Vol., 8, 101–148. Vancouver, British Columbia. Misch, P. (1977) Bedrock geology of the North Cascades. In: Geological Excursions in the Pacific Northwest (Eds Brown, H.E. & Ellis, R.C.), Geological Society of America Field Guide, Annual Meeting, Seattle, WA, 1– 62.
Mullineaux, D.R. (1974) Pumice and other Pyroclastic Deposits in Mount Rainier National Park, Washington. US geol. Surv. Bull., Denver, CO, 1326. O’Sullivan, P.E. (1983) Annually laminated lake sediments and the study of Quaternary environmental changes: a review. Quat. Sci. Rev., 1, 245 –313. Reading, H.G. (1978) Sedimentary Environments and Facies. Elsevier, New York. Riedel, J.L. (1990) Existing conditions of reservoir and streambank erosion. Unpubl. rep. for the Federal Energy Regulation Commission for relicensing of the Skagit River Hydroelectric Project, Seattle City Light and the National Park Service. Rittenhouse, Zeeman & Associates (1977) Waste water, water supply, roads, and building foundations. In unpubl. rep. for the Newhalem Creek Campground, US Department of the Interior, National Park Service. Royse, C.F. (1967) Mazama Ash from the continental slope of Washington. Northwest Sci., 4, 103 –109. Rupke, N.A. (1978) Deep clastic seas. In: Sedimentary Environments and Facies (Ed. Reading, H.G.), pp. 372– 415. Elsevier, New York. Spears Engineering and Technical Services (1996) Computer program for particle size analysis. Stuiver, M. & Reimer, P.J. (1993) Extended 14C database and revised CALIB 3.0 14C age calibration program. Radiocarbon, 35, 215 –230. Sturm, M. & Matter, A. (1978) Turbidites and varves in Lake Brienz, Switzerland: deposition of clastic detritus by density currents. In: Modern and Ancient Lake Sediments (Eds Matter, A. & Tucker, M.E.), Spec. Publs int. Assoc. Sediment., No. 2, pp. 147–168. Blackwell Scientific Publications, Oxford. Tabor, R.W., Haugerud, R.A. & Miller, R.B. (1989) Accreted Terranes of the North Cascades Range, Washington: Overview of the Geology of the North Cascades. 28th Int. Geol. Congr. Field Trip Guidebook, T307. Washington, DC. Waldron, H.H. (1967) Debris flow and erosion control problems caused by the ash eruptions of Irazu Volcano, Costa Rica. US geol. Surv. Bull., Denver, CO, 1241-I, 11–137. White, J.D.L., Houghton, B.F., Hodgson, K.A. & Wilson, C.J.N. (1997) Delayed sedimentary response to the ad 1886 eruption of Tarawera, New Zealand. Geology, 25, 459 – 462. Williams, H. (1942) The Geology of Crater Lake National Park, Oregon, with a Reconaissance of the Cascade Range Southward to Mount Shasta. Carnegie Inst., Washington, Publ. 540.
VSID01 9/12/08 12:57 PM Page 299
Index
Numbers in bold refer to tables; those in italics refer to figures. Acacia Bay Road, Lake Taupo 171–2 strata onlap Taupo ignimbrite 171 swash zone deposits 172 accommodation space, provision of 280–1 accretionary lapilli bed, Taupo–1.8ka eruption 121, 126, 131 achneliths 32 Akademiya Nauk caldera crossed by major fault 36, 37 filled by Karymskoye lake 36 rhyolitic bombs in pyroclastic flow deposits 47–8, 47 Antarctica, subglacially erupted volcanoes 9–10 anticlines, Saar–Nahe Basin 265, 267 subsidence less near to 275 ash brown silty, Lake Ksnea 293–4 grey–white 211 Lake Hestvatn above primary fallout tephra 239 tephra layers in 233 silt-grade 170 tectonic control on preservation of 275–81 unconsolidated, easily moved and modified 263–4 see also lapilli ash; sideromelane ash; vitric ash ash beds Pahvant Butte 62–3, 68, 69, 77 Taupo–1.8ka eruption co-ignimbrite (accretionary lapilli bed) 126 vitric 121, 125 ash grains, flakes adhering to 48 ash lapilli, Karymskoye Lake 42, 55 ash-tuff beds, Saar–Nahe Basin 271, 272 extrabasinal sources 266 preservation of volcanic ash 272–81 hanging-wall position favoured 275 offshore-lacustrine facies 273 prodelta and delta front settings 274 reworking of the ash 274, 275 autobreccia 99, 105 Klondike Mountain Formation 208 poorly-porphyritic rhyolite 97 Sanpoil Volcanics 203–4, 207 Bacon Creek Fault 201, 204, 207, 208 subsidence along 205, 216, 217
ballistic material, Karymskoye Lake 46–50 blocks of hydrothermally altered breccia 47 cauliflower type 46 blocks of ice 47 bombs juvenile basalt 46 remelted old rhyolite 47 ballistic showers, Pahvant Butte 63–4, 65, 78 basalt Bunga Beds 99–101 Icefall Nunatak 17, 20 Lachlan Fold Belt 101 basaltic pods, cryptodomal 99, 100 base surges, Karymskoye Lake 57–8 distal base-surge deposits 40, 44, 45 transforms into fallout deposit 45–6 second prehistoric eruption deposits from 56–7 gas–pyroclastic 58 and surge cloud 38 underflow at base of 54, 58 water-rich 54, 58 and hyperconcentrated water–pyroclastic mixture flows 44–5 basin tectonics, Bunga Beds 89, 95 beaches Lake Taupo, wave energy and processes on 173 new formed by Tsunamis, Karymskoye Lake 50–1 bedform migration 118, 119 Bega Batholith 84, 84 biogenic silica 233, 240, 241 and diatom productivity 229 block-and-ash-flow deposits, upper Lerma basin charcoal within 250 older, dense dacite clasts 250 younger 250–2 blue–grey pyroclastic unit 250 bombs juvenile basalt 47 ovoid 31, 32 remelted rhyolite 47 rhyolitic 47–8, 47 scoria-crust 46, 55, 57 Bonneville, Lake 2 eruption of Pahvant Butte beneath 61–2 lake levels and platform evolution 74 box canyons 113
Volcaniclastic Sedimentation in Lacustrine Settings. Edited by James D. L. White and Nancy R. Riggs T © 2001 Blackwell Science Ltd. ISBN: 978-0-632-05847-1
299
Boyd Volcanic Complex (BVC) geological setting and age 84–6 see also Bunga Beds outlier braided stream deposits 131 breccia breccia bodies 207 Bunga Beds conglomerate–breccia, basal facies 91–2 hyaloclastite 97, 98 in situ fragments 97, 98 marginal to basaltic intrusions 99–101 peperite 100 pumice–sediment 99 rhyolite 97–8 rhyolite–sediment 99 hyaloclastite 14, 24, 32 Icefall Nunatak 17, 18, 29 dominance of juvenile lithic clasts 22 formed by resedimentation? 24 gravelly 20, 21, 24 ovoid bombs 31, 32 scoria breccia 20–1, 31 sideromelane 30 Unit II 26 Pahvant Butte, lapilli ash 63–4 quench-fragmented 99 Bunga Beds outlier 84 basal facies, conglomerate–breccia facies 91–2 basin analogues 101–4 contact with Ordovician basement 91, 91, 95 extralacustrine volcaniclastic epiclastic volcanic debris 105 geology 85, 86 influence of palaeoenvironment on eruption style 104 intrabasinal magmatism basaltic facies 99–101 rhyolitic facies 95–9 intralacustrine sources of volcaniclastic debris 105 limited occurrence of in situ pyroclastic deposits 104 lithofacies and petrofacies, diverse 104–5 provenance affinities and sediment classification in volcanically active lakes 104–5 sedimentary facies of 86–95 basin-centre facies association 93–4, 105
VSID01 9/12/08 12:57 PM Page 300
300 Bunga Beds outlier (cont.) depositional environment and basin character 94–5 northern basin-margin facies association 86–91, 105 southern basin-margin succession 90, 91–3, 105 tectonics, volcanism and basin formation 101 coeval emplacement of basalt and rhyolites 101 tuff–pumice cone complex 99, 104 caldera collapse, lakes formed by 111 calderas 36 Cascade River Schist 287 cauliflower clasts 46, 63, 65 Challis volcanic field basin palaeogeomorphology 180–1 centre of Eocene volcanism 180 channel migration 187, 189, 196 channel mouth bars 90, 166 see also mouth bar deposits Chapala, Lake 249 charcoal 250, 255 Taupo–1.8ka material 114–15 Chelan Mountains Terrane 287 chemical weathering 189 Chigonahuapan, Lake 249 cores from (La Isla I and II) 249 deposition of pumice gravel 258 La Isla II core low magnetic susceptibility 255 organic-rich silts with pumice and ash-fall layers 254–5 stratigraphy 255, 256 lake-level fluctuations 249 magnetic and microfossil analyses 254–7 position of Metepec 258 post-ash flow changes 258 Tres Cruces Tephra (TCT) stratigraphic marker 259 chilling, by rapid convective cooling 31 chronostratigraphic markers permit new constraints 131 regional, Taupo accretionary lapilli bed 126 climate change, central Mexico 247–8 climatic cooling, following eruptions 226, 242 cluster fabrics, pumice 148 coarse clastic wedge, southern basinmargin, Bunga Beds 91–3 coarsening-upward megasequence Klondike Mountain Formation 217 Gilbert-type delta 213–14 coarsening-upward sequence, Lake Ksnea 296 coarsening-upward trend, Klondike Mountain Formation, facies B 213 cock’s tail jets 68 Karymskoye Lake 38, 54 condensed intervals 272
Index conglomerate debris-flow origin, Bunga basal facies 91 East Fork area 185, 187, 189, 196 interbedded sandstone 189 records development of a river 189 Klondike Mountain Formation 214 pebble, Bunga Beds 90, 92 crevasses, formation of in glacier ice 14 cryptodome intrusion, Bunga Beds 94, 95, 97, 99 crystals, Taupo–1.8ka material 114 current activity, Icefall Nunatak 24, 27 cypressoid jets 38 Damnation Creek landslide 287, 288 age of 295, 297 debris falls 72 debris flows 77, 113, 128, 131 cohesionless 72, 216, 218 East Fork area 187, 189, 194 Klondike Mountain Formation, grain-enriched 208 secondary 258 debris-flow deposits 170, 175 cohesive 209 identifiable features 115 on low-energy shorelines 173 Malm Gulch, East Fork area 193 rotated blocks in 68 Taupo–1.8ka eruption 124–5, 125, 160 stacked units 124 decompression-fragmentation of country rock 53 deformation 208 hydroplastic deformation structures 274 regional 95 soft-sediment 94, 95, 125, 230, 271 Lake Ksnea 293, 294, 297 syn-eruptive, Pahvant Butte cone 69–70 slip surfaces demarcate facies assemblages 69, 77–8 syndepositional, Icefall Nunatak 29–30 volcanic mechanism for 209 degassing 53 Icefall Nunatak lavas 23–4 delta plain environment, moderate preservation potential for tuffs 274–5 deltaic deposits, Pahvant Butte platform 71–2, 74 beach and spit 72 delta 71–2 deltaic environments, Bunga Beds 95 deltas fluvial, Saar–Nahe Basin, progradation of 266 Gilbert-type Klondike Mountain Formation 213–16, 218 topset deposits, Pahvant Butte platform 71–2, 78 tributary-inflow, Reporoa basin 128
density currents and development of seasonal rhythmites 295 dry low-energy, deposition from 78 river-fed 90 depositional environment, and basin character, Bunga Beds 94–5 desiccation cracks 272 Devils Elbow granodiorite 201, 204 dewatering structures 125, 271, 275 dewatering zones 26, 26 diamict, pumiceous 115, 116, 117, 124 diamictite assemblage East Fork area 187–9 accumulated during first period of volcanism 193 accumulation in migrating channels 187 sheets deposited by lahars 193 distal base-surge effects and deposits, Karymskoye Lake 40, 44, 45 distal co-surge fallout, Karymskoye Lake 41, 43, 44, 45–6 no accretionary lapilli 46 Donnersberg Formation 266 dropstones, pumice 116, 120, 121 duneforms, surge deposits, Karymskoye Lake 57 duplex structures 67, 68, 68 dykes along Bunga Beach 91 basaltic Bunga Beds 91 effects of intrusion, Karymskoye Lake 53, 54–5 rhyolitic, Bunga Beds 97 sedimentary, Bunga Beds 98 earthquakes, pre-eruption swarm, Karymskoye Lake 38, 53 East Africa Rift Zone lakes lacustrine analogues for Bunga Beds 103 Lake Malawi 103 Lake Tanganyika 104 East Fork area, Challis Volcanic field 180–1 effects of Van Horn Peak cauldron explosions 180, 194, 195 effects of volcanism 181 Ellis Creek and Eightmile eruptions, effects of 194 landscape evolution 193–6 downstream initiation of degradation event 194 erosive event following lake fill 194 final history of lake a problem 195 first fluvial period 193–4 first lacustrine period 194 increase in basin subsidence rate 195 rapid drowning of palaeoriver valley 194 second fluvial period 195 third fluvial period 196 valley flooding and second lacustrine period 195
VSID01 9/12/08 12:57 PM Page 301
Index major pyroclastic-flow deposits 181, 191 sedimentology 181–90 fluvial deposits 187–90 lacustrine deposits 182, 185–7 lithofacies assemblages, transition intervals 181, 185, 196 lithofacies, volcaniclastic deposits 181, 183–4 mudrock intervals, correlation of 181–2 sensitivity to changes in extrinsic variables 196–7 stratigraphy 190–3 fluvial to lacustrine surfaces 186, 188, 190, 190, 192 fluvial to lacustrine surfaces 186, 188, 190–1, 192 transitions within lacustrine deposits 193 Eastern Volcanic Belt of Kamchatka 36 en masse freezing and deposition 209, 216 englacial see page 13 environmental changes central Mexico, and volcanism 247, 258–9, 259 Iceland available records 240 due to continued volcanigenic input into lake sediments 242 due to H4 tephra deposition 241–2 and observed changes in Zone II 242–3 eruption/eruptive columns Icefall Nunatak 29 Karymskoye Lake 54, 57 Pahvant Butte, effective sorting in 69 eruptions basaltic, influence of shallow water on style of 57 caldera, sedimentary responses to 133–5 explosive East Fork area 194, 195 features of fallout deposits 263 from rhyolitic calderas 141 Iceland 226 phreatomagmatic 99, 152 fissure, Iceland 239 Hawaiian 32 hydrothermal, Waiotapu system, Reporoa basin 130 ignimbrite significant topographic modification by 135 Taupo Volcanic Zone 111, 113 Mount Murphy, subglacial and subaerial 10–11 Plinian 152 Nevado de Toluca 249, 258, 259 shallow-water 36 ‘sheet-flow eruptions’ 32 Strombolian 31 sub-glacial 2, 11 effects of glacier physics on 13–17 subaerial phreatomagmatic 35
subaqueous 35–6, 104 see also Pahvant Butte, Lake Bonneville sublacustrine, Karymskoye Lake 38–40, 39, 53–5 ascent of magma 53 end of eruption 54–5 Surtseyan activity 53–4 vent-clearing explosion(s) 53 Surtseyan 2, 10, 29, 36, 53–4 volcano types associated with thicker ice 32 Vulcanian, Nevado de Toluca 249 Eureka Fault 204, 207, 208 subsidence along 205, 216, 217 explosions Karymskoye Lake ejection and deposition due to 54 expansion of gas–pyroclastic mixture 54 from vesiculation of magma and water–magma interaction 53–4 vent-clearing 53 phreatic 53 secondary 125, 131 extension Republic Basin 217 due to uplift of metamorphic core complexes 201 Variscan Mountain Belt, and development of Saar–Nahe Basin 264 extensional basin systems 101, 201 fall deposits Pahvant Butte 69 Taupo volcano 152, 154 fan deposits from Karymskoye Lake lahars 52–3 Lake Taupo break-out 129–30 multiple terraces cut 129–30 faults/faulting 194 Icefall Nunatak 25–6, 29 syndepositional 24 synsedimentary 30 Pahvant Butte duplex slip 67, 68, 68 normal 68 repeated movements may cause deformation 280 Republic Basin extension along detachment faults 201 normal 210 Saar–Nahe Basin effects of movements 275, 280 rejuvenation of 275, 279 transfer faults, Saar–Nahe Basin 265 syndepositionally active, influence of 275, 277, 278 fining upwards East Fork area conglomerate bodies 189 sheet-sandstone assemblage 189–90 wedge-sandstone assemblage 187
301 fining-upwards sequences East Fork area, due to avulsion and channel migration 196 Lake Taupo 172 Five Mile Beach 161 Five Mile Beach, Lake Taupo 160–4, 173 grey pumiceous blocks 161, 162, 163–4 highstand terrace 161 outer-shoreface deposit 164 persistent deepening 174 pumice clasts at storm-surf site 164 pumice gravel capping 161 shoreface reconstruction 161, 162 storm deposits 164 transgressive deposits 161 flood deposits, Lake Taupo break-out 129–30 flow banding 95, 96, 97, 99 flow-layering Bunga Beds rhyolite 95, 98 in clasts 97 flows high-sediment-concentration 154 hyperconcentrated 44–5, 124–5, 125, 131, 189 fluidization, of wet sediments by rhyolite 98–9 fluoride, volcanigenic, toxic 226 fluvial deposits, East Fork area 187–90 conglomerate assemblage 185, 187, 189 effects of climatic variation on 196 features for interpretation of 196 diamictite assemblage 185, 187–9 sheet-sandstone and mudrocksandstone assemblages 185, 189–90 deposition by sandy bedload rivers 190 fluvio-lacustrine sub-environments, preservation of fallout ash in 280, 281 folding, intraformational 209 fossils Klondike Mountain Formation 210, 211 plant fossil assemblages 206, 211 and plant fragments, Bunga Beds 86, 89, 95 Pleistocene megafauna, upper Lerma basin 251, 259 gas-escape pipes 250, 251 Gaugrehweiler Tuff 268 glacier hydrology 13–14 glacier thermal regimes 13 glaciers aquifers and aquicludes 13 layered structure 14, 14, 15 structure of 14 thick, eruptions beneath 14, 16 thin, vault/lake cannot form beneath 14–15, 16
VSID01 9/12/08 12:57 PM Page 302
302 grading inverse 207 coarse-tail 116, 117, 118, 128, 129 normal 116, 117, 118, 146, 148 see also reverse grading grain size, Karymskoye Lake deposits first eruption 55 tuff ring 41, 42–3, 44 grainflows, Pahvant Butte deposition from 66–7, 68–9 fall-avalanche 76, 78 gravel 170 alluvial, Lake Ksnea 294 cross-bedded, Taupo–1.8ka eruption 116, 117, 118 lithic-clast 165 pumice gravel, Taupo–1.8ka eruption 116, 117, 118, 158, 164, 172 accumulated from saturated pumice rafts 158 open-work, strandline deposits 126 gravel bedload transport 196 gravitational settling 54, 58 gravity sorting 271 ground water, and pumice saturation 142 heavy metals 268 Hedin Nunatak, multiple superimposed hyaloclastite deltas 12 Hekla volcano 239 Herron Creek quartz monzonite 201, 204 Hesselberg Tuff 268, 275 Hestvatn, Lake, Iceland changes due to Norse settlement (Vötephra layer) 243 core 94-HV01 dominant magnetic minerals 231, 244 higher tephra frequencies in Zone II puzzling 234, 239 identification of tephra layers and chronology 228, 233–4, 238 major and trace element concentrations 229–30, 232, 233 other sedimentological, chemical and magnetic analyses 229–30, 230–3 physical/chemical characteristics across the H4 and KN tephra layers 234–8 sediment accumulation rates 238–9 tephra mainly from four volcanoes 239 variabilities caused by main processes 244 depositional processes and impact of tephra 239–40 deposition of tephra into the lake system 239–40 frequency and origin of tephra 239 downcore changes 240–3 magnetic zone I 232, 240–1, 244 magnetic zone II 232, 241–3, 244 magnetic zone III 232, 243, 244 magnetic zone IV 232, 243, 244
Index geological and geographical setting 227–8 major change in many measured variables, Zones I and II 241 Nesje core chronology 229 tephra deposition impact on biological and physical processes 226 interrupts normal chemical lacustrine processes 4 tephra layers in sediment core 227 downcore alteration in iron oxides 230 magnetite concentration 230 main processes in downcore changes 227 susceptibility peaks, useful for layer correlation 231 variation in variables in Zone I 240–1 zones I and II boundary, changes at and H4 tephra layer 241 zones II and III boundary, abrupt change in chemical and magnetic variables 243 Hingapo, Lake Taupo 170–1, 171 beach and shoreface sands 170 sediment-starved setting 171 suspension deposits 170–1 Hoferhof Tuff 275 Huka Group 111, 114, 133 spillways through 114, 129 Humberg Black Shale 267 Humberg Tuff 268, 269, 273, 275 Hunsrück Boundary Fault 264 hyaloclastite deltas 12, 14, 15 hyaloclastites 14, 24, 32 Bunga Beds 97, 98, 105 hydraulic theory, and tuya volcanoes 14–15, 16 hydrothermal activity 205 eruptive 208, 209, 210, 217 hydrothermal alteration 95 Eureka fault zone 217 Klondike Mountain Formation 208, 209 Republic Basin 218 hydrovolcanism, Icefall Nunatak 27, 29 Icefall Nunatak, Antarctica cinder cone remnant, Stage III 31 constructive stages separated by unconformities 32 evidence for glacial setting for eruptions 12–13 lava (Stage III), possible interaction with thin glacier 31 lava-orthobreccia cogenetic relationship 22, 23 major lithofacies distribution and architecture 12 not emplaced in a dry state 30 possible development during Stage I 23 possible development during Stage II 28
some alteration (Stage II) 30 summary and interpretation 17, 17–18 polygenetic volcanic centre 10, 17, 32 subglacial eruption 11 volcanic evolution of 17, 20–32 major stages of development 12, 17, 20, 20 production of muddy fines 29 single continuous growth cycle 20 Stage 1I (pillow volcano stage) 20–4 Stage 2II 24–30 Stage 3III (lava effusion) 30–2 Iceland use of tephrachronology 226 volcanic zones and volcanic systems 226 ignimbrites absent from Bunga Beds 104 Taupo eruption 113, 130, 152 effects of 1–2, 3, 111 impacts of emplacement 113–14 Taupo ignimbrite 165 distribution of 154 Five Mile Beach 161 layers of 154 reflective surface of 113, 154 impact sag structures 57, 63, 65, 67, 68 inflation, post-emplacement 30 intraclasts, mudstone 89 intrusions basaltic, Bunga Beds 101 brecciated margins 100 Saar–Nahe Basin 266 Jeckenbach Tuff 269–70 joints/jointing columnar 31, 95 polygonal 95 jökulhaups (catastrophic floods) 14, 15 juvenile clasts, Karymskoye Lake, vesicularity/morphology of 48, 49, 50 Kaiapo, Lake Taupo 165–6, 172 highstand shoreline notch 165 lower shoreface to offshore deposits 165 storm deposits, lithic-rich 164 Kaingora ignimbrite 114 Kappeln Tuff 268, 274 Karymskoye caldera, activity at 2 Karymskoye Lake, Russia, eruptions in 35–60 1996 eruption composition of ejected material 47–8 observation of 38–40 stratigraphy and characteristics 41–50 succession of events and interpretation of eruption processes 53–5 Surtseyan eruption 38 wet eruption 44
VSID01 9/12/08 12:57 PM Page 303
Index deposits and styles of prehistoric eruptions 44–7 deposits of first eruption 55–6 surge and fall deposits of second eruption 56–7 geographical/geological setting 36–8 tsunamis and lahars, effects and deposits of 50–3 tuff ring morphology 40–1 Karymsky stratovolcano andesite and dacite eruption 36 eruption from 38 Katla volcano 239 Kettle metamorphic core complex 201 Kinloch, Lake Taupo, lagoonal setting 168, 170 Klondike Mountain Fault 217 downward motion on 216 Klondike Mountain Formation 201, 205–6 accumulation in restricted volcanotectonic depressions suggested 206 anoxic bottom waters 211 breccia bodies 207 origin of 209 facies associations 206, 206 Gilbert-type delta 213–16, 218 hydrothermally influenced lacustrine basin margin 207–10 lacustrine basin plain deposits 210–13 gravel facies 207, 208 facies association C 214–15 massive matrix-supported beds 209 sand-enriched, clast-rich beds 209 heterolithic facies cosets commonly deformed 208 from facies association C 215 mud-dominated 207–8, 209 sand-dominated 208 late-stage volcanogenic sedimentation, tectono- sedimentary setting 216 restricted to small basins in Republic graben 205 seasonal stratification 211 sinter deposits 210 stratigraphy 202 two-tiered division 205 Tom Thumb Tuff Member 205 Ksnea, Lake cyclical sedimentation 295 Damnation Creek deposition of complex delta 295–6 deposits exposed near 290–1, 291 Damnation Creek delta change in sedimentation during Unit–4b deposition 296 cross-bedded tephra deposited by turbidity currents 296 deposition by suspension settlement 297 deposition of prograding delta 296 impounded by Damnation Creek landslide dam 287, 289
Mazama tephra physical differences, primary water settlement and fluvial transportation 294, 295 primary fall deposit 295 organic material 292, 295, 296, 297 original depth at landslide dam 290 predates cataclysmic eruption of Mount Mazama 295 seasonal floods and thermal stratification 295 tephra deposition 287 sites I and II, time taken 296–7 texture of sediments from 291 Unit–1 291–2 information on sedimentation at time of cataclysmic eruption 295 laminated beds 291 seasonal rhythmites 295 lacustrine environment information at Mazama tephra time 295 younger age bracket 291, 295 Unit–2 292, 297 Unit–3 ash-sized orange-tan tephra 292 glass composition matches Mazama Tephra standard 292 Unit–4 292–4 diffuse boundary with primary Mazama tephra layer 292 glass chemistry 293 sedimentary units in 293 Unit–4a cross-bedding 293 mainly tephra 293 rhythmite bedding 293 Unit–4b brown silty ash interbedded with cross-bedded deposits 293–4 found within Unit–4a 293, 293 Unit–5 295 Unit–6, sand capping unit–4a 294 Unit–7, Site 2, alluvial fan deposits capping Unit–6 294 Kuratau, Lake Taupo 167–70 La Isla I core, upper Lerma basin ash-flow deposit 252 Upper Toluca Pumice (UTP), lies over silty organic-rich lacustrine sediments 253 La Isla II core, upper Lerma basin 254–7 ash-flow deposit 252, 255, 257 diatom assemblages 251 Tres Cruces Tephra 257 Unit III low magnetic susceptibility and high LOI values 255, 257 lower interval, diatom record 257 upper interval, shallow freshwater lake indicated 257 Unit IVa, freshwater pond assemblage followed by shallow, freshwater lake 257 Unit IVb, lacustrine sedimentation 257
303 Upper Toluca Pumice (UTP) 255, 257 lies over silty organic-rich lacustrine sediments 253 Lachlan Fold Belt 84, 84 volcanism 101 lacustrine deposits East Fork area 182, 185–7 created by lava dam 194 mudrock assemblage 182, 185, 186, 195 wedge-sandstone assemblage 185–6, 186 Klondike Mountain Formation basin plain 210–13 hydrothermally influenced basin margin 207–10 Lake Taupo, differences from nonvolcanic deposits 175–6 Republic Basin, preservation of 217–18 Sanpoil volcanics 204–5, 216 lacustrine environments 295 impacts on of minor tephra fallout 297 lacustrine eruptions 2–3 lagoonal deposits, Lake Taupo 160 Lake Taupo break-out flood facies 122, 129–30 lakes diatom production and tephra fall 227 englacial and active volcanoes 9–10, 32 Icefall Nunatak, development of Stage II lithofacies 28 ephemeral 135 and stream erosion 113–14 eruption-impounded 3 resedimentation in 3 as tephrostratigraphic repositories 4 volcanic 1–2 in volcanic environments 2 see also Ksnea, Lake; Taupo, Lake landslides 287 lapilli accretionary 67, 68, 121, 126, 131, 208, 209 armoured 57, 63, 67 openwork 68, 69 lapilli ash basaltic, Karymskoye Lake 40, 41, 42, 43, 45 Pahvant Butte 63–4 Las Cruces volcanic range 249 lava deltas 29, 30 lava flows andesitic and dacitic, Sanpoil Volcanics 202, 207 basaltic andesitic, Saar–Nahe Basin 266 damming streams 1 lavas Icefall Nunatak 17, 19, 22 Stage I 22–4 Stage II 26–7 Stage III 30–1 Lebach Group, Jeckenbach and Odernheim Units 267 tephrostratigraphic markers 281
VSID01 9/12/08 12:57 PM Page 304
304 limestone, Saar–Nahe Basin biogenic 271 micritic, wavy lamination 269 liquefaction features 72 lithic fragments, Taupo–1.8ka material 114 lithophysae 95, 96 load structures 271 Bunga Beds 89, 95, 101 Klondike Mountain Formation 208, 209, 211, 212 longshore drift, Lake Taupo 159 Lower Toluca Pumice (LTP) 249 rich in ochre pumice 252 surface disrupted by debris flow emplacement 258 magma Bunga Beds, volatile-poor 104 Karymskoye Lake ascent 53 fragmentation 53–4 magma clots 54–5 magmatism, intrabasinal 266 Bunga Beds basaltic facies 99–101 rhyolitic facies 95–9 Mallacoota Beds 84 mantle plumes 10 marker beds see chronostratigraphic markers; tephrostratigraphic marker beds mass-flow events subaqueous, Bunga Beds 89–90 Taupo–1.8ka eruption 131, 133 multiple 125 see also debris flows; sediment gravity flows Mazama tephra Damnation Creek landslide 287, 288 deposition Lake Ksnea, influence of physical features 294 in landslide–dammed lake 285–98 secondary subaqueous 285 time of 296, 297 geological setting 287 lithostratigraphic units 290 low silica, high iron and titanium 287 radiocarbon dating 287, 290 determining age of Lake Ksnea and Damnation Creek landslide 295 meltwater flow towards eruption site 13–14 and subglacial eruptions 13 Merimbula Group 86 metamorphism, low-grade 95 Metepec, upper Lerma basin 249 ash layer from Nevado de Toluca flow 258 block-and-ash deposits, older and younger 250, 251, 252 good palaeoenvironmental information 259 LOI values very low 253 Lower Toluca Pumice (LTP) 252 magnetic/microfossil analyses 253–4
Index Isöetes mexicana pollen 253–4 lahar and fluvial deposits 254 magnetic susceptibility 254 microfaulting 211, 215 microturbidites 269, 273, 274 Mount Mazama, eruption of 285 Mount Murphy, Marie Byrd Land 11 described 10 interactions between magma and former ice 10–11 shield succession erupted in association with ‘thin’ ice 12 mouth bar deposits 272 mudrock, Challis area 181–2 mudrock-sandstone assemblage, East Fork area 190, 195 structureless 195 mudstone pyritic 208 generation of 210 Saar–Nahe Basin grey lacustrine 266, 272 hummocky cross-stratified units 271 planar laminated 272 mudstone-siltstone couplets, Klondike Mountain Formation 211 mudstone/siltstone, Bunga Beds 88–9 black, representing anoxic sedimentation 91, 93–4 deltaic setting 89 deposition from clay and silt-rich water 89 Nevado de Toluca volcano 4 carbonized wood dated in two outcrops 252 catastrophic eruptions, Pleistocene 259 dacite dome development and destruction 258 largest eruption in Late Pleistocene history 257–9 Lower Toluca Pumice (LTP) 258 Upper Toluca Pumice (UTP) 252, 258 Vulcanian- and Plinian-type eruptions 249 O’Brien Creek Formation 201, 202 Odernheim Tuff 269–70 offshore-lacustrine facies association, Saar–Nahe Basin 268–72, 270 delta plain facies association 271–2 laminated black shales 268–9 micritic limestones 269 sandstones and tuffites, fine-grained 269 tuff layers, transgressive, highest preservation potential 273 tuffs, normally or reverse-graded 269, 271 Okanogan Highlands 200 extensional basins in 201 major lithological elements 201 volcanogenic sediments accumulation 199, 201 climate and lake sediment accumulation 217
Okanogan metamorphic core complex 201 Omineca crystalline belt 210 Orakei Korako blockage 112, 128, 131 organic matter Klondike Mountain Formation 210 La Isla cores 253 Lake Ksnea 292, 295, 296, 297 Sanpoil volcanics 205 orthobreccia 20, 22, 26 Pahvant Butte, Lake Bonneville the cone 65–70, 76–7 anticlinal features of western rim 66, 66 depositional features 66–7 form 65–6 lithofacies UCd 68–9 lithofacies UCw 67–8 palagonite 70 syn-eruptive deformation 69–70 eruption, emergence and construction of the volcano 76–8 history of illustrated 75 morphology and structural elements 62 the mound 62–5, 72, 73, 76 broadly cross-stratified ash 64 coarse ash and lapilli, well-bedded 62–3 defining characteristics 62 lapilli ash, massive to weakly bedded 63–4 thick-bedded ash with rotated tuff blocks 64–5 the platform 70–4, 77 beach ridge 70–1 built by littoral drift 76–7 characteristics of platform foresets 78 contact relationships 72–4 lake levels and platform evolution 74 main platform topset 71–2, 78 origin as a syn-eruptive feature 74 platform foreset 72 post-eruptive evolution 76 progradational beach deposits, local delta 70–1 shifting vent sites 76 stacked foreset units, record smallscale lake-level changes 74 a syn-eruptive shoreline suggested 73–4 palaeoenvironment, influence of on eruption style 104 palaeoshoreline, Lake Reporoa highstand 122, 126, 131, 155, 165 transgressive deposits 122, 126, 127, 128, 155, 161 reworked material trapped in swash zone 126 palaeosols beneath Lower Toluca Pumice (LTP) 252 beneath Upper Toluca Pumice (UTP) 252
VSID01 9/12/08 12:57 PM Page 305
Index palagonite alteration/ palagonitization 31 characteristic of hydrovolcanism 31–2 Pahvant Butte 67, 70, 72 Pappelberg Tuff 268, 275 paraconformity 30 pedogenic carbonate nodules 272 peperite textures 96, 97, 98, 99 peperites 105 Klondike Mountain Formation 208 Sanpoil Volcanics 204, 205, 207 peperitic mixing 98 perturbation, of sedimentary systems and sediment yields 135 Pfalz Anticline 267 phenocrysts, Icefall Nunatak 17, 20 pillow lavas 20, 21, 29 platform foresets, Pahvant Butte, sedimentary features and sedimentation 72 pollen Lake Hestvatn tephra layers 241 declined due to climatic cooling 242 increase in Zone III 243 Metepec, upper Lerma basin, Isöetes mexicana and Pinus 253–4 Zacango Quarry, Upper Lerma basin 253 suggests water body with grassland 258, 259 pore-fluid pressure, excess 30 Poukura Pa, Lake Taupo 167, 170, 172–3 deep water facies association of a low energy environment 170 shoreface sequence 170 swash deposits, deepen to shoreface deposits 170 wave-base storm deposit 170 preservation potential, of volcanic ash 272–81 provenance influences, in volcanically active basins 105 pumice aqueous deposition of 141 coarser fragments in shoreface deposits 159 depositional features 147–8, 147 cluster fabrics 148 lenticular bodies fining-upwards 148 recessional beach berms 149, 161, 163 sequential settling of clasts 147 thick beds, formation of 148 a low-density particle 148 Lower Toluca Pumice 252–3 rich in ochre pumice 252 pumice gravel, Taupo–1.8ka eruption 116, 117, 118, 126, 128, 158, 164, 172 accumulated from saturated pumice rafts 158 as a strandline deposit 161 rounded granules and pebbles 160
settling and deposition of in lacustrine and associated environments 141–50 summary of behaviour 149 Taupo–1.8ka material 114, 125 Upper Toluca Pumice (UTP) 252–3 pumice beds/deposits coarse, Lake Taupo 174 openwork 165 pumice blocks, grey, Lake Taupo 152, 154, 161, 163 pumice plain, Lake Taupo, effect on local weather 113 pumice rafts 147 ‘beaching’ affected by wind direction 155 containing floating bushes 51–2 stranded 165 suspension sedimentation from 164 waterlogging and sinking 118, 129 of individual clasts 121, 141 pumice saturation behaviour 142–4 ‘time to sink’ trend 143–4, 143 two-stage saturation 144 pyrite, authigenic 211 pyroclastic deposits 1996, Karymskoye Lake, tuff ring deposits 41–5 Saar–Nahe Basin 269, 271 pyroclastic deposits/material 1996, Karymskoye Lake 41–50 ballistic material 46–50 distal base-surge effects 45 distal co-surge fallout 45–6 preservation as ‘primary’ and ‘reworked’ in lakes 4 pyroclastic flows secondary 131 generation of 124 pyroclastic lithofacies, Saar–Nahe Basin, interrelationships 281, 281, 282, 283 pyroclastic sediments, reworked, Taupo–1.8ka eruption 122, 123–6 accretionary lapilli bed 126 debris- and hyperconcentrated-flow deposits 124 –5, 125 deposits of secondary phreatic eruptions 125–6 deposits of secondary pyroclastic flows 124 pyroclastic-fall deposits, East Fork area 185 pyroclasts Icefall Nunatak 17 sedimentation and resedimentation of 3 pyrophyllite–quartz–sericite alteration assemblage 95 quench fracturing, in situ 98 Raumberg Tuff 268, 274, 275 splits into two horizons 275
305 remobilization processes early 122, 123–6, 131 rapid 133 Reporoa, Lake, Taupo Volcanic Zone 109–40 basin physiography 114 chronology and depositional model 131–3 breach-outlet channel, quick growth of 133 depositional model 132, 132 early remobilization processes 131 estimates of peak discharge 129, 133 highstand period appears brief 132–3 refilling of intracaldera Lake Taupo 133 reworking of Taupo ignimbrite 131 lithofacies associations and depositional environments 121–30 basinal lacustrine deposits 122, 128–9 early remobilization deposits 122, 123–6 highstand palaeoshoreline 112, 122, 126 Lake Taupo break-out flood facies 112, 122, 129–30 post-Lake Reporoa to present 122, 130–1 transgressive shoreline deposits 122, 126–8, 155, 161 tributary inflow deltas and basinfloor fans 122, 128 lithofacies and petrofacies 114–21 distinctive accretionary lapilli bed 121, 126 lithofacies descriptions/ interpretations 115–21 sedimentary responses to caldera eruptions 133–5 Te Toke sub-basin 114, 128 fan opening into 130, 133 two separate falls in base level recorded 135 volcanism/sedimentation/basin subsidence coincidence 205 Republic Basin, USA 199–222 basin evolution 216–18 effects of reduction in basin subsidence rates 217 basin margins and Gilbert-type delta development 215 breccia bodies interstratified with carbonaceous mudstone 209 ungraded, hydrothermal eruption deposits 209–10 controls on pattern of sedimentation 202 development of 201 extensional basin 201 hydrothermal eruption crater 209 infilling similar to that of half-graben models 202, 218
VSID01 9/12/08 12:57 PM Page 306
306 Republic Basin, USA (cont.) lacustrine sedimentation, influences on character of 218 sedimentary–tectonic evolution of 201–2 sedimentology and stratigraphy 202–16 sedimentary and volcanic facies 203–6 sedimentary, volcanic and hydrothermal facies and facies associations 206–16 stratigraphic and sedimentary relations 216 tectonic tilting 218 volcanogenic sedimentation 199, 207–13, 216, 217 reverse grading 94, 272 Pahvant Butte 64, 66, 68 Taupo–1.8ka eruption 141, 147, 148 rhyolite 36, 47 Bunga Beds 88, 95–9 chemistry has A-type affinities 105 hydrous magmas 104 lavas, cryptodomes and feeder dykes 95–7 poorly porphyritic 95, 97, 98–9 quench fragmentation of 98 rhyolite breccias 97–8 rhyolite–sediment contact relations 98–9 stratified juvenile volcaniclastic associations 99, 100 rhythmite beds, Ksnea Lake 293, 295 rhythmites, seasonal formation of 295 rill erosion 67, 68, 71–2 ripples, on Karymskoye Lake tuff ring 45 Road Creek, East Fork area, diamictite assemblage 193 Rubielos de Mora Basin, analogue for the Bunga Beds 103 Saar–Nahe Basin asymmetry of depositional cycles 280 basin fill 264 basin structure 264–5 anticlines, synsedimentary formation of 265 extensional plate boundary forces 264 half-graben structure 264 transfer faults, effects of 265 geological setting and structural framework 264 preservation of volcanic ash 272–81 tectonic control on 275–81 syn-rift evolution of 265–7 main tectono-sedimentary phases 265 pre-volcanic syn-rift phase 265–6 volcanic syn-rift phase 266–7 transgressive–regressive cycles 272 upper Stephanian depositional environments and volcanism 267–72
Index lithofacies 268–72 petrography 268 tephrostratigraphy 267–8 Westphalian–Stephanian erosional unconformity 265–6 St Alban Tuff 268, 274, 274, 275 San Antonio volcano 249 sand crystal lithic, Taupo–1.8ka eruption 116, 118–19, 118, 125, 128, 130, 164, 165, 170, 172 as surf zone deposits 161 crystal–lithic–vitric, Lake Taupo 165, 166 crystal–vitric, Lake Taupo 159 fine, black, Tres Cruces Tephra 253 fine-sand beds, Lake Taupo 160, 170 ripple-laminated 166, 168 rippled, Taupo–1.8ka eruption 116, 119, 120, 128, 160, 164 pumice, Taupo–1.8ka eruption 116, 119, 120, 128 vitric 164 vitric ash 158 Sand Hollow, East Fork area conglomerate assemblage, altered sandstone near base 189 wedge-sandstone assemblage 185–7, 186 sand spits 70–1, 72, 187 sandstone Bunga Beds Bouma turbidite divisions 91 coarse, pebbly 90, 92 fine, bioturbated 89 interbedded with black mudstone 94 interbedded with grey mudstone 85, 86, 87, 88, 89 litharenites 94 pebbly basaltic 92, 93, 94 pebbly volcanic 99 tabular 87, 88, 89–90, 90, 91–2 East Fork area composition associated with transition intervals 193 sheet-sandstone, low diversity 187, 189 wave ripples 182, 185 wedge-sandstone assemblage 185–7, 195 gravelly, Icefall Nunatak 24, 29 Klondike Mountain Formation fine, fossiliferous 210 Gilbert-type delta 215 Saar–Nahe Basin fine-grained 266, 269, 271, 272 pebbly, trough cross-bedded 272 ripple cross-bedded 271, 272, 274 sandstone-mudstone beds Icefall Nunatak 29 Klondike Mountain Formation, unusually thick, products of catastrophic events 211, 213 sandstone-mudstone couplets, Klondike Mountain Formation 210 Sanpoil Volcanics 201, 202, 203–5
autobreccia 203–4 calc-alkaline andesite and dacite lavas 203, 207 hypabyssal feeders to 204 organic and coalified material in 205 peperites 204 sedimentary interbeds 204 mainly lacustrine 204–5 stratigraphy 202 saturated pumice behaviour 145–8 saturation of cold/cooling pumice 142 of hot pumice 142 saturation grading 148 ‘saturation-graded’ fabric 118, 129 Scatter Creek Fault 201 Scatter Creek rhyodacite 204 schist clasts, Lake Ksnea 294 scoria fragments, Bunga Beds 94 scoria-crust bombs 46, 55, 57 scour-and-fill structures, Pahvant Butte 62, 65, 68, 71 sediment gravity flows 29 Icefall Nunatak 25 Klondike Mountain Formation 208 cohesionless 209, 215 Pahvant Butte, emplacement of blocks in 64 Republic Basin, clay contributes to cohesiveness of 217 subaqueous 128 see also debris flows; debris-flow deposits sediment liquefaction 230 sedimentation climate and tectonism as primary controls 3 Lake Taupo influenced by available ignimbrite material 154 influx of fine-grained material 152 kept pace with lake-level rise 174 shoreface–offshore, sediment availability and storm deposit development 173–4 Republic Basin and basin evolution 216–18 sedimentary and volcanic facies 203–6 sedimentary, volcanic and hydrothermal facies and facies associations 206–16 shoreface–offshore, Lake Taupo, and wave energy 173–4 subaqueous, in quiet water 86–91 seismic shock, from faulting 280 shales, black 266, 272 laminated 268, 272 sheetwash, ephemeral 131 Sherman Fault 201 shoreface deposits, Lake Taupo 170, 172 plane and cross-lamination 159 storm influence in 159, 160
VSID01 9/12/08 12:57 PM Page 307
Index shoreline deposits East Fork area progradation shown by wedgesandstone 195 rare 187 Lake Taupo 155, 159–60 surf zone trough cross-bedded deposits 159, 160 swash deposits 159, 160 well developed 175 sideromelane ash Pahvant Butte beds with reverse graded bases 64–5 dune-like bedforms 64, 65 sideromelane clasts 20, 21, 25, 26, 27 sideromelane tuffs, palagonitization of 70 silt Lake Ksnea mud 291 vitric, Taupo–1.8ka eruption 116, 118, 119, 120, 121, 128, 160 Skagit River carried Mazama tephra into Lake Ksnea 294 dammed by Damnation Creek landslide material 287, 288 upper 296 current watershed precipitation 287 Skagit River Gorge mouth, subsurface geology 294 slide bodies, Bunga Beds 89, 94, 95, 99 slip-failure, Pahvant Butte 69–70, 77–8 slope instability, Bunga Beds 94 slump folds 94 intraformational 89, 94 slumping Bunga Beds 94, 99 syneruptive 57 snow densification affecting glacier permeability 13 division of glaciers into zones 14 snow, ice and firn defined 14 sole structures 94 spherulites 95 spit complex, Pahvant Butte, formation of 70–1, 72 steam cupola 24 storm deposits/layers, Lake Taupo 166, 172, 173, 175 controlling factors 175 lithic-rich 161, 164 storm scouring 172 storm waves, action of 158–9 stranding, of pumice 148 stratovolcanoes 36, 193–4 streams, dammed by lava flows and pyroclastic eruptions 1–2 see also East Fork area, Challis Volcanic field; Ksnea, Lake subsidence along faults, Republic Basin 205, 216, 217 basin subsidence rate, East Fork area 195 regional tectonic, Reporoa basin 126
subsidence pits, Karymskoye Lake 37, 40–1 and magma ascent 53 surf zone sediments 159, 160, 161, 166 Surtseyan eruptions 2, 10, 29, 36, 53–4 suspension deposits 164–5, 170–1, 175 Saar–Nahe Basin 268–9 suspension settling 69, 91, 121, 128, 129, 158, 218, 297 from pumice rafts 164–5 offshore in a large lake 182 swash zone sediments 126, 150, 159, 170 syneruptive redeposition, Icefall Nunatak 17 table-mountain volcanoes 10, 32 caused by overflowing 15, 17 and hydraulic theory 14, 15 tachylite 21, 24, 25 Taupo–1.8ka eruption 113, 152, 154 ignimbrite-veneer deposit (IVD) 113, 115, 126, 154 erosional trimlines in 129 lithofacies and petrofacies 114–21 accretionary lapilli bed 116, 121, 121 cross-bedded gravel 116, 117, 118 crystal–lithic sand 116, 118–19, 118 openwork pumice gravel 116, 117, 118, 126 pumice dropstones 116, 120, 121 pumice gravel, inversely or normally graded 116, 117, 118 pumice sand, massive–laminated 116, 119, 120 pumiceous diamict 115, 116, 117 rippled sands 116, 119, 120 vitric ash, vesiculated 116, 121 vitric silt, massive– laminated 116, 119, 120, 121, 128 post-eruptive environment 113–14 sedimentary responses to caldera eruptions 133–5 valley-pond ignimbrite (VPI) 113, 124, 131, 154 spillways through 129 Taupo–1.8ka lacustrine sediments lagoonal deposits 160 offshore deposits 148–9 shoreface deposits 159, 160 shoreline deposits 159–60 summary of lithofacies 156–7 Taupo AD–181 eruption, settling and deposition of pumice 141–50 pumice saturation behaviour 142–4 saturated pumice behaviour 145–8 Taupo, Lake break-out flood facies 122, 129–30 break-out flood re-established Waikato river 133 effects of lake draining 174 filling of, problems in calculation 154–5 formation of 111, 133 homogeneity of lithofacies 175 patterns in lake sedimentation 173–4
307 lake draining 174 transgressive surfaces 173 typical deepening succession 174 wave energy and processes on the beach 173 wave energy and shoreface–offshore sedimentation 173–4 patterns in sedimentation 173–4 post-eruption sedimentary record, shows record of reworking 175 post-Taupo–1.8ka eruption deposits, depositional settings 155, 155, 156–7, 158 lagoonal deposits 160 offshore deposits 158–9 shoreface deposits 159, 160 shoreline deposits 159–60 prevailing wind direction 155 pumice deposits in a variety of depositional environments 147–8 sedimentation regimes 160–73 high energy and low sediment influx 164–6 high-energy and high sedimentinflux rate 160–4 low energy and high energy influx rate 166–70 low-energy and low sediment-influx rate 170 variations in lithofacies associations round 172–3 unusual sedimentary system 174 Taupo Volcanic Zone (TVZ) 109–10, 141, 152 geological setting 111–13 pumice saturation behaviour 142–4 saturated pumice behaviour 145–8 subdivision of the chronological framework 111 Taupo volcano 152–4 tectonic activity and lake-level rise, genetic link between 280–1 tectonic ponding 103 tectonics control on ash preservation 275–81 extensional, volcanic syn-rift phase, Saar–Nahe Basin 266 syn-depositional, Bunga Beds 95, 101 tilting of Republic Basin 218 tectonism, East Fork area 195, 197 tephra frequency and origin of, Iceland 239 see also tephra layers, Lake Hestvatn high preservation of deposits in the lacustrine record 273 unconsolidated, preservation of primary deposits 264 see also ash; bombs; Mazama tephra; pumice tephra jets 68, 76, 77 tephra layers, Lake Hestvatn 228 boundary, zones I and II and H4 tephra 241 chemical composition, tephra layers in 94-HV01 core 235–6
VSID01 9/12/08 12:57 PM Page 308
308 tephra layers, Lake Hestvatn (cont.) chronology 228, 238 based on correlation between identified tephra layers 238 concentrations of major and trace elements 237, 239–40, 240–1, 241 return to pre-H4 eruption values 242 from at least five volcanic systems 234 Hekla layers 228 components of 237, 239 H4 layer, physical and chemical characteristics 234, 236, 244 Katla layers 228 deposition into the lake 239–40 KN layer, physical and chemical characteristics 234, 235, 237 radiocarbon ages 229 radiocarbon dating difficult in core 94-HV01 238 sediment accumulation rates 238–9, 240 Settlement layer (Vö) 228, 243, 244 T-tephra 234, 244 carbon increase after deposition of 241 tephra characteristics and concentration 233 tephra particles, microprobe analysis of 233–4 variable carbon content, H4 and KN layers 237–8 tephra stratigraphy, Holocene, Iceland 226 tephrostratigraphic marker beds 266, 267 Humberg marker horizon 273 St Alban tephrostratigraphic marker 274, 274 tephrostratigraphy, Upper Stephanian deposits, Saar–Nahe Basin 267–8 thermal quenching 30 ‘tonsteins’ 264 topographic blocking, of base surge, Karymskoye Lake 54 traction carpet collapse 209 traction deposition, during storms 158 tractional reworking, cross-bedded sandstone, Bunga Beds 90 transgressive deposits Lake Reporoa 122, 126–8, 155, 161 Lake Taupo 161 Saar–Nahe Basin 273 transgressive surfaces, Lake Taupo 173 transgressive–regressive cycles, Saar–Nahe Basin 272 Tres Cruces Tephra (TCT) 257 black massive sand layer 253 product of monogenetic volcanic activity 259 shows signs of weathering 259 Tres Cruces volcano 249 tsunamis, Karymskoye Lake backwash deposits 50 breaking and redistributing bombs 46
Index deposits floated material 51–2 non-floating material 51 effects and deposits of 38, 50–2 formation of new beaches and cliffs 50–1 runup height of 52 timing of largest 52, 55 tuff beds, Saar–Nahe Basin 271 assemblage of juvenile magmatic components 268 characterized by sharp planar contacts 269, 271 concentration in transgressive parts of depositional cycles 267, 269, 273 heavy mineral assemblage 268 as tephrostratigraphic markers 267–8 tuff cones lacustrine, low gradients on flanks 29 Pahvant Butte syn-eruptive outward slips 78 upper cone facies 67–9, 77 sub-aqueous, Icefall Nunatak 29 slope failure on southern flank 29 tuff rings 2, 36 Karymskoye Lake 37, 39, 40 ash lapilli and lapilli ash 42 deposits 41–5 each bed represents deposit of single explosion 43–4 formed by multiple flows of hyperconcentrated water– pyroclastic mixture 44–5 grain size of deposits 41, 42–3, 44 maximum thickness 42, 42 morphology of 40–1 not of fall origin 44 older rings buried by following explosions 56 pits 37, 40–1 surface layer 45 vertical fissures 37, 41 see also ballistic material subaqueous succession, Bunga Beds 99, 100, 104 tuffites 269 turbidite facies association, basin-centre, Bunga Beds 93–4 turbidites Bunga Beds 89 Bouma divisions 91, 94 extralacustrine syn-volcanic resedimented pyroclastic in origin 105 high-concentration 91 slope 91–2 Icefall Nunatak, Stage II 27, 29 Reporoa basin 128 Saar–Nahe Basin 271 turbidity currents 29, 194, 211, 213, 215, 218 ‘continuous-feed’ 3 grain-enriched 208 and heterolithic mud-dominated facies 209 high-concentration 72, 94, 115 low-density 296, 297
tuya volcanoes see table-mountain volcanoes underflows bottom of base surges, Karymskoye Lake 54, 58 and deposition in Bunga Beds 89 Lake Taupo 172 upper Lerma basin, Mexico described 249 diatom assemblages acid pond assemblage 249–50, 251 freshwater pond assemblage 250, 251 shallow freshwater lake assemblage 250, 251 fluctuations in lacustrine levels 259–60 magnetic and microfossil analyses 253–7 palaeoenvironmental history 257–9 volcanic stratigraphy 250–3 ash-flow deposit 252 block-and-ash-flow deposit (older) 250 block-and-ash-flow deposit (younger) 250–2 Lower Toluca Pumice 252 Tres Cruces Tephra 253 Upper Toluca Pumice 252–3 upper Stephanian depositional environments and volcanism lithofacies 268–72 delta plain facies association 271–2 off-shore lacustrine facies association 268–71, 270 prodelta to delta front facies association 271 petrography 268 tephrostratigraphy 267–8 Upper Toluca Pumice (UTP) 249, 252–3 use as a stratigraphic marker 252 upward fining/coarsening /fining trends, basin margin successions, Bunga Beds 95 upward-coarsening and shallowing cycles, Saar–Nahe Basin 266, 271–2 upward-deepening succession, Lake Taupo 174 Van Horn Peak cauldron complex 180–1 explosive volcanism 180, 194, 195 varves, non-glacial 268 vaults, englacial 14–15, 16, 22, 27, 32 and formation of tuya volcanoes 15, 16 vegetation affected by distal base surges 45 effects of destruction in Taupo–1.8ka eruption 113, 154 effects of tephra deposition 226 fluorine damage 242, 244 may increase erosion in affected watersheds 227, 242 killed by lahars 52
VSID01 9/12/08 12:57 PM Page 309
Index Lake Hestvatn, change initiated prior to H4 tephra 241–2 vesicularity of juvenile clasts, Karymskoye Lake 48, 49, 50 prehistoric eruptions 48, 55, 57 vitric ash, fine, Taupo–1.8ka material 114, 158, 160 vitric fragments 152 volcaniclastic sediments, postTaupo–1.8ka eruption 128 volcanism bimodal, southeast Australia 101 Bunga Beds basin 104 Eocene, Challis volcanic field 180 influencing basin drainage patterns 194, 196 in lacustrine depositional record 1 Marie Byrd Land 10 Quaternary, central Mexico 347 Republic Basin 218 Sanpoil Volcanics 201, 202, 203–5, 207 waning at time of extension 217, 218
signature in Lake Taupo lacustrine sediments 175 silicic 111 Taupo Volcanic Zone, related to subduction of Pacific Plate 152 volcanism-lacustrine sedimentation interaction, recording of 175 volcanoes basaltic englacial lacustrine see Icefall Nunatak, Antarctica rhyolitic caldera 133–5 shield volcanoes 10 stratovolcanoes 36, 193–4 see also named volcanoes volcanogenic sediments defined 199 lacustrine deposits, Republic Basin 207–13, 216, 217 Waikato River modern, is underfit 130 outflow from Lake Taupo 114 Waipehi, Lake Taupo 164–5 suspension deposits 164–5
309 Waitahanui, Lake Taupo 166, 167, 168, 173 erosion of beach swash deposits 166 high sediment input inferred 166 shoreface deposits 166 storm deposits 166 surf zone sediments 166 wave energy, and sediment reworking, Lake Taupo 174–5 Yalwal–Comerong–Eden Rift Zone 84, 101 Zacango quarry, upper Lerma basin 249 block-and-ash deposits, older, gas escape pipes 250 block-and-ash deposits, younger, three flow units 250, 251 good palaeoenvironmental information 259 Lower Toluca Pumice (LTP) 252 magnetic and microfossil analyses 253 pond established after Nevado de Toluca eruption 258 Upper Toluca Pumice (UTP) 252–3