Ophiolites in Earth History
Geological Society Special Publications Society Book Editors R. J. PANKHURST (CHIEF EDITOR) P. DOYLE F. J. GREGORY J. S. GRIFFITHS A. J. HARTLEY R. E. HOLDSWORTH
J. A. HOWE P. T. LEAT A. C. MORTON N. S. ROBINS J. P. TURNER
Special Publication reviewing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society's Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society has a team of Book Editors (listed above) who ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Society. The referees' forms and comments must be available to the Society's Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. Geological Society Special Publications are included in the ISI Index of Scientific Book Contents, but they do not have an impact factor, the latter being applicable only to journals. More information about submitting a proposal and producing a Special Publication can be found on the Society's web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways: DILEK, Y. & ROBINSON, P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218. BAZYLEV, B. A., KARAMATA, S. & ZAKARIADZE, G. S. (2003). Petrology and evolution of the Brezovica ultramafic massif, Serbia. In: DILEK, Y. & ROBINSON, P. T. (eds) Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 91-108.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 218
Ophiolites in Earth History EDITED BY Y. DILEK Miami University, USA and
P. T. ROBINSON Dalhousie University, Canada
2003 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society's fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society's international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the American Geological Institute (AGI), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists' Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies' publications at a discount. The Society's online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J OBG: Tel. +44 (0)20 7434 9944; Fax +44 (0)20 7439 8975; E-mail:
[email protected]. For information about the Society's meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society's Corporate Affiliates Scheme, write to
[email protected]. Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN, UK (Orders: Tel. +44 (0)1225 445046 Fax+44 (0)1225 442836) Online bookshop: http://bookshop.geolsoc.org.uk The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. © The Geological Society of London 2003. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 03058719/03/$15.00. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 1-86239-145-9 Typeset by Keytec Typesetting Ltd, Bridport, UK Printed by Cromwell Press, Trowbridge, UK.
Distributors USA AAPG Bookstore PO Box 979 Tulsa OK 74101-0979 USA Orders: Tel. + 1 918 584-2555 Fax+1 918560-2652 E-mail
[email protected] India Affiliated East-West Press PVT Ltd G-l/16 Ansari Road, Daryaganj, New Delhi 110002 India Orders: Tel. +91 11 2327-9113 Fax+91 11 2326-0538 E-mail
[email protected] Japan Kanda Book Trading Company Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-shi, Tokyo 206-0034 Japan Orders'. Tel. +81 (0)423 57-7650 Fax+81(0)423 57-7651 Email
[email protected] Contents Preface
ix
Introduction DlLEK, Y. & ROBINSON, P. T. Ophiolites in Earth history: introduction
1
DILEK, Y. Ophiolite pulses, mantle plumes and orogeny
9
Tethyan ophiolites in the Alpine-Himalayan orogenic system FLOWER, M. F. J. & DILEK, Y. Arc-trench rollback and forearc accretion: 1. A collisioninduced mantle flow model for Tethyan ophiolites
21
DILEK, Y. & FLOWER, M. F. J. Arc-trench rollback and forearc accretion: 2. A model template for ophiolites in Albania, Cyprus, and Oman
43
MUNTENER, O. & PICCARDO, G. B. Melt migration in ophiolitic peridotites: the message from Alpine-Apennine peridotites and implications for embryonic ocean basins
69
BAZYLEV, B. A., KARAMATA, S. & ZAKARIADZE, G. S. Petrology and evolution of the Brezovica ultramafic massif, Serbia
91
SACCANI, E., PADOA, E. & PHOTIADES, A. Triassic mid-ocean ridge basalts from the Argolis Peninsula (Greece): new constraints for the early oceanization phases of the NeoTethyan Pindos basin
109
SARKARINEJAD, K. Structural and microstructural analysis of a palaeo-transform fault zone in the Neyriz ophiolite, Iran
129
AITCHISON, J. C, DAVIS, A. M., ABRAJEVITCH, A. V, An, J. R., BADENGZHU, Liu, J., Luo, H., McDERMlD, I. R. C. & ZIABREV, S. V Stratigraphic and sedimentological constraints on the age and tectonic evolution of the Neotethyan ophiolites along the Yarlung Tsangpo suture Zone, Tibet
147
HEBERT, R., HUOT, F, WANG, C. & Liu, Z. Yarlung Zangbo ophiolites (Southern Tibet) revisited: geodynamic implications from the mineral record
165
MALPAS, I, ZHOU, M.-F, ROBINSON, P. T. & REYNOLDS, P. H. Geochemical and geochronological constraints on the origin and emplacement of the Yarlung Zangbo ophiolites, Southern Tibet
191
Magmatic, metamorphic and tectonic processes in ophiolite genesis HARPER, G. D. Tectonic implications of boninite, arc tholeiite, and MORB magma types in the Josephine Ophiolite, California-Oregon
207
SCHROETTER, J. M., PAGE, P., BEDARD, J. H., TREMBLAY, A. & BECU, V Forearc extension and sea-floor spreading in the Thetford Mines Ophiolite Complex
231
RAYMOND, L. A., SWANSON, S. E., LOVE A. B. & ALLAN, J. F Cr-spinel compositions, metadunite petrology, and the petrotectonic history of Blue Ridge ophiolites, Southern Appalachian Orogen, USA
253
HIRANO, N., OGAWA, Y, SAITO, K., YOSHIDA, T, SATO, H. & TANIGUCHI, H. Multi-stage evolution of the Tertiary Mineoka ophiolite, Japan: new geochemical and age constraints
279
TAKAHASHI, A., OGAWA, Y, OHTA, Y & HIRANO, N. The nature of faulting and deformation in the Mineoka ophiolite, NW Pacific Rim
299
vi
CONTENTS
STAKES, D. S. & TAYLOR, H. P. Jr Oxygen isotope and chemical studies on the origin of large plagiogranite bodies in northern Oman, and their relationship to the overlying massive sulphide deposits
315
Hydrothermal and biogenic alteration of oceanic crust as recorded in ophiolites GREGORY, R. T. Ophiolites and global geochemical cycles: implications for the isotopic evolution of seawater
353
GIGUERE, E., HEBERT, R., BEAUDOIN, G., BEDARD, J. H. & BERCLAZ, A. Hydrothermal circulation and metamorphism in crustal gabbroic rocks of the Bay of Islands ophiolite complex, Newfoundland, Canada: evidence from mineral and oxygen isotope geochemistry
369
MUEHLENBACHS, K., FURNES, H., FONNELAND, H. C. & HELLEVANG, B. Ophiolites as
401
faithful records of the oxygen isotope ratio of ancient seawater: the Solund-Stavfjord Ophiolite Complex as a Late Ordovician example FURNES, H. & MUEHLENBACHS, K. Bioalteration recorded in ophiolitic pillow lavas
415
Ophiolite emplacement: mechanisms and processes WAKABAYASHI, J. & DILEK, Y. What constitutes 'emplacement' of an ophiolite?: Mechanisms and relationship to subduction initiation and formation of metamorphic soles
427
GRAY, D. R. & GREGORY R. T. Ophiolite obduction and the Samail Ophiolite: the behaviour of the underlying margin
449
SEARLE, M. P., WARREN, C. I, WATERS, D. J. & PARRISH, R. R. Subduction zone polarity in the Oman Mountains: implications for ophiolite emplacement
467
Regional occurrence of ophiolites and geodynamics HARRIS, R. Geodynamic patterns of ophiolites and marginal basins in the Indonesian and New Guinea regions
481
MlLSOM, J. Forearc ophiolites: a view from the western Pacific
507
SPAGGIARI, C. V, GRAY, D. R. & FOSTER, D. A. Tethyan- and Cordilleran-type ophiolites of eastern Australia: implications for the evolution of the Tasmanides
517
ZHANG, Q., WANG, Y., ZHOU, G. Q., QIAN, Q. & ROBINSON P. T. Ophiolites in China: their distribution, ages and tectonic settings
541
SPADEA, P., ZANETTI, A. & VANNUCCI, R. Mineral chemistry of ultramafic massifs in the Southern Uralides orogenic belt (Russia) and the petrogenesis of the Lower Palaeozoic ophiolites of the Uralian Ocean
567
ISHIWATARI, A., SOKOLOV, S. D. & VYSOTSKIY S. V Petrological diversity and origin of ophiolites in Japan and Far East Russia with emphasis on depleted harzburgite
597
SOKOLOV, S. D., LUCHITSKAYA, M. V, SILANTYEV, S. A., MOROZOV, O. L., GANELIN, A. V, BAZYLEV, B. A., OSIPENKO, A. B., PALANDZHYAN, S. A. & KRAVCHENKO-BEREZHNOY, I. R. Ophiolites in accretionary complexes along the Early Cretaceous margin of NE Asia: age, composition, and geodynamic diversity
619
STERN, C. R. & DE WIT, M. J. Rocas Verdes ophiolites, southernmost South America: remnants of progressive stages of development of oceanic-type crust in a continental margin back-arc basin
665
DILEK, Y. & AHMED, Z. Proterozoic ophiolites of the Arabian Shield and their significance in Precambrian tectonics
685
Preface
This book is derived from the interdisciplinary, contemporary work of the international ophiolite community in a most up-to-date treatment of process-oriented problems and questions on the generation and evolution of ophiolites. It is a large collection of research papers from a wide range of international contributors. Some of these papers were presented in thematic ophiolite sessions at the 2001 Annual Meeting of the Geological Society of America (Boston) and the 2001 Fall Meeting of the American Geophysical Union (San Francisco). The 32 papers here examine the mode and nature of igneous, metamorphic, tectonic, sedimentological, and biological processes associated with the evolution of oceanic crust in different tectonic settings in Earth history as revealed in various ophiolites and ophiolite belts around the world, and the geodynamic significance of these ophiolites in the evolution of different orogenic systems. Divided into six thematic sections, the book presents a wealth of new data and syntheses from mainly Phanerozoic ophiolites around the world. We would like to express our thanks to the contributors to this book for their time and effort. We also would like to extend our sincere appreciation and gratitude to Angharad Hills (Staff Editor) and Andy Morton (Book Series Editor) for their help and advice at review stages, and to the Geological Society Publishing House staff for their support in the publication process. Diligent work by Senior Production Editor Sarah Gibbs at all stages throughout the preparation and reproduction of this book contributed to its success. Cathy Edwards in the Geology Department at Miami University helped with manuscript preparation and proofreading of the chapters. The Office of Advancement of Research and Scholarship, the College of Arts and Science, and the Department of Geology at Miami University provided partial financial support for the preparation of the book that we gratefully acknowledge. We wish to thank the following colleagues for their timely and thorough reviews of the manuscripts that helped us maintain the high scientific standards for which we have striven: James Allan (Appalachian State University, USA); Jeffrey C. Alt (University of Michigan, USA); Shoji Arai (Kanazawa University, Japan); Neil Banerjee (University of Alberta, Canada); Asish Basu (University of Rochester, New York, USA); Jean Bebien (Universite de Paris-Sud, Orsay, France); Manuel Berberian (New Jersey, USA); Sherman Bloomer (Oregon State University, USA); Craig Buchan (Curtin University of Technology, Australia); Sun-Lin Chung (National Taiwan University, Taiwan); Ian WD. Dalziel (University of Texas at Austin, USA); Hugh Davies (University of Papua New Guinea);
Yildirim Dilek (Miami University, USA); Jaroslav Dostal (St. Mary's University, Canada); Grenville Draper (Florida International University, USA); Stephen Edwards (University of Greenwich, England); Don Elthon (University of Houston, USA); John Encarnacion (St. Louis University, USA); Martin Fisk (Oregon State University, USA); Martin F.J. Flower (University of Illinois at Chicago, USA); Gretchen Frueh-Green (ETH-Zentrum, Switzerland); Ulrich Glasmacher (Germany); David Gray (University of Melbourne, Australia); Ron Harris (Brigham Young University, USA); Kendall Hauer (Miami University, USA); James W. Hawkins (Scripps Institution of Oceanography, California, USA); Rejean Hebert (Universite Laval, Quebec, Canada); Rod Holcombe (University of Queensland, Australia); Paul Holm (Earlham College, USA); Francois Huot (Universite Laval, Quebec, Canada); Akira Ishiwatari (Kanazawa University, Japan); Barbara John (University of Wyoming, USA); Thierry Juteau (IUEM, Plouzane, France); Ade Kadarusman (Tokyo Institute of Technology, Japan); Andrew Kerr (Cardiff University, UK); Elena Konstantinovskaia (Russian Academy of Sciences, Moscow-Russia); John Malpas (University of Hong Kong, China); Catherine Mevel (Institute de Physique du Globe, Paris-France); Calvin Miller (Vanderbilt University, USA); John Milsom (University College London, UK); Eldridge Moores (University of California at Davis, USA); Kula Misra (University of Tennessee, USA); Karlis Muehlenbachs (University of Alberta, Canada); Christopher Parkinson (University of New Orleans, USA); Gene Perry (Northern Illinois University, USA); Tjerk Peters (Universitat Bern, Switzerland); Ali Polat (University of Windsor, Canada); Elisabetta Rampone (University of Genova, Italy); Paul T. Robinson (Dalhousie University, Canada); Andrew Saunders (University of Leicester, UK); Peter Schiffman (University of California at Davis, USA); Richard Sedlock (San Jose State University, California, USA); John Shervais (Utah State University, USA); Eli Silver (University of California at Santa Cruz, USA); Ian Smith (The University of Auckland, New Zealand); Piera Spadea (Universita di Udine, Italy); Catherine Spaggiari (Monash University, Australia); Debra Stakes (MBARI, California, USA); Hubert Staudigel (Scripps Institution of Oceanography, California, USA); Charles Stern (University of Colorado, USA); Mohamed Sultan (University at Buffalo, New York, USA); Damon A. H. Teagle (University of Southampton, UK); David Vanko (Towson University, Maryland); John Wakabayashi (Hayword, USA); and Steve Wojtal (Oberlin College, USA). Yildirim Dilek Oxford, USA, October 2003
This page intentionally left blank
Ophiolites in Earth history: introduction YILDIRIM DILEK 1 & PAUL T. ROBINSON 2 Department of Geology, Miami University, Oxford, OH 45056, USA (e-mail:
[email protected]) 2 Department of Earth Sciences, Dalhousie University, Halifax, N.S. B3H 3J5, Canada 1
Ophiolites record significant evidence for tectonic and magmatic processes from rift-drift through accretionary and collisional stages of continental margin evolution in various tectonic settings. Structural, petrological and geochemical features of Ophiolites and associated rock units provide essential information on mantle flow field effects, including plume activities, collision-induced aesthenospheric extrusion, crustal growth via magmatism and tectonic accretion in subductionaccretion cycles, changes in the structure and composition of the crust and mantle reservoirs through time, and evolution of global geochemical cycles and seawater compositions. Ophiolite studies over the years have played a major role in better understanding of mid-ocean ridge and subduction zone processes, mantle dynamics and heterogeneity, magma chamber processes, fluid flow mechanisms and fluid-rock interactions in oceanic lithosphere, the evolution of deep biosphere, the role of plate tectonics and plume tectonics in crustal evolution during the Precambrian and the Phanerozoic, and mechanisms of continental growth in accretionary and collisional mountain belts. Through multi-disciplinary investigations and comparative studies of Ophiolites and modern oceanic crust and using advanced instrumentation and computational facilities, the international ophiolite community has gathered a wealth of new data and syntheses from Ophiolites around the world during the last 10 years. The purpose of this book is to present the most recent data, observations and ideas on different aspects of 'ophiolite science' through case studies and to document the mode and nature of igneous, metamorphic, tectonic, sedimentological and/or biological processes associated with the evolution of oceanic crust in different tectonic settings in Earth's history. It comprises 32 papers collected in six sections on temporal relations amongst ophiolite genesis, mantle plume events and orogeny in Earth history; Tethyan ophiolites in the Alpine Himalayan orogenic system; magmatic, metamorphic and tectonic processes in ophiolite genesis; hydrothermal and biogenic alteration of oceanic crust; mechanisms of ophiolite emplace-
ment; and regional occurrences of ophiolites and their geodynamic implications.
Ophiolites, mantle plumes and orogeny Ophiolite occurrences around the world are not a random geological phenomenon. Ophiolites with certain age groups in different orogenic belts characterize distinct ophiolite pulses, which mark times of enhanced ophiolite genesis and emplacement. Examining the geological record of mountain-building episodes and related events, Dilek shows that ophiolite pulses overlap significantly with the timing of major collisional events during the assembly of supercontinents, their break-up and increased mantle plume activities that developed extensive large igneous provinces (LIPs). These global events have been involved in the Wilson cycle evolution of ancient ocean basins that in turn contributed to ophiolite genesis in diverse tectonic settings. Suprasubduction zone ophiolites represent anomalous oceanic crust generation in subduction rollback cycles during the closing stages of basins prior to terminal continental collisions. Accelerated LIP formation associated with superplume activities may have facilitated both the generation and tectonic emplacement of ophiolites at global scales. These spatial and temporal relations suggest that ophiolite pulses, mantle plume activities and orogenic events have been closely linked through complex mantle dynamics in Earth history.
Tethyan ophiolites in the AlpineHimalayan orogenic system Papers in this section present diverse data from Tethyan ophiolites and provide refined geodynamic models for their evolution. Flower & Dilek examine the processes of arc-trench rollback and forearc accretion, and present an 'actualistic' model for ophiolites based on recent observations of forearc evolution in western Pacific and Mediterranean marginal basins. Collision-induced mantle flow and 'slab-pull' forces may result in rapid
From: DILEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 1-8. 0305-8719/03/$15 © The Geological Society of London 2003.
2
Y. DILEK & P. T. ROBINSON
arc-trench rollback pulses and associated extensional episodes (splitting of nascent volcanic proto-arcs), producing proto-ophiolites in arc-forearc settings. These ophiolites commonly include hightemperature metamorphic soles, boninitic rocks, juxtaposed refractory peridotites and high-temperature epidosites that are generally absent in mid-ocean ridge, normal arc and back-arc basin environments. As subduction rollback continues, arc-forearc complexes become increasingly heterogeneous, displaying significant internal age and structural discrepancies, a common feature both in the SW Pacific subduction zone environments and Tethyan ophiolites. When an arc-trench rollback cycle is terminated by a collision, heterogeneous forearc lithosphere is accreted as ophiolites in the initial stages of the evolution of collisional orogenic belts. This model demonstrates the apparent correspondence of subduction nucleation and mantle flow to plate collisions at regional and global scales. In a companion paper, Dilek & Flower explore the application of the arc-trench rollback and forearc accretion model to Neo-Tethyan ophiolites, specifically to the Mirdita (Albania), Troodos (Cyprus) and Semail (Oman) ophiolites. NeoTethyan oceans evolved as east-west-oriented basins separated by discrete continental fragments, which were rifted off from the northern edge of Gondwana beginning in the Triassic. Triassic rift assemblages containing within-plate-type alkaline basalt to transitional (T-MORB) and mid-ocean ridge basalt (MORE) are spatially associated with ophiolites in the eastern Mediterranean region and may represent the precursor of Late Triassic oceanic crust, which was subsequently consumed to produce the suprasubduction zone ophiolites. The three ophiolites examined here include a basement of typical 'oceanic' lithosphere intruded and overlain by boninitic (ultra-refractory) to calcalkaline series rocks that formed in a proto-arcforearc setting. This progression was a result of upper plate extension and further melting of previously depleted asthenosphere that occurred in response to successive stages of slab rollback. This igneous evolution of the ophiolites involved subduction initiation and one or more episodes of proto-arc splitting before the termination of slab rollback cycles as a result of trench-continent collisions. Miintener & Piccardo examine the Lanzo and Corsica ophiolitic peridotites in the Alpine-Apennine mountain system that are interpreted as remnants of the Ligurian Tethys. The texture, geochemistry and petrology of these peridotites suggest that they represent exhumed subcontinental lithospheric mantle, which was modified and refertilized by migrating melts during opening of
the embryonic Piedmont-Ligurian Ocean. Pervasive melt infiltration and melt-rock reaction produced gabbroic intrusions with a wide range of compositions characteristic of the melting column beneath mid-ocean ridges. These observations are critical to better understand the effects of melt percolation and impregnation in development of plagioclase-enriched peridotites. The Ligurian ophiolites clearly do not represent a typical Penrose-type, idealized oceanic crust. Bazylev et al. present mineral and bulk-rock chemistry data from the Jurassic Brezovica ultramafic massif (Serbia) in the Dinarides and show that its petrogenetic evolution involved two distinct magmatic stages. A suite of spinel harzburgites was produced during the first stage as a result of partial melting of the mantle and segregation of tholeiitic melts. Percolation of melt through these spinel harzburgites and melt-rock reaction produced dunites and refractory harzburgites during the second stage and generated highCa boninitic melt. The authors conclude that the second magmatic stage had to occur in a suprasubduction zone setting. Saccani et al. present new field and geochemical constraints from the Western Hellenides in Greece, documenting that initial stages of seafloor spreading and oceanic crust formation in the Pindos basin probably occurred in the Mid- to Late Triassic, earlier than previously thought. Pillow lavas from the Argolis Peninsula have MORB trace element characteristics and are divided into T-MORB and normal MORB (NMORB). These are the oldest unequivocally dated oceanic crust in the Hellenide sector of the Pindos Basin. Early Triassic rifting produced shoshonitic and calc-alkaline lavas derived from a mantle source that was previously contaminated by subduction components. Associated alkaline basalts were derived from ocean island basalt-type (OIB) mantle source. Mixing of mantle sources produced enriched MORB (E-MORB) and T-MORB, and then N-MORB lavas were erupted in Mid(?)- to Late Triassic, suggesting that sea-floor spreading had reached a steady state. The authors cite the Red Sea as a modern analogue with along-strike chemical variations for the Pindos Basin. Sarkarinejad describes the internal structure of the Cretaceous Neyriz ophiolite in southern Iran, and presents structural and microstructural observations for the existence of a NW-trending palaeotransform fault zone within this Neo-Tethyan ophiolite. Fabric analysis of mylonitic rocks (including hornblende and plagioclase textures and chemistry) suggests that the plastic deformation of mafic-ultramafic rocks occurred at amphibolitefacies conditions within a dextrally slipping oceanic transform fault zone. The author infers that
INTRODUCTION the Neyriz transform fault separated ENE-trending spreading centre segments within a Neo-Tethyan basin. The last three papers in this section present diverse stratigraphic, petrological, geochemical and geochronological data from the YarlungTsangpo suture zone ophiolites in southern Tibet. Aitchison et al. define several discrete terranes along the suture zone and use their sedimentological and biostratigraphic data to constrain the timing of ophiolite formation and terrane accretion within this segment of the HimalayanTibetan orogenic belt. Different ages of ophiolitic assemblages from Xigaze, Jungwa and Zedong indicate that the suture zone may contain remnants of multiple (two?) island arc complexes that had evolved within the same branch of Neo-Tethys. Hebert et al. report mineral chemistry data and petrological findings from mafic-ultramafic rocks of the Yarlung Tsangpo ophiolites. Mantle peridotites were exhumed from depths of more than 50 km and underwent 10-40% partial melting and melt percolation within a suprasubduction zone wedge. The Yarlung Tsangpo ophiolites represent a heterogeneous collage of arc, forearc and backarc oceanic lithosphere developed in a NeoTethyan basin south of the active continental margin of Eurasia. Malpas et al. present new geochronological data from the Yarlung-Tsangpo ophiolites and a refined geodynamic model for their evolution. The new sensitive high-resolution ion microprobe date of 126 Ma for the Dazhuqu massif indicates that the Xigaze ophiolite is significantly younger than the Loubusa ophiolite and Zedong island arc complex (c. 175 Ma). These findings are consistent with the geochemical interpretations of Hebert et al. Basaltic rocks from all ophiolites are composed of island arc tholeiites, and the peridotites show textural and chemical evidence for percolation of boninitic melts through the upper mantle at later stages of magmatism. The Yarlung-Tsangpo ophiolites may have formed at different times in suprasubduction zone environments and were subsequently juxtaposed during the collision of the Indian continental margin with the arc-trench system around 90-80 Ma.
Magmatic, metamorphic and tectonic processes in ophiolite genesis The six papers in this section present processoriented case studies of oceanic crust evolution from the Appalachian, Cordilleran, Tethyan and Japanese ophiolites. Harper demonstrates that the extrusive sequence and sheeted dyke complex in the Jurassic Josephine ophiolite in CaliforniaOregon (USA) display chemical evidence for a
3
wide range in magma types and degree of fractionation. New discoveries of Fe-Ti-rich and Ti-poor (boninitic) magmas in the Josephine ophiolite illustrate its compositional complexity and provide new constraints on its tectonic environment of formation. The Fe-Ti lavas imply formation along a propagating rift, whereas the low-Ti lavas suggest a forearc environment of their origin. The Lau Basin is cited as a likely modern analogue because the available geochemical data from several environments within this modern back-arc basin are consistent with the new chemical data and interpretations from the Josephine ophiolite. Northern Tonga and the Andaman Sea may also be plausible analogues for the Josephine ophiolite. Schroetter et al. examine the internal structure and stratigraphy of the Ordovician Thetford Mines ophiolite in Quebec (Canada). The discovery of a locally well-developed sheeted dyke complex, combined with other structural data, indicates that the Ordovician oceanic crust was developed at a slow-spreading centre, where faulting and magmatism were coeval, keeping pace with crustal extension. The boninitic affinity of cumulate rocks and lavas suggests that the Thetford Mines ophiolite probably formed in a forearc setting. This is one of the best-documented cases of well-established pre-collisional extensional tectonics in a palaeoforearc environment. Raymond et al. investigate the occurrence and petrogenesis of ultramafic rock bodies in the Southern Appalachian (USA) orogenic belt. These ultramafic rocks are part of dismembered Ordovician ophiolites, which probably formed in a slowspreading centre within a subduction zone setting. A suprasubduction zone environment of origin is supported by the existence of metadunites representing sublithospheric melt channels and zones of high melt flux. The authors suggest that the Taconic subduction zone that was responsible for the formation of the Southern Appalachian ophiolites may have been west-directed, rather than east-directed as previous models have inferred. Hirano et al. show that the Tertiary Mineoka ophiolite in central Japan had a multi-stage tectonic evolution prior to its emplacement onto the Japanese continental margin. It occurs near a trench-trench-trench triple junction and contains tholeiitic pillow basalts and dolerites, calc-alkaline plutonic rocks and alkali-basaltic sheet flows. The sea-floor spreading stage of the ophiolite probably occurred during the generation of an oceanic Mineoka Plate in the Eocene. Subduction of the Pacific Plate beneath the Mineoka Plate produced island arc volcanism during 40-25 Ma (second stage). Eruption of the within-plate-type alkali basalts (WPB) during the third stage occurred around 20 Ma, shortly before the emplacement of
4
Y. DILEK & P. T. ROBINSON
the polygenetic Mineoka ophiolite onto the continental margin. The ophiolite was derived from the Mineoka Plate, not from the Philippine Sea or Pacific Plates as previous models suggest. The companion paper by Takahashi et aL examines the internal structure of the Mineoka ophiolite and reports three main phases of deformation recorded by ophiolitic rocks. The first deformation phase was manifested in oblique normal faults and associated vein systems, and was associated with extensional tectonics at a palaeo-spreading centre. The second phase of deformation, characterized by thrust faults and strike-slip shear zones, was related to the emplacement of the ophiolite. The third phase of deformation is represented by transpressional dextral faults, manifestation of the modern tectonic regime in a trench-trench-trench triple junction. The last paper in this section, by Stakes & Taylor, documents the occurrence of large plagiogranite intrusions in the northern part of the Semail ophiolite (Oman) and their spatial and temporal association with the formation of massive sulphide deposits. Chemical, isotopic and field relations indicate that plagiogranite bodies near the overlying sheeted dykes formed through a complex process of combined assimilation and fractional crystallization, and recharge by injection of basaltic magma in open-system magma chambers. These plagiogranites were clearly late-stage magmatic products postdating the formation of the main ophiolitic crust and acted as shallow point sources of heat and metals for development of the overlying economic massive sulphide deposits.
Hydrothermal and biogenic alteration of oceanic crust as recorded in ophiolites The four papers in this section examine the nature, mechanisms and products of hydrothermal and biogenic alteration of oceanic crust and their implications for geochemical cycles in Earth history. Gregory demonstrates that the hydrothermal alteration history of ophiolites has major implications for the isotopic evolution of seawater. Isotopic profiles through ophiolites (e.g. Semail) show completely different characteristics depending on the element involved (Nd, Sr and O) and its residence time in the ocean. Oxygen isotopes are perhaps the most useful indicators of geochemical cycles and seawater-rock interaction. The mean value of altered oceanic crust is close to its primary 18O/16O ratio, which means that there must be complementary reservoirs of 18Odepleted and -enriched rocks in the altered ocean crust. Ophiolites are particularly useful because they are pieces of oceanic lithosphere that have
escaped recycling. Ophiolite studies show that oxygen isotopic composition of seawater resides at near steady-state conditions over Earth history. Giguere et aL present mineral and oxygen isotope geochemistry data from gabbroic rocks of the North Arm Mountain massif in the Bay of Islands ophiolite in Newfoundland (Canada) to constrain the chronology and temperature conditions of fluid circulation with respect to the timing and nature of deformation as recorded in these lower-crustal rocks. With continued cooling of gabbroic rocks, amphibole compositions changed as temperatures of amphibole formation fell steadily. Early amphiboles show near igneous oxygen isotope compositions typical of MORB or backarc basin basalt (BABB). Seawater infiltration into the lower crust occurred along listric shear zones under low fluid/rock ratios during the initial stages of deformation and metamorphism. Further cooling facilitated brittle deformation and greater seawater penetration at depth with increased fluid/ rock ratios, as suggested by very low 618O values. Field relations suggest that late-stage trondhjemitic intrusions may have provided heat and convective circulation of hydrothermal fluids causing high- T alteration superimposed on earlier stage of lower-T alteration. These relations clearly show that successive episodes of hydrothermal alteration of fossil lower crust in the Bay of Islands ophiolite were entirely intra-oceanic in origin. Muehlenbachs et aL use the hydrothermal alteration history of the Ordovician SolundStavfjord Ophiolite Complex (SSOC) in western Norway to examine the oxygen isotope ratio of ancient seawater. Similar to most ophiolites, the SSOC shows enrichment of 18O in the lavas altered at low temperatures and depletion in the dykes and gabbros altered at higher temperatures; this is also compatible with the alteration profile of 5.9 Ma in situ oceanic crust drilled in Ocean Drilling Program Hole 504B south of the Costa Rica Rift. Ophiolites can reflect the isotopic composition of ancient seawater. There is no observable secular trend in the 618O of seawater, and hence the mode and scale of seawatersea-floor interaction has not changed with time. The 618O of sediments and fossils may not record true values but rather owe their compositions to isotopic resetting, warmer oceans or biased sampling of restricted basins. Thus, models of ancient climates and ocean volumes determined from such data may be incorrect. Furnes & Muehlenbachs examine the nature and extent of bioalteration in fossil oceanic crust with different ages. Bioalteration of volcanic glass has been demonstrated in in situ oceanic crust but is not yet well documented from ophiolites. The authors have looked for evidence of bioalteration
INTRODUCTION
5
in glassy pillow lavas from four major ophiolites: Cretaceous Troodos (Cyprus), Jurassic Mirdita (Albania), Ordovician Solund-Stavfjord (western Norway) and early Proterozoic Jormua (Finland). Bioalteration may be recognized from textural evidence, organic remains, chemical fingerprints (C, N, S and P) and carbon isotopic signatures. Textural evidence in the form of coalesced spheres and tubes is present only in Troodos and Mirdita, the youngest of the ophiolites investigated. Some textural features in the SSOC resemble biogenerated textures, but rocks metamorphosed to amphibolite facies grade lack any evidence of bioalteration. Organic remains, in the form of twisted filaments, have been found only in Troodos. Probable organic carbon has been found in rocks from Troodos and the SSOC. Carbon isotope data in glassy samples are shifted to lower values and have a pattern very similar to that for in situ oceanic lavas. Evidence of bio-alteration appears to survive low-grade greenschist-facies metamorphism but is generally destroyed at higher grades of metamorphism.
structural data from rocks beneath the ophiolite nappe suggesting that there was an earlier period of underthrusting-subduction beneath the Arabian continental margin prior to its formation and obduction. Therefore, emplacement of the Semail nappe cannot simply be linked to a single subduction zone dipping away from the continent during the evolution of the ophiolite. The age of eclogite metamorphism in the lower-plate rocks beneath the ophiolite nappe (Saih Hatat Window) is crucial in testing this and other existing models. Searle et al. dispute this model by Gray & Gregory and discuss whether all structures and metamorphism observed in northern Oman are related to a single, prolonged episode of ophiolite emplacement, lasted for c. 27 million years and associated with a subduction zone dipping away from the Arabian continent. Suprasubduction zone origin of the ophiolite, metamorphic sole generation and eclogite formation were all linked to this subduction zone. Clearly, more precise age dating of the highpressure rocks beneath the ophiolite is needed to resolve the current debate.
Ophiolite emplacement: mechanisms and processes
Regional occurrence of ophiolites and geodynamic implications
Emplacement of ophiolites into continental margins is a first-order tectonic problem in plate tectonics and a significant phase in the evolution of orogenic belts. Ever since their recognition as on-land fragments of ancient oceanic lithosphere, mechanisms and processes involved in incorporation of ophiolites into continents have been a subject of discussion amongst researchers. The three papers in this section evaluate the existing models and ideas on ophiolite emplacement mechanisms with a focus on the Cretaceous Semail ophiolite in Oman. Wakabayashi & Dilek discuss the mechanisms and significance of subduction initiation and metamorphic sole formation in ophiolite emplacement and define four prototype ophiolites based on their emplacement mechanisms. Tethyan ophiolites are collisional-type emplaced over passive continental margins, whereas Cordilleran ophiolites are emplaced over subduction complexes through accretionary processes. Emplacement of ridge-trench intersection (RTI) ophiolites occurs through complex processes resulting from interaction of a spreading ridge and a subduction zone. Macquarie Island-type ophiolite represents oceanic crust exposed as a result of shifts in plate boundary configurations (i.e. spreading ridge segments converting into a diffuse transpressional plate boundary). Gray & Gregory review emplacement models for the Semail ophiolite in Oman and present
The papers in this section involve the regional occurrence of ophiolite belts on different continents and provide new petrological, geochemical and geochronological data and syntheses to better constrain their geodynamic evolution. Harris explores the spatial, temporal, geological and geochemical patterns of ophiolites in the Indonesian and New Guinea region (ING) in the first paper. ING is a repository of island arcs, marginal basins, continental fragments and ophiolites amalgamated by repeated plate boundary reorganizations. Major plate boundary reorganizations in the ING region coincide with global plate motions and there is a strong correlation in space and time between ophiolite genesis and collisional events. Opening of basins and suprasubduction zone generation of ophiolites are likely to have been 'enhanced' by extrusion of aesthenosphere escaping collisional zones in the region. Ophiolites forming in these suprasubduction zone environments display age and compositional heterogeneity, indicating their composite nature. Milsom examines the New Caledonia region in the SW Pacific to determine the spatial relations between forearc ophiolites and their volcanic arc systems. Repeated episodes of collisional events, postcollisional faulting and magmatism, and sea-floor spreading appear to have displaced and separated forearc tectonic assemblages from their respective volcanic arc systems in the New Caledonia-New
6
Y. DILEK & P. T. ROBINSON
Guinea region. This complex history may be responsible for the apparent lack of volcanic arc edifices associated with other forearc ophiolites (e.g. Troodos in Cyprus) around the world. Spaggiari et al. provide an overview of the Neoproterozoic to Cambrian ophiolites of the Tasmanides in eastern Australia, and examine the differences in their emplacement styles and tectonic settings. Eastern Australian ophiolites fall into Tethyan- and Cordilleran-type categories depending on their relationship to 'continental basement', and they appear to have developed in various suprasubduction zone environments (arc, forearc, back-arc) along the eastern Gondwana margin. Their age progression and geochemistry, combined with regional structural and tectonic constraints, suggest that they evolved in a complex rifted arc-back-arc system during 530-485 Ma, and that the collapse of this system into the continental margin of East Gondwana resulted in their emplacement. This event might have been related to far-field stresses associated with the collisional assembly of greater Gondwana in the early Palaeozoic. Zhang et al. summarize the regional distribution, ages and inferred tectonic settings of ophiolites in China. The Chinese ophiolites fall into four major age groups, Proterozoic, early Palaeozoic, late Palaeozoic and Mesozoic-Cenozoic, and they mainly occur along suture zones separating different tectonic blocks. They have a melange character in general and display structural and metamorphic evidence for multiple episodes of collisional events. The majority of the Chinese ophiolites are compositionally heterogeneous, containing mixtures of island arc tholeiite and boninite with lesser amounts of MORE and OIB. Palaeo-Tethyan ophiolites mostly have MORBtype rocks and may have formed in small intracontinental basins. Spadea et al. investigate the pyroxene and amphibole compositions of various mantle peridotites, particularly the Nurali and Mindyak massifs in the Southern Uralides in Russia. The Ural Mountains are a fold mountain system that records a Late Paleozoic arc-continent collision along the eastern European palaeomargin of Baltica. The Main Uralian Fault marks the related suture zone that consists of a melange composed of arc fragments and dismembered ophiolites. The Nurali and Mindyak peridotites have several anomalous features for abyssal peridotites: fertile composition; internal zoning from Iherzolite to dunite to harzburgite; anomalous crust-mantle transition with amphibole-bearing, plagioclase-free, ultramafic cumulates; lack of associated crustal section; and intrusion of late (400 Ma) gabbro-diorite plutons. These peridotite bodies underwent multi-
stage igneous events including porous flow, and rock-melt interaction involving pyroxene dissolution and plagioclase precipitation. They thus show some similarities to peridotites of subcontinental mantle and/or continent-ocean transition zone mantle. The authors present two explanations for the origin of these peridotite massifs in the Southern Uralides: (1) the anomalous features (for abyssal peridotites) reflect modification of normal MORE peridotites formed beneath a spreading axis by large volumes of island arc melts; or (2) the peridotites were originally part of subcontinental mantle, which underwent modification by dominantly tholeiitic melts causing plagioclase precipitation. Ishiwatari et al. discuss the petrological diversity and origin of ophiolites in Japan and Far East Russia, and distinguish highly depleted mantle harzburgite (DH) massifs in them. These ophiolites range in age from Early Palaeozoic to Cenozoic and are tectonically underlain by blueschist-bearing rocks and accretionary complexes that are generally younger in age. The majority of the ophiolites probably formed intra-oceanic island arc systems, as their petrological and geochemical characteristics suggest, and were incorporated into the Eurasian continental margin through repeated episodes of Mariana-type nonaccretionary subduction zone processes over time. There is little in the English literature on the ophiolite complexes of NE Asia. Sokolov et al. present new data on the age, structure and composition of ophiolites in the West Koryak fold belt in Far East Russia. The region consists chiefly of a variety of accreted terranes of different age and character. The ophiolites fall into two main categories. Palaeozoic ophiolites are primarily oceanic (MORB) in character and are viewed as fragments of the Panthalassa Ocean. Mesozoic ophiolites typically have an SSZ signature. In general, the ophiolites become younger towards the Pacific Ocean in the east. Accretionary prisms contain terrigeneous melanges similar to those of the Shimanto Belt of SW Japan. Stern & De Wit describe the geology and geochemistry of the Mesozoic Rocas Verdes ophiolites in the southernmost Andes (South America) and show that these ophiolites evolved in a Late Jurassic-Early Cretaceous intra-arc basin along the southern edge of Gondwana. Primary crosscutting relations of ophiolitic dyke swarms with the surrounding crystalline basement rocks of the Andean magmatic arc indicate that Rocas Verdes basin was an ensialic small ocean that opened up by 'unzipping' from the south to the north, synchronously with the onset of seafloor spreading in the South Atlantic at c. 132 Ma. Thus the Rocas Verdes ophiolites provide a unique
INTRODUCTION opportunity to investigate the mode and nature of igneous, metamorphic and tectonic processes associated with continental rifting, sea-floor spreading and tectonic collapse of a back-arc basin in an Andean-type active continental margin. Finally, Dilek & Ahmed present an overview of the Proterozoic ophiolites in the Arabian Shield and discuss their significance in Precambrian tectonics. The Arabian Shield ophiolites range in age from c. 870 Ma to c. 627 Ma and display a record of rift-drift, sea-floor spreading and collision tectonics during the evolution of the East African Orogen in the aftermath of the break-up of Rodinia. Ophiolites in the western part of the shield were part of ensimatic are terranes, which were sutured through a series of collisional events. Younger ophiolites in the eastern Arabian Shield were incorporated into accretionary complexes through offscraping and collisional events during continued subduction, similar to the accretionary history of those Phanerozoic ophiolites in NE Asia as reported by Sokolov et al. The youngest ophiolites in the shield (Nabitah-Hamdah fault zone ophiolites) are post-collisional in origin and they represent Ligurian-type oceanic crust developed in an intracontinental para-rift basin. The Arabian shield ophiolites are clearly diverse in origin and provide a great opportunity to investigate oceanic and juvenile crust evolution in the latest Precambrian.
Concluding remarks Ophiolites are critical windows into Earth history to examine the mode and nature of and the interplay between various igneous, metamorphic, sedimentological, hydrothermal and tectonic processes during generation of oceanic lithosphere. They also provide essential information on the mechanics and kinematics of mountain building episodes, as their incorporation into continental margins involved major tectonic events in orogenesis. New data and observations presented in different papers in this book clearly show that there is not a single, unique tectonic environment of ophiolite formation, and that ophiolites are diverse in origin, representing fragments of fossil oceanic lithosphere formed in various tectonic settings and in different stages of Wilson cycle evolution of ancient ocean basins. Most ophiolites are heterogeneous in lithological make-up, internal architecture and alteration history, indicating that their formation involved complex and multiple phases of magmatism, metamorphism and tectonism. Precise radiometric, isotopic and biostratigraphic age dating is needed to better constrain the timing of different evolutionary phases in ophiolite generation.
7
Some ophiolites contain peridotites that may represent exhumed subcontinental lithospheric mantle. It is particularly interesting that this appears to be the case for those ophiolitic assemblages in the Alps and Apennines, where the ophiolite concept was born and first developed through keen observations by influential researchers such as Alexandre Brogniart (1740-1847) and Gustav Steinmann (1856-1929). The existence of these subcontinental lithospheric mantle peridotites suggests that some ophiolites may record the initial stages of rift-drift evolution of small ocean basins in Earth history. Detailed petrological studies of some of the peridotite massifs (i.e. Miintener & Piccardo; Spadea et al.) indicate that pervasive melt migration through these ultramafic rocks resulted in extensive melt-rock reaction, precipitation of plagioclase-enriched peridotites and generation of gabbroic intrusions during the early stages of oceanic lithosphere formation. Late-stage and off-axis(?) magmatism that produced large plagiogranite-trondhjemite intrusions into the pre-existing oceanic crust was responsible for extensive hdyrothermal alteration and mineralization in some ophiolites (Semail, Oman, Stakes & Taylor; Bay of Islands, Newfoundland, Giguere et al.). These intrusive bodies provided the local heat source that set up convective circulation of high-temperature fluids reacting with the host rocks and precipitating in due course epidosites and economic massive sulphide deposits. These spatial and temporal links between late plagiogranite intrusions and alteration-mineralization indicate that magmatism in oceanic crust generation is commonly episodic and multi-stage. Mantle dynamics and heterogeneity at regional and global scales appear to have played a critical role in the evolution of small ocean basins (mostly back-arc and/or marginal basins) and their lithosphere. Collision-induced mantle extrusion and flow strongly affected arc-trench rollback mechanisms, melt flow patterns and thermal state in subduction environments that collectively controlled ophiolite-forming processes (Dilek & Flower; Flower & Dilek). Some ophiolites and related tectonic units (i.e. rift assemblages as precursors to ophiolite generation) display geochemical evidence for mantle source(s), which were contaminated by previous subduction events in the region (e.g. Saccani et al.). These observations and interpretations from ophiolites, coupled with isotopic signatures of oceanic basalts, suggest that the mantle is heterogeneous at all scales mainly as a result of subduction of sediments, hydrothermal alteration of oceanic crust and melting-induced differentiation. Emplacement of ophiolites in collisional orogenic belts involves underplating of less-dense
8
Y. DILEK & P. T. ROBINSON
crustal material beneath displaced oceanic lithosphere in subduction zone environments. The arrival of, and attempted partial subduction of, passive continental margins and/or island arc complexes at oceanic trenches provides the necessary physical conditions for this type of ophiolite emplacement. In accretionary-type orogenic belts (such as in Japan, Far East Asia and late Mesozoic-Cenozoic western North American Cordillera), continued consumption of ocean floor at active continental margins facilitates progressive ophiolite emplacement through tectonic incorporation of stranded slabs of oceanic crust, abyssal peridotites and seamounts into the subduction-accretion complexes. These kinds of ophiolites (defined as 'Cordilleran' by Wakabayashi & Dilek) are commonly spatially associated with blueschist-bearing tectonostratigraphic units and subduction melanges. 'Ophiolite science' is a dynamic, evolving and interdisciplinary enterprise that is at its best
through international collaboration. Future international ophiolite studies, focusing on: (1) careful and systematic documentation of primary (seafloor spreading and/or igneous accretion stage) and secondary (emplacement and post-emplacement) structures within different ophiolitic subunits and of contact relations between them; (2) precise and systematic radiometric and isotopic dating of igneous and metamorphic rocks in ophiolites, and biostratigraphic dating of overlying sedimentary cover and underlying melange units; (3) isotopic analysis of ophiolite peridotites to delineate the mantle composition and signatures of their melt source, and mantle domains; and (4) combined geochemical, petrological and structural studies of ophiolites and associated tectonic units to differentiate tectonic settings of their origin and evolution, will help us better understand the Earth history and the processes involved in its evolution through time.
Ophiolite pulses, mantle plumes and orogeny YILDIRIM DILEK Department of Geology, Miami University, Oxford, OH 45056, USA (e-mail:
[email protected]) Abstract: Ophiolites show a wide range of internal structure, pseudostratigraphy and chemical fingerprints suggesting various tectonic settings of their origin. In general, they are characterized as mafic-ultramafic assemblages and associated sedimentary and metamorphic rock units that formed during different stages of the Wilson cycle evolution of ancient oceans, and that were subsequently incorporated into continental margins through collisional and/or accretionary erogenic events. Distributions of ophiolites with certain age groups in different erogenic belts define distinct ophiolite pulses, times of enhanced ophiolite genesis and emplacement, in Earth history. These pulses coincide with the timing of major collisional events during the assembly of supercontinents (i.e. Rodinia, Gondwana and Pangaea), dismantling of these supercontinents, and increased mantle plume activities that formed widespread large igneous provinces (LIPs). Suprasubduction zone ophiolites in orogenic belts signify oceanic crust generation in subduction rollback cycles during the closing stages of basins prior to terminal continental collisions. Both collision-driven assembly of supercontinents and deep penetration of subducted slabs into the lower mantle may produce plumes that in turn facilitate continental rifting, sea-floor spreading and oceanic plateau generation, all of which seem to have contributed to ophiolite genesis. Accelerated LIP formation and seafloor spreading that are associated with superplume events are likely to have caused widespread collisions and tectonic accretion of ophiolites at global scales. Together, these spatial and temporal relations suggest close links between ophiolite pulses, mantle plumes and orogenic events in Earth history.
The traditional definition of ophiolites as on-land fragments of fossil oceanic lithosphere developed at palaeo-spreading centres (Gass 1990) has played an important role on the formulation and advancement of the plate tectonic theory (Coleman 1977, and references therein), and ophiolites have been used extensively to make palinspastic reconstructions of ancient ocean basins and mountain belts (i.e. Dewey et al 1973; Dercourt et al 1986; Lemoine et al. 1986). Exposures of ophiolite complexes along curvilinear fault zones in orogenic belts have been interpreted to represent suture zones, where plate collisions (commonly involving continents and island arcs) occurred in the past (Burke et al 1977). The 1972 Penrose definition of an ophiolite suite having a layer-cake pseudostratigraphy, complete with a sheeted dyke complex, resulting from sea-floor spreading has been central to the ophiolite studies and palaeogeographical reconstructions (Anonymous 1972). Although this ophiolite-oceanic crust analogy and the mid-ocean ridge origin of ophiolites were challenged early on, mainly by geochemists (e.g. Miyashiro 1973), it has been assumed that, in general, ophiolites represent the beginning stages (rift-drift and seafloor spreading) of Wilson cycles.
This traditional view of ophiolites was modified in the mid- to late 1970s, when researchers recovered ophiolitic rocks from the Lau and Mariana back-arc basins, the inner trench walls of the Yap and Mariana trenches, and the Mariana forearc (Hawkins 1977). A new paradigm for ophiolite genesis emerged in the early 1980s, asserting that most ophiolites had developed in suprasubduction zone (SSZ) environments at convergent plate boundaries (Pearce et al. 1984). The widely accepted association of ophiolite genesis with subduction zone settings has shifted the temporal position of ophiolites within Wilson cycles from the beginning to the closing stages. This inference suggests that many ophiolites were produced in the closing stages of ocean basins prior to continental collisions. Dilek & Flower (2003) and Flower & Dilek (2003) have shown, based on actualistic models from the Western Pacific and Eastern Asia and their Tethyan examples, that mantle flow and slab rollback may have played a major role in the formation of SSZ ophiolites in the Alpine-Himalayan orogenic system during the final stages of the evolution of Tethyan basins. Slab rollback is driven by trenchward mantle flow and slab buoyancy forces and
From: DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 9-19. 0305-8719/037$ 15 © The Geological Society of London 2003.
10
Y. DILEK
results in lithospheric-scale extension and associated magmatism in the upper plate that collectively play a major role in the formation of proto-arc, arc and back-arc 'oceanic crust'. The Penrose definition of an ophiolite suite does not prescribe a specific tectonic setting of its genesis and states that 'the use of the term should be independent of its supposed origin' (Anonymous 1972). Extensive research by the international scientific community during the last 30 years has shown that individual ophiolites differ significantly in terms of their structural architecture, chemical fingerprints and evolutionary paths, indicating different tectonic environments of origin, even within the same orogenic belt (Nicolas 1989; Dilek et al 2000). Ophiolites in the Alpine-Himalayan orogenic belt, for example, range from relics of intracontinental rift basins and embryonic normal oceanic crust with mid-ocean ridge basalt (MORE) affinity (Ligurian-type) to protoarc-forearc-back-arc assemblages with SSZ affinities (Mediterranean-type) (Dilek 2003). The peri-Caribbean ophiolites include tectonically emplaced fragments of oceanic crust, which in part represent a large igneous province (LIP), whereas some of the Pacific Rim ophiolites may have had protracted and polygenetic igneous histories that involved the evolution of ensimatic arc terranes through multiple episodes of magmatism, rifting and tectonic accretion, as documented from the Mesozoic ophiolites in the Philippines and the Sierra Nevada foothills (Sierran-type) (Dilek 2003). Ophiolites situated within the accretionary complexes of ancient active margins are commonly associated with melanges and high-pressure metamorphic rocks and may represent fragments of abyssal peridotites and ocean island basalts (OIB), seamounts, island arcs and/or mid-ocean ridge crust scraped off from downgoing plates. These kinds of ophiolites (Franciscan-type) in ancient accretionary complexes do not show genetic and temporal relations (i.e. no melt-residua relationship or chronostratigraphic order) and commonly display diverse chemical affinities and metamorphic grades. Thus, ophiolites are highly diverse in terms of their tectonic origin and emplacement mechanisms (Wakabayashi & Dilek 2003). Despite significant differences in their origin and emplacement mechanisms, ophiolites around the world appear to show distinct patterns of distribution through time and space, suggesting that their evolution may have been linked to some first-order global tectonic events. In this paper I present an overview of the spatial and temporal occurrences of major ophiolite belts in the Earth's history and discuss the possible causes of this 'ophiolite pattern' in a global tectonic framework.
However, examples and the discussion in this paper are constrained to the Neoproterozoic and Phanerozoic occurrences of ophiolites because our knowledge of the Archaean ophiolites, specifically their igneous and emplacement ages and tectonic environment of origin, is still limited.
Distribution of ophiolite belts in space and time Pan-African and Brasiliano ophiolites Figure 1 shows the distribution of major ophiolites with certain age groups in semi-continuous, curvilinear belts around the world. The Late Proterozoic (2 vs. FeO* (in wt%) for eruptive and intrusive lithologies sampled from typical Izu-BoninMariana intra-oceanic forearcs and the Troodos (Cyprus) and Semail (Oman) ophiolites. (a) Bonin Islands and Sumisu Rift (Izu-Bonin-Mariana system) (Pearce et al. 1992a; Taylor et al. 1994). (b) Mariana Islands (Reagan & Meijer 1984; Stern et al. 1989) and Mariana Trough (Gribble et al. 1998). (c) Troodos ophiolite, Cyprus (Malpas et al. 1984; Flower & Levine 1987; Gibson et al. 1987; Rogers et al. 1989; Taylor et al. 1992; Bednarz & Schmincke 1994; Portnyagin et al. 1996, 1997). (d) Semail ophiolite, Oman (Umino et al. 1990; Lachize et al. 1996; Pezard et al. 2000; Ishikawa et al. 2002).
MANTLE FLOW MODEL FOR TETHYAN OPHIOLITES
23
Fig. 2. MORB-normalized incompatible element distributions for eruptive lithologies sampled from typical forearcs and ophiolites. (a) Troodos ophiolite, Cyprus: Upper Pillow Lavas (series 1 and 3), and Lower Pillow Lavas (Flower & Levine 1987; Gibson et al. 1987; Rogers et al. 1989; Bednarz & Schmincke 1994; Portnyagin et al 1996, 1997). (b) Semail ophiolite, Oman (Umino et al 1990; Lachize et al 1996; Pezard et al 2000; Ishikawa et al 2002). Alley Series volcanic rocks (calc-alkaline and boninitic), Geotimes Series volcanic rocks (MORB-type volcanic rocks), (c) Izu-Bonin arc-forearc, and backarc Sumisu Rift (Izu-Bonin-Mariana system) (Pearce et al 1992a; Taylor et al 1994). (d) Mariana arc-forearc (Reagan & Meijer 1984; Stern et al 1989) and backarc Mariana Trough (Gribble et al 1998). such processes with distal plate kinematic changes suggests a fundamental connection between ophiolite genesis and global-scale plate tectonics (Flower et al. 2001; Flower 2003). The causes of arctrench rollback may therefore be crucial to our understanding of ophiolites and their geodynamic significance in Earth history. Here, following the 'Tectonic Facies' approach of Hsu (1994) and Hsu (1997), we present an 'actualistic' model for ophiolites based on processes of forearc evolution in western Pacific and Mediterranean marginal basins. In a companion paper (Dilek & Flower this volume) the model is adopted as a template for interpreting three well-studied Tethyan ophiolites.
A brief history of Tethys Tethyan orogens mark a succession of continental plate collisions that followed breakup of the Gondwana continent and repeated cycles of ocean basin opening and closure. Although the relevant plate kinematic reconstructions are controversial,
there is a general agreement that the northward drift of Gondwana fragments involved three or more such cycles of opening (e.g. Dercourt et al. 1986; Audley-Charles & Harris 1990; Ustaomer & Robertson 1993; Metcalfe 1996; Stampfli & Borel 2002). These cycles were commenced with diachronous 'unzipping' of the northern margin of Gondwana, and produced Tethyan basins evolving as triangular inlets that propagated westward from the proto-Pacific Ocean. According to the majority of views, Palaeo-Tethys was initiated in the Late Devonian with the separation of continental blocks that later amalgamated as the North China, South China, Iran, Kazakhstan, Indochina, Qaidam, Tarim, and Hainan blocks (Audley-Charles & Harris 1990; Metcalfe et al, 1999). Meso-Tethys probably began to open in the Early Permian with detachment of the Cimmerian and other microcontinents, and Neo-Tethys began opening between the Late Triassic and Late Jurassic with the separation of what later became the Pelagonia, Tauride-Anatolide, Lhasa, West Burma, and Woyla blocks (Dilek et al 1999; Metcalfe et a/.,1999).
24
M. F. J. FLOWER & Y. DILEK
Fig. 3. Plate boundary evolution and arc—trench rollback in active Tethyan domains, (a) The circum-Mediterranean region (after Wortel & Spakman 2000). Arrows indicate directions of probable slab tearing beneath the ApennineCalabrian, Hellenic, and Carpathian arcs. Adr, Adriatic Sea; Aeg, Aegean Sea; Alb, Alboran Sea; Ap, Apennines; AP, Algero-Provencal Basin; Cal, Calabria; Car, Carpathians; Co, Corsica; Cr, Crete; Cyp, Cyprus; Hel, Hellenic arctrench; Ion, Ionian Sea; Lev, Levantine Basin; Mag, Maghrebides (from the Rif to Sicily); NAF, North Anatolian Fault; Sa, Sardinia; Si, Sicily; Tyr, Tyrrhenian Sea. Barbs indicate subduction or thrusting vergence, black suggesting a continuous slab, and white, possible post-collision breaker!, (b) Western Pacific region (after Flower et al. 1998, 2001). Arrows indicate directions of probable slab tearing beneath the Himalayas, Sunda-Banda arcs, and Northern Luzon-Taiwan. TS, Tien Shan; ATF, Altyn Tagh Fault; And-Nic, Andaman-Nicobar Islands; 1C, Indochina; Ma, Malay Peninsula; Sm, Sumatra; Bo, Borneo; Ja, Java; Su, Sunda; Ba, Banda; Sw, Sulawesi; Phil, Philippines; Tw, Taiwan; Izu-Bon, Izu-Bonin Islands; Ma, Mariana Islands; Wma, West Mariana arc; PKR, Palau-Kyushu ridge; Ru, Ryukyu Islands; SCS, South China Sea; SS, Sulu Sea; CS, Celebes Sea; MS, Molucca Sea; WPSB, West Philippine Sea Basin; BS, Banda Sea; SB, Shikoku Basin; SR, Sumisu Rift; PVB, Parece Vela Basin.
Palaeo-Tethys The closure of Palaeo-Tethys and the corresponding inception of Meso-Tethys were marked by the accretion of Kunlun, Qaidam and Ala Shan Ter-
ranes to Kazakhstan-Siberia in the Early Permian, followed in the Late Permian to Early Triassic by suturing of Sibumasu and Qiangtang to Cathaysialand as Palaeo-Tethys was finally consumed by subduction (Metcalfe et al., 1999). Meso-Tethys
MANTLE FLOW MODEL FOR TETHYAN OPHIOLITES closure between the Late Triassic and Late Jurassic was accompanied in the east by diachronous accretion of the Lhasa, West Burma, and Woyla micro-continents (Metcalfe et al., 1999), and in the west, Cimmeria, Iran, Pelagonia, and others (Dercourt et al. 1986; Ustaomer & Robertson 1993; Stampfli & Borel 2002) to Eurasia. Finally, as the African, Arabian, and Indian plates collided with Eurasia, and Australia collided with newly accreted Sunda, remnants of the Neo- and MesoTethyan lithosphere were being progressively consumed by subduction (Dercourt et al. 1986; Audley-Charles & Harris 1990; Metcalfe et al., 1999; Stampfli & Borel 2002). Although successive Tethyan basins were more or less separated by micro-continents throughout much of the Triassic and Jurassic, they may have remained connected at their western extremities, between the Mediterranean and Caucasus. This interpretation is supported by evidence suggesting that Mid-Jurassic remnant basins were being consumed by subduction as collisions between retreating arc-forearc complexes and continents prevented further extension (Stampfli & Borel 2002). For example, Paleocene closure of the Liguria-Piedmont basin was coeval with the Betic-Rif, western-northern Alpine, and Carpathian orogenies and, as younger basins collapsed, it was followed by the NeogenePleistocene Apennine, Maghrebe, Dinaride, and Hellenide orogenies (Faccenna et al. 1997; Jolivet et al. 1999; Jolivet & Faccenna 2000; Stampfli & Borel 2002).
Neo-Tethys The closure of Neo-Tethys coincided with opening of the North Atlantic Ocean that began at c. 180 Ma and was followed by the separation of East and West Gondwana at c. 158 Ma. By c. 130 Ma East Gondwana (Africa-India-SeychellesMadagascar-Australia-Antarctica- South America) had also begun to sunder as opening of the South Atlantic commenced at c. 110-100 Ma and the North Atlantic opening continued. By the Mid-Cretaceous, a block comprising Africa and India-Seychelles-Madagascar began to detach from Australia-Antarctica, followed shortly by the separation of Australia and initiation of seafloor spreading at the Southeast Indian Ridge (Metcalfe 1996). At c. 98 Ma, the India-Seychelles block separated from Madagascar and by the Late Cretaceous, along with Africa-Arabia and Australia, was moving rapidly northwards towards accreting Eurasia. Finally, following separation from the Seychelles at c. 65 Ma (Gnos et al. 1997), India collided with Eurasia between c. 50 and 45 Ma (Lee & Lawver 1994). After
25
separating at c. 40 Ma, Arabia and Africa collided with accreting Eurasia at c. 30 Ma and 25 Ma, respectively (Dewey et al. 1989; Jolivet & Faccenna 2000). The record of Gondwana disaggregation and (partial) reassembly offers a potential rationale for the timing and location of 'spontaneous' subduction nucleation and, in turn, the processes giving rise to ophiolites. However, the causes of subduction nucleation remain unclear. Are such events determined by global-scale plate kinematics (as suggested by Gnos et al. 1997) or do they reflect viscous mantle instabilities caused by density and thermal heterogeneities (e.g. Toth & Gurnis 1998; Faccenna et al. 1997)? These questions bear, in turn, on whether ophiolites represent a global-scale plate tectonic phenomenon (e.g. determining where they are initiated) or local phenomena related to an imminent plate collision (determining both where they are initiated and where they are emplaced). Although this latter question is beyond the scope of the present paper, we will attempt to provide a basis for its future consideration.
Towards an actualistic model Today, Tethyan tectonic and magmatic activity is dominated by effects of the African, Arabian, Indian, and Australian collisions concomitant with continued basin opening in parts of the Mediterranean Sea and western Pacific (e.g. Fig. 3). On a global scale, subduction zones may remain static for lengthy periods and evolve as simple linear orogens. Others, notably in the regions discussed here, are observed to migrate oceanward at rates exceeding 100m ma"1, often developing into spectacular oroclines (Fig. 3). Such rapid 'arctrench rollback' processes are expected to continue indefinitely unless they are terminated by collisions of their retreating arc-forearc complexes with mid-ocean ridges, continental plates, or other migrating subduction systems.
Forearc complexes as lithological 'high-tide marks' In the Mediterranean, rollback cycles have mostly been interrupted by forearc collisions and the ensuing consumption (or 'collapse') by subduction of short-lived backarc basins (Dalziel 1989; Clift & Dixon 1998; Jolivet & Faccenna 2000; Robertson 2000). In contrast, arc-trench rollback in the western Pacific has been relatively unconstrained with continuing eastward propagation of forearcs, free from the 'jaws' of an impending plate collision (Karig 1971; Hussong et al. 1981; Karig et al. 1986; Jolivet et al. 1991b; Tamaki & Honza
26
M. F. J. FLOWER & Y. DILEK
1991). On the other hand, these regions show strong similarities, arc-trench rollback cycles in both cases being initiated by splitting of nascent volcanic 'proto-arcs' into active and remnant segments, the associated refractory magmas indicating unusual thermal conditions (Pe-Piper & Piper 1989, 1994; Stern & Bloomer 1992; Hawkins & Castillo 1998; Insergueix-Filippi et al. 1998, 2000). The net effect of this type of process, sometimes compounded by additional arc splitting events (Hussong et al. 1981; Ishii et al 1995; Fassoulas 1999), is to produce an evolving forearc terrane that potentially includes the igneous and metamorphic products of successive 'proto-arc', arc, and backarc episodes (Bloomer 1983; HickeyVargas 1989; Johnson et al. 1991, 1992; Giaramita et al. 1992; Ishii et al. 1992, 1995; Marlow et al. 1992) along with characteristically high-temperature hydrothermal deposits (Banerjee et al. 2000; Fryer et al. 2000; Gillis & Banerjee 2000; Banerjee & Gillis 2001; Gillis 2002). The presence of allochthonous continental and oceanic lithosphere fragments may represent crustal features prior to the inception of rollback (Johnson et al. 1991; Giaramita et al. 1992; Parkinson et al. 1998; Parkinson & Arculus 1999). As they evolve, therefore, arc-forearc complexes progressively resemble lithological 'high tide marks' (HTMs), increasingly heterogeneous, accreted assemblages of proto-arc, arc, and backarc crust exhibiting significant internal age and structural discrepancies (Flower et al. 1998, 2001; Flower 2003). Subduction nucleation At least two lines of evidence highlight the anomalous thermal character of asthenospheric mantle associated with subduction nucleation events that appear to trigger rollback cycles: the presence of boninite and high-magnesium andesites (HMA) in proto-arcs (Casey & Dewey 1984; Stern et al. 1989; Stern & Bloomer 1992; Clift & Dixon 1998; Wallin & Metcalf 1998) and the inflected P-T-t histories of sub-ophiolitic metamorphic 'soles' interpreted from thermobarometric studies (Wakabayashi & Dilek 1988, 2000; Gjata et al. 1992; Insergueix-Filippi et al. 1998, 2000; Searle & Cox 1999; Bebien et al 2000; Dimo-Lahitte et al. 2001). Boninite-HMA volcanism is relatively rare in modern settings. A notable exception, however, is the Hunter Ridge, between the southernmost New Hebrides and Fiji islands, where boninite magmatism marks a locus of incipient subduction along an active transform fracture zone (Falloon & Crawford 1991; Danyushevsky et al., 1995; Crawford et al. 1997). Here, the eastern extremity of the New Hebrides subduction system appears to
be nucleating along a transform fault that is linked to the southward-propagating North Fiji spreading centre (Monzier et al. 1993a, 1993b, 1997). Examples of coeval, if now-extinct, boninite and HMA volcanism also characterize forearc-remnant arc pairs in the Mediterranean and western Pacific regions, confirming the unusual thermal character of subduction nucleation events. The best-documented example of subduction nucleation occurs in the Izu-Bonin-Mariana (IBM) 'subduction factory' (MARGINS) where Mid-Eocene subduction nucleation was followed by rapid backarc basin opening and rollback of the IBM arc-forearc terrane (Karig 1971; Hussong et al. 1981; Stern & Bloomer 1992; Bloomer et al. 1995; Hawkins & Castillo 1998). The locus of subduction inception is marked by the Palau-Kyushu ridge, a boninite-bearing 'proto-arc' remnant that dissects the Philippine Sea Plate. Subduction probably began at c. 50 Ma, shortly before the India-Asia collision (c. 45-40 Ma) and reorientation of Pacific Plate motion (c. 43 Ma) (Bloomer et al. 1995; Hawkins & Castillo 1998) (Fig. 4) with underthrusting along a transform fracture zone of the West Philippine Sea Basin Plate, either by the Pacific Plate or a younger, hypothetical, North New Guinea Plate (Stern & Bloomer 1992). As subduction continued, splitting of the Palau-Kyushu proto-arc led to opening of the Parece Vela Basin (c. 40-25 Ma), with concomitant rollback of the newly active West Mariana arc. Splitting of the latter gave way to further rollback of the active Mariana arc with opening of the still-active Mariana Trough (Karig 1971; Hussong et al. 1981). Mariana Trough opening continues today accompanied by northward 'unzipping' of the West Mariana-Mariana arc, offering a diachronous, actualistic model for protoophiolite genesis. According to such a model, the present-day Mariana forearc includes the accumulated products of West Philippine Sea sea-floor spreading, 'proto-arc' boninitic and calc-alkaline activity, and subsequent (West Mariana, Mariana) arc volcanism. Boninites dredged from the Palau-Kyushu ridge match those in lower horizons of the Mariana forearc, the latter feature conforming in these and other respects to an in situ 'protoophiolite' (Ishii et al. 1988; Ogawa & Taniguchi 1989). Mid-Miocene HMA volcanic rocks (c. 13 Ma) are likewise preserved in central and southern parts of the Ryukyu forearc and are matched by analogous activity in the FujianTaiwan region (Shinjo 1999), suggesting subduction nucleation, triggered by collision of the Luzon arc with Eurasia, prior to opening of the Okinawa Trough (Shinjo 1999). Other examples include Mid-Miocene (c. 14-15 Ma) HMA vol-
MANTLE FLOW MODEL FOR TETHYAN OPHIOLITES
27
Fig. 4. Arc-trench rollback model, from Bloomer et al. (1995), based on evolution of the Izu-Bonin-Mariana forearc and eastern part of the Philippine Sea Plate (following Karig 1971; Hussong et al. 1981; Stern & Bloomer 1992). (a) 50-40 Ma. Subduction nucleation beneath the West Philippine Sea Basin Plate either by the Pacific Plate or (hypothetical) young North New Guinea Plate along a transform fracture zone in proto- West Philippine Sea Basin spreading centre. Boninite melt genesis accompanies early forearc development with inception of the calc-alkaline arc forming the Palau-Kyushu ridge, (b) 40-25 Ma. Continued subduction with slab steepening of the Pacific Plate, splitting of the Palau-Kyushu arc, Parece Vela Basin opening, and rollback of the active West Mariana arc. (c) 250 Ma. Continued subduction, Mariana Trough opening by splitting of the West Mariana arc, and rollback of the active Mariana arc. (d) 0 Ma. Subduction beneath the modern Mariana arc-trench system with continued Mariana Trough opening by 'unzipping' of the West Mariana-Mariana arc to the north (Iwo Jima). PKR, Palau-Kyushu ridge; PVB, Parece Vela Basin; WMA, West Mariana arc; MT, Mariana Trough; MA, Mariana arc; IBM, Izu-Bonin-Mariana forearc; PAC, Pacific Plate; WPSB, West Philippine Sea Basin; WPAC, Western Pacific; NNG, 'North New Guinea' Plate.
canism recorded from islands between the Hellenic forearc and Cycladean metamorphic core complexes (Smith & Spray 1984; Pe-Piper 1994; Forster & Lister 1999; Migiros et al. 2000) that corresponds to HMA-bearing ophiolite fragments in the Hellenic forearc (Fortuin et al. 1997; Clift 1998), and Oligo-Miocene HMA (c. 18 Ma) in Sardinia and Corsica, and in Calabrian forearc ophiolite fragments (Beccaluva 1982; Delaloye et al. 1984; Compagnoni et al. 1989; Beccaluva et al. 1994), which pre-date opening of the Tyrrhenian Sea (Morra et al. 1991 \ Padoa 1999). Experimental studies of boninite melts suggest that they result from the combined effects of mantle decompression and slab-derived H2O-rich fluid (Umino & Kushiro 1989; van der Laan
et al. 1989; Falloon & Danyushevsky 2000), interpreted by some as an indication of ocean ridge spreading conditions (Gjata et al. 1992; Stern & Bloomer 1992; Peacock 1994; Peacock et al. 1995). However, comparison of experimental data for basalts, boninites, and variably fertile peridotites suggests that three additional conditions are required for boninites to form: (1) anomalous asthenospheric potential temperatures (Tp > c. 1400°C); (2) significant lithospheric extension (stretching factors, /?, > c. 3) (McKenzie & Sickle 1988; Latin & White 1990); (3) refractory (previously melt-depleted) peridotite sources (e.g. van der Laan et al. 1989; Hirose & Kawamoto 1995; Hirose 1997; Falloon & Danyushevsky 2000) (Fig. 5). In other words, boni-
28
M. F. J. FLOWER & Y. DILEK
Fig. 5. Partial melting models for (a) fertile and (b) refractory peridotite in the presence of H2O and CO2 between pressures of 0 and 5 GPa, interpolated from published experimental data. Geotherms (bold dashed curves) are calculated for asthenospheric potential temperatures (7J,) of 1280 °C and 1440 °C, respectively, for stretching factors (/?) of c. >3 and >5, assuming 'pure shear' lithosphere extension (McKenzie & Sickle 1988; Latin & White 1990). The solidus curve (bold line) for 'fertile' peridotite (pyrolite)- C-H-O is taken from Wyllie (1990) for the condition X = CO2/(CO2 + H2O) = 0.8 along the bold grey curve marking equilibrium between amphibole, carbonate, peridotite, and vapor; based on sources given by Wyllie et al., (1990). The anhydrous solidus for fertile peridotite (HK-66) is from Hirose & Kushiro (1993), with a hypothetical H2O-undersaturated (1 wt% H2O) solidus interpolated from hydrous experiments on refractory Iherzolite (KLB-1) (Hirose 1997). Anhydrous, f^O-saturated, and H2Oundersaturated (1 wt% H2O) solidi for the refractory Iherzolite (KLB-1) are from Hirose & Kawamoto (1995). Predicted melt segregation conditions agree with those determined experimentally for primitive MORB and high-Ca boninite (HCB) (e.g. van der Laan et al. 1989; Falloon & Danyushevsky 2000). MBL, mechanical boundary layer; BON, boninite; LVZ, low-velocity zone; PIC, picrite; NE, nephelenite; BAS, basanite.
nite liquidus temperatures appear to exceed those of 'normal' mid-ocean ridge magmas by c. 150200 °C and require a source that is significantly less fertile than that producing normal MORE (van der Laan et al. 1989; Falloon & Danyushevsky 2000).
These observations concur with predictions from 2D numerical models (Gjata et al. 1992; Insergueix-Filippi et al. 1998, 2000) and are also supported by anomalous P-T-t histories recorded from sub-ophiolitic metamorphic soles (Fig. 6). These appear to record patterns of cold thrusting
MANTLE FLOW MODEL FOR TETHYAN OPHIOLITES
29
Dilek et al 1997; Hacker & Gnos 1997; Searle & Cox 1999; Dimo-Lahitte et al 2001), suggesting that metamorphic 'soles' may be best interpreted as MORB-like relics of incipient subduction (e.g. Fig. 6; Wababayashi & Dilek, this volume). Although some ophiolites slightly post-date their high-temperature soles, most are coeval or slightly older, consistent with interpretations that sundered forearc and remnant 'proto-arc' components were single entities prior to splitting and the onset of rollback. Incipient subduction relics are thus preserved a priori in forearc 'proto-ophiolites' (Ishii 1989; Ogawa 1995).
Arc-trench rollback: endogenous vs. exogenous causes? Plate kinematic
Fig. 6. Interpolated P-T-t histories for sub-ophiolitic metamorphic soles and other metamorphic lithologies. (a) Mirdita ophiolite, Albania (Gjata et al. 1992; Robertson & Shallo 2000; Dimo-Lahitte et al 2001). (b) Troodos-Mamonia complexes, Cyprus (Malpas et al. 1992; Bailey et al. 2000). (c) Semail ophiolite, Oman. 1 and 2, Saih Hatat; 3, Saih Hatat-Wadi Tayin; 4, Asimah-Bani Hamid; 5, Hacker & Gnos (1997); 6, As Sitah eclogites -> Hulw blueschist (Searle et al. 1994; Searle & Cox 1999). Pressure-temperature equilibration conditions for ophiolitic blueschist-bearing units from the Shuksan, and Franciscan terranes and Cyclades (Aegean) remnant arc are also shown. Clockwise (CW) and counter-clockwise (CCW) P-T-t paths appear to distinguish 'high temperature' from 'high pressure' sole types.
to at least c. 60 km depth with temperatures rising to well over 700 °C, a stage of near-isobaric cooling, and (eventual) exhumation (Wakabayashi & Dilek 1988, 2000; Hacker 1991; Gjata et al 1992; Beccaluva et al 1994; Searle et al 1994; Wakabayashi & Unruh 1995; Dilek & Whitney 1997;
effects
Although reliable age data for ophiolites are sometimes difficult to acquire and have proved contradictory, a clearer spatial-temporal picture is emerging of relations between active Tethyan forearc complexes and their respective remnant proto-arcs. For example, boninite-bearing 'protoarcs' (c. 49-45 Ma), preserved in the IBM forearc and its remnants (Hawkins & Castillo 1998) and the Zambales (Philippines) ophiolite (Encarnacion 1997), are nearly coeval with the initiation of Celebes Sea opening (c. 48 Ma) (Beiersdorf et al. 1997), whereas they slightly predate the 'hard' collision of India with Eurasia (c. 45-40 Ma) and the corresponding change in Pacific Plate motion from NNW to NW. The Palawan and Mindoro ophiolites (c. 34 Ma) also appear to mark a change from sea-floor spreading to convergence, coeval with the inception of Alao Shan-Red River leftlateral shearing (c. 33 Ma) and sea-floor spreading in the South China Sea (c. 32 Ma). Both were terminated by a collision between the Sulu Ridge arc remnant and the West Philippine arc (c. 17 Ma) (Rangin et al. 1995), which triggered the initiation of Sulu Sea opening (c. 17 Ma) (Rangin et al. 1995; Yumul et al. 1998, 2001). New subduction that produced the Ryukyu proto-arc probably occurred between c. 21 and 18 Ma, preceding collisions of Taiwan (c. 15-12 Ma) (Chung et al. 1994), and other micro-continents with the Luzon arc, at c. 17-16 Ma (Rangin et al. 1985; Pubellier & Cobbold 1996; Pubellier et al. 1996) and 12-6 Ma (Sibuet & Hsu 1997). In western Tethys, the Late Cretaceous Semail and Troodos ophiolites (c. 98-75 Ma) (Urquhart & Banner 1994; Hacker & Mosenfelder 1996; Hacker et al. 1996) and those exposed in the Zagros and Tauride belts (c. 98-75 Ma) (Dilek et al. 1999; Parlak & Delaloye 1999; Ghazi & Hassanipak 2000; Parlak et al. 2000; Babaie et al.
30
M. F. J. FLOWER & Y. DILEK
2001a, 2001b; Ghasemi et al. 2002) and Hellenic arc (c. 98-75 Ma) (Langosch et al. 1999, 2000), were formed shortly before collisions of the Iranian (c. 75-70 Ma), Apulian (c. 70-65 Ma), Pelagonian (c. 72 Ma), and other micro-continents (Robertson & Shallo 2000; Stampfli et al. 2001; Stampfli & Borel 2002) with accreting Eurasia. However, younger ophiolitic remnants (c. 12— 10 Ma) in Crete and the Aegean Cyclades are near-coeval with both Aegean continental collapse (Lee et al. 1990) and the initiation of westward escape by Anatolia (c. 13-5 Ma) (Le Pichon 1982; Le Pichon et al. 1995). Thus, although rollback of Hellenic subduction may have continued since the Paleocene, as inferred from seismic tomography (Spakman et al. 1992; Spakman & Bijwaard 1998), it was probably interrupted by the effects of regional microplate collisions in the Late Cretaceous and Pliocene (Le Pichon et al. 1995).
'Slab pull' and extrusion tectonics Encarnacion et al. (2001) proposed a linkage between genesis of the South Palawan (Philippines) ophiolite and the coupled inception of South China Sea spreading and left-lateral motion on the Red River fault (Lee et al. 2000; Wang et al. 2000). According to the classic extrusion tectonics model (Tapponnier et al. 1982, 1986), marginal basin opening and arc-trench rollback are linked responses to collision-induced lithosphere 'escape', as interpreted for opening of the Aegean and South China Sea basins and seaward escape, respectively, of Anatolia and Indochina (Briais et al. 1993; Le Pichon et al. 1995; Lundgren et al. 1996). However, as a general explanation of marginal basin opening, extrusion tectonics seems unable, at least in these cases, to account for the observation that basin opening commenced prior to, and proceeded at a faster rate than, the escape of their respective conjugate blocks (Chung et al. 1997; Lee et al. 2000; Wang et al. 2000; Le Pichon et al. 2002). Given the apparent linkage of continental escape and marginal basin opening, we need to consider the extent to which backarc basin opening is intrinsic to subduction and what, if any, exogenous factors play a role. As already noted, arc-trench rollback has usually been ascribed to slab buoyancy forces, assuming these to exceed those of the convecting asthenosphere (Isacks & Molnar 1971; Uyeda & Kanamori 1979). Most studies of the mechanical interactions between subducting and overriding plates suggest that where backarc basin opening is passive, the seismicity associated with subduction shows 'down-dip compression' (e.g. Fig. 7a). If horizon-
Fig. 7. Hypothetical effects of 'slab pull' and 'mantle extrusion', (a) Slab force model, showing interaction between subducting slab and overriding plate (from Seno & Yamanaka 1998). Where backarc basin opening is a passive response to slab compression, slabs show 'downdip compression'. However, down-dip tension accompanies backarc opening in some cases (e.g. the Mariana, Kyushu and Hellenic arcs) (Seno & Yamanaka 1998), suggesting slab retreat is driven by trenchward mantle flow. The dip of the subducting plate is q, AB is the trench axis and CE the aseismic front. PS' is the effective ridge push, FS the slab pull, and PC the collision force. FS' is the horizontal component of the traction on CD', t is the shear stress at the thrust zone (see Seno & Yamanaka 1998). Where rollback is slow, FS' is negative, in which case Fc could be positive or negative depending on whether Ps' is smaller or larger than FS'. If FS' is negative and opposed to FP', FC is likely to be very small, resulting in back-arc extension, (b) Two-dimensional mantle flow model (after van Keken et al. 2002); the slab is assumed to be subducting at constant speed. Two flow components are shown: (1) 'endogenous' (slab-induced) flow; (2) exogeneous (e.g. collision-induced) flow.
tal traction is negative, backarc extension will be relatively modest (Seno & Yamanaka 1998). In some cases (e.g. the Mariana, Kyushu and Hellenic arcs), however, down-dip tension accompanies backarc opening, suggesting that slab retreat is more rapid, and driven by trenchward mantle flow (Seno & Yamanaka 1998). These relationships are shown in Figure 7b, indicating the possibility of two potential flow field components: 'endogenous' (slab-induced) and 'exogeneous' (mantle-driven).
MANTLE FLOW MODEL FOR TETHYAN OPHIOLITES
Collision-induced mantle extrusion Thus although slab pull can explain the dynamics of many subduction zones, it is probably unable to account for those cases, such as the Hellenic and Mariana arc systems (Seno & Yamanaka 1998), where slab steepening, arc bending, and accelerated basin opening coincide (McCabe & Uyeda 1983; Hynes & Mott 1985; Dvorkin et al 1993; Bevis et al. 1995). Slab pull is probably subsidiary to exogenous mantle flow (Seno & Yamanaka 1998), a mechanism that can potentially reconcile marginal basin and continental escape kinematics with the accretionary build-up of forearc complexes. During early stages of typical Wilson Cycles, plate motions are probably driven by a combination of mantle upwelling ('ridge push'), downwelling ('slab pull'), and the effects of lateral impingement (on continental cratonic keels) (Forsyth & Uyeda 1975; Russo & Silver 1996). For example, the correspondence of arc-trench rollback in the Caribbean and South Scotia Sea regions (Dalziel et al. 2001) to accelerated westward motion of South America (Russo et al. 1993; Russo & Silver 1996) may be a far-field mantle flow response to the 30-25 Ma Africa-Eurasia collision (Silver & Russo 1996). In response to the latter, westward migration of the Mid-Atlantic Ridge and the corresponding eastward offset of major Mid-Atlantic hotspots (Iceland, St. Helena, Tristan da Cunha, the Azores, and Bouvet) would have been immediately translated to South American plate motion. At later stages, asthenosphere flow is likely to be displaced by thick continental plates as they approach each other and, eventually, collide. For example, Tamaki (1995) suggested that lateral displacement of asthenosphere prior to and following the India-Asia collision led to rapid eastward propagation of Western Pacific marginal basins. Such a process, broadly consistent with the timing and kinematics of basin opening (Hall 2002), also offers a plausible explanation for widespread intra-plate volcanism that characterizes much of east and SE Asia, contamination of the upper mantle beneath eastern Eurasia and western Pacific basins (attributed to delamination of the Sino-Korean craton), and the sharp boundary separating DUPAL-like (contaminated) from N-MORB Pacific mantle. Accordingly, if HTM ('high-tide mark') forearc assemblages (the accreted igneous and metamorphic products of arc-trench rollback) are a valid analogue for ophiolite, such features can be taken to represent distal mantle flow boundaries (Flower et al. 1998, 2001; Flower 2003). This is not to say that arc-trench rollback is exclusively triggered by lateral mantle flow produced by plate
31
collisions. As already noted, mantle flow fields giving rise to arc-trench rollback and protoophiolite genesis may be contingent on other modes of differential plate motion. However, the notion of collision-induced mantle extrusion as the driver of Tethyan ophiolite genesis appears able to reconcile coeval continental escape (Armijo et al. 1989; Jolivet et al. 1991a, 1991b), postcollision lithosphere stretching (England & Molnar 1997a, 1997b; Ren et al. 2002), and arc-trench rollback (Hussong et al. 1981; Tamaki & Honza 1991) (Fig. 7), along with regional mantle attributes cited by Flower et al. (1998, 2001) and Flower (2003).
From proto-ophiolite to ophiolite: a preliminary Tethyan verdict Studies of metamorphic soles record an unambiguous pattern of cold thrusting to at least c. 60 km depth, temperatures rising to well over 700 °C, a stage of near-isobaric cooling, and (eventual) exhumation (Wakabayashi & Dilek 1988, 2000, this volume; Hacker 1991; Gjata et al. 1992; Beccaluva et al. 1994; Searle et al. 1994; Wakabayashi & Unruh 1995; Dilek & Whitney 1997; Dilek et al. 1997; Hacker & Gnos 1997; Searle & Cox 1999; Dimo-Lahitte et al. 2001) (e.g. Fig. 5). The only serious alternative to subduction nucleation as a trigger for ophiolite formation is the proposal that ophiolites result from the consumption of recently active spreading centres at preexisting subduction zones (e.g. Hacker & Mosenfelder 1996; Hacker et al. 1996). This rests on the assumption that newly formed ( clinopyroxene, and covariations of forsterite (fo) content in olivine, anorthite (an) content in plagi-
ALPINE-APENNINE PERIDOTITES oclase and Mg-number in clinopyroxene, which are typical of low-pressure crystallization of olivine tholeiites. Clinopyroxenes of primitive cumulates have rather flat heavy REE (HREE) to middle REE (MREE) patterns, at about (910) X Cl, and LREE depletion (CeN/SmN = 0.21-0.29). Calculated liquid compositions from the most primitive samples indicate a clear MORE affinity, in agreement with the Sr and Nd isotope ratios of some ol-gabbros and their clinopyroxenes (Rampone et al 1998; Bill et al 2000). U-Pb ages of zircons from highly differentiated Fe-Ti gabbros exhibit a surprisingly narrow window of crystallization ages from 166 to 160 Ma (Schaltegger et al. 2002), whereas some plagiogranites of the Western Alps and Apennines are distinctly younger (153-148 Ma, Borsi et al. 1996; Costa & Caby 2001). This suggests that regional-scale upwelling and partial melting of a MORB-type asthenospheric source started in the Mid-Jurassic.
Extrusive rocks Basaltic rocks are common in Alpine ophiolites and occur as pillows or massive flows and as discrete dykes intruding deformed gabbros and mantle peridotites (Fig. 2). Petrological and geochemical studies have provided evidence of their overall tholeiitic composition and MORB affinity, ranging from T-MORB to N-MORB (Venturelli et al. 1981; Beccaluva et al. 1984; Ottonello et al. 1984; Rampone et al. 1998; Bill et al. 2000; Desmurs et al. 2002). The most primitive basalts show either moderate LREE fractionation (CeN/SmN = 0.6) or almost flat to slightly LREEenriched REE spectra, and HREE abundances at about 10 X Cl. They have fairly homogeneous Nd isotopic ratios, consistent with their MORB affinity, but variable Sr isotopic ratios (up to 0.7085), which are related to oceanic sea-water alteration (Rampone et al. 1998; Bill et al. 2000; Schaltegger et al. 2002). Geochemical modelling indicates that the most primitive T-MORB and NMORB-type basalts are consistent with melts generated by low to moderate degrees of fractional melting of a MORB-type asthenospheric mantle source (Vannucci et al. 1993a); however, the source of some basalt is enriched in incompatible elements (Desmurs et al. 2002). This compositional variation seems to correlate with the spatial distribution of the mafic rocks within the oceancontinent transition whereby mafic rocks with TMORB signatures occur close to the continental margin whereas N-MORB signatures are predominantly found oceanwards (Desmurs et al. 2002).
73
Observations from the Lanzo and Corsica peridotites Field observations The Lanzo and Monte Maggiore (Corsica) peridotites comprise the mantle section of partially dismembered ophiolites exposed in the Western Alps and the Northern Apennines (Fig. 1). The peridotites are composed mainly of massive plagioclase peridotites, and minor spinel peridotites, harzburgites and dunites. At Lanzo, plagioclase peridotites were thought to be formed either as the residuum of low degrees of partial melting (Bodinier 1988) or as a result of melt formation and incomplete melt extraction and crystallization in an upwelling diapir (Boudier & Nicolas 1972; Boudier 1978; Nicolas 1986). The peridotite is rich in plagioclase and clinopyroxene, and spinels commonly have high Cr/(Cr + Al) ratios. These characteristics are similar to some plagioclase peridotites dredged from slow-spreading ridges and along fracture zones (Dick 1989; Cannat et al. 1997; Seyler & Bonatti 1997; Tartarotti et al. 2002). Isotopic studies of the Lanzo peridotites have pointed out important differences between the Northern and Southern bodies (Bodinier et al. 1991). The Southern body of the Lanzo Massif has been interpreted as an asthenospheric diapir that rose from the garnet stability field and was emplaced in the early Mesozoic, during the opening of the Piedmont Ligurian basin. The Northern body has been considered a fragment of the subcontinental lithosphere that became isolated by the convective mantle at 400-700 Ma (Bodinier et al. 1991). In Lanzo, the peridotites are cut by an older suite of spinel websterites and a younger suite of discordant dunite, followed by various sets of gabbroic veins and dykes. Dunite cuts and locally replaces earlier spinel websterites (Fig. 3a), as indicated by trains of Cr-spinel that are continuous with the surrounding spinel websterites (see also Boudier & Nicolas 1972; Boudier 1978). New field observations show that some discordant dunites locally contain small interstitial clinopyroxene (Fig. 3b), and large clinopyroxene megacrysts (crystals of more than 1 cm in diameter, Fig. 3d) sometimes associated with plagioclase (Fig. 3c). In places, large, euhedral clinopyroxenes form aggregates a few millimetres wide or (deformed) gabbroic veinlets (Fig. 3e and f), similar to the 'indigenous' dykes described by Boudier & Nicolas (1972) and Boudier (1978). Locally, medial pyroxenite dykes in dunite have also been observed. The gabbroic dykes can be separated into two groups (Boudier & Nicolas 1972; Boudier 1978). Type 1 is an older 'indigenous'
74
O. MUNTENER & G.B. PICCARDO
Fig. 3. Field aspects of plagioclase peridotites and dunites from Lanzo South (Northern Italy), showing plagioclase peridotites, cut by spinel dunites. (a) Harzburgite-dunite contact (southern flank of Monte Musine'). It should be noted that the foliation is discordant to the dunite-harzburgite contact, (b) Interstitial clinopyroxene (green Crdiopside) within dunite. (c) Clinopyroxene (cpx) + plagioclase (pig) cluster in dunite (Monte Musine'). (d) Clinopyroxene megacryst in dunite (Monte Musine'). (e) 'Indigenous' microgabbroic vein and a clinopyroxene megacryst within dunite (Monte Arpone). (f) Weakly deformed 'indigenous' Mg-gabbro dykelet cutting dunite (Mt Arpone). At the lower right, the gabbro is discordant to the peridotite-dunite contact. Subsequent high-temperature ductile deformation formed dunite mylonite. du, dunite; hz, harzburgite; pig 1hz, plagioclase Iherzolite.
group of olivine gabbronorite, frequently occurring in en echelon fuzzy contacts with the surrounding Iherzolites and dunites. This type is restricted to the southwestern part of the massif
(Compagnoni et al 1984; Pognante et al. 1985). Type 2 is an intrusive group of troctolites, (olivine) gabbros, gabbronorites to oxide gabbros, with sharp contacts and chilled margins towards
A L P I N E - A P E N N I N E PERIDOTITES
the peridotite. These dykes can be found all over the massif (Compagnoni et al. 1984; Pognante et al. 1985), cut across dunites and plagioclase peridotites, and are generally undeformed. Thus it seems that penetrative high-temperature deformation of the peridotite ceased between the formation of type 1 and 2; however, both are locally mylonitized and partially hydrated under upper amphibolite- to granulite-facies conditions (Compagnoni et al. 1984; Pognante et al. 1985). The geochemistry of the mafic rocks reveals that most gabbros represent cumulates with little or no trapped liquid, indicating efficient extraction of derivative liquids (Bodinier et al. 1986). In addition, late porphyritic basaltic dykes of N-MORB affinity (Pognante et al. 1985) cut the peridotites and gabbros. The extracted melts had a T-MORB and a T- to N-MORB composition in the north and south, respectively, (Bodinier 1988), similar to basalts from the Ligurian Alps (Beccaluva et al. 1984). In Corsica, the Monte Maggiore peridotites are strongly depleted, with a spinel-facies granular assemblage: they are clinopyroxene-poor, refractory spinel Iherzolites, which are interpreted as mantle residua after MORB-type partial melting processes (Jackson & Ohnenstetter 1981; Rampone et al. 1997). Preliminary Sm/Nd isotope data provide a mid-Jurassic (165 Ma) DM (depleted mantle) model age of depletion (Rampone 2002). In places these peridotites contain plagioclase and show evidence for trapped melt crystallization (Rampone et al. 1997), In the Monte Maggiore region, peridotites with oriented and diffuse impregnation (Fig. 4a and b) are cut by discordant dunites (Fig. 4e), followed by the intrusion of gabbroic veins (Fig. 4c) and metrescale pockets of mafic-ultramafic cumulates, composed mainly of olivine gabbronorites (Fig. 4d). The cumulates cut dunite-peridotite contacts and the existing oriented impregnation. Locally these pockets dominate volumetrically and the former peridotite is completely replaced by gabbronorite mineral assemblages consisting of euhedral ortho- and clinopyroxene and interstitial plagioclase (Fig. 4d). Another common feature at Monte Maggiore is the formation of coarsegrained and undeformed late gabbroic dykes (with crystals exceeding 5 cm in diameter), which crosscut deformed peridotites and gabbronorite cumulates (Fig. 4f).
Residual mantle mineral assemblages The studied spinel- and plagioclase-bearing Iherzolites are mainly porphyroclastic peridotites comprising a deformed mantle assemblage and less deformed or undeformed interstitial igneous as-
75
semblages. Rare samples are nearly plagioclasefree spinel Iherzolites and show textures typical of common spinel peridotites. Olivine occurs in large grains (up to 1 cm) and pyroxenes form millimetre- to centimetre-scale porphyroclasts. Spinel is brown and Al-rich, and commonly shows hollyleaf shapes (Fig. 5a). Deformation-induced undulatory extinction and gliding in olivine is common. Pyroxenes are commonly deformed and show fine exsolution lamellae of the complementary pyroxene. In samples not affected by melt impregnation, Al-rich spinel is locally intergrown with orthopyroxene, producing a texture similar to that of garnet breakdown (Vannucci et al. 1993b). In the same sample, a small rim of olivine + plagioclase is locally developed between orthopyroxene and spinel, according to the reaction orthopyroxene + spinel —> olivine + plagioclase (Fig. 5b) This microstructure probably formed during decompression from the spinel peridotite to the plagioclase peridotite field before melt impregnation and melt-rock reaction, and provides rare evidence for a metamorphic origin of the plagioclase in the Lanzo peridotite.
Impregnation textures The sequence of igneous microstructures in the spinel peridotite is well established. Early meltrock reaction dissolved clinopyroxene along grain boundaries and precipitated orthopyroxene + plagioclase around and within clinopyroxene (Fig. 5c and d). Textural relationships indicate cotectic crystallization of plagioclase + orthopyroxene (Fig. 5d). These intergrowths are not deformed, contrary to the original mantle clinopyroxene. A similar structure can be observed in spinel websterites. These features indicate that the migrating liquid crystallized clinopyroxene-free, orthopyroxene-rich gabbronoritic microgranular aggregates. Orthopyroxene partially replaced mantle minerals, showing concave contacts against the peridotite clinopyroxene (Fig. 5e). However, in many samples crystallization of two pyroxenes and plagioclase is also common. This is illustrated in Figure 5e and f, where undeformed and interstitial clinoand orthopyroxene crystallized between large mantle minerals. Large kinked mantle olivine is recrystallized close to the interstitial orthopyroxene (Fig. 5f), supporting the general observation that the impregnating assemblages are less deformed than the precursor mantle assemblage. In places the igneous domains consist of plagioclase patches, replacing spinel, together with granular orthopyroxene + olivine ± clinopyroxene (Fig. 5g). These microgabbroic aggregates form xenomorphic granoblastic mosaic textures between mantle minerals generally several millimetres to
76
O. MUNTENER & G.B. P I C C A R D O
Fig. 4. Field aspects of plagioclase peridotite and gabbronorite from the Monte Maggiore peridotite (Corsica), (a) Impregnated peridotites with interstitial plagioclase surrounding mantle minerals. Locally plagioclase (pig) + orthopyroxene (opx) coalesce, forming gabbronorite veinlets. (b) Oriented plagioclase-rich impregnation in peridotite discordantly cut by a cpx-rich gabbronorite dykelet, related to the cumulate pods. It should be noted that in the upper portion of the outcrop the gabbronorite dyke ends in a millimetre-scale veinlet. (c) Impregnated peridotite intruded by irregular gabbroic veins or pods related to the cumulate suite. Euhedral green cpx and interstitial pig in the gabbroic veins or pods should be noted, (d) Euhedral olivine + opx + cpx and anhedral plagioclase in a gabbronorite cumulate, (e) Discordant dunite-plagioclase peridotite contact cut by a gabbro dykelet. The euhedral clinopyroxene megacrysts in the dykelet should be noted. The foliation in both peridotite and dunite runs approximately perpendicular to the contact. Locally, millimetre-scale plagioclase seams of the gabbro intrude the surrounding dunite and plagioclase peridotite. (f) Coarse-grained Mg-gabbro dyke with chilled margins cuts spinel peridotite.
A L P I N E - A P E N N I N E PERIDOTITES
77
Fig. 5. Selected thin section specimens from the Lanzo South area, (a) Intergrowth of orthopyroxene (opx) and spinel (spl) in a spinel peridotite from Mt Arpone. Of particular note is the overall shape of this intergrowth, recalling former garnet. It should be noted also that spl-opx contacts are fresh without any sign of reaction, (b) Opx-spl contacts separated by a small rim of olivine + plagioclase (completely altered). This indicates that orthopyroxene and spinel are unstable and transform into a (metamorphic) assemblage of olivine (ol) + plagioclase (pig), (c) Corrosion of exsolved and deformed mantle clinopyroxene by opx + pig intergrowths. (d) Intergrowth of opx + pig indicating cotectic crystallization of the two phases, (e) Undeformed interstitial opx + cpx separating large, kinked mantle olivine. (f) Rim of interstitial opx replacing mantle cpx. Of particular note are the concave opx-cpx contacts. It should be noted also that kinked olivine is replaced by undeformed olivine in the left part of the photomicrograph. This demonstrates that cpx + liquid reacted to form ol + opx. (g) Undeformed micro-gabbronorite aggregate between deformed mantle minerals. The anhedral shape of pig and opx between granular euhedral olivine should be noted. Cpx forms small interstitial grains. These textural relationships demonstrate the crystallization sequence ol —> opx + plag —> cpx and indicate that migrating liquids were saturated in opx before cpx. (h) anhedral cpx along triple point of olivine in discordant dunite.
78
O. MUNTENER & G.B. PICCARDO
centimetres wide. The granoblasts do not show a preferred orientation and form troctolitic to gabbronoritic assemblages. Interstitial pyroxenes do not show exsolution of the complementary pyroxene, are virtually undeformed and show concave contacts to the host peridotite minerals (Fig. 5e and f). In other samples olivine, clinopyroxene and orthopyroxene form euhedral crystals surrounded by interstitial plagioclase. Mantle clinopyroxene is seemingly unreacted. Microstructures in the dunites are characterized by coarse-grained olivine (up to 2 cm in size) and more or less rounded spinels, as described previously (Boudier & Nicolas 1972; Boudier 1978; Nicolas 1986). In addition, many dunites contain interstitial clinopyroxene, which is exclusively found along olivine triple junctions (Fig. 5h).
Geochemical data We analysed crystals of peridotite and gabbroic samples by electron microprobe and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Preliminary results are shown in Tables 1 and 2 and illustrated in Figures 6 and 7. In terms of major element compositions, the main variation in clinopyroxenes in peridotites is reflected in the Al and Ti contents, with high Al and low Ti in spinel peridotites, and low Al and higher Ti in plagioclase peridotite. Mg numbers (molar Mg/(Mg + Fetot)) range from 0.89 to 0.92 (Table 1). Na2O contents are invariably low (1 and a weak to significant negative Eu anomaly (Fig. 6), which increases from core to rim, indicating equilibration with plagioclase. In addition, most of the trace ele-
ments (i.e. Ti, Sc, V, Zr, Y) in clinopyroxene from plagioclase peridotites are enriched with respect to the precursor clinopyroxene in spinel Iherzolites (Table 1). Orthopyroxene follows the trends given by clinopyroxene. Plagioclase from both Lanzo and Monte Maggiore peridotites show similar REE chondrite-normalized patterns with significant LREE fractionation (CeN/NdN < 0.5) and very low Sr (2 ratio in bulk-rock compositions demonstrate a dual distribution of the investigated rocks (Fig. 8). Harzburgites of the first two groups (excluding sample BR-3) show relatively low FeO/SiO2 ratios, which fall mainly within the range typical for residual mantle, and which tend to decrease slightly with an increase of spinel Cr number. Harzburgite BR-3, harzburgites of the third group, and dunites have elevated FeO/ SiC>2 ratios. This compositional shift may reflect
Fig. 7. Variations of bulk-rock Cr2O3/SiO2 ratios for Brezovica ultramafic rocks. Symbols are the same as in Fig. 3.
Fig. 8. Variations of bulk-rock FeO/SiO2 ratios for Brezovica ultramafic rocks. Symbols are the same as in Figure 3.
100
B. A. BAZYLEV ETAL.
olivine crystallization, dissolution of orthopyroxene, or influx of iron during re-equilibration of the peridotite with a fractionated melt. Anomalous harzburgites of the third group also differ from abyssal residual spinel peridotites in their elevated Ti/Al ratios (Fig. 9), which can reflect interaction of these rocks with a fractionated melt.
Thermal history as a reflection of magmatic processes The geothermometry of spinel peridotites provides important information about the cooling history of these rocks, as the equilibrium temperatures, estimated by mineral geothermometers, may be treated as the closure temperatures of the corresponding exchange reactions. It was recognized that abyssal spinel peridotites have higher equilibration temperatures than alpine-type (suprasubduction) peridotites, which argue for a faster cooling of the mantle rocks in a mid-ocean ridge (MOR) geodynamic setting (Dick & Fisher 1984; Parkinson & Pearce 1998). The closure temperatures of both two-pyroxene and olivine-spinel reactions in mantle spinel peridotite were modelled by Bazylev & Silantyev (1999), assuming a fixed value for the average size of the spinel grains (0.3 mm) and for the distance of the analysed points from the orthopyroxene-clinopyroxene contact (20 urn). It was established that continuous cooling of the rocks after melt segre-
Fig. 9. Variations of chondrite-normalized bulk-rock Ti/ Al ratios with Cr number of spinels. Chondrite composition is after Anders & Grevesse (1989). Field of spinel peridotites from Mid-Atlantic Ridge is after Bazylev et al. (1999). PM is the composition of the primitive mantle (McDonough & Sun 1995); arrow indicates the compositional changes during mantle partial melting. Symbols are the same as in Figure 3.
gation in a MOR setting should produce a relatively large difference (150-200 °C) in the closure temperatures of the two-pyroxene and olivine spinel reactions as a result of different diffusion parameters. Based on the model of adiabatic melting below mid-ocean ridges (Langmuir et al. 1992), we infer that the spinel peridotites in a single area should be heated to similar maximum temperatures during their ascent, resulting in a similar degree of partial melting and a similar cooling rate after melt segregation. Thus, rocks from a restricted MOR locality should demonstrate similar values for both the spinel Cr number and the exchange reaction closure temperatures. This was found to be the case for all MOR spinel peridotites for which representative analytical data are available (Bazylev & Silantyev 2000a), including areas where spinel dunites compose a significant part of the mantle, such as at Ocean Drilling Program Site 895 in the Hess Deep (Arai & Matsukage 1996; Dick & Natland 1996) or at the southern inner corner high of the 15°20'N Fracture Zone in the Atlantic Ocean (Bazylev & Silantyev 2000a). The three harzburgite groups within the Brezovica massif, which are different both in their mineral compositions and in their structural position within the ultramafic body, are characterized by different average closure temperatures of twopyroxene and olivine-spinel exchange reactions. Generally, these temperatures are 894 °C and 714 °C for the first group, 831 °C and 755 °C for the second group, and 916 °C and 726 °C for the third group. Calibrations are from papers by Wells (1977) and Ballhaus et al. (1991). Generally, significant variations in the two-pyroxene and olivine-spinel temperatures within the spinel peridotites from a restricted area are not consistent with continuous rock cooling after melt segregation and with a MOR setting (Bazylev & Silantyev 2000b). Both the mineral chemistry and the geothermometry of these rocks are consistent with a model of locally repeated heating of the rocks to solidus temperatures (i.e. a second magmatic episode for a part of these rocks; Bazylev & Silantyev 2000b). Because the process of mantle melting is manifested in an increase of the primary spinel Cr number (Dick & Bullen 1984; Hellebrand et al., 2001), the second magmatic episode should evidently be related to the formation of rocks with the most chromian spinels that are represented in the case of the Brezovica by dunites and the third group harzburgites. The difference between the two-pyroxene and olivine-spinel temperatures varies for the Brezovica rock groups generally because of variations in the two-pyroxene temperatures; the olivine-
BREZOVICAULTRAMAFIC MASSIF, SERBIA spinel temperatures are rather similar for all the rock groups (735 ± 20 °C). This observation is consistent with a variant of modelled spatial distribution of the mineral exchange closure temperatures in peridotites located at an isothermal level within the lithosphere after repeated heating by an infiltrated (percolated) melt, as presented in Figure 10. The values of various parameters (e.g. the temperature of the percolating melt, the duration of percolation, the initial temperature of the peridotites and initial two-pyroxene and olivinespinel temperatures in these rocks before percolation) were chosen to explain the phenomena of spatially associated spinel peridotites with different closure temperatures for exchange reactions (Bazylev & Silantyev 2000b); the values for the Brezovica rocks may have been somewhat different. Nevertheless, this model reproduces well the main features of the Brezovica peridotite petrologyThe large difference (190°C) between the Opx-Cpx and Ol-Spl temperatures for the third group of harzburgites indicates continuous cooling after melt segregation (or crystallization) to a temperature lower than about 726 °C (which is the
101
estimated average Ol-Spl temperature for these rocks). These rocks demonstrate the highest twopyroxene temperatures and the highest spinel Cr number, both indicating their initial position near a zone of magmatic channels (Fig. 10) that is marked now by dunite dykes and bodies (Kelemen et al. 1995). The second group of harzburgites, located adjacent to the third group, was heated during the second magmatic event to temperatures lower than the solidus (and lower than the temperatures to which the third group harzburgites were heated). As a result of this event, their cooling after the cessation of the melt percolation was slower, resulting in lower closure temperature of the two-pyroxene reaction (Fig. 10). This process provided the small difference between the average values of the two-pyroxene and olivinespinel temperatures shown by these rocks (c. 80 °C). The harzburgites of the first group appear to have been even farther from the heat source (Fig. 10). They were heated (if heated at all) to temperatures lower than 830°C, and the time of heating was too short to provide pyroxene reequilibration at this temperature. Therefore, the
Fig. 10. An example of the calculated distribution of closure temperatures of exchange reactions in mantle peridotites as a result of their secondary heating by the percolated melt (Bazylev & Silantyev 2000b). Tr, maximum temperature of the rocks; 7pp, final closure temperature of the two-pyroxene reaction; 7^s, final closure temperature of the olivine-spinel reaction. The parameters for this model are: percolating melt temperature is assumed to be 1250 °C, the duration of percolation is assumed to be 75 ka, the assumed initial temperature of the peridotites is 800 °C, the adopted initial closure temperature in peridotites (before the melt percolation) are 950 °C and 820 °C for the two-pyroxene exchange reaction and for the olivine-spinel reaction, respectively. 1, outer zone, where rpp is not affected by heating (corresponds to the location of the Brezovica first group of harzburgites); 2, intermediate zone (corresponds to the location of the Brezovica second group of harzburgites); 3, inner zone adjacent to the magmatic channel; the temperature during melt percolation exceeds 1200 °C in its innermost part, causing peridotite melting or melt-rock interaction. This zone corresponds to the location of Brezovica dunites and third group of harzburgites.
102
B. A. BAZYLEV ETAL.
two-pyroxene temperatures in these rocks are relict and rather reflect their cooling conditions after the previous magmatic episode. Based on the model, the final olivine-spinel temperatures for all the Brezovica ultramafic rocks should be similar and equal to the initial temperature of the rocks at the level where melt percolation took place; this temperature is estimated to be 735 ± 20 °C. Thus, continuous cooling of the Brezovica ultramafic rocks after segregation of a melt produced during the first magmatic stage is reflected in the two-pyroxene reaction closure temperatures in the first group of harzburgites. Continuous cooling after the second heating event and segregation of melt during the second magmatic stage are reflected in the closure temperatures of the third group of harzburgites. During continuous cooling, the closure temperatures of both the two-pyroxene and olivine-spinel reactions depend on cooling rate. It was established by a physical modelling that the main factor influencing the mantle rock cooling rate is the thickness of the lower lithospheric level where conductive heat transfer occurs (Bazylev & Silantyev 1999, 2000a). The thickness of this level (a09 km) can be estimated directly from the calculated average temperatures of the two-pyroxene (7J,p, °C) and olivine-spinel (7J>S, °C) equilibria using the equations of Bazylev & Silantyev (2000a):
and
where T^ (°C) is the maximum temperature of the hydrothermal fluid penetrating the rocks (i.e. the temperature at the upper surface of the zone of conductive heat transfer in the lithosphere), which can be estimated from the data on the metamorphic mineralogy. An average value of 550 °C can be adopted if mineral data are lacking. It is also possible to evaluate the pressure at which the last melt fractions were segregated from the mantle peridotites from the thickness of the zones of conductive and convective heat transfer in the lithosphere by using the equation (Bazylev & Silantyev 2000a)
where a0 is an average of estimates from both the two-pyroxene and olivine-spinel geothermometry, 40 (deg km~ l ) is an appropriate average geothermal gradient below an axial part of a mid-ocean ridge, 3.2 is a coefficient reflecting the average density of the lithosphere, and 0.3 is a correction for the average water depth.
The calculations performed for the Brezovica ultramafic rocks allow us to conclude that in the first magmatic stage, the last melt segregation occurred at 29 km, or at 9.3 kbar. During the second magmatic stage, the last melt segregation occurred at about 25 km, a depth roughly equivalent to a pressure of 8.2 kbar. A conductive layer of 12 km thickness is estimated for this magmatic event. Because lateral heat transfer may have occurred during cooling after cessation of melt percolation, the reported depth and pressure estimates for the second magmatic stage should be considered minimum values. The time necessary for the closure of the twopyroxene reaction in the first group of harzburgites at 900 °C indicates the minimum amount of time between the two melt stages during the evolution of the Brezovica ultramafic rocks; this time is estimated to be c. 0.5 Ma (Bazylev & Silantyev 2000b).
Discussion Magmatic evolution of the Brezovica ultramafic rocks The mineral chemistry, bulk-rock chemistry and thermal history of the Brezovica ultramafic rocks provide evidence for two separate stages of melting with a cooling phase between them. The spinel harzburgites of the first group and probably the harzburgites of the second group were formed during the first melting stage. Spinel compositions of these rocks lie outside the midocean peridotite field (Fig. 3), but they are comparable with those for suprasubduction zone peridotites (Ishii et al. 1992). Relatively large variations in spinel Cr number in these rocks (0.46-0.64) indicate large local variations in the degree of partial melting (Hellebrand et al. 2001). Such variations cannot be produced by decompression melting in a MOR setting (e.g. Langmuir et al. 1992) and are not exhibited by MOR spinel harzburgites (Dick & Natland 1996). However, they can be produced by melting induced by fluid or melt input in a hot mantle in a suprasubduction zone setting (Ishii et al. 1992; Parkinson & Pearce 1998). Nevertheless, the obtained mineral and bulk-rock chemistry does not seem to have been influenced significantly by the influx of such subduction zone component. In the harzburgites of the first and the second groups, the increase in spinel Cr number is accompanied by an increase in the Mg number of olivine and orthopyroxene (Fig. 4), an increase in the olivine nickel content, an increase in the bulk-rock Cr number (Fig. 6), and a decrease in the bulk-rock FeO/SiO2 (Fig. 8) and Ti/Al (Fig. 9) ratios. These
BREZOVICAULTRAMAFIC MASSIF, SERBIA features suggest simple partial melting of a mantle that was not affected by melt-rock interaction. Therefore, the harzburgites of these two groups represent different stages of mantle partial melting, with the second group harzburgites corresponding to higher degrees of melting. Additional evidence for a suprasubduction setting of the first magmatic stage is provided by the estimated pressure at which the last melt fraction was segregated. This value (9.3 kbar), coupled with the spinel Cr number in these rocks, indicates deeper conditions of melting than those established for MOR settings (Bazylev & Silantyev 2000b). Knowing the pressure at which the last melt fraction was segregated from the mantle and the spinel Cr number in these rocks, we can estimate the aluminium content of the melt using the expression of Bazylev (1996):
This expression yields 14.3 wt% A12O3 for the melt segregated from the first group of harzburgites during the first magmatic stage. This value is consistent with the tholeiitic nature of the melt. The oxygen fugacity during the first magmatic stage, calculated using the method of Ballhaus et al. (1991), is on average 1.7 log units below the fayalite-magnetite-quartz buffer (FMQ) for the first group of harzburgites and 0.3 log units below FMQ for the second group. These values are generally within the ranges typical for both MOR and suprasubduction zone spinel peridotites (Parkinson & Pearce 1998). Thus we suggest that the first magmatic stage in the Brezovica ultramafic rocks occurred in a suprasubduction zone setting and resulted in the segregation of melts with a tholeiitic affinity. During the second magmatic stage, the harzburgites of the third group and the dunites were formed in the Brezovica massif. As inferred from the thermal history of these rocks, this stage occurred after they cooled to temperatures of about 900 °C as a result of local heating to solidus temperatures by the percolating melt. The ranges of spinel Cr number in the third group of harzburgites (0.62-0.72) and dunites (0.69-0.80) indicate a suprasubduction setting of this magmatic stage (Dick & Bullen 1984; Arai 1994). The aluminium content in the melt segregated from harzburgites of the third group during the second magmatic stage is estimated to be 11.1 wt% A12O3. Such a low aluminium content is indicative of melts that are intermediate in composition between tholeiite and high-Ca boninite, although it is more typical of boninitic melts.
103
The oxygen fugacity during the second magmatic stage is calculated to be 0.7 log units above FMQ. This value lies outside the range typical for MOR spinel peridotites but is consistent with the values for suprasubduction zone peridotites (Parkinson & Pearce 1998). However, the compositional features of the primary minerals and rocks formed at this stage are not consistent with a simple partial melting process. In particular, the Mg number of olivines and pyroxenes in harzburgites of the third group and in harzburgite BR-3 is significantly lowered compared with the trend of mantle partial melting (Fig. 4), the nickel content in olivines from these rocks is lower than for the first and second group of harzburgites, the titanium contents in the Crrich spinels are elevated (Fig. 5), the sodium content in the clinopyroxenes is higher than for the first and second group harzburgites, and the bulk-rock FeO/SiO2 and Ti/Al ratios are significantly higher than those of the first and second group harzburgites (Figs 8 and 9). All these features are typical for melt-rock interaction. This process is related to silica undersaturation of deep mantle melts during their ascent and results in dissolution of orthopyroxene from the wall rocks and crystallization of olivine (and sometimes of Cr-rich spinel), accompanied by partial chemical re-equilibration between the melt and wall rocks (Kelemen et al 1992, 1995). The dissolution of orthopyroxene in the Brezovica harzburgites is manifested by an elevated bulk-rock Cr number (Fig. 6) and FeO/SiO2 ratio (Fig. 8), whereas crystallization of olivine is indicated by its low Mg number (Fig. 4) and nickel content. Chemical re-equilibration with the melt is manifested by a lower orthopyroxene Mg number (Fig. 4), an elevated sodium content in clinopyroxene, and an elevated titanium content in both spinel and bulkrock compositions. Dunite formation indicates a prograde stage of melt-rock interaction manifested by the complete dissolution of orthopyroxene (Kelemen et al. 1995). The melt evolution accompanying this process results in clinopyroxene crystallization (Dick & Natland 1996) and locally in the crystallization of hornblende in dunites (Arai & Matsukage 1996). These features are found in the dunites of the Brezovica massif. A coupled increase in the titanium content and degree of iron oxidation in the Cr-rich spinel within the rock suite and within a single rock (dunite BR-12) (Fig. 5) is also indicative of melt-rock interaction (Allan & Dick 1996). In most of the Brezovica dunites, olivine has high Mg number and high nickel contents, which are also indicative of melt-rock interaction rather than crystallization of dunite from a melt (Dick & Natland 1996). Finally, most of the
104
B. A. BAZYLEV ETAL.
investigated Brezovica dunites have bulk-rock Cr2O3/SiO2 ratios similar to those of the residual harzburgites (Fig. 7). This observation is easily explained by orthopyroxene dissolution in the initial harzburgite, but would not be compatible with direct crystallization of the dunites from a melt. The titanium contents in the magmatic hornblende from the dunite BR-12 (0.63 wt% TiO2) and in associated spinel (0.46 wt% TiOa) are significantly lower than the titanium contents in these minerals from most refractory abyssal spinel peridotites formed in MOR segments (Arai & Matsukage 1996; Bazylev et al. 2001). Because water and titanium are both incompatible during melt fractionation, the titanium content of the first hornblende to crystallize from a melt should reflect the titanium content of the melt at the time of its water saturation. For melts with similar major element compositions, which are differentiated in similar conditions, the titanium content in hornblende reflects the water content (water/titanium ratio) in an initial melt: the higher the water content in the primary melt, the earlier melt saturation in water occurs, and the lower the titanium contents in the hornblende and related melt. Hence, we infer that the melt that percolated through the Brezovica ultramafic rocks during the second magmatic stage was significantly richer in water than primary MOR melts and was characterized by a very high water/titanium ratio. This is consistent with the boninite affinity of this melt and its suprasubduction zone affinity. Therefore, we infer that the second magmatic stage recorded in the Brezovica ultramafic rocks occurred in a suprasubduction geodynamic setting and that it resulted in the segregation of melts with a high-Ca boninite affinity Interaction between spinel harzburgites and percolating melt, rather than partial melting, took place at this stage. As mentioned above, basalts are not preserved in the Brezovica massif possibly because of erosion. Rare gabbros (fresh and rodingitized) from the upper part of the Brezovica massif in the Livad area show extremely low titanium contents, below 0.15wt% TiOi (Batocanin & Memovic 1996), and from this parameter we infer that the gabbro may be cogenetic with melts of boninite affinity segregated from the Brezovica harzburgites.
Metamorphic evolution of the Brezovica ultramafic massif It is surprising that the Brezovica harzburgites show no signs of a thermal influence from the
overlying cumulate rocks. This can be explained either by a genetic link between the dunite bodies in the mantle harzburgites and the dunites and peridotites in the cumulate sequence, or by a tectonic contact between the mantle and the crustal (cumulate) parts of the ultramafic sheet. In the first case, repeated heating of the mantle harzburgites is related not only to melt penetration, but also to the crystallization of the cumulate rocks that should have taken place at c. 25 km depth. We do not yet have any direct evidence for such deep crystallization. On the other hand, the nature of the contact between the mantle and the cumulate rock sequences in Brezovica is obscured by intensive serpentinization, obscuring a possible tectonic contact. Another unexpected result of our investigation is the absence of any influence of the metamorphism related to the metamorphic sole below the ultramafic rocks on the closure temperatures of exchange reactions between the primary minerals in peridotites. Such an influence should be evident if the ultramafic rocks were as hot at the time of their tectonic juxtaposition with cold olistostromal rocks as was proposed earlier, 900 °C or even higher (Karamata 1968a, 1968b; Djordjevic et al. 1987). As is indicated by numerical physical and thermodynamic modelling (Bazylev & Silantyev 2000b), the difference between the two-pyroxene and olivine-spinel temperatures for all the Brezovica harzburgites should be similar and large (c. 150-200 °C), and the calculated temperatures in the harzburgites should fall regularly away from the contact zone. However, the observed distribution of the closure temperatures of exchange reactions is consistent with a local secondary heating process, as demonstrated above. Preservation of this distribution requires that the initial temperature of the Brezovica ultramafic rocks at the time of their tectonic juxtaposition with the olistostromal melange should not exceed the minimal value of the olivine-spinel, i.e. about 735 ± 20 °C. This temperature is close to the maximum metamorphic temperature of the olistostromal rocks in the underlying sole (600-700 °C). This problem disappears if the hot ultramafic rocks were not instantaneously juxtaposed with the cold olistostromal rocks (where the maximal contact temperature should be an average of the two initial temperatures (Jaeger 1968)). Beginning with development of the thrust fault, the olistostromal rocks were progressively heated by every new part of the moving ultramafic slice coming into contact with the structurally lower rocks. The time required to heat the olistostromal rocks is estimated to be 0.5-5 Ma assuming an initial depth of thrusting of 25 km, a thrust rate of 1-10 cm a"1
BREZOVICA ULTRAMAFIC MASSIF, SERBIA and a fault zone dip of 30°. This time seems to be sufficient to provide the necessary heat for the olistostromal rocks. Therefore, it can be expected that the maximum contact temperature would be only c. 50-100 °C lower than the initial temperature of the ultramafic rocks that is consistent with the data from the metamorphic sole. The metamorphic mineral chemistry and the estimated temperatures of the amphibolite- to greenschist-facies metamorphism of the ultramafic rocks indicate that the temperature of the peridotite metamorphism falls from the bottom to the top of the ultramafic slice. This raises the question of whether the metamorphism was related to fluid circulation in the lithosphere or to a high-gradient metamorphic aureole similar to that described in the olistostromal rocks at the sole of the ultramafic body. Although the data presented in this paper are not sufficient to resolve this problem, we believe that they will stimulate continuing investigations. The last metamorphic stage recognizable in the Brezovica ultramafic rocks was serpentinization. Both the isochemical character of the serpentinization and the presence of brucite associated with serpentine in these rocks argue against a MOR setting for this process. In fact, the serpentinization of abyssal peridotites is not an isochemical process, and brucite is generally absent in these rocks (Snow & Dick 1995). Both the isochemical character of the serpentinization and the presence of brucite are typical of alpine-type peridotites (Wicks & Plant 1979), which seem to be serpentinized in the upper crust by meteoric water (Wenner & Thaylor 1974), or in a frontal part of the mantle wedge in a suprasubduction zone setting by fluid derived from subducted sediments (Dmitriev et al. 1999). In any case, serpentinization of the Brezovica ultramafic rocks appears to have occurred after they were technically juxtaposed with the olistostromal rocks.
Initial geological setting of the Brezovica massif To identify the Brezovica massif as part of the Central Dinaridic ophiolite belt or the Vardar zone we need to compare the Brezovica ultramafic rocks with possible analogues in the neighbouring areas. The Brezovica peridotites are most similar geologically and petrographically to the harzburgite massifs of the eastern belt of Albania (eastern-type massifs: Kukes, Bulqiza, Shebenic). Spinel compositions in harzburgites and dunites from all these massifs (Bebien et al. 1998) are close to those of the Brezovica body, with Cr number 0.4-0.8. In the Bulqiza massif, harzbur-
105
gites with relatively low-chromian spinels (0.55) occur in the lowest part, harzburgites with subordinate dunites with high-chromian spinels (0.76-0.80) in the middle part, dunites with subordinate harzburgites (Spl Cr number 0.810.835) in the upper part, and cumulates with basal dunites (Spl Cr number 0.82-0.835) in the uppermost part (Alliu et al. 1994). The possible initial presence of very low-Ti basalts and boninites in the Brezovica massif is inferred from the mineralogy and chemistry of the ultramafic and gabbroic rocks, and similar volcanic rocks are widespread in the Albanian massifs (Bebien et al. 2000). Similar to the Brezovica massif, the Albanian harzburgite massifs are allochthonous slices underlain by metamorphic soles that originated under similar conditions. The thickness of the mantle rock sequence in the Djakovica area (which is immediately adjacent to Albania; Fig. 1) is >7 km (Roksandic 1974), similar to that of the Albanian ultramafic massifs (Alliu et al. 1994). Considering these data, as well as the existence of the numerous peridotite outcrops between Djakovica and Brezovica, the decrease in the thickness of the ultramafic rocks and of the metamorphic sole eastward, and the higher temperature of metamorphism beneath the ultramafic slices westward, the Brezovica ultramafic massif can be interpreted as the frontal, easternmost relic of suprasubduction lithosphere emplaced during and after the closure of the Central Dinaridic basin. Therefore, the hypothesis of Nicolas & Boudier (1999) that the eastern-type Albanian massifs formed at a fast-spreading mid-ocean ridge is not supported by the data presented in this paper.
Conclusions (1) The Brezovica ultramafic rocks in southern Serbia underwent two separate magmatic stages, as is inferred from the mineral and bulk-rock chemistry and their thermal history. In the first stage, a suite of spinel harzburgites was formed as a result of partial melting to produce tholeiitic melts. During the second stage, these spinel harzburgites were repeatedly heated and affected by percolating melt. This process resulted in the formation of dunites and anomalous spinel harzburgites by melt-harzburgite interaction. The melt that segregated from these rocks during the second magmatic stage was high-Ca boninite. Both magmatic stages occurred in a suprasubduction geodynamic setting at a relatively deep level (2528km). (2) At the time of tectonic juxtaposition of the Brezovica massif with underlying olistostromal rocks that produced a metamorphic sole beneath
106
B. A. BAZYLEV ETAL.
the massif, the Brezovica ultramafic rocks were cooled to temperatures not exceeding 735 ± 20 °C. (3) The initial geological setting of the Brezovica ultramafic rocks is related to the Central Dinaridic-Mirdita basin. The Brezovica ultramafic massif in its present position is interpreted as the easternmost part of the eastward-emplaced Mirdita-Djakovica-Orahovac-Brezovica 'nappe' (wedge), which represents the obducted fragments of a suprasubduction oceanic-type lithosphere formed in the Central Dinaridic-Mirdita basin. The authors thank N. N. Kononkova (Vernadsky Institute, Moscow) for providing the electron microprobe data, and I. A. Roshchina and T. V Romashova (Vernadsky Institute, Moscow) for providing bulk-rock analyses. J. Allan, E. Rampone, J. Bebien, and Y. Dilek are thanked for their constructive criticism and suggestions for improvements of the manuscript. The work was supported by Russian Foundation of Basic Research grant 01-05-64288.
References ALIAJ, SH. & MECO, S. 1994. Mirdita oceanic basin was located westwards of Korab zone (=Pelagonian zone). Ofioliti, 19, 97-103. ALLAN, J.F. & DICK, H.J.B. 1996. Cr-rich spinel as a tracer for melt migration and melt-rock interaction in the mantle: Hess Deep, Leg 147. In: MEVEL, C., GILLIS, K.M., ALLAN, J.F. & MEYER, P.S. (eds) Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, 147, 157-172. ALLIU, I., BECCALUVA, L. & f INA, A. ET AL.1994. The Skanderbeg and Bulqiza mafic-ultramafic ophiolitic complexes and their relationships with chromititic ore deposits (Working Group Meeting of IGCP Project 256—Field Trip B). Ofioliti, 19, 27-55. ANDERS, E. & GREVESSE, N. 1989. Abundances of the elements: meteoritic and solar. Geochimica et Cosmochimica Acta, 53, 197-214. ARAI, S. 1994. Characterization of spinel peridotites by olivine-spinel compositional relationships: review and interpretation. Chemical Geology, 113, 191 — 204. ARAI, S. & MATSUKAGE, K. 1996. Petrology of gabbrotroctolite-peridotite complex from Hess Deep, equatorial Pacific: implications for mantle-melt interaction within the oceanic lithosphere. In: MEVEL, C., GILLIS, K.M., ALLAN, J.F. & MEYER, P.S. (eds) Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, 147, 135-156. BALLHAUS, C., BERRY, R.F. & GREEN, D.H. 1991. High pressure experimental calibration of the olivineorthopyroxene-spinel oxygen geobarometer: implications for the oxidation state of the upper mantle. Contributions to Mineralogy and Petrology, 107, 27-40. BATOCANIN, D. & MEMOVIC, E. 1996. Two members of a dismembered oceanic crust, the deepest and the
highest, at Budanac, Brezovica. In: KNEZEVIC, V. & KRSTIC, B. (eds) Terranes of Serbia. Barex, Belgrade, 265-268. BAZYLEV, B.A. 1996. Compositions of clinopyroxene and spinel in restite hyperbasites as indicators of the generation conditions and compositions of coupled primary mantle magmas. Geochemistry International, 33, 115-124. BAZYLEV, B.A. 2000. Appearance of the awaruitebearing mineral assemblage in peridotites from 15°20'N FZ (Atlantic Ocean) as one of manifestations of the oceanic metamorphism. Russian Journal of Earth Sciences, 2(3/4), 279-293 (in Russian). BAZYLEV, B.A. & SILANTYEV, S.A. 1999. Geodynamic interpretation of subsolidus metamorphism of mantle spinel peridotites. Ofioliti, 24, 61-62. BAZYLEV, B.A. & SILANTYEV, S.A. 2000a. Geodynamic interpretation of subsolidus recrystallization of mantle spinel peridotites: 1. Mid-oceanic ridges. Petrology, 8, 201-213. BAZYLEV, B.A. & SILANTYEV, S.A. 2000b. Geodynamic interpretation of subsolidus recrystallization of mantle spinel peridotites: 2. Ophiolites and xenoliths. Petrology, 8, 311-331. BAZYLEV, B.A., MAGAKYAN, R., SILANT'EV, S.A., IGNATENKO, K.I., ROMASHOVA, T.V. & XENOPHONTOS, C. 1993. Petrology of peridotites from Mamonia complex, southwestern Cyprus. Petrology, 1, 302-330. BAZYLEV, B.A., ZAKARIADZE, G.S., ZHELYAZKOVAPANAYOTOVA, M.D., KOLCHEVA, K., OBERHAENSLI, R.E. & SOLOV'EVA, N.V. 1999. Petrology of ultramafic rocks from the ophiolite association in the crystalline basement of the Rhodope massif. Petrology,!, 192-211. BAZYLEV, B.A., SILANTYEV, S.A., DICK, H.J.B. & KONONKOVA, N.N. 2001. Magmatic amphiboles and micas in oceanic peridotites and some specific features of the related magmas: 15°20'N MAR Fracture Zone. Russian Journal of Earth Sciences, 3(3), 219-234. BEBIEN, J., SHALLO, M., MANIK, K. & GEGA, D. 1998. The Shebenik massif (Albania): a link between MOR- and SSZ-type ophiolites. Ofioliti, 23, 7-16. BEBIEN, J., DIMO-LAHITTE, A., VERGELY, P., INSERGUEix-FiLippi, D. & DUPEYRAT, L. 2000. Albanian ophiolites. I—Magmatic and metamorphic processes associated with the initiation of a subduction. Ofioliti, 25, 39-46. BECCALUVA, L., COLTORTI, M., PREMTI, L, SACCANI, E., SIENA, F. & ZEDA, O. 1994. Mid-ocean ridge and suprasubduction affinities in the ophiolitic belts of Albania. Ofioliti, 19, 77-96. BIZIMIS, M., SALTERS, V.J.M. & BONATTI, E. 2000. Trace and REE content of clinopyroxenes from supra-subduction zone peridotites. Implications for melting and enrichment processes in island arcs. Chemical Geology, 165, 67-85. CAROSI, R., CORTESOGNO, L., GAGGERO, L. & MARRONI, M. 1996. Geological and petrological features of the metamorphic sole from the Mirdita nappe, northern Albania. Ofioliti, 21, 21-40. CIRIC, A. & ERIC, V. 1996. Contact metamorphism
B R E Z O V I C A U L T R A M A F I C MASSIF, SERBIA beneath the peridotite of Borov Vrh-Brezovica (Serbia). In: KNEZEVIC, V. & KRSTIC, B. (eds) Terranes of Serbia. Barex, Belgrade, 259-264. DICK, H.J.B. 1989. Abyssal peridotites, very slow spreading ridges and ocean ridge magmatism. In: SAUNDERS, A.D. & NORRY, M.J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications, 42, 71-105. DICK, H.J.B. & BULLEN, T. 1984. Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology, 86, 54-76. DICK, H.J.B. & FISHER, R.L. 1984. Mineralogic studies of the residues of mantle melting: abyssal and alpine-type peridotites. In: KORNPROBST, J. (ed.) Kimberlites II. The Mantle and Crust-Mantle Relationships. Elsevier, Amsterdam, 295-308. DICK, H.J.B. & NATLAND, J.H. 1996. Late-stage melt evolution and transport in the shallow mantle beneath the East Pacific Rise. In: MEVEL, C., GILLIS, K.M., ALLAN, J.F. & MEYER, P.S. (eds) Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, 147, 103-134. DICK, H.J.B., FISHER, R.L. & BRYAN, W.B. 1984. Mineralogical variability of the uppermost mantle along mid-ocean ridges. Earth and Planetary Science Letters, 73, 91-104. DJORDJEVIC, P., JOVANOVIC, L., KARAMATA, S., KURAT, G., MAJER, V. & NTAFLOS, TH. 1987. Thermal and tectonic history of some ultramafic rocks of Brezovica—the Livad mass, Yugoslavia. Ofioliti, 12, 245. DMITRIEV, L.V., BAZYLEV, B.A., SILANTYEV, S.A., BORISOV, M.V., SOKOLOV, S.Yu. & BOUGAULT, H. 1999. Hydrogen and methane formation during serpentinization of mantle ultramafic rocks of ocean and oil origin. Russian Journal of Earth Sciences, 1(6), 511-519 (in Russian). EVANS, B.W. 1977. Metamorphism of alpine peridotite and serpentinite. Annual Review of Earth and Planetary Sciences, 5, 397-447. HELLEBRAND, E., SNOW, J.E., DICK, H.J.B. & HOFMANN, A. 2001. Coupled major and trace elements as indicators of the extent of melting in mid-oceanridge peridotites. Nature, 410, 677-681. ISHII, T., ROBINSON, P.T., MAEKAWA, H. & FISKE, R. 1992. Petrological studies of peridotites from diapiric serpentinite seamounts in the Izu-Ogasawara— Mariana forearcs. In: FRYER, P., PEARCE, J.A., STOKKING, L.R. ET AL. (eds) Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, 125, 445486. JAEGER, J.C. 1968. Cooling and solidification of igneous rocks. In: HESS, H.H. & POLDERVAART, A. (eds) Basalts. Interscience, New York, 2, 503-536. KARAMATA, S. 1968a. Ultrabasic (ultramafic) intrusions of Dinarides. Zapiski Vsesoyuznogo Mineralogicheskogo Obschestva, XCVII, 49-54 (in Russian). KARAMATA, S. 1968b. Zonality in contact metamorphic rocks around the ultramafic mass of Brezovica (Serbia, Yugoslavia). In: Proceedings of the XXIIIth International Geological Congress, Prague. 1, 197207.
107
KARAMATA, S. 1985. Metamorphism in the contact aureole of Brezovica (Serbia, Yugoslavia) as a model of metamorphism beneath obducted hot ultramafic bodies. Bulletin XCX de I'Academic Serbe des Sciences et des Arts, Classe des Sciences Naturelles et Mathematique, 26, 51-68. KARAMATA, S. & KRSTIC, B. 1996. Terranes of Serbia and neighbouring areas. In: KNEZEVIC, V. & KRSTIC, B. (eds) Terranes of Serbia. Barex, Belgrade, 25-40. KARAMATA, S. & MILOVANOVIC, D. 1990. Metamorphism beneath the Brezovica ultramafics in the Ostrovica area. In: XII Kongres geologia Jugoslavije, II, Ohrid. 290-300 (in Serb.-Croat.). KARAMATA, S., MAJER, V. & PAMIC, J. 1980. Ophiolites of Yugoslavia. Ofioliti, Tethian ophiolites (special issue), 1, 105-125. KARAMATA, S., KORIKOVSKIY, S.P., BORONIHI, V.A. & ZINOVYEVA, N.G. 1991. Equilibria in metapsammites and mica gneisses from a contact aureole in a basement of ophiolitie ultramafic massif Brezovica (south Serbia). Izvestiya Akademii Nauk SSSR, Seriya Geologicheskaya, 10, 58-70 (in Russian). KARAMATA, S., DIMITRIJEVIC, M.D. & DIMITRIJEVIC, M.N. 2000. A correlation of ophiolitie belts and oceanic realms of the Vardar zone and the Dinarides. In: Proceedings of the International Symposium 'Geology and Metallogeny of the Dinarides and the Vardar Zone'. Academy of Sciences and Arts of the Republic of Srpska, Collections and Monographs, Department of Natural, Mathematical and Technical Sciences, 1, 191-194. KELEMEN, P.B., DICK, H.J.B. & QUICK, J.E. 1992. Formation of harzburgite by pervasive melt/rock reaction in the upper mantle. Nature, 358, 635-641. KELEMEN, P.B., SHIMIZU, N. & SALTERS, VJ.M. 1995. Extraction of mid-ocean ridge basalts from the upwelling mantle by focused flow of melt in dunite channels. Nature, 375, 747-753. LANGMUIR, C.H., KLEIN, E. & PLANK, T. 1992. Petrological systematics of mid-ocean ridge basalts: constraints on melt generation beneath ocean ridges. In: PHIPPS MORGAN, J., BLACKMAN, D.K. & SINTON, V.M. (eds) Mantle Flow and Melt Generation at Mid-Ocean Ridges. Geophysical Monograph, American Geophysical Union, 71, 183-280. LUGOVIC, B., ALTHERR, R., RACZEK, I., HOFMANN, A.W. & MAJER, V. 1991. Geochemistry of peridotites and mafic igneous rocks from the Central Dinaric Ophiolite Belt, Yugoslavia. Contributions to Mineralogy and Petrology, 106, 201-216. McDoNOUGH, W.F. & SUN, S.S. 1995. The composition of the Earth. Chemical Geology, 120, 223-253. NICOLAS, A. & BOUDIER, F. 1999. Slow spreading accretion and mantle denudation in the Mirdita ophiolite (Albania). Journal of Geophysical Research, 104, 15155-15167. PAMIC, J. 1983. Considerations on the boundary between Iherzolite and harzburgite subprovinces in the Dinarides and northern Hellenides. Ofioliti, 8, 153-164. PARKINSON, I.J. & PEARCE, J.A. 1998. Peridotites from the Izu-Bonin-Mariana Forearc (ODP Leg 125): evidence for mantle melting and melt-mantle interaction in a supra-subduction zone setting. Journal
108
B. A. BAZYLEV ETAL.
of Petrology, 39, 1577-1618. ROBERTSON, A.H.F. & KARAMATA, S. 1994. The role of subduction-accretion processes in the tectonic evolution of the Mesozoic Tethys in Serbia. Tectonophysics, 234, 73-94. ROBERTSON, A. & SHALLO, M. 2000. Mesozoic-Tertiary tectonic evolution of Albania in its regional Eastern Mediterranean context. Tectonophysics, 316, 197— 254. ROKSANDIC, M. 1974. Application of results of geophysical explorations in studying of structural features and evolution of Dinarides. Glas SANU CCXCII, Prirodny i Matematicny Nauke, Beograd, 38, 61-95 (in Serb.-Croat.). SHALLO, M. 1996. Relationships between Albanian ophiolites and Korab-Pelagonian microplate. In: KNEZEVIC, V. & KRSTIC, B. (eds) Terranes of
Serbia. Barex, Belgrade, 231-236. SHIBATA, T. & THOMPSON, G. 1986. Peridotites from the Mid-Atlantic Ridge at 43°N and their petrogenetic relations to abyssal tholeiites. Contributions to Mineralogy and Petrology, 93, 144-159. SNOW, I.E. & DICK, H.J.B. 1995. Pervasive magnesium loss by marine weathering of peridotite. Geochimica et Cosmochimica Ada, 59, 4219-4235. WELLS, P.R.A. 1977. Pyroxene thermometry in simple and complex systems. Contributions to Mineralogy and Petrology, 62, 129-139. WENNER, D.B. & THAYLOR, H.P. JR 1974. D/H and O18/ O16 studies of serpentization of ultramafic rocks. Geochimica et Cosmochimica Acta, 38, 1255-1286. WICKS, F.J. & PLANT, A.G. 1979. Electron-microprobe and X-ray microbeam studies of serpentine textures. Canadian Mineralogist, 17, 785-830.
Triassic mid-ocean ridge basalts from the Argolis Peninsula (Greece): new constraints for the early oceanization phases of the Neo-Tethyan Pindos basin EMILIO SACCANI 1 , ELISA PADOA 2 & ADONIS PHOTIADES 3 1
Dipartimento di Scienze della Terra, Universita di Ferrara, C.so Ercole I d'Este 32, 44100 Ferrara, Italy (e-mail:
[email protected]) 2 Dipartimento di Scienze della Terra, Universita di Firenze, Via G. La Pira 4, 50121 Florence, Italy 3 'Institute of Geology and Mineral Exploration (IGME), 70 Messoghion Str., 11527 Athens, Greece Abstract: The Middle Unit of the central-northern Argolis Peninsula, in NE Peloponnesus (Greece), is composed of several tectonic slices, locally including intact sequences of mafic volcanic rocks topped by radiolarian cherts. Although some of these sequences are Jurassic in age, many of them display a Triassic age based on biostratigraphical evidence. The petrological studies presented in this paper indicate that the Triassic volcanic rocks were generated in a mid-ocean ridge setting, and that they represent the oldest remnants of the Pindos oceanic crust so far recognized in the Subpelagonian zone. On the basis of immobile trace element analyses, two chemically distinct groups of Triassic lavas can be recognized in the various volcanic sequences. One group is represented by transitional-type mid-ocean ridge basalts (T-MORBs) displaying moderate light rare earth element (LREE) enrichment, and incompatible element abundances very similar to those observed in present-day T-MORBs. The other group exhibits a range of characteristics typical of many normal-type MORBs: that is, variable LREE depletion and flat N-MORB normalized patterns of incompatible element abundance. Moreover, many geochemical characteristics indicate that the various N-MORB type volcanic sequences originated from chemically distinct (heterogeneous) sub-oceanic mantle sources. Analogous to similar basalts from ophiolitic melanges of the Dinaride-Hellenide belt, the T-MORBs from the Argolis Middle Unit are interpreted as having originated from a primitive mantle source variably enriched by an ocean-island basalt (OIB)-type component. In contrast, the contemporaneous occurrence of N-MORBs implies that, during the Mid-Late Triassic, oceanic spreading of the Pindos basin had already reached, at least in some sectors, a quasi-steady state involving only sub-oceanic mantle sources and their partial melt derivatives. Our model for the Triassic opening of the Pindos oceanic basin and its related tectonomagmatic evolution is largely supported by comparison with the Red Sea embryonic ocean, a modern analogous setting.
The Pindos ocean is one of the Neo-Tethyan basins in the Eastern Mediterranean that developed during the early Mesozoic along the northern margin of Gondwanaland (Robertson et al 1991). Remnants of the Pindos basin are widely preserved in late Mesozoic-Cenozoic accretionary complexes in the Mirdita-Subpelagonian zone of the Dinaride-Hellenide belt and are mainly represented by complex tectonosedimentary associations of Triassic rift-related volcanic rocks (Pe-Piper 1998, and references therein), Middle Jurassic mid-ocean ridge (MOR) and suprasubduction-zone (SSZ) ophiolitic sequences (Jones & Robertson 1991; Capedri et al. 1996), Permian-Jurassic marginal and platform-related sedimentary rocks and Juras-
sic-Tertiary trench-type sedimentary rocks (Clift & Robertson 1989; Robertson 1994; Degnan & Robertson 1998). The nature and composition of these various rock types suggest a general geodynamic evolution of the Pindos basin characterized by Late Permian-Triassic rifting phases between the Apulian and Pelagonian microplates, followed by Jurassic spreading of the Pindos Neo-Tethyan oceanic basin and subsequent development of intra-oceanic convergent zones, as well as MidLate Jurassic generation of oceanic lithosphere in a suprasubduction setting (Jones & Robertson 1991; Doutsos et al 1993). A Triassic age for the beginning of spreading in the Hellenide sector of the Pindos ocean has been
From: DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 109-127. 0305-8719/037$ 15 © The Geological Society of London 2003.
110
E. SACCANI ETAL.
proposed by some workers (Ferriere 1982; Robertson et al 1991; Robertson et al 1996; PePiper 1998). However, this conclusion is generally based on limited data regarding Triassic radiolarites, which are commonly tentatively associated with basaltic sequences mainly showing withinplate (alkaline) affinity or, subordinately, ranging from transitional within-plate to transitional midocean ridge basalt (T-MORB) compositions. Moreover, a number of available biostratigraphical and radiometric ages (Spray et al. 1984; Bebien et al. 2000) indicate that the normal mid-ocean ridge magmatism largely developed during the Jurassic. Consequently, the possible Triassic beginning of oceanic development within the Hellenide sector of the Neo-Tethys is still poorly constrained. The Middle Unit of the central-northern Argolis (eastern Peloponnesus, Greece) is represented by a composite tectonic association of various types of thrust sheets, some of which include coherent sequences of basalts topped by radiolarian cherts. Cherts indicate two age ranges: MidLate Triassic (Bortolotti et al. 2001, 2002) and Mid-Late Jurassic (Baumgartner 1985). The Jurassic volcanic rocks are represented by MORB and are related to the magmatic activities that developed during the evolution of the Jurassic Pindos oceanic lithosphere (Baumgartner 1985; Dostal et al. 1991). By contrast, the nature of the Triassic basalts is still unknown. The main purpose of this paper is to present the petrological and geochemical characteristics of the Triassic basalts from the Middle Unit of the central-northern Argolis Peninsula to constrain their tectonomagmatic implications during the early stages of oceanic spreading in the Pindos basin.
Geology of the central-northern Argolis Peninsula The Argolis Peninsula (Fig. 1), in the southern sector of the Dinaride-Hellenide belt, is represented by a composite tectonic complex that was assembled during the closure of the Pindos basin, mainly from the Late Jurassic to the Miocene. Three main tectonostratigraphic units can be recognized in the central-northern Argolis Peninsula. The Lower Unit is represented by Mesozoic continental sequences (mainly made up of Middle Triassic-Early Jurassic carbonate successions), which represent both the subsiding continental platform of Apulia (Pantokrator Unit: Baumgartner 1985; Clift & Robertson 1990) and riftrelated intra-platform basins (Asklipion Unit: Clift & Robertson 1990). These sequences are overlain
by Oxfordian-Kimmeridgian siliceous mudstones and radiolarian cherts (Angelokastron and Koliaki Cherts: Baumgartner 1980, 1985, 1987), which, in turn, are stratigraphically followed by highly sheared ophiolite-derived sedimentary rocks (Lower Ophiolitic Unit or 'volcano-sedimentary ophiolitic succession': Photiades 1986, 1989). These rocks consist of coarsening upward turbidites, as well as ophiolitic sandstones and breccias (Late Oxfordian-Early Kimmeridgian Dhimaina Fm: Baumgartner 1985), followed by disorganized ophiolitic olistostromes (Potami Fm: Baumgartner 1985). The olistostromes include rounded fragments of boninitic lavas and coarse-grained boninitic-type rocks, set in an arenitic matrix consisting of various ophiolitic clasts, as well as fragments originating from the underlying limestones and cherts. The boninitic lavas and boninitic-type rocks originated in a subduction-related environment, possibly in an intra-oceanic islandarc setting (Dostal et al. 1991; Capedri et al. 1996). The Middle Unit consists of several imbricated tectonic slices, which can be roughly subdivided into two main types: (1) tectonic sheets mainly consisting of basic volcanic rocks, which locally have serpentinite slivers at their base and are associated with radiolarian cherts (Jurassic Migdhalitsa ophiolite unit: Baumgartner 1985); (2) tectonic sheets preserving well-developed stratigraphic successions, which consist of Albian-Cenomanian neritic limestones, talus-breccias rich in clasts of basalt and chert cemented by Campanian-Maastrichtian hemipelagic limestones, PaleoceneMiddle Eocene pelagic to reefal limestones ('mesoautochthonous sequence': Baumgartner 1985; Photiades & Skourtsis-Coroneou 1994), and finally a post-Ypresian flysch. The Upper Unit tectonically overlies the Middle Unit, and is almost always found in thrust contact over the post-Ypresian flysch. It is composed of a polymictic ophiolitic melange at its base, and Cretaceous limestones in its upper part (Photiades 1986). The ophiolitic melange contains metresized lens-shaped blocks, mainly including subduction-related ophiolitic rocks (serpentinized harzburgites, dunites, subordinate basalts and boninites: Dostal et al. 1991), cherts, carbonates, quartz greywackes, marbles, amphibolites and micaschists.
Structure and stratigraphy of the Middle Unit The Middle Unit of the central-northern Argolis Peninsula crops out mainly in the synclinal structures of Dhimaina and Lygourio-Palea Epidavros
TRIASSIC BASALTS OF PINDOS BASIN, GREECE
111
Fig. 1. Structural zones of the Albanide-Hellenide Alpine orogenic belt (modified after Robertson & Shallo 2000). The locations of the study areas expanded in Figure 2 are also indicated.
(Fig. 2a), and in the Vothiki-Radho graben (Fig. 2b). The tectonic emplacement of the Middle Unit over the Lower Unit is generally represented by a refolded thrust surface, probably related to a Late Jurassic-Early Cretaceous tectonic phase (Aubouin et al 1970). By contrast, the overthrusting of the Upper Unit onto the Middle Unit marks a subsequent, post-Eocene tectonic phase (Photiades 1986). The Middle Unit is a composite
tectonic unit consisting of several imbricated thrust sheets separated by shear zones. The basal and central parts of the Middle Unit consist predominantly of sheets of volcanic rocks, whereas wedges of Cretaceous-Eocene sedimentary successions are prevalent at the top of the unit. Contacts between the volcanic rocks and the Cretaceous-Eocene sedimentary sequences generally appear tectonic; no clear evidence of strati-
112
E. SACCANI ETAL.
Fig. 2. Simplified geological maps showing the main outcrops of the Middle Unit of the northern (a) and central (b) Argolis Peninsula.
TRIASSIC BASALTS OF PINDOS BASIN, GREECE graphic relationships can be observed in the field. None the less, previous workers (Baumgartner 1985; Photiades & Skourtis-Coroneou 1994) have suggested that the Cretaceous-Eocene sequence represents a diachronous, transgressive (i.e. 'mesoautochthonous') sedimentary package, deposited upon the volcanic sequences after their early emplacement onto the continental Lower Unit. In the lower and central parts of the Middle Unit, the intense fragmentation resulted in development of tectonic slivers of large blocks of various lithotypes, including mafic volcanic rocks, cherts and subordinate serpentinites, as well as marbles and limestones. These limestone blocks have been considered 'exotic' by Baumgartner (1985), and their occurrence further supports the hypothesis of a tectonic nature for the Middle Unit. The volcanic sequences include pillow lavas, massive lavas and pillow breccias, and are generally dismembered by several major thrust sheets that are, in turn, affected by second-order internal thrusts, shear zones and overturned folds. Many outcrops of pillow lavas and pillow breccias are spatially associated with ore deposits of Mn, FeCu and Ba, probably generated during sea-floor hydrothermal metamorphism (Photiades 1986; Robertson et al. 1987; Photiades & Economou 1991). Relatively continuous volcanic sequences, locally topped by radiolarian cherts, are preserved in places. Some of these basaltic sequences display a MORB magmatic affinity (Baumgartner 1985; Dostal et al. 1991) and are associated with Kimmeridgian-Tithonian radiolarian cherts (Baumgartner 1985). Consequently, the overall lower and central part of the Middle Unit has been regarded as a Jurassic ophiolitic unit (i.e. Migdhalitsa Unit: Baumgartner 1985). None the less, a number of basaltic sequences topped by Ladinian-Norian ribbon-radiolarian cherts have recently been found (Bortolotti et al. 2001, 2002), commonly in an overturned setting, as slices in the Middle Unit. Although the nature of these volcanic sequences is poorly understood, the occurrence of tectonic sheets composed of both Triassic volcanic chert and Jurassic ophiolitic sequences further supports the tectonic nature of the Middle Unit.
Sampling and methods Location of samples Analyses were carried out on the basaltic rocks from the Middle Unit of the northern (Fig. 2a) and central (Fig. 2b) Argolis areas. In particular, sampling was performed in those outcrops where
113
clear stratigraphic relationships between volcanic rocks and Triassic radiolarian cherts (Bortolotti et al. 2001, 2002) were observed. Various samples were collected from each outcrop to achieve maximum diversity between samples along the volcanic sections. The sampled stratigraphic sequences commonly appear partially dismembered by local shear zones. None the less, samples were also taken from tectonized sections, but only where small degrees of displacement were recognized, and where field relationships were clearly established to make sure that the whole sequence could be ascribed to a single thrust sheet. According to these criteria, seven sections were chosen (Figs 2 and 3) for sampling the mafic extrusive rocks overlain by Triassic radiolarian cherts. These sections are located: (1) to the south of Dhimaina, along the road to Lyghourio (sections GR195, GR56 and GR51); (2) to the west of Palea Epidavros, along the road to Lyghourio (sections GR47 and GR50); (3) in the southern part of Vothiki village (sections GR71 and GR181). The lithostratigraphy of the measured logs and the position of the collected samples inside each volcanic sequence are indicated in Figure 4, together with the position of the dated radiolarian cherts. The sampled volcanic facies include mafic pillowed (samples GR 47b, 51a, 51b, 56a, 195a, 71a, 71b, 71c, 71d, 181a, 181b, 181d) and massive flows (samples GR 56b, 56c, 181c), as well as pillow breccias and hyaloclastites (samples GR 47a, 50c, 50d, 50e, 56d). According to Bortolotti et al. (2001, 2002), sections GR71, GR181 and GR195 cannot be assigned to a precise age, but the radiolarian fauna are sufficiently preserved to allow their attribution to a Triassic sensu latu age. By contrast, the ages of sections GR47, GR50, GR51 and GR56 are well constrained by radiolarian dates, and range from Ladinian to Norian.
Analytical methods Samples were analysed for major and some trace elements (Ni, Co, Cr, y Rb, Sr, Ba, Th, Nb, Zr, Y, Zn) by X-ray fluorescence (XRF) using pressedpowder pellets (Table 1). The matrix correction methods proposed by Franzini et al. (1972) were applied. Accuracy is better than 2% for major oxides, and better than 5% for trace element determinations. The detection limits for trace elements range from 1 to 2 ppm. Both accuracy and detection limits were determined using results from international standards. Thirteen representative samples (Table 1) were chosen for additional trace-element analyses, including rare earth elements (REE), Sc, Nb, Hf, Ta,
114
E. SACCANI ETAL.
Fig. 3. Geological cross-sections of the northern and central Argolis Peninsula. No vertical exaggeration. (See Fig. 2 for legend and locations.)
Th and U, using inductively coupled plasma-mass spectrometry (ICP-MS). The precision and accuracy of the data were evaluated using results for international standard rocks, duplicate runs of several samples, and the blind standards included in the sample set. Accuracy varies from 1 to 8%, whereas detection limits (in ppm) are: 0.29 for Sc;
0.02 for Nb, Hf and Ta; 3500 km2) ultramafic massif. Harzburgite and cpx-poor Iherzolite are the dominant lithologies and isotopic data for these rocks (Miller et al. 2003) suggest that peridotite melting occurred during the Early Jurassic. Ultramafic rocks in this region are locally cut by pegmatitic gabbronorite and rare normal mid-ocean ridge basalt (N-MORB) type tholeiitic basalt dykes for which Miller et al. (2003) reported a Late Jurassic 40Ar/39Ar age of 152 ±33 Ma. Additional outcrops of ultramafic rocks occur further NW in the Kiogar region, where no reports of the upper levels of any ophiolitic succession exist and the most detailed descriptions of these rocks are from reconnaissance studies only (Gansser 1964). All the ophiolitic rocks lie in the hanging wall of southdirected thrusts, which transported them over rocks of Indian affinity (Gansser 1964). The Jungbwa ophiolite has been thrust 30-40 km south of the YTSZ (sensu stricto) and lies atop passive margin sediments of the Indian terrane (Murphy & Yin, 2003). The Kiogar ophiolite has
been thrust a similar distance southwards over the Indian terrane and development of extensive mudmatrix melange (Gansser 1964) may have accompanied its emplacement. The northern side of the YTSZ (sensu stricto) in the Kailas region is marked by the north-directed South Kailas thrust along which zones of sheared serpentinite and ophiolitic melange occur (Yin et al. 1999).
Dazhuqu terrane: Renbung-Quxu Outcrops of ophiolitic rocks, which occur as blocks in a serpentinite-matrix melange, can be traced eastwards intermittently from Renbung to west of Lhasa airport. The largest zone of outcrop is located on the SW side of the Yarlung Tsangpo to the south of a 6126 m peak on which Miocene Gangrinboche facies conglomerates lie unconformably upon Lhasa terrane rocks (Aitchison et al. 2002b). South of Quxu, serpentinite-matrix melange with large blocks of ultramafic rocks crops out together with diabase dykes and basalts where the YTSZ passes through a narrow col (Jiangdanyako, 29°18'N, 090°43'E) between Lhasa terrane granites and the Indian terrane. The ophiolitic rocks lie within shear zones associated with the north-directed Renbu-Zedong thrust system.
Ophiolitic rocks: Zedong-Luobusa A further major occurrence of ophiolitic rocks along the YTSZ crops out to the SE of Lhasa near
YARLUNG TSANGPO SUTURE ZONE OPHIOLITES Zedong and Luobusa. The largest chromite deposit (Fig. 4g) in China is currently being worked at Luobusa and rocks there have been the subject of numerous investigations (Badengzhu 1979, 1981; Huang et al. 1981; Bai et al. 1993, 2000; Zhou & Robinson 1994; Zhou 1995; Zhou et al, 1996, 2002; Griselin et al 1999; Hu 1999; Hebert et al. 2000, 2001). The Luobusa ophiolite is dominated by a harzburgitic mantle section that has experienced Late Miocene India-Asia collision-related north-directed thrusting over a dunite transition zone and ophiolitic melange (Zhou et al. 1996). Lenses and pods of dunite are abundant within the ultramafic section and these rocks contain the chromite mineralization. Restricted occurrences of gabbros are also known but no diabase dykes have been reported from Luobusa. Together, the entire package has been thrust northwards over Lower Miocene Luobusa conglomerates (Gangrinboche facies) on the southern margin of the Lhasa terrane (Aitchison et al. 2002b) and is itself overthrust by Triassic Indian terrane rocks. The original structural relationship between the ophiolitic rocks and the Indian terrane is not preserved. The geochemistry and petrology of the Luobusa ophiolite are distinctive. Spinels from the Luobusa massifs have Cr numbers as high as 0.69-0.94 and olivines are more forsteritic than elsewhere along the YTSZ (Wang et al. 1999; Hebert et al. 2000, 2001). Such values are similar to those for rocks produced from boninitic melts and are considerably more refractory than for rocks from elsewhere along the suture. One of the most unusual features of the Luobusa ophiolite is the reported occurrence of small amounts of microdiamonds, graphite, SiC and other rare minerals found in association with chromitites (Bai et al. 1993, 2000; Hu 1999). Ophiolitic rocks at Luobusa and Zedong are intimately associated with island arc tholeiitic rocks of the Zedong terrane. At Luobusa, this terrane is limited to a small zone of overturned variolitic pillow basalts, which are stratigraphically overlain by red mudstones. Outcrop is much more extensive near Zedong, where the terrane consists of a sequence in which island arc tholeiitic pillow basalts are overlain by a thin succession of red radiolarian cherts, which are themselves overlain by an up to 1 km thick pile of shoshonitic autoclastic breccias (McDermid 2002). The exact nature of the original relationship between these two terranes remains indeterminate, as all contacts are faulted. Nevertheless, limited age data from the Luobusa ophiolite (Sm-Nd age of 177 ± 31 Ma on gabbro dykes given by Zhou et al. 2002) are similar to U-Pb, Ar-Ar and fossilbased ages for rocks of the Zedong terrane (McDermid 2002; McDermid et al 2002). The
159
relationship of ophiolitic rocks in the ZedongLuobusa district to other ophiolitic rocks along the suture remains ambiguous. Although ophiolitic rocks in the Zedong to Luobusa area lie in an identical structural position along the YTSZ, both the compositions (Wang et al 1999; Hebert et al 2000, 2001) and ages of these rocks (Ziabrev 2001; McDermid 2002; McDermid et al 2002; Ziabrev et al 2003b) differ from those of other ophiolitic rocks of the YTSZ. Further work is clearly required to resolve this enigma. We note reports of similar ages for the oldest ophiolitic rocks of the Spontang ophiolite in Ladakh, NW India (Pedersen et al 2001), and suggest that it is entirely possible that remnants of more than one intra-oceanic island arc are preserved between India and Asia. Emplacement
Kinematic indicators along the southern margin of the Dazhuqu terrane ophiolites near Bainang, SE of Xigaze, indicate its southward emplacement onto the feather-edge of the northern Indian continent (Girardeau et al 1985c; Ratschbacher et al. 1994). Ophiolites further west along the suture have similarly been emplaced southwards onto the northern Indian passive margin. Precise constraints on the timing of this emplacement event have proved difficult to obtain with seemingly contradictory datasets. Traditionally the ages of amphibolite-facies metamorphic rocks within melange zones at the base of ophiolitic successions have been interpreted as indicative of emplacement events. In this case, Late Cretaceous ages for amphibolites from near Bainang (Wang et al. 1987) would seemingly indicate obduction at that time. A tectonic event of this nature, however, seems inconsistent with any events recorded in regional sedimentary successions, which should reflect such an event. An alternative interpretation of the Late Cretaceous metamorphic ages known from various localities along the suture within Tibet (Zhou cited by Aitchison et al 2000) and further afield in NW India and Pakistan (Searle et al 1999) is that they represent an ophiolitespecific thermal event such as the subduction of a spreading centre (Shervais 2001). The timing of emplacement of ophiolitic rocks onto the northern margin of India is perhaps better constrained from the sedimentary record within, and south of, the suture. Distal Indian passive margin sediments in the northern Tethyan Himalayan zone, north of Qomolangma (Mt. Everest) record a significant tectonic disturbance and the development of a major disconformity around the CretaceousPaleogene boundary (Wan et al 2002) before resumption of normal carbonate accumulation in a
160
J. C. AITCHISON £T,4L.
quiescent environment until the time of IndiaAsia collision. Along the suture itself, ophiolitic rocks as well as strata of the Indian and Bainang terranes are unconformably overlain by Paleogene Liuqu conglomerates. These coarse clastic sediments are interpreted to have accumulated during the collision of the ophiolitic terrane and the Indian continental margin, which lay to its south (Davis et al. 2002). Other Lower Miocene conglomerates along the suture contain clasts from all nearby terranes, including those north of the suture, and constrain the timing of collision between India and Asia (Aitchison et al. 2002b). Regional development of mud-matrix melange was extensive along the northern Indian margin in the Paleogene (Gansser 1964; Shackleton 1981; Liu & Einsele 1996; Liu 2001). These melanges possibly developed in response to ophiolite obduction and are analogous to similar rocks in contemporary arc-continent collision zones such as Timor (Barber et al. 1986) and Taiwan (Chang et al. 2000).
Discussion As a result of detailed investigations along the YTSZ over the past few years, it is now possible to provide a better understanding of ophiolites in the region. Improved precision for Upper Mesozoic radiolarian biostratigraphy (O'Dogherty 1994; Baumgartner et al. 1995) allows more precise age constraints to be placed on any interpretations of the development of this region. Radiolarian ages from sediments immediately overlying the ophiolite constrain the timing and duration of the ophiolite generation event. Biostratigraphic investigations of numerous sections indicate that the Dazhuqu terrane ophiolite formed in the Barremian (mid-Cretaceous) with accumulation of volcaniclastic sediments continuing into the Aptian (Zyabrev et al. 1999; Ziabrev 2001; Ziabrev et al. 2003b). Results of palaeomagnetic investigations indicate generation of the Dazhuqu terrane ophiolite at equatorial to low northern latitudes at least 1000-1500 km south of Asia's margin (Abrajevitch et al. 2001, 2003; Abrajevitch jevitch 2002). Some fragments of the ophiolite have experienced a counter-clockwise rotation and individual ophiolitic massifs appear to have different travel histories. Although many workers had previously inferred depositional continuity between ophiolitic rocks of the Dazhuqu terrane and volcaniclastic turbidites of the Xigaze terrane, detailed investigations of all known localities where this relationship has been inferred unequivocally reveal that this contact is tectonic. Sedimentary rocks overlying the ophiolite are volcaniclastic, as are those in the Xigaze terrane,
but the sources of the two suites of rocks were different. Furthermore, the structural evolution of both terranes was remarkably different and there is no evidence that these two entities shared a common history before India-Asia collision. Present interpretation of the Dazhuqu terrane ophiolite as having an origin within a suprasubduction zone setting is supported by detailed mineralogical and geochemical studies in the Xigaze area (Wang et al. 1987; Hebert et al. 2000, 2001). The close spatial association of these rocks with the subduction complex assemblage of the Bainang terrane suggests the existence of a southfacing intra-oceanic island arc at near equatorial latitude within Neotethys during the mid-Cretaceous (Aitchison et al. 2000). As more data become available from along the YTSZ, interpretation of this zone becomes increasingly complicated. Ages for ophiolitic assemblages in essentially the same structural positions at Xigaze, Jungwa and Zedong are distinctly different. It now seems that the YTSZ contains remnants of at least one, and possibly two, intra-oceanic island arc systems. Given the complexity of analogous modern settings such as the western Pacific, it is clear that early interpretations were oversimplified. Multiple arcs may have coexisted in a wide Tethys with no single simple convergent margin. It is also probable that plate boundaries were not necessarily parallel to one another. The suture extends from one side of Tibet to the other and beyond. Despite a few detailed studies of isolated occurrences, the Yarlung Tsangpo ophiolites have received remarkably little attention relative to their significance in interpretation of the evolution of Tethys and the India-Asia collision. Continuing Tibet research at the University of Hong Kong is supported by grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project Nos. HKU7102/98P, 7299/99P and 7069/0IP). We thank colleagues in the Geological Society of Tibet for their assistance in arranging the logistics related to this research. The authors gratefully acknowledge the constructive reviews of Y. Dilek and P. Robinson, which helped to improve the manuscript.
References ABRAJEVITCH, A. 2002. Paleomagnetism of the Dazhuqu Terrane, Yarlung Zangbo Suture Zone, Southern Tibet. MPhil thesis, University of Hong Kong. ABRAJEVITCH, A., AITCHISON, J.C. & ALI, J.R. 2001. Paleomagnetism of the Dazhuqu Terrane, Yarlung Zangbo Suture Zone, Southern Tibet. EOS Transactions, American Geophysical Union Fall Meeting Supplement, GP11A-0188. ABRAJEVITCH, A., AITCHISON, J.C. & ALI, J.R. 2002. Paleomagnetism of the Dazhuqu Terrane, Yarlung Zangbo Suture Zone, Southern Tibet. In: Tectonic
YARLUNG TSANGPO SUTURE ZONE OPHIOLITES Processes in the Evolution of China, Croucher Advanced Studies Institute Abstracts. University of Hong Kong, Hong Kong. ABRAJEVITCH, A., AITCHISON, J.C., ALI, J.R., BADENGZHU,, DAVIS, A.M., Liu, J.B. & ZIABREV, S.V. in press. Neotethys and the India-Eurasia collision— new insights following a palaeomagnetic investigation of the Dazhuqu Terrane (southern Tibet). Geophysical Journal International. AITCHISON, J.C. & DAVIS, A.M. 2001. Orogenic conglomerates indicate timing of collision in Tibet. Journal of Asian Earth Sciences, 19(3 A), 1-2. AITCHISON, J.C., BADENGZHU, & DAVIS, A.M. ET AL. 2000. Remnants of a Cretaceous intra-oceanic subduction system within the Yarlung-Zangbo suture (southern Tibet). Earth and Planetary Science Letters, 183, 231-244. AITCHISON, J.C., ABRAJEVITCH, A. & ALI, J.R. ET AL. 2002a. New insights into the evolution of the Yarlung Tsangpo suture zone, Xizang (Tibet), China. Episodes, 25(3), 90-94. AITCHISON, J.C., DAVIS, A.M., BADENGZHU, & Luo, H. 2002b. New constraints on the India-Asia collision: The Lower Miocene Gangrinboche conglomerates, Yarlung Tsangpo suture zone, SE Tibet. Journal of Asian Earth Sciences, 21(3), 253-265. AITCHISON, J.C., DAVIS, A.M., BADENGZHU, & Luo, H. 2003. The Gangdese thrust: a phantom structure that did not raise Tibet. Terra Nova, 15, 155-162. ALLEGRE, C.J., COURTILLOT, V. & TAPPONNIER, P. ET AL. 1984. Structure and evolution of the Himalaya-Tibet erogenic belt. Nature, 307(5946), 17-22. BADENGZHU (Compiler) 1979. Xizang Autonomous Region Zhanang—Sangri Regional Geology Reconnaissance Map 1:50,000. Xizang Geological Survey Geological Team No. 2, Group 2, Lhasa. BADENGZHU (Compiler) 1981. Xizang Autonomous Region Sangri—Jiacha Regional Geology Reconnaissance Map 1:50,000. Xizang Geological Survey Geological Team No. 2, Lhasa. BAI, W., ZHOU, M.F. & ROBINSON, P.T. 1993. Possibly diamond-bearing mantle peridotites and podiform chromitites in the Luobusa and Donqiao ophiolites, Tibet. Canadian Journal of Earth Sciences, 30(8), 1650-1659. BAI, W., ROBINSON, P.T. & FANG, Q. ET AL. 2000. The PGE and base-metal alloys in the podiform chromitites of the Luobusa ophiolite, southern Tibet. Canadian Mineralogist, 38(3), 585-598. BARBER, A.J., TJOKROSAPOETRO, S. & CHARLTON, T.R. 1986. Mud volcanoes, shale diapirs, wrench faults, and melanges in accretionary complexes, eastern Indonesia. AAPG Bulletin, 70(11), 1729-1741. BASSOULLET, J.P., COLCHEN, M., MASCLE, G., MATTE, P. & WANG, N.W. 1984. Les ensembles sedimentaires de la zone du Tsangpo (Lhaze, Lhasa, Linzhu). In: MERCIER, J.L. & Li, G.C. (eds) Mission Franco-Chinoise au Tibet 1980. Etude geologique et geophysique de la croute terrestre et du manteau superieur du Tibet et de I'Himalaya. CNRS, Paris, 133-153. BAUMGARTNER, P.O., O'DOGHERTY, L., GORICAN, S., URQUHART, E., PILLEVUIT, A. & DE WEVER, P.
161
(eds) 1995. Middle Jurassic to Lower Cretaceous Radiolaria of Tethys: Occurrence, Systematics, Biochronology. Memoires de Geologic, Universite de Lausanne, 23. Bureau of Geology and Mineral Resources of Xizang Autonomous Region 1993. Regional Geology of Xizang (Tibet) Autonomous Region. Peoples Republic of China, Ministry of Geology and Mineral Resources Geological Memoirs Series, Geological Publishing House, Beijing, 1, 31. Bureau of Geology and Mineral Resources of Xizang Autonomous Region Geological Team No. 2. 1979. Regional Geological Report for Zhanang-Sangri County, Xizang (Tibet) Autonomous Region, 10, 188. BURG, J.P. 1983. Tectogenese comparee de deux segments de chaine de collision:—Le Sud du Tibet (Suture du Tsangpo)—La Chaine Hercynienne en Europe (sutures du Massif Central). Doctorat d'Etat thesis, Universite des Sciences et Techniques du Languedoc, Montpellier. BURG, J.P. & CHEN, G.M. 1984. Tectonics and structural zonation of southern Tibet, China. Nature, 311(5983), 219-223. CHANG, C.P., ANGELIER, J. & HUANG, C.Y. 2000. Origin and evolution of a melange: the active plate boundary and suture zone of the Longitudinal Valley, Taiwan. Tectonophysics, 325, 43-62. COULON, C, MALUSKI, H., BOLLINGER, C. & WANG, S. 1986. Mesozoic and Cenozoic volcanic rocks from central and southern Tibet: 39Ar-40Ar dating, petrological characteristics and geodynamic significance. Earth and Planetary Science Letters, 79, 281-302. DANELIAN, T. & ROBERTSON, A.H.F. 1997. Radiolarian evidence for the stratigraphy and palace-oceanography of the deep-water passive margin of the Indian Plate (Karamba Formation, Indus suture zone, Ladakh Himalaya). Marine Micropaleontology, 30(1-3), 171-195. DAVIS, A.M., AITCHISON, J.C., BADENGZHU,, Luo, H., MALPAS, J. & ZYABREV, S. 1999. Eocene obliqueslip basin development, Tibet: terrane tracks on the roof of the world. In: EVENCHICK, C.A., WOODSWORTH, GJ. & JONGENS, R. (eds) Terrane Paths 99. Circum Pacific Terrane Conference, Okanagan Valley, B.C., Canada. Circum Pacific Terrane Conference Abstracts and Program, 28. DAVIS, A.M., AITCHISON, J.C., BADENGZHU, & Liu, H. 2001. Late Cretaceous-Paleocene island arc collision related conglomerates, Yarlung-Tsangpo suture zone, Tibet. Journal of Asian Earth Sciences, 19(3 A), 13. DAVIS, A.M., AITCHISON, J.C., BADENGZHU,, Luo, H. & ZYABREV, S. 2002. Paleogene island arc collisionrelated conglomerates, Yarlung-Tsangpo suture zone, Tibet. Sedimentary Geology, 150(3/4), 247273. DURR, S.B. 1996. Provenance of Xigaze fore-arc basin clastic rocks (Cretaceous, South Tibet). Geological Society of America Bulletin, 108(6), 669-684. EINSELE, G., Liu, B. & DURR, S. ET AL. 1994. The Xigaze forearc basin; evolution and facies architecture (Cretaceous, Tibet). Sedimentary Geology, 90,
162
J. C. AITCHISON £T,41.
1-32. GAETANI, M. & GARZANTI, E. 1991. Multicyclic history of the northern India continental margin (Northwestern Himalaya). AAPG Bulletin, 75(9), 14271446. GANSSER, A. 1964. The Geology of the Himalayas. Wiley-Interscience, New York. GIRARDEAU, J. & MERCIER, J.C.C. 1988. Petrology and texture of the ultramafic rocks of the Xigaze Ophiolite (Tibet); constraints for mantle structure beneath slow-spreading ridges. Tectonophysics, 147(1-2), 33-58. GIRARDEAU, J., MARCOUX, J. & ALLEGRE, CJ. ET AL. 1984a. Tectonic environment and geodynamic significance of the Neo-Cimmerian Donqiao Ophiolite, Bangong-Nujiang suture zone, Tibet. Nature, 307(5946), 27-31. GIRARDEAU, J., MARCOUX, J. & ZAO, Y. 1984b. Lithologic and tectonic environment of the Xigaze Ophiolite (Yarlung Zangbo suture zone, southern Tibet, China), and kinematics of its emplacement. Eclogae Geologicae Helvetiae, 77(1), 153-170. GIRARDEAU, J., MERCIER, J.C. & XIBIN, W. 1985a. Petrology of the mafic rocks of the Xigaze ophiolite, Tibet; implications for the genesis of the oceanic lithosphere. Contributions to Mineralogy and Petrology, 90(4), 309-321. GIRARDEAU, J., MERCIER, J.C.C. & ZAO, Y.G. 1985b. Origin of the Xigaze Ophiolite, Yarlung Zangbo suture zone, southern Tibet. Tectonophysics, 119, 407-433. GIRARDEAU, J., MERCIER, J.C.C. & ZAO, Y.G. 1985c. Structure of the Xigaze Ophiolite, Yarlung Zangbo suture zone, southern Tibet, China; genetic implications. Tectonics, 4(3), 267-788. GIRARDEAU, J., NICOLAS, A. & MARCOUX, J. ET AL. 1984c. Les ophiolites de Xigaze et la suture du Yarlung Zangbo (Tibet). In: MERCIER, J.L. & Li, G.C. (eds) Mission Franco-Chinoise au Tibet 1980. Etude geologique et geophysique de la croute terrestre et du manteau superieur du Tibet et de I'Himalaya. CNRS, Paris, 189-197. GRADSTEIN, P.M., AGTERBERG, P.P., OGG, J.G., HARDENBOL, J., VAN VEEN, P., THIERRY, J. & HUANG, Z.
1994. A Mesozoic time scale. Journal of Geophysical Research, 99B, 24051-24074. GRISELIN, M., DA VIES, G.R. & PEARSON, D.G. 1999. REE and Sr-Pb-Nd isotope geochemistry of Tibetan peridotites: implications for melting processes. Ofioliti,24(lA), 102-103. HARRISON, T.M., YIN, A., GROVE, M., LOVERA, O.M., RYERSON, FJ. & ZHOU, X. 2000. The Zedong Window: a record of superposed Tertiary convergence in southeastern Tibet. Journal of Geophysical Research, 105(B8), 19211-19230. HEBERT, R., BEAUDOIN, G., VARFALVY, V., HUOT, P., WANG, C.S. & Liu, Z.F. 2000. Yarlung Zangbo ophiolites, southern Tibet revisited. Earth Science Frontiers, 1, 124-126. HEBERT, R., WANG, C., VARFALVY, V., HUOT, P., Liu, Z., BEAUDOIN, G. & DOSTAL, J. 2001. Yarlung Zangbo Suture Zone ophiolites and their suprasubduction zone setting. Journal of Asian Earth Sciences, 19(3A), 27-28.
Hu, X.F. 1999. Origin of diamonds in chromitites of the Luobusa Ophiolite, southern Tibet, China. MSc thesis, Dalhousie University, Halifax, N.S. HUANG, Y.H., ZHON, X.z. & WANG, X.B. 1981. Pumpellyite from the contact metamorphic zone of ultramafic bodies of Luobusha Ophiolite suite, South Tibet. Bulletin de Mineralogie, 104(4), 441-444. JADOUL, P., BERRA, P. & GARZANTI, E. 1998. The Tethys Himalayan passive margin from Late Triassic to Early Cretaceous (South Tibet). Journal of Asian Earth Sciences, 16, 173-194. Li, H.S. & Wu, H.R. 1985. Radiolaria from the Cretaceous Congdu Formation in southern Xizang (Tibet). Acta Micropalaeontologica Sinica, 2(1), 61-76. Liu, G. 1992. Permian to Eocene sediments and Indian passive margin evolution in the Tibetan Himalayas. Tubinger Geowissenschaftliche Arbeiten, 13, 1268. Liu, G.H. & EINSELE, G. 1996. Various types of olistostromes in a closing ocean basin, Tethyan Himalaya (Cretaceous, Tibet). Sedimentary Geology, 104, 203-226. Liu, J.B. 2001. Yamdrok Melange, Gyantze district, Xizang (Tibet), China. MPhil thesis, University of Hong Kong. Liu, J.B. & AITCHISON, J.C. 2002. Upper Paleocene radiolarians from the Yamdrok melange, south Xizang (Tibet), China. Micropaleontology, 48(Supplement 1), 145-154. Liu, J.B., AITCHISON, J.C., BADENGZHU,, DAVIS, A.M. & ZIABREV, S.V. 2002. Origin and tectonic significance of the Yamdrok Melange, South Xizang (Tibet), China. In: Tectonic Processes in the Evolution of China, Croucher Advanced Studies Institute Abstracts. University of Hong Kong, Hong Kong. MALUSKI, H., PROUST, P. & XIAO, X.C. 1982. 39Ar/40Ar dating of the trans-Himalayan calc-alkaline magmatism of southern Tibet. Nature, 298(5870), 152154. MARCOUX, J., DE WEVER, P. & NICOLAS, A. ET AL. 1982. Preliminary report of depositional sediments on top of volcanic member; Xigaze Ophiolite (Yarlung Zangbo suture zone); South Xingang (Tibet). Ofioliti, 2(3), 395-396. MATSUOKA, A., KOBAYASHI, K., NAGAHASHI, T., YANG, Q., WANG, Y.J., TAKEI, M. & ZENG, Q.G. 2001a. Early Middle Jurassic (Aalenian) radiolarian fauna from the Xialu chert in the Yarlung Zangbo Suture Zone, southern Tibet. In: METCALFE, I., SMITH, J.M.B., MORWOOD, M. & DAVIDSON, I. (eds) Faunal and Floral Migrations and Evolution in SE Asia-Australasia. Balkema, Lisse, 105-110. MATSUOKA, A., YANG, Q., KOBAYASHI, K., TAKEI, M., NAGAHASHI, T., ZENG, Q.G. & WANG, YJ. 2001b. Jurassic-Cretaceous radiolarian biostratigraphy and sedimentary environments of the Ceno-Tethys: records from the Xialu Chert in the Yarlung-Zangbo Suture Zone, southern Tibet. Journal of Asian Earth Sciences, 20(3), 277-287. MCDERMID, I., AITCHISON, J.C., BADENGZHU, & DAVIS, A.M. 200la. The Zedong Terrane: an intra-oceanic magmatic arc assemblage, Tibet. Journal of Asian Earth Sciences, 19(3A), 44.
YARLUNG TSANGPO SUTURE ZONE OPHIOLITES MCDERMID, I., AITCHISON, J.C., DAVIS, A.M., HARRISON, T.M. & GROVE, M. 200Ib. The Zedong Terrane: a Jurassic intra-oceanic magmatic arc within the Yarlung-Zangbo Suture Zone of Southeastern Tibet. Journal of Asian Earth Sciences, 19(3A), 44-45. MCDERMID, I.R.C. 2002. Zedong terrane, south Tibet. PhD thesis, University of Hong Kong. MCDERMID, I.R.C., AITCHISON, J.C., DAVIS, A.M., HARRISON, T.M. & GROVE, M. 2002. The Zedong terrane: a Late Jurassic intra-oceanic magmatic arc within the Yarlung-Zangbo suture zone, southeastern Tibet. Chemical Geology, 187(3-4), 267-277. MILLER, C, THONI, M., FRANK, W., SCHUSTER, R., MELCHER, F., MEISEL, T. & ZANETTI, A. 2003. Geochemistry and tectonomagmatic affinity of the Yungbwa ophiolite, SW Tibet. Lithos, 66(3-4), 155-172. MURPHY, M.A. & YIN, A. 2003. Structural evolution and sequence of thrusting in the Tethyan foldthrust belt and Indus-Yalu suture zone, southwest Tibet. Geological Society of America Bulletin, 115(1), 21-34. MURPHY, M.A., YIN, A. & HARRISON, T.M. ET AL. 1997. Did the Indo-Asian collision alone create the Tibetan plateau? Geology, 25, 719-722. MURPHY, M.A., YIN, A. & KAPP, P. ET AL. 2002. Structural evolution of the Gurla Mandhata detachment system, southwest Tibet: implications for the eastward extent of the Karakoram fault system. Geological Society of America Bulletin, 114(4), 428-447. NICOLAS, A., GIRARDEAU, J. & MARCOUX, J. ET AL. 1981. The Xigaze ophiolite (Tibet); a peculiar oceanic lithosphere. Nature, 294(5840), 414-417. O'DOGHERTY, L. 1994. Biochronology and Paleontology of Mid-Cretaceous Radiolarians from Northern Apennines (Italy) and Betic Cordillera (Spain). Memoires de Geologic, Universite de Lausanne, 21, 1-415. PEARCE, J.A. & DENG, W.M. 1988. The ophiolites of the Tibet Geotraverses, Lhasa to Golmud (1985) and Lhasa to Kammandu (1986). Philosophical Transactions of the Royal Society of London, Series A, 327(1594), 215-238. PEDERSEN, R.B., SEARLE, M.P. & CORFIELD, R.I. 2001. U-Pb zircon ages from the Spontang Ophiolite, Ladakh Himalaya. Journal of the Geological Society, London, 158, 513-520. PINI, G.A. 1999. Tectonosomes and Olistostromes in the Argille Scagliose of the Northern Apennines, Italy. Geological Society of America, Special Papers, 225, 1-70. Pozzi, J.P., WESTPHAL, M., GIRARDEAU, J., BESSE, J., YAO, X.Z., XIAN, Y.C. & Li, S.X. 1984. Paleomagnetism of the Xigaze ophiolite and flysch (Yarlung Zangbo suture zone, southern Tibet); latitude and direction of spreading. Earth and Planetary Science Letters, 70(2), 383-394. PROUST, F., BURG, J.P., MATTE, P., TAPPONNIER, P., Li, T.D., Li, G.C. & CHEN, G.M. 1984. Succession des phases de plissement sur une transversale du Tibet meridional, implication geodynamiques. In: MERCIER, J.L. & Li, G.C. (eds) Mission Franco-Chinoise
163
au Tibet 1980. Etude geologique et geophysique de la croute terrestre et du manteau superieur du Tibet et de I'Himalaya. CNRS, Paris, 385-392. RATSCHBACHER, L., FRISCH, W. & Liu, G. 1994. Distributed deformation in southern and western Tibet during and after the India-Asia collision. Journal of Geophysical Research, 99(B10), 1991719945. ROBERTSON, A.H.F. 1998. Rift-related sedimentation and volcanism of the North Indian margin inferred from a Permian-Triassic exotic block at Lamayuru, Indus suture zone (Ladakh Himalaya) and regional comparisons. Journal of Asian Earth Sciences, 16(2-3), 159-172. ROBERTSON, A.H.F. & SHARP, I. 1998. Mesozoic deepwater slope rise sedimentation and volcanism along the North Indian passive margin; evidence from the Karamba Complex, Indus suture zone (western Ladakh Himalaya). Journal of Asian Earth Sciences, 16(2-3), 195-215. SEARLE, M.P., WINDLEY, B.F. & COWARD, M.P. ET AL. 1987. The closing of Tethys and the tectonics of the Himalaya. Geological Society of America Bulletin, 98(6), 678-701. SEARLE, M.P., KHAN, M.A., FRASER, J.E., GOUGH, S.J. & JAN, M.Q. 1999. The tectonic evolution of the Kohistan-Karakoram collision belt along the Karakoram Highway transect, north Pakistan. Tectonics, 18, 929-949. SHACKLETON, R.M. 1981. Structure of southern Tibet; report on a traverse from Lhasa to Khatmandu organised by Academia Sinica. Journal of Structural Geology, 3(1), 97-105. SHERVAIS, J.W. 2001. Birth, death, and resurrection: the life cycle of suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems, 2, 2000GC000080. TAPPONNIER, P., MERCIER, J.L. & PROUST, F. ET AL. 1981. The Tibetan side of the India-Eurasia collision. Nature, 294(5840), 405-410. WAN, X., WANG, L., WANG, C. & JANSA, L. 1998. Discovery and significance of Cretaceous fossils from the Xigaze forearc basin, Tibet. Journal of Asian Earth Sciences, 16(2-3), 217-223. WAN, X.Q., JANSA, L.F. & SARTI, M. 2002. Cretaceous and Paleogene boundary strata in southern Tibet and their implication for the India-Eurasia collision. Lethaia, 35, 131-146. WANG, C.S. & Liu, Z.F. ET AL. 1999. Xigaze Forearc Basin and Yarlung Zangbo Suture Zone, Tibet. Geological Publishing House, Beijing. WANG, C.S., Li, X.H., Hu, X.M. & JANSA, L.F. 2002. Latest marine horizon north of Qomolangma (Mt Everest): implications for closure of Tethys seaway and collision tectonics. Terra Nova, 14(2), 114-120. WANG, X.B., BAG, P.S. & XIAO, X.C. 1987. Ophiolites of the Yarlung Zangbo (Tsangbo) River, Xizang (Tibet). Publishing House of Surveying and Mapping, Beijing. WEI, B.J. & PENG, Y.H. (Compilers) 1984. Xizang Autonomous Region Xigaze-Quxu Geological Traverse Map 1:200,000. Xizang Geological Survey Geological Team No. 2, Lhasa. WIEDMANN, J. & DURR, S.B. 1995. First ammonites
164
J. C. AITCHISON ETAL.
from the Mid- to Upper Cretaceous Xigaze Group, South Tibet, and their significance. Newsletters on Stratigraphy, 32(1), 17-26. Wu, H.R. 1986. Some new genera and species of Cenomanian Radiolaria from southern Xizang (Tibet). Acta Micropalaeontologica Sinica, 3(4), 347— 360. Wu, H.R. 1993. Upper Jurassic and Lower Cretaceous radiolarians of Xialu chert, Yarlung Zangbo ophiolite belt, southern Tibet. In: BLUEFORD, J.R. & MURCHEY, B.L. (eds) Radiolaria of Giant and Subgiant Fields of Asia, Nazarov Memorial Volume. Micropaleontology Special Paper, 6, 115-136. XU, R.H.,
SCHAERER, U. &
ALLEGRE, CJ.
1985.
Magmatism and metamorphism in the Lhasa Block (Tibet); a geochronological study. Journal of Geology, 93(1), 41-57. YIN, A. & HARRISON, T.M. 2000. Geologic evolution of the Himalayan-Tibetan orogen. Annual Review of Earth and Planetary Sciences, 28, 211-280. YIN, A., HARRISON, T.M., RYERSON, F.J., CHEN, W.J., KIDD, W.S.F. & COPELAND, P. 1994. Tertiary structural evolution of the Gangdese thrust system in southeastern Tibet. Journal of Geophysical Research, 99, 18175-18201. YIN, A., HARRISON, T.M. & MURPHY, M.A. ET AL. 1999. Tertiary deformation history of southeastern and southwestern Tibet during the Indo-Asian collision. Geological Society of America Bulletin, 111(11), 1644-1664. YIN, J., Xu, J., Liu, C. & Li, H. 1988. The Tibetan Plateau; regional stratigraphic context and previous work. Philosophical Transactions of the Royal Society of London, Series A, 327(1594), 5-52. ZHANG, S.M. & Fu, X.L. (Compilers) 1982. Xizang Autonomous Region Xigaze—Saga Geological Traverse Map 1:200,000. Xizang Geological Survey Geological Team No. 2, Lhasa. ZHOU, M.F. 1995. Petrogenesis of the podiform chromitites in the Luobusa ophiolite, southern Tibet. PhD thesis, Dalhousie University, Halifax, N.S. ZHOU, M.F. & ROBINSON, P.T. 1994. Petrogenesis of podiform chromitites in the Luobusa Ophiolite, Tibet;
a melt-rock reaction model. Geological Association of Canada, Program with Abstracts, 19 125. ZHOU, M.F., ROBINSON, P.T., MALPAS, J. & Li, Z. 1996. Podiform chromitites in the Luobusa Ophiolite (southern Tibet); implications for melt-rock interaction and chromite segregation in the upper mantle. Journal of Petrology, 37(1), 3-21. ZHOU, S., Mo, X.X., MAHONEY, J.J., ZHANG, S.Q., Guo, TJ. & ZHAO, Z.D. 2002. Geochronology and Nd and Pb isotope characteristics of gabbro dikes in the Luobusha ophiolite, Tibet. Chinese Science Bulletin, 41(2}, 143-146. ZIABREV, S. 2001. Tectonic evolution of Dazhuqu and Bainang terranes, Yarlung Zangbo suture, Tibet as constrained by radiolarian biostratigraphy. PhD thesis, University of Hong Kong. ZIABREV, S., AITCHISON, J., BADENGZHU,, DAVIS, A.M., Luo, H. & Liu, J. 2000. Tethyan relics in the Yarlung-Tsangpo suture, Tibet: structural setting, radiolarian ages and their tectonic significance. In: CARTER, E.S., WHALEN, P., NOBLE, P.J. & CRAWFORD, A.EJ. (eds) Ninth Meeting, The International Association of Radiolarian Paleontologists INTERRAD 2000, Blairsden, CA. Program with Abstracts. 72. ZIABREV, S.V., AITCHISON, J.C., ABRAJEVITCH, A., BADENGZHU, DAVIS, A.M. & Luo, H. 2003a. Bainang Terrane, Yarlung-Tsangpo suture, southern Tibet (Xizang, China): a trace of intra-Tethyan subduction on the Roof of the World. Journal of the Geological Society, London. ZIABREV, S.V., AITCHISON, J.C., ABRAJEVITCH, A., BADENGZHU, DAVIS, A.M. & Luo, H. 2003b. Precise radiolarian age constraints on the timing of ophiolite generation and sedimentation in the Dazhuqu terrane, Yarlung-Tsangpo suture zone, Tibet. Journal of the Geological Society, London, 160, 591-599. ZIABREV, S.V., AITCHISON, J.C., BADENGZHU, ., DAVIS, A.M., Luo, H. & MALPAS, J. 1999. Radiolarian biostratigraphy of supra-ophiolite sequences in the Xigaze area, Yarlung-Tsangpo suture, Southern Tibet (preliminary report). Radiolaria, 17, 13-19.
Yarlimg Zangbo ophiolites (Southern Tibet) revisited: geodynamic implications from the mineral record REJEAN HEBERT 1 , FRANCOIS HUOT 1 , CHENGSHAN WANG 2 & ZHIFEI LIU 3 1
Departement de Geologic et de Genie Geologique, Universite Laval, Sainte-Foy, Quebec, Canada, G1K IP4 (e-mail:
[email protected]) 2 Institute of Sedimentary Geology, Chengdu University of Technology, Chengdu, Sichuan, 610059, PR. China 3 Department of Marine Geology and Geophysics, Tongji University, 1239 Siping Road, Shanghai 200012, P.R. China Abstract: We present mineral chemistry data and petrological evidence from the Yarlung Zangbo suture zone ophiolites (Southern Tibet) suggesting that they represent a collage of heterogeneous massifs. Mantle sections in these ophiolites consist of harzburgite and Iherzolite cut by several generations of gabbroic to diabasic intrusions, all affected by high-temperature deformation. Pyroxenitic bands are parallel to the mantle foliation. Crustal plutonic sections, consisting of dunite, wehrlite and gabbro, are thin or absent and have been observed only in the Dazhuqu massif. Plagioclase is an additional phase associated with crustal peridotites. The mineral chemistry of silicate minerals and spinel in the mantle and crustal rocks varies widely and is believed to reflect complex melt percolation and reaction. The massifs record polybaric exhumation steps from at least 50 km depth to the plagioclase stability field. Pyroxene has reequilibrated compositions from 1200°C down to medium-grade metamorphic conditions. The mantle peridotites are interpreted as the residues of 10-40% partial melting of a fertile Iherzolitic source. High Cr number, low TiC>2 content and relatively high Fe3+ number of spinels suggest that the ophiolitic massifs were generated in a suprasubduction zone (arc or back-arc) environment.
The Yarlung Zangbo suture zone (YZSZ) represents one of the major tectonic features of the Tibetan Plateau and records the collisional event between the Indian and Eurasian plates (Molnar & Tapponnier 1975; Gansser 1974; Fig. 1). Vestiges of the Tethyan oceanic domain, such as ophiolitic massifs, are partly preserved along the YZSZ. A better understanding of their origin and evolution from intra-oceanic to collisional settings is critical for development of plausible geodynamic models for the Mesozoic and Tertiary tectonics of the Tibetan Plateau. The common characteristics of the YZSZ ophiolitic massifs (Fig. 2), as defined by the collective results of the studies following the 1980 Sino-French Cooperative Investigation of the Himalayas (Wu & Deng 1980; Nicolas et al. 1981; Mercier & Li 1984; Girardeau et al. 1985a, 1985b), include: (1) a thin (2-4 km) crustal section; (2) the absence or low volume of crustal gabbroic components; (3) the occurrence of multiple gabbroic and diabasic dykes and sills that cut both the mantle and the crustal sections (Fig. 3).
The Xigaze ophiolites were interpreted to have formed at a slow-spreading mid-ocean ridge in a small basin located near the Eurasian palaeocontinent (Nicolas et al. 1981; Girardeau et al. 1984, 1985a, 1985c; Pozzi et al. 1984; Xiao 1980). The ophiolite belt formed some 120 ± 10 Ma ago as inferred from a U-Pb age (Gopel et al. 1984) from the Xigaze massif. Marcoux et al. (1982) published an Albian-Cenomanian age for radiolarian cherts overlying the Xigaze massif, whereas Ziabrev et al. (2000) reported an age of 121 Ma (Barremian) on Bainang radiolarian cherts. The ophiolite is believed to have been initially obducted southward during the Late Cretaceous-Eocene periods and backthrust in OligoMiocene times (Gansser 1974; Nicolas et al. 1981; Allegre et al. 1984). Subduction of the NeoTethys caused the development of the Gangdese continental arc on the southern margin of the Lhasa block at least from 95 to 40 Ma (Allegre et al. 1984; see also Wang et al. 2000 for a review) and possibly as early as 153 ± 6 Ma (Murphy et al. 1997). Part of the Gangdese arc was built on
From: DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 165-190. 0305-8719/03/$15 © The Geological Society of London 2003.
166
R. HEBERT ETAL.
Fig. 1. Location map of the Yarlung Zangbo ophiolite massifs, (a) Western part; (b) eastern part. Modified after Gansser (1991) and Wang et al (1999). QT, Qiahgtang Terrane; LT, Lhasa Terrane; BNSZ, Baugong-Nujrang Suture Zone; YZSZ, Varlung Zangbo Suture Zone; MCT, Main Central Thrust; MET, Main Boundary Thrust.
an older metamorphosed ophiolitic crust (Proust et al. 1984) possibly accreted by the Late Jurassic to Early Cretaceous (Wang et al. 2000). As many as four compressional phases together with younger strike-slip and east-west extensional episodes have been recorded (Tapponier et al. 1981), all of which result from the closing of the Tethyan and Neo-Tethyan domains, and subsequent collision between Eurasia and India. These deformational episodes led to the partial dismemberment, faulting and shortening of the ophiolitic sequences (Girardeau et al. 1984). We conducted fieldwork in the eastern segment of the YZSZ during 1998 and 1999 to reassess its geology and to shed new light on the preserved fragments of Neo-Tethyan ocean floor (Hebert et al. 1999, 2000, 2001b, 2001c; Huot et al. 2002). Progress in ophiolite research and developments from modern oceanic crust studies guided us in the fieldwork. The investigated segment extends from Jiding to Luobusa, which comprises a strip of 250 km length. This work significantly extends the area of previous investigations on the Xigaze ophiolite, which herein we call the YZSZ ophiolites. The aim of this paper is to present the mineral chemistry of peridotites and some mafic plutonic rocks of the Yarlung Zangbo ophiolites and to discuss their petrogenetic and geodynamic significance. We investigated six lithologically and chemically distinct massifs along the YZSZ; these are, from west to east: Jiding, Beimarang, Qunrang,
Dazhuqu, Jinlu and Luobusa (Fig. 1). These massifs are named according to the nearest village and/or river valley to provide a geographical reference for each studied section.
Overview of the crustal unit of the YZSZ ophiolites The ophiolitic belt is clearly heterogeneous along strike and reveals a complex geological history. No single stratigraphic column is representative of the belt as a whole. The massifs are overprinted by late-stage deformational structures such as thrust, backthrust and strike-slip faults, folds and shear zones that make primary features difficult to decipher. Here we present important features of two highly distinct massifs (Fig. 2). The uppermost portion of the crustal unit, which crops out on the northern part of the ophiolites, is represented by a volcanic sequence and associated radiolarites. According to Tapponier et al (1981) and Marcoux et al. (1982), the ophiolitic volcano-sedimentary unit is stratigraphically overlain by the Xigaze Group, a sequence consisting of turbidites deposited in a forearc basin south of the Gangdese arc. On the other hand, Aitchison et al. (2000) strongly favoured a fault contact between these units. The volcanic products in most massifs include massive and pillowed basalts together with their autobrecciated facies. Volcaniclastic deposits are only locally
Fig. 2. (continued
overleaf).
Fig. 2. (continued overleaf}.
MINERAL CHEMISTRY OF TIBETAN OPHIOLITES
169
Fig. 2. Geological maps of the studied ophiolite massifs showing sample locations, (a) Segment from Tiding to Beimarang massifs; (b) Segment from Xialu to Dazhuqu massifs. Modified after Wang et al. (1984); (c) Jinglu (Zedang) massif. Modified after Aitchson et al (2000).
present in the ophiolitic sequence but, as reported by Aitchison et al. (2000), are dominant in the Zedong terrane. This Early Cretaceous terrane is composed of island arc volcanic and volcaniclastic rocks having compositions ranging from basaltic andesite to rare dacite. The Zedong terrane is thought to be technically juxtaposed (Aitchison et al. 2000) with the clinopyroxene-phyric basalts belonging to the Zedang ophiolitic volcanic sec-
tion (Jinlu massif in this paper). In the YZSZ ophiolitic belt, diabasic sills and rare dykes locally intrude the volcanic pile and generally increase downward, passing into a sill complex. Among the studied massifs, Qunrang has relatively thick volcanic and diabasic sections (1-2 km). The plutonic section is rather thin and even absent in the YZSZ ophiolitic belt (e.g. Qunrang). According to Nicolas et al. (1981), the thin plutonic
170
R. H E B E R T ^ r ^ L .
Fig. 3. Idealized cross-sections of the Beimarang and Dazhuqu massifs showing similarities and differences along the YZSZ ophiolite belt. Modified after Girardeau et al. (1985a, 1985b).
section reflects a low magma supply and is not a consequence of tectonic thinning. In the study area, the Dazhuqu massif is the only one with a plutonic section (0.5 wt.%; and ±5% for trace elements. The ICP-MS analyses yield accuracies better than ±5%. Major
oxides are normalized to 100% on a volatile-free basis.
Major element oxides and trace elements The basaltic volcanic rocks from the various ophiolite localities show a number of similar geochemical features. They all exhibit relatively low contents of Ni, Cr, Ti, Zr, Nb and Hf, but high, although variable, alkalis, Ba, Sr, U, Th and V relative to mid-ocean ridge basalt (MORB) (Table 2). They plot in an AFM diagram with an arc-tholeiite trend (Fig. 4) and in the island-arc tholeiite field on the Zr v. Zr/Y diagram of Pearce & Norry (1979) (Fig. 5). Their chondrite-normalized REE show moderate light REE (LREE) depletion (Fig. 6), spanning the range typical of MORB (Saunders 1984). There are, however, some differences between the lavas from Luobusa and the other massifs. The pillow lava blocks found in the melange at the base of the Luobusa massif show a more gentle negative slope in the LREE than the lavas and dykes from the other ophiolites, suggesting derivation from a less depleted mantle source. In the trace element spider diagrams (Fig. 7), the ophiolitic basaltic rocks from Zedong, Dazhuqu and Xigaze show concentrations of large ion lithophile elements up to 10 times MORB, clear depletion of Nb, and high field strength element concentrations of 0.3-1 times MORB. These patterns indicate that these basalts are island-arc tholeiites generated in a suprasubduction environment. The Luobusa samples, in contrast, have a much reduced Nb anomaly and generally higher contents of high field strength elements, supporting their generation from a less depleted source. All of these basalts are very different from the analysed samples from the Zedong terrane (Table 2), which are typical intra-oceanic arc andesites, with distinct trace element signatures as shown in Figures 6 and 7.
Geochronology Previous studies of the ophiolites along the IYS suggest that subduction took place during the Mid-Cretaceous (Coulon et al. 1986) or Late Cretaceous (Allegre et al. 1984; Dewey et al. 1989; Yin et al. 1994), and that the Neo-Tethyan ocean was finally closed during Paleogene continental collision (Molnar & Tapponnier 1975; Patriat & Achache 1984; Rowley 1996). Biostratigraphic dates have been obtained from supraophiolitic radiolarites in Xigaze (Zyabrev et al. 2002) suggesting this ophiolite formed in the MidCretaceous (Barremian). Radiometric ages of 177 ±31 Ma from diabase dykes cutting the cumulate section and Rb-Sr ages of 173 ± 11 Ma
Table 2. Major oxides (wt.%) and trace elemental abundances (ppm) of rocks from the Indus-Yarlung Zangbo Suture Zone Zedong ophiolite
Dazhuqu ophiolite
Xigaze ophiolites
Luobusa ophiolite
Zedong ;indesites
Sample
MRZ4
MRZ5
MRZ9
MRZ10
Bl
B2
B3
B4
Y7
Y10
Yll
Y15
Y16
Y17
D2
D9
MZ3
MZ8
SiO2 TiO2 A1203 Fe203 MnO MgO CaO Na20 K2O P205 Rb Ba Th U Nb Ta Sr Zr Hf Y V Cr Ni Cu Zn Sc La Ce Pr Nd Sm Eu Gd Tb
52.5 0.39 16.8 10.5 0.21 6.9 6.1 5.18 0.29 0.03 1.7 13 0.04 0.02 0.18 0.09 123 13 0.48 11 211 21 22 132 111 42 0.44 1.41 0.27 1.77 0.86 0.4 1.36 0.25 1.91 0.43 1.30 0.19 1.38 0.2
51.1 0.78 16.9 9.6 0.17 8.5 6.3 4.26 0.57 0.06 3.7 14 0.01 0 0.19 0.11 153 31 1.22 18 211 286 86 92 66 35 0.61 2.65 0.63 4.45 1.87 0.58 2.76 0.49 3.23 0.71 2.07 0.30 2.05 0.28
50.0 0.48 17.0 9.6 0.16 8.9 8.0 4.20 0.06 0.06 0.31 5 0.02 0.02 0.32 0.05 81 23 0.74 11 217 164 58 95 66 38 0.72 2.44 0.49 2.96 1.07 0.5 1.54 0.28 1.97 0.44 1.29 0.19 1.38 0.18
46.1 0.42 14.6 9.9 0.25 11.0 7.2 3.68 0.49 0.04 5.3 30 0.01 0.01 0.12 0.02 151 8 0.57 10 212 158 54 97 49 40 0.55 1.86 0.37 2.25 0.83 0.32 1.25 0.22 1.64 0.37 1.05 0.16 1.10 0.17
46.9 2 15.2 12.8 0.21 5.5 14.2 2.56 0.67 0.2 40 57 0.41 0.18 5.39 1.88 132 58 2.20 32
47.0 1 16.4 10.1 0.19 5.0 16.2 3.27 0.27 0.1 4.6 16 0.14 0.06 1.7 1.18 73 48 1.86 30
45.8 1 16.6 8.2 0.15 6.4 21.8 0.11 0.00 0.1 0.17 16 0.10 0.05 5.8 2.98 109 51 1.54 18
49.7 2 14.5 11.5 0.18 6.8 11.0 3.85 0.56 0.2 11 35 0.19 0.09 2.1 1.06 57 56 2.23 35
6.7 17 2.74 13.4 4.09 1.59 5.61 0.90 6.06 1.25 3.68 0.51 3.34 0.49
3.3 9.8 1.72 9.31 3.29 1.07 4.86 0.84 5.42 1.16 3.68 0.51 3.28 0.47
1.9 6.0 1.05 5.57 2.16 0.91 2.95 0.51 3.42 0.75 2.24 0.31 1.94 0.32
4.4 13 2.36 12.6 4.17 1.56 6.02 0.99 6.58 1.38 4.15 0.57 3.69 0.53
50.6 1.04 13.6 10.7 0.18 8.4 7.1 3.23 1.07 0.11 7.0 14 0.06 0.02 0.83 0.18 164 72 1.63 24 263 117 44 53 62 36 2.62 8.68 1.59 8.56 3.11 1.04 3.96 0.69 4.64 0.94 2.84 0.39 2.75 0.38
50.6 0.72 13.9 9.1 0.17 10.8 9.2 2.66 0.88 0.07 5.4 7 0.03 0.01 0.58 0.18 195 34 1.03 16 226 346 114 102 57 34 1.23 4.13 0.78 4.68 1.74 0.62 2.42 0.43 2.94 0.65 1.88 0.27 1.91 0.26
50.2 0.94 13.0 10.2 0.17 10.0 8.7 3.04 0.22 0.11 1.1 4 0.04 0.03 0.72 0.14 169 59 1.73 22 270 350 104 10 44 35 2.19 7.06 1.28 7.46 2.61 0.88 3.47 0.6 3.98 0.86 2.6 0.37 2.64 0.37
48.2 1.70 14.7 9.8 0.17 9.3 9.2 5.42 0.75 0.19 8.9 45 0.13 0.35 2.02 0.23 220 137 3.82 43 247 31 17 71 1001 31 4.40 14.35 2.73 15.23 5.28 1.73 6.67 1.15 7.65 1.61 4.71 0.69 4.77 0.66
50.8 0.97 16.2 9.5 0.15 5.4 12.2 4.03 0.64 0.10 10.1 23 0.16 0.1 0.93 0.31 255 59 1.94 21 224 244 51 52 69 35 2.05 6.42 1.21 6.98 2.47 0.93 3.24 0.56 3.75 0.8 2.35 0.34 2.36 0.34
51.5 1.05 16.3 9.7 0.13 7.0 6.9 5.11 0.18 0.12 1.7 40 0.07 0.06 1.11 0.19 180 74 2.3 24 224 182 54 31 58 38 2.54 8.23 1.49 8.31 2.9 1.04 3.77 0.65 4.49 0.97 2.73 0.39 2.71 0.37
51.1 0.86 16.1 9.2 0.14 7.8 7.5 4.69 0.02 0.10 0.09 3 0.06 0.03 0.75 0.12 98 59 1.51 21 211 107 47 70 74 34 1.90 6.33 1.15 6.38 2.28 0.84 3.10 0.55 3.77 0.8 2.32 0.33 2.27 0.33
51.7 0.49 16.3 6.1 0.12 9.4 11.1 2.42 1.57 0.04 5.7 12 0.02 0.02 0.36 0.1 272 23 0.68 12 196 647 107 84 35 45.3 0.95 2.73 0.57 3.46 1.35 0.54 1.88 0.35 2.35 0.49 1.4 0.21 1.38 0.19
57.3 0.69 17.2 7.3 0.09 4.3 6.1 3.96 3.64 0.39 46 868 3.88 0.86 2.68 0.41 712 151 1.98 17 197 60 19 52 29 21 24.2 57.4 8.19 39.5 7.73 1.74 5.63 0.64 3.42 0.61 1.63 0.21 1.45 0.18
57.8 0.84 15.8 9 0.14 5.5 7.2 2.89 1.84 0.35 27 226 2.83 0.65 2.59 0.56 619 107 1.33 19 257 54 20 115 59 29 14.8 34.9 4.99 22.6 5.2 1.34 4.53 0.64 3.82 0.76 2.09 0.29 1.99 0.28
Dy Ho Er Tm Yb Lu
198
J. MALPAS ETAL.
Fig. 4. AFM diagram of the basaltic rocks from Dazhuqu terrane ophiolite massifs. FeO*, total iron as FeO.
Fig. 5. Plot of Zr v. Zr/Y for basaltic rocks from the Dazhuqu terrane ophiolites. Fields of basalts from the various tectonic settings are from Pearce & Norry (1979). Symbols as in Figure 4.
for the pillow basalts of the Luobusa ophiolite (Zhou et al 2002), indicate that this massif may be somewhat older than Xigaze. We here report a number of additional age dates from a variety of rocks along the suture zone.
Analytical methods Dating of zircon was carried out by SHRIMP techniques whereas Ar/Ar analyses were carried out on hornblende and biotite. For the SHRIMP analyses, zircons were separated using conventional heavy liquid and magnetic techniques, and cathodoluminescence images were obtained using
Fig. 6. (a, b) Chondrite-normalized REE distribution patterns of basaltic rocks from the Dazhuqu terrane ophiolites. Symbols as in Figure 4. (c) Andesites from the Zedong terrane.
a Philips XL30 scanning electron microscope to investigate their internal structures. The instrumental techniques for isotopic analysis of zircons using the SHRIMP II ion microprobe at Curtin University of Technology are similar to those of Compston et al. (1984). Pb/U ages are based on a value of 564 Ma determined by conventional U-Pb analysis of the standard zircon CZ3. Uncertainties of 207Pb/206Pb ages are independent of the standard analyses, but
ORIGIN AND EMPLACEMENT OF YARLUNG ZANGBO OPHIOLITES
199
and uncertainties in mean ages are quoted at the 95% confidence level (2o). For the Ar/Ar dating, hornblende and biotite were handpicked from heavy liquid separates. The samples were irradiated in the McMaster University nuclear reactor and argon isotopic analyses were undertaken using a VG 3600 mass spectrometer coupled to an internal tantalum resistance furnace of the double-vacuum type, at Dalhousie University, Canada. Hornblende MMhb-1, with an assumed age of 520 ± 2 Ma (Samson & Alexander 1987) was used as a standard for all analyses. Other experimental procedures follow those described by Muecke et al. (1988). Errors are the 2o analytical uncertainties.
Dating results
Fig. 7. Spider diagrams of MORE-normalized trace element abundances in basaltic rocks from the Dazhuqu terrane (a, b) and andesites from the Zedong terrane (c). Symbols as in Figure 6.
are sensitive to the common Pb correction in lowU zircons that have been calculated for zircons that are < 1000 Ma. Thus, the 206pb_238u age is normally preferred. Common Pb was corrected using the 204 method discussed by Compston et al. (1984). U, Th and Pb concentrations were calculated using the methods given by ClaoueLong et al. (1991). Individual analyses (Table 3) are presented as lo error boxes on concordia plots
A quartz diorite, D13, from the cumulate section of the Dazhuqu massif, contains zircons with a variety of textures and morphologies. Most grains show distinct zoning but their internal complexity does not affect the ages obtained from different grains, whether from cores and rims, or from high- and low-U regions. All crystals, even those of different shape, give the same age within the uncertainties. The mean 206pb/238U age is 126 ±2 Ma, where the uncertainty is the 2a error on the mean. The results are concordant, with the mean 207Pb/235U age being 126 ± 3 Ma. The 207 Pb/206Pb age, which is very poorly defined in such young zircons, is 122 ± 47 Ma. The data are consistent with a single age population of zircons from D13. The x2 value for the 206Pb/238U age is 3.34, and for the 207Pb/235U age it is 0.46. The former value suggests there may be a very small amount of geological scatter in the 206Pb/238U ages. This is a result of the lowest-U zircons having 206Pb/238U ages around 107 Ma (see Table 3). These ages occur within both the cores and rims of the zircons analysed. The younger ages may reflect the difficulty of making 204Pb corrections for low-Pb samples as the mean 204Pb is likely to be an overestimate as a Poisson statistic. This problem would result in slightly lower 206Pb/ 238 U ages, and very young or negative 207pb/206Pb ages, which is the case for these few analyses. The rare low-U analyses do not perturb the mean age obtained for the high-U analyses. Based on the low x2 value of the 207Pb/235U age, it is assumed that the zircons are from a single Gaussian distribution. Figure 8 gives concordia plots of the data and shows a single group of concordant zircons. The low-U zircons have been removed so that the data for the high-U, lower uncertainty grains can be more easily seen. There is no evidence of any resetting of the ages since 126 Ma, thus the
Table 3. SHRIMP zircon analytical results for zircons from quartz diorite D13, from Dazhuqu massif, southern Tibet (uncertainties are lo) Spot
1 2 3 5 6 7 8 10 11 12 13 14 15 16 17 18 19
U (ppm)
341 50 288 221 565 92 82 56 430 830 795 149 483 130 76 124 264
Th (ppm)
854 31 1109
282 1308
202 205 172 2005 2080 1551
585 830 235 179 186 683
Pb
11 2 11 6 17 3 3 2 18 26 23 6 13 6 3 3 8
% cone.
94 0 53 126 78 0 0 0 143 179 315 0 102 109 0 0 0
Ages (Ma) 206pb_238U
±
207pb/235U
±
207pb/206pb
±
208pb/232U
±
129 107 124 126 127 119 108 108 127 130 130 128 125 124 117 127 123
2 10 2 2 2 3 5 5 2 2 2 3 2 5 5 3 2
130 66 129 125 129 83 59 38 125 127 125 115 124 124 48 113 101
9 159 3 17 9 40 77 83 3 5 6 29 3 71 72 39 21
138 0 232 100 162 0 0 0 89 72 41 0 122 114 0 0 0
155 46 49 309 164 82 53 92 38 95 92 51 39
125 63 121 125 124 112 100 110 122 123 127 133 118 120 106 130 123
3 105 2 6 3 8 14 13 2 2 2 5 2 17 14 11 4
1012
73 68 54
ORIGIN AND EMPLACEMENT OF YARLUNG ZANGBO OPHIOLITES
201
were formed at a convergent plate margin during the closure process. The formation age of the ophiolites in Xigaze was previously reported as 120 ± 10 Ma using conventional 238U/206Pb methods (Gopel et al. 1984) and 109 ± 21 Ma by the Nd/Sm method (Prinzhofer 1987, cited by Nicolas 1989). The SHRIMP date of 126 Ma for the Dazhuqu massif reported here is considered the most reliable age of formation of the Xigaze ophiolite. It is in accord with the Barremian biostratigraphic age reported by Zyabrev et al. (2002). However, evidence from the Luobusa and Zedong massifs suggests that these ophiolites are older (c. 175 Ma). Fig. 8. SHRIMP U-Pb zircon concordia plot for sample D13, a quartz diorite from the Dazhuqu massif. Amphibolites in the melange zones occur as large rafts directly below the ultramafic sections of the ophiolites, are basaltic in character, and are complex zonation of the zircons is interpreted as similar to metamorphic soles beneath ophiolites deuteric interaction of zircon and fluids during elsewhere. Because there is no other known source cooling and crystallization rather than later meta- for these rocks and they appear, particularly near morphic processes. Xigaze, to be disrupted locally, we interpret them A hornblende andesite from the Zedong terrane, to be fragments of once coherent metamorphic MZ2, has an Ar/Ar plateau age of 127.9 ± soles. Ar/Ar ages from amphiboles and biotites of 0.33 Ma (Fig. 9). When 40Ar/36Ar is plotted v. 94.8 Ma and 79.3 Ma, respectively, from amphibo39 Ar/36Ar, the age is 126.5 ± 0.35 Ma. lite blocks at Luobusa, and 87.9 Ma from the Samples BO-17 and LW1-2 are amphibolites amphibolites at Xigaze suggest that initial displafrom the melange zone at the base of the Luobusa cement of the ophiolites (Fig. lOc) was approxiophiolite. An amphibole separate from BO-17 has mately consanguineous along this part of the IYS. an Ar/Ar age of 85.7 Ma, whereas biotite in LW1The collision of India with Eurasia began in the 2 has an age of 80.6 Ma (Fig. 9). An amphibole Cenozoic at c. 55 Ma as shown by palaeomagnetic separate, A13, from a large knocker of meta- and other data (Nicolas et al. 1981; Allegre et al. morphic rocks at the base of the Xigaze ophiolite 1984; Patriat & Achache 1984), and the Neoyields a plateau age of 87.9 ± 0.4 Ma. A plot of Tethyan ocean basin was completely closed and 40Ar/36Ar v ^Ar/36Ar yielded an age of the ophiolites obducted before the end of the 88.0 ± 0.3 Ma. We tentatively interpret the amphi- Eocene (before 40 Ma) (Tapponnier et al. 1981). bolites as remnants of metamorphic soles and This is confirmed by the Ar/Ar biotite date of suggest that the dates obtained from these rocks 41 Ma for the Gangdese batholith near Luobusa, mark the age of initial displacement of the which probably represents waning arc magmatism ophiolites beneath the Zedong arc (Fig. lOc). in response to the final stages of subduction along A biotite separate from sample L36, a granite the southern Eurasian margin, and which is in from the Gangdese batholith near Luobusa, yields accord with granodiorites dated at 41 Ma (Scharer an Ar/Ar age of 41 Ma (Fig. 9). This is not unlike et al. 1984). In addition, a time gap of 30-40 Ma the ages previously obtained for the Gangdese between ophiolite displacement and final emplacemagmatism (Yin et al. 1994) and is much younger ment exists, during which there would have been than the interpreted displacement age of the accretion of ophiolite and arc terranes before their ophiolite. juxtaposition with the Eurasian margin.
Discussion Ophiolite formation, the closure of the NeoTethyan ocean and the collision between Eurasia and India A number of age determinations can be used to constrain the history of closure of the NeoTethyan ocean (Table 1). The suprasubduction zone signature of the ophiolites indicates that they
Tholeiitic and boninitic magmatism in the ophiolite suites The basaltic rocks from the ophiolite suites all show suprasubduction signatures. They are essentially island-arc tholeiites. In addition, accessory chromites in associated peridotites, particularly in Xigaze, are generally high-Al varieties (Cr numbers 22-65) indicating a tholeiitic affinity (see Dick & Bullen 1984). However, investigation of
202
J. MALPAS ETAL.
Fig. 9. Ar/Ar spectra of amphibole and biotite from rocks of the Yarlung Zangbo suture zone. See text for explanation.
the mineralogy and chemistry of the mantle peridotites and their podiform chromitite deposits from the Luobusa massif indicates that in addition to the arc moleiitic magmatism there was subsequent production of more depleted magmas (Zhou
et al. 1996). These chromitites have high-Cr chromite (Cr numbers 82-85) and low TiO2 (0.10.2 wt.%), suggesting crystallization from boninitic melts. Minor interstitial orthopyroxene in the chromitites also has low A^Os (1.0-1.2 wt.%),
ORIGIN AND EMPLACEMENT OF YARLUNG ZANGBO OPHIOLITES
203
Fig. 10. Schematic diagram of the plate tectonic evolution of the Neo-Tethyan ocean from c. 170 Ma to c. 26 Ma. (a, b) Northward intra-oceanic subduction of oceanic lithosphere initiates the formation of the Zedong island arc at c. 160 Ma. Farther to the north, northward subduction beneath the Eurasian continental margin (Lhasa Block) produces the Gangdese continental arc commencing at c. 155 Ma. (c) The Dazhuqu terrane ophiolites are thrust southward onto the Indian continental margin as the Indian Block collides with the intra-oceanic arc complex, (d, e) Continued subduction of Neo-Tethyan lithosphere to the north eventually juxtaposes the Indian Block and the Lhasa Block, and results in the northward emplacement of the ophiolites and associated arc rocks by 'flake tectonics'. B, Bainang terrane; D, Dazhuqu terrane; Z, Zedong terrane; X, Xigaze terrane; G, Gangdese arc (after Aitchison et al. 2000).
204
J. MALPAS ETAL.
supporting the view that they formed from boninitic melts. Percolation of these boninitic melts through the upper mantle and their focus along downward propagating cracks allowed reaction between the melts and the host peridotites. This reaction removed pyroxenes from the host rock, forming dunite dykes and dunite envelopes around the podiform chromitites. Modification of the melts by this process is believed to have triggered the precipitation of chromite in melt pockets now represented by the podiform bodies (Zhou et al. 1996). The boninitic melts reflect hydrous mantle melting in a suprasubduction zone environment (Dick & Bullen 1984) and formed either by high degrees of partial melting or by remelting of a progressively depleted source above a subduction zone (Crawford et al. 1989, and references therein). This suggests that the Luobusa podiform chromitites probably formed at depth beneath an island arc. The pyroxenite dykes associated with the podiform bodies show a progressive gradation from diopsidite to orthopyroxenite. These types of pyroxenite dykes occur in many erogenic Iherzolite massifs, such as that at Ronda, Spain, and indicate high-pressure fractionation, likewise suggesting formation beneath an island arc. The extensive melt-rock interaction reflects both disequilibrium between the melts and their host peridotites and the thickened lithosphere beneath an island arc. Passage of these melts through the Luobusa mantle section is believed to be the cause of high-temperature recrystallization. In simple terms, the available evidence indicates the presence of two magmatic suites in Luobusa, an older arc tholeiitic suite represented by the pillow lavas and a boninitic suite that gave rise to the podiform chromitites. Similar associations are common in many ophiolites produced in suprasubduction zone environments. The island arc originally above the Luobusa ophiolite is probably represented by the Zedong terrane, which includes andesites, dacites and boninites (McDermid et al. 2002), indicating the development of a mature arc sequence. The dates available for the Luobusa massif and the Zedong terrane appear to support this model. The Luobusa oceanic crust originally formed at c. 175 Ma and provided the basement for the development of the Zedong volcanic arc between the Mid-Jurassic and Mid-Cretaceous, during which time the mantle was modified by boninitic magmatism.
Tectonic model It has commonly been assumed that the ophiolites in the IYS represent lithosphere from the Neo-
Tethyan ocean (Nicolas et al. 1981; Allegre et al. 1984; Xiao 1988), and that they are the same age everywhere. However, Hsu et al. (1995) suggested that the IYS ophiolites are remnants of a number of collapsed back-arc basins, a view supported by our evidence that the ophiolites are of different ages. The difference in geochemistry between the lavas of Luobusa and the other ophiolites (Figs 6 and 7) also supports this view. We believe that the mantle peridotites in Luobusa were originally chemically similar to those in the other ophiolites, such as Xigaze, which are now found in the same suture zone, i.e. they had the composition of depleted MORB mantle trapped in suprasubduction wedges. Remelting of depleted peridotites in the wedge beneath the island arc of the Zedong terrane produced the boninitic magmas that formed the mantle chromitite deposits at Luobusa and the effusive volcanic rocks near Zedong (Fig. 10). To the west, no island-arc edifices were constructed and the ophiolitic rocks of Dazhuqu and Xigaze represent lithosphere extension above the subduction zone, contemporaneous with late arc volcanism at Zedong. The northward subduction of Neo-Tethyan oceanic lithosphere between the Indian margin and the intra-oceanic subduction complexes led to the accretion of the Bainang terrane to the south of the Dazhuqu terrane (Fig. lOa and b). Following the disappearance of the Neo-Tethyan oceanic lithosphere to the south of the subduction complex, continuous northward movement of the Indian Plate beneath the island arc provided a mechanism for the uplift of the Dazhuqu terrane ophiolites onto the Indian continental margin (Fig. lOc). Blocks of amphibolites in the melange zones beneath the ophiolites, probably derived from a metamorphic sole, suggest that the initial displacement of the ophiolites occurred at c. 80-90 Ma (i.e. Late Cretaceous) (Fig. lOc). This significantly predates India-Eurasia collision at c. 55 Ma (Eocene), which resulted in the closure of the NeoTethyan ocean and the formation of the Gangdese batholith as a result of northward subduction beneath the Eurasian margin (Fig. lOd). This suggests that the ophiolites were first accreted onto the Indian continent as its margin collided with and was partially subducted northward beneath the suprasubduction zone lithosphere. This modified Indian margin, including the accreted ophiolite terrane, was later juxtaposed against the Eurasian continental margin as a result of northward subduction of the remaining Neo-Tethyan lithosphere (Fig. lOd). Because of the subduction polarity, emplacement of the ophiolite onto the Eurasian continent must have been facilitated by a 'flake tectonic' mechanism resulting in northwarddirected thrusting (Fig. lOe).
ORIGIN AND EMPLACEMENT OF YARLUNG ZANGBO OPHIOLITES
Conclusions Formation of the Dazhuqu terrane ophiolites appears to have occurred at different times in suprasubduction zone environments associated with the collapse of the Neo-Tethyan ocean basin. Intra-oceanic subduction was an important component of this process of ocean closure, which was clearly more complicated than previously thought and perhaps longer lived. Given that the Luobusa massif shows a suprasubduction geochemical signature at c. 175 Ma, and the collision of India and Eurasia occurred at c. 55 Ma, there is a span of at least 120 Ma during which subduction of NeoTethyan oceanic lithosphere was taking place. Intra-oceanic subduction appears to have produced a mature island arc built on oceanic crust in some places (Luobusa), but resulted in lithosphere extension and the formation of ophiolites elsewhere (Xigaze), perhaps at the same time. The newly formed island arcs were probably the regions in which podiform chromitites formed as a result of melt-rock interaction. The Indian continent eventually collided in the Late Cretaceous with the subduction zone, resulting in the accretion of the Bainang, Dazhuqu and Zedong terranes to the continental margin and the displacement of the ophiolitic rocks. This juxtaposition of the intra-oceanic subduction-related terranes with the Indian continental margin significantly predated the final closure of the NeoTethyan ocean to the north. The collision of India and Eurasia resulted in emplacement of ophiolitic rocks onto the Eurasian margin and the production of the Gangdese batholith in the Eocene. This study was supported by research grants from the Research Grant Council of the Hong Kong SAR, China (HKU7086/01P to J.M.) and NSERC of Canada to P.T.R. and P.H.R. Fieldwork in Tibet in the last few years was assisted by H. Wu and Badengzhu from the Tibetan Geological Survey.
References ABRAJEVITCH, A., AITCHISON, J.C. & ALI, J. 2001. Paleomagnetism of the Dazhuqu Terrane, YarlungZangbo Suture Zone, Southern Tibet. EOS Transactions, American Geophysical Union Fall Meeting Supplement, Abstract GP11A-0118. AITCHISON, J.C., BADENGZHU, & DAVIS, A.M. ET AL. 2000. Remnants of a Cretaceous intra-oceanic subduction system within the Yarlung-Zangbo suture (southern Tibet). Earth and Planetary Science Letters, 183,231-244. AITCHISON, J.C., ABRAJEVITCH, A. & ALI, J.R. ET AL. 2002. New insight into the evolution of the Yarlung-Tsangpo suture zone, Xizang (Tibet), China. Episodes, 25, 90-94. ALLEGRE, C.J., COURTILLOT, V., TAPPONNIER, P. ET AL.
205
1984. Structure and evolution of the HimalayaTibet erogenic belt. Nature, 307, 17-22. BAI, W.-J., ZHOU, M.-F. & ROBINSON, P.T. 1993. Possible diamond-bearing mantle peridotites and chromitites in the Luobusa and Donqiao ophiolites, Tibet. Canadian Journal of Earth Science, 30, 1650-1659. BAI, W.-J., ROBINSON, P.T., FANG, Q., Hu, X., ZHOU, M.-F. & MALPAS, J. 2000. Origin of PGE and base metal alloys in podiform chromitites of the Luobusa ophiolite, southern Tibet. Canadian Mineralogist, 38, 585-598. CLAOUE-LONG, J.C., ZHANG, Z.C., MA, G.G. & Du, S.H. 1991. The age of the Permian-Triassic boundary. Earth and Planetary Science Letters, 105, 182-190. COMPSTON, W., WILLIAMS, I.S. & MEYER, C. 1984. UPb geochronology of zircons from Lunar Breccia 73217 using a sensitive high mass-resolution ion microprobe. Journal of Geophysical Research, 89, 525-534. COULON, C., MALUSKI, H., BOLLINGER, C. & WANG, S. 1986. Mesozoic and Cenozoic volcanic rocks from central and southern Tibet: 39Ar/40Ar dating, petrological characteristics and geodynamical significance. Earth and Planetary Science Letters, 79, 281-302. CRAWFORD, A.J., FALLOON, T.J. & GREEN, T.H. 1989. Classification, petrogenesis and tectonic setting of boninites. In: CRAWFORD, A.J. (ed.) Boninites and Related Rocks. Unwin Hyman, London, 1-49. DEWEY, J.F., CANDE, S. & PITMAN, W.C. 1989. Tectonic evolution of the India-Eurasia collision zone. Eclogae Geologicae Helvetica, 82, 717-734. DICK, HJ.B. & BULLEN, T. 1984. Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology, 86, 54-76. GIRARDEAU, J., MARCOUX, J. & ZHAO, Y. 1984. Lithologic and tectonic environment of the Xigaze ophiolite (Yarlung Zangbo suture zone, Southern Tibet, China), and kinematics of its emplacement. Eclogae Geologicae Helvetica, 77, 153-170. GIRARDEAU, J., MERCIER, J.C.C. & WANG, X.B. 1985. Petrology of the mafic rocks of the Xigaze ophiolite, Tibet. Implications for the genesis of the oceanic lithosphere. Contributions to Mineralogy and Petrology, 90, 309-321. GOPEL, C., ALLEGRE, CJ. & Xu, R. 1984. Lead isotope study of the Xigaze ophiolite (Tibet): the problem of the relationship between magmatites (gabbros, dolerites, lavas) and tectonites (harzburgites). Earth and Planetary Science Letters, 6, 301-310. Hsu, K.J., PAN, G., SENGOR, A.M.G. ET AL. 1995. Tectonic evolution of the Tibetan Plateau: a working hypothesis based on the Archipelago model of orogenesis. International Geological Review, 37, 473-508. TENNER, G.A., LONGERICH, H.P., JACKSON, S.E. & FRYER, BJ. 1990. ICP-MS—a powerful tool for high precision trace element analysis in earth sciences: evidence from analysis of selected USGS reference samples. Chemical Geology, 83, 133-148. MCDERMID, I.R.C., AITCHISON, J.C., DAVIS, A.M.,
206
J. MALPAS ETAL.
HARRISON, T.M. & GROVE, M. 2002. The Zedong terrane: a Late Jurassic intra-oceanic magmatic arc within the Yarlung-Tsangpo suture zone, southern Tibet. Chemical Geology, 187, 267-277. METCALFE, I. 1999. The ancient Tethyan oceans of Asia: How many? How old? How deep? How wide? UNEACAsia Papers, 1, 1-9. MOLNAR, P. & TAPPONNIER, P. 1975. Cenozoic tectonics of Asia; effects of a continental collision. Science, 189, 419-426. MUECKE, G.K., ELIAS, P. & REYNOLDS, P.H. 1988. Hercynian/Alleghanian overprinting of an Acadian terrane: Ar/Ar studies in the Meguma zone, Nova Scotia, Canada. Chemical Geology, 73, 153-167. NICOLAS, A. 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Dordrecht. NICOLAS, A., GIRARDEAU, J. & MARCOUX, J. ET AL. 1981. The Xigaze ophiolite (Tibet): a peculiar oceanic lithosphere. Nature, 294, 414-417. PATRIAT, P. & ACHACHE, J. 1984. India-Eurasia collision chronology has implications for crustal shortening and driving mechanisms of plates. Nature, 311,615-621. PEARCE, J.A. & NORRY, MJ. 1979. Petrogenic implications of Ti, Zr, Y and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology, 69, 33-47. Pozzi, J.P., WEATPHAL, M., GIRARDEAU, J., BESSE, J., YAO, X.Z. & Li, S.X. 1984. Paleomagnetism of the Xigaze ophiolite and flysch (Yarlung Zangbo suture zone, southern Tibet): latitude and direction of spreading. Earth and Planetary Science Letters, 70, 383-394. ROWLEY, D.B. 1996. Age of initiation of collision between India and Asia: a review of stratigraphic data. Earth and Planetary Science Letters, 145, 1-13. SAMSON, S.D. & ALEXANDER, E.G. JR 1987. Calibration of the interlaboratory Ar/Ar dating standard, Mmhb-1. Chemical Geology, 66, 27-34. SAUNDERS, A.D. 1984. The rare earth element characteristics of igneous rocks from the ocean basins. /W.HENDERSON, P. (ed.) Rare Earth Element Geochemistry. Developments in Geochemistry, 2, 205-236. SCHARER, U., Xu, R.H. & ALLEGRE, CJ. 1984. U-Pb
geochronology of the Gangdese (Transhimalaya) plutonism in the Lhasa-Xigaze region, Tibet. Earth and Planetary Science Letters, 69, 311-320. TAPPONNIER, P., MERCIER, J.L., PROUST, F. ET AL. 1981. The Tibetan side of the India-Eurasia collision. Nature, 294, 393-399. WANG, C., Liu, Z. & HEBERT, R. 2000. The YarlungZangbo paleo-ophiolite, southern Tibet: implications for the dynamic evolution of the YarlungZangbo suture zone. Journal of Asian Earth Sciences, 18,651-661. XIAO, X.C. 1988. General Review of the Tectonic Evolution of the Crustal-Mantle of the QinghaiXizhang (Tibet) Plateau. Geological Publishing House, Beijing (in Chinese). YIN, A. & HARRISON, T.M. 2000. Geologic evolution of the Himalayan-Tibetan orogen. Annual Review of Earth and Planetary Sciences, 28, 211-280. YIN, A., HARRISON, T.M., RYERSON, F.J., CHEN, W., KIDD, W.S.F. & COPELAND, P. 1994. Tertiary structural evolution of the Gangdese thrust system, southeastern Tibet. Journal of Geophysical Research,^, 18175-18201. ZHOU, M.-F., ROBINSON, P.T., REYNOLDS, P.H., et al. 1995. Supra-subduction zone origin of the Luobusa ophiolite, Indus-Yarlung Zangbo Suture zone, Tibet. GAC/MAC Abstract, 20, A-114. ZHOU, M.-F., ROBINSON, P.T., MALPAS, J. & Li, Z. 1996. Podiform chromitites of the Luobusa ophiolite (Southern Tibet): implications for melt/rock interaction and chromite segregation in the upper mantle. Journal of Petrology, 37, 3-21. ZHOU, M.-F., MALPAS, J., ROBINSON, P.T. & REYNOLDS, P. 1997. Dynamothermal aureole of the Donqiao ophiolite, Northern Tibet. Canadian Journal of Earth Sciences, 34, 59-65. ZHOU, S., Mo, X.X., MAHONEY, J.J., ZHANG, S.Q., Guo, TJ. & ZHAO, Z.D. 2002. Geochronology and Nd and Pb isotope characteristics of gabbro dikes in the Luobusha ophiolite, Tibet. Chinese Science Bulletin, 47, 143-146. ZYABREV, S.V., AITCHISON, J.C., BADENGZHU,, DAVIS, A.M., Luo, H. & Liu, J.B. 2002. Radiolarian biostratigraphy of supra-ophiolites sequences in the Xigaze area, Yarlung-Tsangpo suture, Southern Tibet (preliminary report). Radiolaria, 17, 13—19.
Tectonic implications of boninite, arc tholeiite, and MORE magma types in the Josephine Ophiolite, California-Oregon GREGORY D. HARPER Department of Earth and Atmospheric Sciences, State University of New York at Albany, Albany, NY 12222, USA (e-mail:
[email protected]) Abstract: The Josephine Ophiolite is a large complete ophiolite flanked by arc complexes, including rifted arc fades, and overlain by volcanopelagic and volcaniclastic sedimentary rocks. The extrusive sequence and sheeted dyke complex record a wide range in magma types and degree of fractionation. The upper part of the extrusive sequence, as well as late dykes in the ophiolite, have mid-ocean ridge basalt (MORB) affinities and include unusual highly fractionated Fe-Ti basalts. The sheeted dyke complex and lower pillow lavas are dominated by transitional island-arc tholeiite (IAT) to MORB, but about 10% consist of low-Ti, high-Mg basalts and andesites. Whole-rock chemistry and Cr-spinel compositions indicate that the lowTi rocks range from boninite (BON) to primitive arc basalt. The low-Ti samples have trace element characteristics indicating a greater subduction component than the IAT-MORB or MORB samples, as well as derivation from a wide range of sources ranging from depleted to enriched relative to an average N-MORB mantle source. Mixing of low-Ti and MORB magmas may have produced the IAT-MORB magma type that is most characteristic of the ophiolite. Podiform chromites and late magmatic features in the mantle peridotite, described by previous workers, appear to have been formed from the low-Ti magmas. Regional geological relationships and the presence of boninitic magmas suggest that arc rifting and initial sea-floor spreading to form the Josephine Ophiolite occurred in the forearc of a west-facing arc built on edge of the North American plate. Arc magmatism appears to have jumped westward, at which time the Josephine basin became situated in a back-arc setting, analogous to the inferred evolution of the modern Lau back-arc basin. Alternatively, the Josephine Ophiolite may have formed in a setting analogous to the north end of the Tonga Trench or the south end of the North Fiji basin, both sites of modern boninites, where a back-arc spreading centre has propagated across an arc into the forearc. Rift propagation during formation of the Josephine Ophiolite is consistent with the presence of highly fractionated Fe-Ti basalts.
Ophiolites are generally believed to represent ancient ocean crust and upper mantle, yet the tectonic setting of many ophiolites is equivocal, Whereas some appear to have formed at midocean ridges, the geochemistry, petrology, and sedimentary sequences of many ophiolites suggest they formed above a subduction zone and they have therefore been called 'suprasubduction zone' (SSZ) ophiolites (Pearce et al. 1984). Spreading centres in back-arc basins are a likely tectonic setting for many ophiolites, where geochemistry of magmas varies from mid-ocean ridge basalt (MORB), especially in mature back-arc basins, to transitional between MORB and island-arc tholeiite (IAT; e.g. Hawkins et al. 1990; Pearce et al. 1994; Hawkins 1995). Recently, basalts erupted on segments of the Chile Ridge that are closest to where the ridge is being subducted have some of the geochemical characteristics of magmas erupted in magmatic arcs (Klein & Karsten 1995). Thus a mid-ocean ridge origin of some
ophiolites having arc-like geochemical signatures is possible (Sturm et al. 2000), especially those that show no sedimentary or regional geological evidence for a nearby arc. A number of ophiolites contain boninites (e.g. Cameron et al. 1979; Cameron 1989; Coish 1989; Meffre et al. 1996; Ishikawa et al. 2002), an unusual magma type that appears to be derived by melting of refractory peridotite ('second stage melts'; e.g. van der Laan et al. 1989). Tertiary boninites are common in forearcs of many of the western Pacific intraoceanic arcs (e.g. Umino 1986; Bloomer & Hawkins 1987; Murton et al. 1992). A few modern occurrences of boninites are known, including forearcs where a back-arc spreading centre has propagated across an arc axis (Falloon & Crawford 1991; Sigurdsson et al. 1993) and directly behind the Tonga arc where a spreading centre is propagating into rifted arc crust (Kamenetsky et al. 1997). The unusual conditions needed to produce boninitic magmas provide an
From: DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 207-230. 0305-8719/037$ 15 © The Geological Society of London 2003.
208
G. D. HARPER
important constraint on the tectonic origin of ophiolites that contain boninites. The Josephine Ophiolite is part of a belt of c. 165 Ma ophiolites that extends from southern California to Washington (Fig. 1). It is the largest and most complete of these ophiolites, and is best developed in northwestern California and southwestern Oregon (Fig. 2). A suprasubduction zone setting for generation of the Josephine Ophiolite is indicated by the presence of flanking arc complexes of similar age and overlying tuffaceous
hemipelagic rocks that grade upwards into flysch of largely volcanic arc provenance (Dick 1976, 1977a; Harper & Wright 1984; Harper et al 1985, 1994; Wyld & Wright 1988; Saleeby 1992). Previous studies have shown that dykes and lavas of the Ophiolite have geochemical affinities transitional between IAT and MORB (Harper 1984, 1988; Wyld & Wright 1988). More recently, Harper (2003) has shown that the upper pillow lavas and late dykes in the Ophiolite form a MORB-affinity suite that includes unusual highly fractionated Fe-Ti basalts. This paper focuses on a suite of low-Ti pillow lavas and dykes, some of which are boninitic. A synopsis of geochemical data for all sheeted dykes and pillow lavas is presented for comparison. These data, along with the regional geological setting, are used to reevaluate possible settings for Ophiolite formation, specifically whether arc rifting and initial sea-floor spreading occurred in a forearc. The geochemical data for the Josephine Ophiolite also provide a basis for comparison with other Middle Jurassic ophiolites in the western USA (Fig. 1), including the Ingalls Ophiolite Complex (Metzger et al. 2002) and the Coast Range Ophiolite (e.g. Shervais 1990; Giaramita et al. 1998). In particular, geochemical comparison of the Coast Range Ophiolite with the Josephine Ophiolite may help resolve whether they are related (e.g. Harper et al. 1985; Saleeby 1992) or, as some argue, the Coast Range Ophiolite is exotic with respect to North America (e.g. Dickinson et al. 1996; Godfrey & Dilek 2000; Pessagno et al. 2000).
Geological setting
Fig. 1. Middle Jurassic Ophiolites of the western USA that are similar in age to the Josephine Ophiolite. JODEO, Devils Elbow Remnant of the Josephine Ophiolite (Wyld & Wright 1988). Modified from Metzger et al. (2002).
The Klamath Mountains of northwestern California and southwestern Oregon consist of ophiolitic and island-arc terranes, separated by east-dipping thrust faults, that young to the west (Burchfiel & Davis 1981; Irwin 1994). The Josephine Ophiolite and its overlying sedimentary rocks (Galice Fm) are part of the westernmost of these thrust sheets, the Western Klamath Terrane. The main body of the Josephine Ophiolite (Fig. 2) and remnants exposed farther south (Devils Elbow remnant; DEO in Fig. 2; Wyld & Wright 1988) define an along-strike length of c. 250 km. The roof thrust (Orleans thrust; Fig. 2) is a major crustal boundary; the outcrop pattern indicates greater than 40 km of displacement and geophysical data suggest as much as 100 km (Jachens et al. 1986). The upper plate of the thrust is the ophiolitic Rattlesnake Creek Terrane (Fig. 2). As a result of the underthrusting, the Josephine Ophiolite and overlying Galice Formation were regionally metamorphosed under prehnite-pumpellyite (north) to lower greenschist (south) facies conditions (Harper
BONINITES IN THE JOSEPHINE OPHIOLITE
209
Fig. 2. Generalized geological map of the west-central Klamath Mountains (Snoke 1977; Harper 1984; Yule 1996; and compilation by Irwin 1994).
et al 1988, 1994). The basal thrust (Madstone Cabin thrust, Fig. 2) juxtaposes the Josephine Ophiolite over a mafic intrusive complex of similar age (Chetco Intrusive Complex; Dick 1976, 1977a; Harper et al. 1990, 1996; Yule 1996). Geochronological data show that displacement on both the roof and basal thrusts overlapped in time (Harper et al. 1994). The Josephine Ophiolite is conformably overlain by a hemipelagic sequence consisting of siliceous argillite, chert, and tuffaceous chert (Harper 1984; Pinto-Auso & Harper 1985); similar rocks and locally volcanic-rich greywacke are locally intercalated with pillow lavas of the Ophiolite. The hemipelagic sequence grades upwards into a thick flysch sequence derived largely from an active arc (Snoke 1977; Harper 1983, 1984). The ages of the Josephine Ophiolite (162-
164 Ma), hemipelagic sequence (c. 162-157 Ma), Galice flysch (c. 157-153 Ma), and thrust emplacement (c. 153-150 Ma) are tightly constrained by biostratigraphic (Pessagno & Blome 1990; Pessagno et al. 2000) and radiometric age data (Wyld & Wright 1988; Harper et al. 1994). North of the Josephine Ophiolite is the RogueChetco island-arc complex (Dick 1976, 1977a; Garcia 1982; Harper & Wright 1984; Harper et al. 1994; Yule 1996). The Rogue Formation consists of submarine volcanic breccias, tuffs, volcaniclastic rocks and less abundant flows, and, like the Josephine Ophiolite, is overlain by flysch of the Galice Formation. The Chetco Intrusive Complex lies structurally beneath the Rogue Formation and its c. 200 Ma ophiolitic basement, and represents the core of the island arc. The age of RogueChetco arc magmatism is well constrained at
210
G. D. HARPER
160-153 Ma (Hacker & Ernst 1993; Harper et al 1994; Hacker ef al. 1995; Yule 1996). Structurally above and east of the Josephine Ophiolite and Galice Formation is a 174-159 Ma volcano-plutonic arc complex. The Western Hayfork Terrane (Fig. 2) was affected by a c. 165 Ma erogenic event (Wright & Fahan 1988; Hacker et al. 1995), and was followed by emplacement of the 164-159 Ma Wooley Creek plutonic belt. The basement for this arc complex is the Late Triassic to Early Jurassic Rattlesnake Creek Terrane (RCT) that largely consists of a disrupted ophiolite (Wright & Wyld 1994; Hacker et al. 1995). Ophiolitic rocks virtually identical in lithology and age to the RCT also form the basement for the younger Rogue-Chetco arc (Fig. 2; Yule et al. 1992; Yule 1996). Josephine 'rift facies' (Fig. 2) are present in the RCT structurally above the Josephine Ophiolite (Preston Peak complex; Saleeby et al. 1982) and within the basement for the Rogue-Chetco arc (Fiddler Mountain complex; Yule 1996). These rift facies consist of mafic dyke complexes, talus breccias, minor pelagic rocks, and locally olistostromes built on older ophiolitic (RCT) basement. The Devils Elbow Remnant of the Josephine Ophiolite in the southern Klamath Mountains (Fig. 1) is also considered a rift facies, consisting of sheeted dykes, pillow lavas, and ophiolite breccia overlying older RCT-like ophiolitic basement (Wyld & Wright 1988). U/Pb zircon ages of the Devils Elbow and Preston Peak rift facies are 164 ± 1 Ma (Wyld & Wright 1988; Saleeby & Harper 1993), just older than the single highresolution zircon age of 162 ± 1 Ma for the main body of the Josephine Ophiolite (Harper et al. 1994).
Low-Ti dykes and lavas Occurrence The proportion of low-Ti dykes and lavas in the Josephine Ophiolite is difficult to estimate, but they are clearly a minor component, making up perhaps 10% or less of the sheeted dyke complex and pillow lavas. In general, the low-Ti lavas are restricted to the lower pillow lavas. In most of the area, the upper pillow lavas form a late MORBaffinity suite that includes fractionated Fe-Ti basalts (Harper 2003). At one locality in the southernmost part of the study area, however, this late MORB/Fe-Ti suite is absent and low-Ti lavas occur at the top of the extrusive sequence (samples Z91a and Z91b). In the type extrusive sequence of the Josephine Ophiolite, where the most detailed stratigraphic sampling has been carried out, only a single, 3 m thick low-Ti unit is
present, consisting of a broken pillow breccia (sample Y5). It is situated in the middle of a c. 400m thick section, beneath the MORB/Fe-Ti upper pillow lavas (Harper 2003). One low-Ti pillow lava (sample R20) occurs as a screen in the upper part of the sheeted dyke complex, and lowTi dykes in the sheeted dyke complex are cut by other dykes (i.e. they are not late dykes). The field relationships of the low-Ti pillow lavas and dykes indicate they formed 'on axis'. As discussed below, most of the low-Ti pillow lavas and dykes have primitive compositions. Harper (1988) used this as an argument for periodic freezing of axial magma chambers, which would have allowed for mantle-derived melts to rise to the surface. Oceanic faults and related large-scale tilting of the crustal sequence are postulated to record structural extension during periods when magma chambers were absent (Alexander & Harper 1992).
Petrography Pillow lavas and sheeted dykes of the Josephine Ophiolite show the effects of extensive sub-seafloor hydrothermal alteration as well as subsequent prehnite-pumpellyite to lower greenschist facies regional metamorphism (Harper 1984; Harper et al. 1988). Nevertheless, igneous textures are often well preserved, and relict clinopyroxene and Crspinel are common. All low-Ti samples contain Cr-spinel, which occurs as inclusions in mafic phenocrysts (most common), as microphenocrysts, and/or in the groundmass. The low-Ti volcanic rocks are pillow lavas and pillow breccias. They are distinctive in the field by the presence of light macrovariolites that grade outward into dark, originally glassy pillow rims. Some of the glassy rims are unusually thick (up to 5 cm) and many are vesicular (Table 1). With only two exceptions, neither macrovariolites nor more than 2% vesicles are present in the IAT-MORB or MORB-affinity pillow lavas. Microphenocrysts of olivine (pseudomorphs), containing translucent reddish brown octahedra of Cr-spinel, are present in amounts up to 6%. The olivine has been completely replaced by chlorite, chlorite + quartz or, in sample L10, entirely by quartz. Sample Y5 has rare pseudomorphs that may be altered orthopyroxene based on rectangular and six-sided shapes. Clinopyroxene occurs as microphenocrysts in some samples, and is abundant as a groundmass mineral. Glassy margins of pillows (now mostly chlorite) contain prismatic, skeletal, acicular, or feathery clinopyroxene set in a matrix of altered glass. Variolites consist of dendritic clinopyroxene or radiating clinopyroxene intergrown with plagioclase (albitized). Zierenberg et al. (1988) de-
Table 1. Major (%) and trace element (ppm) analyses for low-Ti dykes and lavas Extrusive sequence
Sheeted dyke complex Sample: % vesicles: % phen: Si02 Ti02 A1203 FeOT MnO MgO CaO Na 2 0 K2O P205 LOI Total Ba Rb Sr Y Zr Nb Ni Cr V Sc Th Hf Ta La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Mg no. Ti/Zr Ti/V Zr/Sm Zr/Y T
A20
A22
A90
E2
F30
RAB1
14
9
17
1
11
2
F88c 11 1
47.93 0.38 11.07 8.12 0.15 18.15 8.77 1.33 0.21 0.05 3.18 99.34 39 4 94 12 34 0.62 503 1224 169 29 0.32 0.99 0.05 2.84 6.52 1.12 4.61 1.29 0.49 1.72 0.30 1.94 0.42 1.22 0.19 1.27 0.20 80 67 13 26 2.8
51.35 0.44 15.14 7.60 0.14 9.20 10.09 2.71 0.10 0.04 1.66 98.48 130 0 115 12 35 0.50 110 444 192 36 0.20 0.98 0.04 1.25 2.97 0.51 2.74 0.91 0.36 1.33 0.25 1.70 0.37 1.04 0.16 1.06 0.17 68 75 14 38 3.0
47.96 0.38 13.27 7.10 0.13 15.16 10.95 0.96 0.10 0.05 3.22 99.28 19 3 114 12 24 0.63 348 848 161 30 0.25 0.85 0.04 2.41 5.68 1.02 4.53 1.40 0.54 1.68 0.33 2.04 0.44 1.26 0.20 1.37 0.20 79 95 14 17 2.0
52.40 0.47 14.62 7.79 0.13 9.57 10.52 2.48 0.16 0.06 2.27 100.47 30 2 169 10 31 0.60 157 519 225 40 0.28 0.83 0.04 1.82 4.28 0.72 3.77 1.15 0.44 1.47 0.26 1.71 0.37 1.10 0.17 1.12 0.18 69 91 13 27 3.0
56.75 0.55 14.29 6.51 0.13 6.63 3.25 6.31 1.64 0.03 3.82 99.92 106 27 158 12 36 0.86 122 374 133 34 0.26 0.99 0.06 2.42 5.02 0.87 4.43 1.45 0.41 1.87 0.33 2.12 0.43 1.19 0.18 1.12 0.16 65 92 24 25 3.0
49.90 0.36 11.60 7.91 0.18 15.50 8.07 1.76 0.19 0.08 4.54 100.09 37 5 71 12 24 0.79 283 959 191 33 0.23 0.77 0.06 1.82 4.06 0.63 2.95 1.03 0.36 1.39 0.28 1.76 0.38 1.22 0.18 1.31 0.19 78 90 11 23 2.0
53.07 0.45 13.95 7.60 0.17 11.06 8.30 3.02 0.42 0.06 n.d. 98.10 52 5 111 12 33 0.41 162 556 203 38 0.19 0.86 0.03 1.47 3.40 0.54 3.00 1.17 0.43 1.67 0.33 2.26 0.49 1.37 0.21 1.34 0.21 72 81 13 28 2.7
2+
G25 11 4
52.70 0.33 13.70 6.45 0.18 11.32 6.52 2.34 2.90 0.02 3.10 99.56 60 49 37 14 28 n.d. 84 444 190 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 76 71 10 2.0 2+
L10 20 2
L4 11 9
R20 10 4
Y29c 19 3
Y5 8%, and TiO2 0.5% (Fig. 8).
Fig. 3. MgO variation diagrams for low-Ti dykes (^) and pillow lavas and breccias (t>) from the Josephine Ophiolite. Oxides were recalculated to 100% anhydrous, (a) MgO v. Cr. (b) MgO v. TiO2. (c) MgO v. A12O3. (d) MgO v. SiO2. SDC, basal sheeted dyke complex. UPL, upper pillow lavas. SDC and UPL trends are from Harper (2003). Fields for boninite and arc picrite are from Le Maitre (2002).
216
G. D. HARPER
Fig. 4. Ti v. V discriminant diagram (Shervais 1982). IAT, island-arc tholeiite; MORE, mid-ocean ridge basalt; WPB, within-plate basalt. Data for Hole 786 boninites from the Izu-Bonin forearc are from Murton et al. (1992).
Boninites are characterized by enrichment of LILE and light REE (LREE) coupled with very low HFSE such as Ti, Y, and heavy REE (HREE). This is reflected, for example, in a La/Sm v. TiO2 plot (Fig. 8), which shows the enrichment of boninites in LREE at very low TiO2 The very low abundance of the HFSE Y and Yb in boninites is evident in Y-Cr and Yb-Cr plots (Figs 5 and 6). Fields for boninites are generally not included in discriminant diagrams, and thus a field denned by boninites from the Izu-Bonin forearc (Ocean Drilling Program (ODP) Hole 786; Murton et al. 1992) was added to the diagrams. These rocks are fairly representative of boninites because they include both low-Ca and high-Ca varieties, the latter characterized by lower SiO2 and higher CaO and FeOT (Crawford et al. 1989). There is a complete gradation from boninites into low-Ti island-arc basalts (Beccaluva & Serri 1988). The Josephine Ophiolite shows an unusually wide range in magma types and degree of fractio-
nation. A summary of the magma types and fractionation for sheeted dykes and lavas, other than the low-Ti samples, is first presented for comparison. Sheeted dyke complex and lower pillow lavas (except low-Ti samples). Fractional crystallization trends can be inferred using MgO, as MgO will decrease with fractionation of mafic phases. Harper (2003) showed that a subset of samples from the basal sheeted dyke complex, considered to have undergone the least element mobility, show increasing FeOT and TiO2 enrichment trends with decreasing MgO typical of tholeiitic suites (SDC in Fig. 3b). The increase in Al2Os with decreasing MgO (Fig. 3c) indicates late crystallization of plagioclase, consistent with the crystallization order in the cumulate sequence (Harper 1984). Plagiogranites, which make up less than 1% of the ophiolite, define the negative sloping part of the sheeted dyke trend in Figure 3b and c.
BONINITES IN THE JOSEPHINE OPHIOLITE
217
Fig. 5. Y v. Cr discriminant diagram (Pearce 1982). (See Fig. 4 for key to symbols.) Arrows A, B, and C represent crystallization paths for magmas fractionating Cr-spinel + olivine + pyroxene from MORE, IAT, and boninite magmas, respectively. IAT, island-arc tholeiite; MORB, mid-ocean ridge basalt. Data for Hole 786 boninites from the Izu-Bonin forearc are from Murton et al. (1992).
The subparallel REE and MORB-normalized patterns for a suite of samples from the sheeted dyke complex and lower pillow lavas (Fig. 9a) are consistent with fractionation from a similar parental magma (Pearce 1982, 1983), with negative Eu and Ti anomalies in the most evolved sample (plagiogranite Z59b) reflecting plagioclase and Fe-Ti oxide fractionation, respectively. Nearly all samples of the sheeted dyke complex and lower pillow lavas show characteristics transitional between IAT and MORB, as recognized in previous studies (Harper 1984; Harper et aL 1985; Wyld & Wright 1988), although a few samples consistently plot in the IAT or MORB fields in discriminant diagrams. Most samples have Ti/V values of 20 or higher, similar to MORB (Fig. 4), but plot in the IAT field in Cr-Yand Th/Yb v. Ta/ Yb diagrams (Figs 5 and 7). They show enrichment in Th and have negative Ta and Nb anomalies in a MORB-normalized diagram (Fig. 9a), both characteristics of arc magmas (e.g. Pearce
1982). Ta/Yb ratios indicate a mantle source that is enriched relative to N-MORB mantle, but less enriched than average E-MORB. Three samples collected from the same section along the southwestern limb of the ophiolite, however, have distinctly higher Ta/Yb; sample Z83a (Table 2) is an E-MORB, and another sample from the sheeted dyke complex and a late dyke in the peridotite have calc-alkaline affinities (Fig. 7). Upper pillow lavas and late dykes. The upper pillow lavas and late dykes show Fe and Ti enrichment trends (Fig. 3b) characteristic of tholeiites, and in contrast to the sheeted dyke complex show decreasing A^Oa with decreasing MgO (UPL in Fig. 3c) suggesting early onset of plagioclase fractionation (Harper 2003). Many samples are unusual highly fractionated Fe-Ti basalts (e.g. sample Qlp, Table 2), defined as having greater than 12% FeOT and 2% TiO2 (e.g. Sinton et al. 1983).
218
G. D. HARPER
Fig. 6. Low-Ti dykes (^) and pillow lavas and breccias (900 ppm) that they plot above modelled partial melting curves (Fig. 6). The majority of the low-Ti samples have SiO2 ^53% (anhydrous; Fig. 3d), and at least some of those having 40% on our map) constitute a sheeted dyke complex, which in turn implies formation by sea-floor spreading. Most dykes are parallel to syn-oceanic normal faults, which implies that they were emplaced into planes of minimum compressional stress during east-west crustal extension. The localized absence of a sheeted dyke complex at the plutonic-lava interface can be explained as the result of intra-oceanic erosion-denudation, compounded by fragmentation by subvolcanic explosions responsible for formation of the brecciated hypabyssal facies.
Volcanic stratigraphy The lava sequences in the Lac de 1'Est and Mt Ham areas are composed of a lower mixed tholeiitic + boninitic volcanic unit, a thin red argillite, and an upper boninitic volcanic unit (Laurent & Hebert 1977; Beulac 1982; Hebert 1983; Oshin & Crocket 1986). However, our mapping (Figs 1 and 3) indicates that the thickness of the lavas is extremely variable, with the greatest accumulations being located within faultbounded basins (graben). Furthermore, the two extrusive units found in the Lac de LEst and Ham areas are not everywhere present. The tholeiites appear to be restricted to the basal sections of the main graben, and we speculate that they might represent local preservation of a tholeiitic protocrust. Preliminary data suggest that a few gabbros in the TMOC also have tholeiitic affinities (though most are boninitic), and may also represent components of a proto-crust. This may explain the more complex structural pattern recorded in the Gabbroic Zone, relative to the Pyroxenitic Zone. Alternatively, the change in fabric orientations may represent a shift from a pattern dominated by crust-mantle shear in the ultramafic rocks, to a more localized control of deformation pattern in the upper crust (gabbros).
244
J.-M. SCHROETTER ETAL.
A preliminary model for the structural and magmatic evolution of the TMOC The style and sequence of syn-oceanic deformation we have documented (Figs 4 and 5) allow us to propose a possible evolutionary scenario for the main, boninitic phase of the TMOC (Figs 7 and 8). The starting point is a layered plutonic crust composed of dunite at the base, pyroxenite in the middle, and gabbro at the top, into which is rooted
a sheeted dyke complex that fed overlying lavas (Fig. 8a). During a first, high-temperature, synmagmatic deformation event, many of the plutonic rocks were transposed and recrystallized along planes subparallel to original bedding planes. The cause of this first event is not certain. It may reflect shear between crust and mantle driven by a diapir (Nicolas 1992; Rabinowicz et al 1993), extensional shear as crust slides away from the ridge axis, or may represent the way deformation
Fig. 7. (a) Interpretative palaeogeographical reconstruction of the Thetford Mines Ophiolitic Complex before its emplacement. Colours are the same as in Fig. 1. (b) Idealized map cross-section of the Adstock-Ham Massif.
THETFORD MINES OPHIOLITE, QUEBEC, CANADA
245
is partitioned in an extending ductile crust at the spreading centre (Tapponier & Francheteau 1978; Harper 1985). The overall regime remains tensional, however, and continuing magmatism is expressed as dykes of peridotite and pyroxenite (E2, Fig. 8b), some of which were probably emplaced as sills within preexisting cumulates. The extent to which rocks of the Dunitic and Pyroxenitic Zones belong to the early cumulate 'stratigraphy', and the extent to which they represent under- and intra-plating intrusions is not certain (seeThy 1990; Bedard 1991, 1993). In many cases, these intra-cumulate intrusions would have been transposed and recrystallized during continuing high-temperature deformation, and so would no longer be recognizable as late intrusions. The presence of pyroxenitic debris within graded layers near the base of the Dunitic Zone implies that some dunites are cumulates formed within intra-plating or under-plating sills, with the pyroxenitic debris being derived from disaggregation of a pre-existing roof. The high-temperature foliation is truncated by the major north-south faults we have documented. On outcrop and map scales, these faults dissect the crust into horst-and-graben structures. Kilometre-scale tilted blocks develop in the rigid upper plutonic crust and overlying extrusive rocks. An upward decrease in throw on these faults indicates that they are growth faults, propagating from bottom to top. Associated intrusions (Figs 3 and 6) imply that these faults were coeval with magmatism and played a role in magma ascent. The Dunitic Zone is only partly affected by these faults, and the lower part of the Dunitic Zone could represent a decollement surface accommodating movement of the tilted blocks (e.g. Harper 1985). The presence of a brecciated hypabyssal facies in areas where the sheeted dyke complex is absent suggests that erosion and/or tectonomagmatic destruction of upper-crustal facies was probably an important part of the geological history of the TMOC (Fig. 8c). Fig. 8. Schematic illustrations of a possible evolutionary scenario for the main, boninitic, crust-forming event of the Thetford Mines Ophiolitic Complex, (a) Event El. Development of a high-temperature foliation in cumulate rocks, (b) Event E2. Synkinematic pyroxenitic intrusions (grey arrows) are emplaced and transposed by continuing high-temperature tectonomagmatic deformation, as are harzburgite and websterite cumulates. Faults begin to form when magmatism wanes. Subvolcanic and talus(?) breccias form in the uppermost crust, (c) Event 3. Main stage of faulting, with only minor late-kinematic intrusions and associated lavas. Crust breaks up into horsts and grabens, and blocks are tilted along a basal decollement. The tops of tilted blocks are eroded, locally removing much of the upper crust.
The TMOC, a slow-spreading environment? The morphology of the axial rift zone of spreading centres is dependent on spreading rate, with prominent fault scarps and a deep axial valley characterizing slow-spreading environments (Macdonald 1982). Seismic data imply that magma chambers are ephemeral at slow-spreading ridges, with the depth of seawater penetration, and the depth of the brittle-ductile transition depending on the presence or absence of magma chambers (Harper 1985; Toomey et al. 1988; Dick et al 1992; Dilek & Eddy 1992; Dilek et al 1998). The deep graben of the TMOC, the deep crustal
246
J.-M. S C H R O E T T E R ^ r ^ L .
Fig. 9. (a) Schematic illustrations of a possible configuration for the genesis of boninites (after Deschamps & Lallemand 2003). (b) Model of boninite formation for southern Quebec Appalachian ophiolites.
THETFORD MINES OPHIOLITE, QUEBEC, CANADA level to which synmagmatic extensional faults penetrate (Figs 1 and 3), and the evidence for near-pervasive lower-crustal hydrothermal metamorphism are all features compatible with a slow spreading rate. An unresolved problem is the nature and location of the accommodation zone (Harper 1985) for the normal faults we have documented (Figs 1 and 3). We have been able to follow the faults down into the Dunitic Zone, but because of the poor exposure and common post-emplacement reactivation of Dunitic Zone rocks, we cannot be certain whether rotation of upper-crustal fault blocks occurred along a decollement zone, whether it penetrated down into the mantle, or whether the deformation was accommodated by a wide, ductile, lower crust. The tectonic exhumation of lower-crustal and mantle rocks along shallowly dipping extensional faults seems to characterize slow-spreading environments (e.g. Dick et al 1992; Dilek et al. 1998). In the TMOC, the deposition of lavas and sediments directly upon mantle or lower-crustal rocks (Fig. 1) could perhaps be explained in this manner.
Formation ofboninitic crust in the northern Appalachians The formation of boninite lavas requires a hot, depleted mantle affected by subduction zone metasomatism (Crawford et al. 1989). Several tectonic environments have been suggested to explain boninite genesis, including: subduction of a spreading ridge; plume-subduction zone interactions; subduction zone initiation; propagation of a backarc spreading centre into a subduction zone; slab rollback caused by a decrease in convergence rate (e.g. Crawford et al. 1989; Stern & Bloomer 1992; Bedard et al. 1998; Macpherson & Hall 2001; Kim & Jacobi 2002; Deschamps & Lallemand 2003). The Taconian ophiolites of the northern Appalachians are unusual in that boninites are extremely abundant, may constitute the dominant magmatic suite over extensive regions and are roughly coeval (490-480 Ma, e.g. Church 1977; Bedard et al. 2000; Bedard & Kim 2002; Kim & Jacobi 2002). Any genetic model must explain the development of extensive boninitic magmatism at this juncture. The identification of sub-north-south extensional oceanic structures in the TMOC suggests that the spreading centre had a north-south trend and that extension and sea-floor spreading took place along an east-west flow direction. Northsouth lineaments have also been reported from other oceanic terranes in southern Quebec and northern New England (Doolan et al. 1982;
247
Tremblay & Malo 1991; Tremblay 1992), suggesting that this may be the orientation of the extensional (spreading) axis at this time. Ordovician palaeogeographical reconstructions (Tremblay 1992) suggest a SE-dipping subduction zone, which would produce a diachronous intersection between spreading centre and subduction zone (Fig. 9), a configuration compatible with the genetic model proposed for boninites by Deschamps & Lallemand (2003). However, this assumes that the north-south trend documented for the TMOC spreading axis has not been affected by rotation of the obducted oceanic terrane, as has been documented in other ophiolites (e.g. Perrin et al. 1993). Another objection to this model is that it requires a very stable geometry to account for development of similar, roughly coeval, boninitic oceanic crust along >1500 km of strike length in the northern Appalachians. In contrast, models involving collision of the Taconic arc against an irregular continental margin and coeval formation of forearc spreading centres in re-entrants (Fig. 9: Cawood & Suhr 1992; Bedard & Kim 2002; Kim & Jacobi 2002) may be more compatible with the apparently synchronous development of anomalous boninitic crust of Betts Cove type (Church 1977) over such an extensive region.
Conclusions Our data highlight the links that exist between the structural and magmatic history of the Thetford Mines Ophiolitic Complex (TMOC; Fig. 7). We have documented the manner in which synmagmatic normal faults dissected the upper crust into tilted fault blocks and controlled deposition of lavas and sediments. These observations imply that the ophiolite was partially dismembered by extensional tectonics before its emplacement onto the continental margin (Fig. 7a). The dominance of a boninitic signature in cumulates and lavas suggests that the TMOC is a forearc-related ophiolite. The evidence for coeval extension and magmatism, and the discovery of a locally well-developed sheeted dyke complex suggest that the TMOC formed when sea-floor spreading was initiated in a forearc, with slow magmatic extension being fed by boninitic, rather than tholeiitic magmas. Emplacement onto the continental margin followed very quickly after the ophiolite formed, and a piggy-back forearc basin was developed. Thanks are due to R. Hebert, R. Laurent and the late P. St-Julien for introducing us to the area and sharing their knowledge during earlier field seasons; to B. Brassard and Ressources Allican Inc. for initiating the project; to N. Pinet, Y. Hebert and F. Huot for numerous
248
J.-M. SCHROETTER ETAL.
discussions; to P. Cousineau and the late lamented G. Kessler for volcanological and sedimentological insights; and to Y. Dilek and P. Robinson for insightful and lightning-fast reviews. This project has been supported by the Geological Survey of Canada, by Canadian National Science and Engineering Research Council grant ESS 233685-99, by a Diversification de 1'Exploration Minerale au Quebec grant provided by ValorisationRecherche Quebec (project 2201-133), and by Ressources Allican Inc. This is Geological Survey of Canada Contribution 2002206.
References BEDARD, J.H. 1991. Cumulate recycling and crustal evolution in the Bay of Islands ophiolite. Journal of Geology, 99, 225-249. BEDARD, J.H. 1993. The oceanic crust as a reactive filter: multiple syn-kinematic intrusion, hybridization and assimilation in an ophiolitic magma chamber. Geology, 21, 77-80. BEDARD, J.H. & KIM, J. 2002. Boninites in NE Appalachian Notre-Dame Subzone ophiolites: tectonic models for lapetus terranes. Geological Society of America, Abstracts with Programs, 34, A60. BEDARD, J.H., LAUZIERE, K., TREMBLAY, A. & SANGSTER, A. 1998. Evidence for forearc seafloor-spreading from the Betts Cove ophiolite, Newfoundland: oceanic crust of boninitic affinity. Tectonophysics, 284, 233-245. BEDARD, J.H., LAUZIERE, K., TREMBLAY, A., SANGSTER, A., DOUMA, S.L. & DEC, T. 2000. Betts Cove Ophiolite and its Cover Rocks, Newfoundland. Geological Survey of Canada Bulletin, 550. BEDARD, J.H., TREMBLAY, A., SCHROETTER, J.-M., BECU, V. & PAGE, P. 2001. Structural and magmatic evolution of the Thetford Mines Ophiolite: preliminary results from new mapping and analysis. Geological Association of Canada & Mineral Association of Canada Joint Annual Meeting. StJohn s, NF, Canada. Program with Abstracts, 26, 11. BERCLAZ, A., BEDARD, J.H. & HEBERT, R. 1998. Importance and significance of deformation processes in the genesis of oceanic crust in a slowspreading environment: evidences from the Bay of Islands Ophiolite, Newfoundland. Geological Association of Canada & Mineralogical Association of Canada Joint Annual Meeting, Quebec, QC, Canada. Program with Abstracts, 23, A-15. BEULAC, R.X. 1982. Etude petrologique du Complexe ophiolitique du lac Nicolet. MSc thesis, Universite Laval, Quebec. BLOOMER, S.H., TAYLOR, B., MACLEOD, C.J., STERN, R.J., FRYER, P., HAWKINS, J.W. & JOHNSON, L. 1995. Early arc volcanism and the ophiolite problem: a perspective from drilling in the western Pacific. In: TAYLOR, B. & NATLAND, J. (eds) Active Margins and Marginal Basins of the Western Pacific. Geophysical Monograph, American Geophysical Union, 88, 1-30. BRASSARD, B. & TREMBLAY, A., 1999. Synthese geologi-
que et metallogenique de la M.R.C. de I'Amiante. Rapport final prepare pour la societe d'aide au developpement de la collectivite de 1'Amiante (SADC), 1/50000 scale map, northern and southern sheets of the Amiante MRC. CANN, J.R. 1974. A model for ocean crustal structure developed. Geophysical Journal of the Royal Astronomical Society, 39, 169-187. CASEY, J.F., KARSON, J.A., ELTON, D.L., ROSENCRANTZ, E. & TITUS, M. 1983. Reconstruction of the geometry of accretion during formation of the Bay of Islands Ophiolite Complex. Tectonics, 2, 509-528. CASTONGUAY, S., RUFFET, G., TREMBLAY, A. & FERAUD, G. 2001. Tectonometamorphic evolution of the southern Quebec Appalachians: 40Ar/39Ar evidence for Middle Ordovician crustal thickening and Silurian-Early Devonian exhumation of the internal Humber Zone. Geological Society of America Bulletin, 113, 144-160. CAWOOD, P.A. & SUHR, G. 1992. Generation and obduction of ophiolites: constraints from the Bay of Islands Complex, western Newfoundland. Tectonics, 11(4), 884-897. CHURCH, W.R. 1977. The ophiolites of southern Quebec: oceanic crust of Betts Cove type. Canadian Journal of Earth Sciences, 14, 1668-1673. CHURCH, W.R. 1978. The ophiolites of southern Quebec: oceanic crust of Betts Cove type: Reply. Canadian Journal of Earth Sciences, 15, 18821883. CHURCH, W.R. 1987. The geochemistry and petrogenesis of ophiolitic volcanic rocks from Lac de 1'Est, Thetford Mines Complex, Quebec, Canada: Discussion. Canadian Journal of Earth Sciences, 24, 1270-1273. CLAGUE, D., FRANKEL, C.S. & EABY, J.S. 1985. The age and origin of felsic intrusions on the Thetford Mines ophiolite, Quebec. Canadian Journal of Earth Sciences, 22, 1257-1261. COLEMAN, R.G. 1971. Plate tectonic emplacement of upper mantle peridotites along continental edges. Journal of Geophysical Research, 76, 1212-1222. COOKE, M.H.-C. 1938. Region de Thetford, de Disraeli et de la Moitie orientate de Warwick (Quebec). Commission Geologique du Canada, Memoire, 211, 176. COUSINEAU, P.A. & ST-JULIEN, P. 1992. The SaintDaniel Melange: evolution of an accretionary complex in the Dunnage terrane of the Quebec Appalachians. Tectonics, 11, 898-909. CRAWFORD, A.J., FALLOON, T.J. & GREEN, D.H. 1989. Classification, petrogenesis and tectonic setting of boninites. In: CRAWFORD, AJ. (ed.) Boninites and Related Rocks. Unwin Hyman, London, 1-49. DAVID, J. & MARQUIS, R. 1994. Geochronologie U-Pb dans les Appalaches du Quebec: application aux roches de la zone de Dunnage. Revue Geologique du Quebec, 1, 16-20. DEROSIER, C. 1971. Etude geologique des breches de la region de Thetford Mines. PhD thesis, Universite de Paris. DESCHAMPS, A. & LALLEMAND, S. 2003. Geodynamic setting of Izu-Bonin-Mariana boninites. In: LAR-
THETFORD MINES OPHIOLITE, QUEBEC, CANADA TER, R.D. & LEAT, P.T. (eds) Infra-Oceanic Subduction Systems: Tectonic and Magmatic Processes. Geological Society of London, Special Publications, 219, 163-185. DICK, H.J.B., ROBINSON, P.T. & MEYER, P.S. 1992. The plutonic foundation of a slow-spreading ridge. In: DUNCAN, R.A. ET AL. (eds) The Indian Ocean: A Synthesis of results from the Ocean Drilling Program. Geophysical Monograph, American Geophysical Union, 70, 1-92. DICKINSON, W.R. & SEELY, D.R. 1979. Structure and stratigraphy of forearc regions. AAPG Bulletin, 63, 2-31. DICKINSON, W.R., KLUTE, M.A., HAYES, M.J., JANECKE, S.U., LUNDIN, E.R., McKiTTRicK, M.A. & OLIVARES, M.D. 1988. Paleogeographic and paleotectonic setting of Laramide sedimentary basins in the central Rocky Mountain region. Geological Society of America Bulletin, 100, 1023-1039. DILEK, Y. & EDDY, C.A. 1992. The Troodos (Cyprus) and Kizildag (S. Turkey) ophiolites as structural models for slow-spreading ridge segments. Journal of Geology, 100, 305-322. DILEK, Y., MOORES, E.M. & FURNES, H. 1998. Structure of modern oceanic crust and ophiolites and implications for faulting and magmatism at oceanic spreading centers. In: BUCK, R., KARSON, J., DELANEY, P. & LAGABRIELLE, Y. (eds) Faulting and Magmatism at Mid-Ocean Ridges. American Geophysical Union, Geophysical Monograph, 106, 219-266. DOOLAN, L.D., GALE, M.H., GALE, P. & HOAR, R. 1982. Geology of the Quebec Re-entrant: possible constraints from early rifts and the Vermont-Quebec serpentine belt. In: ST-JULIEN, P. & BELAND, J. (eds) Major Structural Zones and Faults of the Northern Appalachians. Geological Association of Canada, Special Papers, 24, 87-115. DUNNING, G.R., KROGH, T.E. & PEDERSON, R.B. 1986. U-Pb zircon ages of Appalachian-Caledonian ophiolites. Terra Cognita, 6, L51. FEININGER, T. 1981. Amphibolite associated with the Thetford Mines Ophiolite Complex at Belmina Ridge, Quebec. Canadian Journal of Earth Sciences, 18, 1878-1892. HAMILTON, W.B. 1988. Plate tectonics and island arcs. Geological Society of America Bulletin, 100, 15031527. HAMILTON, W.B. 1995. Subduction systems and magmatism. In: SMELLIE, J.L. (ed.) Volcanism Associated with Extension at Consuming Plate Margins. Geological Society, London, Special Publications, 81, 3-28. HARNOIS, L. & MORENCY, M. 1989. Geochemistry of Mount Orford Ophiolite Complex, Northern Appalachians, Canada. Chemical Geology, 77, 133-147. HARPER, G.D. 1985. Tectonics of slow spreading midocean ridges and consequences of a variable depth to the brittle/ductile transition. Tectonics, 4, 395409. HEBERT, R. 1980. Etude petrologique des roches d'Asbestos et du Mont Ham (Ham Sud), Quebec. MSc thesis, Universite Laval, Quebec. HEBERT, R. 1981. Conglomerats polygeniques ophiolitiques: anciens eboulis de talus de fond oceanique.
249
Canadian Journal of Earth Sciences, 18, 619-623. HEBERT, R. 1985. Petrologie des roches ignees oceaniques et comparaison avec les complexes ophiolitiques du Quebec, de Chypre et de I'Apennin. PhD thesis, Universite de Bretagne occidentale, Brest, France. HEBERT, R. & BEDARD, J.H. 2000. Les ophiolites d'avant-arc et leur potentiel mineral: exemple des complexes ophiolitiques du sud du Quebec. Chronique de la Recherche Miniere, 539, 101-117. HEBERT, R. & LAURENT, R. 1989. Mineral chemistry of ultramafic and mafic plutonic rocks of the Appalachian ophiolites, Quebec, Canada. Chemical Geology, 77, 265-285. HEBERT, Y. 1983. Etude petrologique du Complexe Ophiolitique de Thetford Mines, Quebec. PhD thesis, Universite Laval, Quebec. HUOT, F., HEBERT, R. & TURCOTTE, B. 2002. A multistage magmatic history for the genesis of the Orford Ophiolite (Quebec, Canada): a study of the Mt Chagnon massif. Canadian Journal of Earth Sciences, 39, 1201-1217. KACIRA, N. 1971. Geology ofchromitite occurrences and ultramafic rocks of the Thetford Mines-Disraeli area, Quebec. PhD thesis, University of Western Ontario, London. KARSON, J.A., ELTHON, D. & DELONG, S.E. 1983. Ultramafic intrusions in the Lewis Hills Massif, Bay of Islands Ophiolite Complex, Newfoundland: implications for the igneous processes at oceanic fracture zones. Geological Society of American Bulletin, 94, 15-29. KIM, J. & JACOBI, R.D. 2002. Boninites: characteristics and tectonic constraints, northeastern Appalachians. Physics and Chemistry of the Earth, 27, 109-147. LAURENT, R. 1975. Occurrences and origin of the ophiolites of the Southern Quebec, Northern Appalachians. Canadian Journal of Earth Sciences, 12, 443-455. LAURENT, R. 1978. The ophiolites of southern Quebec: oceanic crust of Betts Cove type: Discussion. Canadian Journal of Earth Sciences, 15, 18801881. LAURENT, R. & HEBERT, Y. 1977. Features of submarine volcanism in ophiolites from the Quebec Appalachians. In: BARAGAR, W.R. (ed.) Volcanic Regimes in Canada. Geological Association of Canada, Special Papers, 16, 91-109. LAURENT, R. & HEBERT, R. 1989. The volcanic and intrusive rocks of the Quebec Appalachian ophiolites (Canada) and their island-arc setting. Chemical Geology, 77, 287-302. LAURENT, R., HEBERT, R. & HEBERT, Y. 1979. Tectonic setting and petrological features of the Quebec Appalachian ophiolites. In: MALPAS, J. & TALKINGTON, R.W. (eds) Ophiolites of the Canadian Appalachians and Soviet Urals. International Geological Correlation Project, 39, 53-77. LYTWYN, J.N. & CASEY, J.F. 1993. The geochemistry and petrogenesis of volcanics and sheeted dikes from the Hatay (Kizildag) ophiolite, southern Turkey: possible formation with the Troodos ophiolite, Cyprus, along fore-arc spreading centers. Tectonophysics, 223, 237-272.
250
J.-M. SCHROETTER £T,4L.
MACDONALD, K.C. 1982. Mid-ocean ridges: fine scale tectonic, volcanic and hydrothermal processes within the plate boundary zone. Annual Review of Earth and Planetary Sciences, 10, 155-190. MACPHERSON, C.G. & HALL, R. 2001. Tectonic setting of Eocene boninite magmatism in the Izu-BoninMariana forearc. Earth and Planetary Science Letters, 186, 215-230. NICOLAS, A. 1992. Kinematics in magmatic rocks with special reference to gabbros. Journal of Petrology, 33, 891-915. OSBERG, P.H. 1978. Synthesis of the geology of northeastern Appalachians. In: TOZER, E.T. & SCHENK, P.E. (eds) Caledonian-Appalachian Orogen of the North Atlantic Region. Geological Survey of Canada, Papers, 78-13, 137-147. OSHIN, I.O. & CROCKET, J.H. 1986. The geochemistry and petrogenesis of ophiolitic volcanic rocks from Lac de 1'Est, Thetford Mines Complex, Quebec, Canada. Canadian Journal of Earth Sciences, 23, 202-213. PAGE, P., BECU, V., TREMBLAY, A., BEDARD, J.H. & BRASSARD, B. 2001. Occurrences of platinumpalladium-enriched chromitites in the Thetford Mines ophiolite, southern Quebec. Geological Society of America, Northeastern Section, Abstracts with Programs, 33(1), 58-59. PEARCE, J.A., LIPPARD, S.J. & ROBERTS, S. 1984. Characteristics and tectonic significance of suprasubduction ophiolites. In: KOKELAAR, B.P. & HoWELLS, M.F. (eds) Marginal Basin Geology. Geological Society, London, Special Publications, 16, 77-94. PERRIN, M., PREVOT, M. & BRUERE, F. 1993. Rotation of Oman ophiolite and initial location of the ridge in the hotspot reference frame. Tectonophysics, 229, 31-42. PINET, N. 1995. Cartographic du secteur Lac BisbyPetit Lac Saint-Francois, Propriete Coleraine. QERDEM B Dossier d'exploration miniere, Ressources Minieres Coleraine Inc., GM 53715, 33. PINET, N. & TREMBLAY, A. 1995. Tectonic evolution of the Quebec-Maine Appalachians: from oceanic spreading to obduction and collision in the north Appalachians. American Journal of Science, 295, 173-200. PINET, N., CASTONGUAY, S. & TREMBLAY, A. 1996a. Thrusting and back thrusting in the Taconian internal zone, southern Quebec Appalachians. Canadian Journal of Earth Sciences, 33, 1283-1293. PINET, N., TREMBLAY, A. & SOSSON, M. 1996b. Extension versus shortening models for hinterland-directed motions in the southern Quebec Appalachians. Tectonophysics, 267, 239-256. RABINOWICZ, M., Rouzo, S., SEMPERE, J.C. & ROSEMBERG, C. 1993. Three-dimensional mantle flow beneath mid-ocean ridges. Journal of Geophysical Research, 98, 7851-7869. ROBINSON, P.T. & MALPAS, J. 1990. The Troodos ophiolite of Cyprus: new perspectives on its origin and emplacement. In: MALPAS, J., MOORES, E.M., PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophiolites, Oceanic Crust Analogues, Proceeedings of Symposium 'TROODOS 1987'. Geological Survey
Department, Nicosia, 13-26. ROBINSON, P.T., MELSON, W.G., O'HEARN, T. & SCHMINCKE, H.-U. 1983. Volcanic glass compositions of Troodos ophiolite, Cyprus. Geology, 11, 400-404. ROBINSON, P., TUCKER, R.D., BRADLEY, D., BERRY, H.N. & OSBERG, P.H. 1998. Paleozoic orogens in New England. Geologiska Foreningens i Stockholm Forhandlingar, 120, 119-148. ST-JULIEN, P. 1987. Geologic des Regions de SaintVictor et de Thetford-Mines (moitie est). Ministere de 1'Energie et des Ressources du Quebec, Quebec, MM86-01. SCHROETTER, J.-M., BECU, V., PAGE, P., TREMBLAY, A.
& BEDARD, J.H. 2000. Chromitites Ophiolitiques de la Region de Thetford MinesRapport des Activites de Terrain de la Saison Ete. Ministere de 1'Energie et des Ressources du Quebec, Quebec, GM 58649. SCHROETTER, J.-M., PAGE, P., TREMBLAY, A. & BEDARD, J.H. 2002. Structural evolution of the ThetfordMines Ophiolitic Complex, Quebec: from synoceanic rifting to obduction and post-obduction deformation. Geological Society of America, Northeastern Section, Abstracts with Programs, 34(1), A21. STERN, R.J. & BLOOMER, S.H. 1992. Subduction zone infancy: examples from the Eocene Izu-BoninMariana and Jurassic California arcs. Geological Society of America Bulletin, 104, 1621-1636. SUHR, G. & CAWOOD, P.A. 2001. Southeastern Lewis Hills (Bay of Islands Ophiolite): geology of a deeply eroded, inside-corner, ridge-transform intersection. Geological Society of America Bulletin, 113, 1025-1038. TAPPONIER, P. & FRANCHETEAU, J. 1978. Necking of the lithosphere and the mechanics of slowly accreting plate boundaries. Journal of Geophysical Research, 83, 3955-3970. THY, P. 1990. Cryptic variation of a cumulate sequence from the plutonic complex of the Troodos ophiolite. In: MALPAS, J., MOORES, E.M., PANAYIOTOU, A. & XENOPHONTES, C. (eds) Ophiolites, Ocean Crustal Analogues, Proceedings of the Symposium 'TROODOS 1987'. Publisher, Nicosia, 165-172. TOOMEY, D.R., SOLOMON, S.C. & PURDY, G.M. 1988. Microearthquakes beneath the median valley of the Mid-Atlantic Ridge near 23°N: tomography and tectonics. Journal of Geophysical Research, 93, 9093-9112. TREMBLAY, A. 1992. Tectonic and accretionary history of Taconian oceanic rocks of the Quebec Appalachians. American Journal of Science, 292, 229252. TREMBLAY, A. & CASTONGUAY, S. 2002. The structural evolution of the Laurentian margin revisited (southern Quebec): implications for the Salinian Orogeny and Appalachian successor basins. Geology, 30(1), 7-82. TREMBLAY, A. & MALO, M. 1991. Significance of brittle and plastic fabrics within the Masawippi Lake fault zone, southern Canadian Appalachians. Journal of Structural Geology, 13(9), 1013-1023. TREMBLAY, A., BEDARD, J.H. & LAUZIERE, C. 1997. Taconian obduction and Silurian exhumation of the
THETFORD MINES OPHIOLITE, QUEBEC, CANADA Betts Cove ophiolite, Canadian Appalachians. Journal of Geology, 105, 701-716. UYEDA, S. & KANAMORI, H. 1979. Back-arc opening and the mode of subduction. Journal of Geophysical Research, 84, 1049-1061. VAN STAAL, C.R., DEWEY, J.F., MACNIOCAILL, C. & McKERROW, W.S. 1998. The tectonic evolution of the Northern Appalachians and British Caledonides: history of a complex, west and southwest Pacifictype segment of lapetus. In: BLUNDELL, D.J. & SCOTT, A.C. (eds) Lyell: the Past is the Key to the Present. Geological Society, London, Special Publications, 143, 199-242. WHITEHEAD, J., REYNOLDS, P.H. & SPRAY, J.G. 1995. The sub-ophiolitic metamorphic rocks of the Quebec Appalachians. Journal of Geodynamics, 19, 325-350. WHITEHEAD, J., DUNNING, G.R. & SPRAY, J.G. 2000.
251
U-Pb geochronology and origin of granitoid rocks in the Thetford Mines ophiolite, Canadian Appalachians. Geological Society of America Bulletin, 112, 915-928. WILLIAMS, H. 1979. Appalachian Orogen in Canada. Canadian Journal of Earth Sciences, 16, 792-807. WILLIAMS, H. & HATCHER, R.D. JR 1983. Appalachian suspect terranes. In: HATCHER, R.D. JR, WILLIAMS, H. & ZIETZ, I. (eds) Geophysics of Mountain Chains. Geological Society of America, Memoirs, 158, 33-53. WILLIAMS, H. & ST-JULIEN, P. 1982. The Baie VerteBrompton Line: Early Paleozoic continent-ocean interface in the Canadian Appalachians. In: STJULIEN, P. & BELAND, J. (eds) Major Structural Zones and Faults of the Northern Appalachians. Geological Association of Canada, Special Papers, 24, 177-207.
This page intentionally left blank
Cr-spinel compositions, metadunite petrology, and the petrotectonic history of Blue Ridge ophiolites, Southern Appalachian Orogen, USA LOREN A. RAYMOND 1 , SAMUEL E. SWANSON 2 , ANTHONY B. LOVE 1 & JAMES F. ALLAN 1 1 Department of Geology, Appalachian State University, Boone, NC 28608, USA (e-mail: raymondla@appstate. edu) 2 Department of Geology, University of Georgia, Athens, GA 30602, USA Abstract: Resolution of the petrotectonic history of Blue Ridge ophiolites of the Southern Appalachian Orogen has remained enigmatic because of metamorphism and tectonic fragmentation of ultramafic bodies. Understanding of this history is confounded by the presence of five partial metamorphic overprints and by similar Ti enrichments in spinels from Blue Ridge and modern mid-ocean ridge basalt ultramafic rocks that result from different processes. Chrome spinels from oceanic ultramafic lithosphere show increases in Ti caused by metasomatism induced by passing mafic melts, which create both dunite melt channels within harzburgite wall rocks and associated troctolite impregnation zones. In the Blue Ridge Belt, the oldest metadunite mineral association generally lacks high-Ti spinel, whereas the higher Ti spinels are relatively low in Al and Mg and occur in three amphibolite- to greenschist-facies retrograde metamorphic associations that occur in deformed, metasomatized ultramafic bodies with high aspect ratios. Some spinel compositions in the oldest mineral association are similar to those from arc-suprasubduction zone ultramafic lithosphere. Together, available data are consistent with the hypothesis that: (1) the Blue Ridge ophiolites are fragmented, metamorphosed, very slow-spreading ridge, Xigaze-type ophiolites, consisting of mafic rocks, minor plutonic rocks, and a sublithospheric ultramafic tectonite base; (2) the metadunites represent sublithospheric melt channels and zones of high melt flux, perhaps formed in a suprasubduction zone setting; (3) pre-Taconic subduction may have been west-directed rather than east-directed. The Taconic orogenesis deformed, fragmented, and metamorphosed the ophiolites; and later Taconic, Acadian, and Alleghenian metamorphism hydrated the bodies, while associated deformation exaggerated their elongation.
More than 200 metamorphosed ultramafic rock bodies occur as isolated, scattered pods, lenses, and blocks forming a crudely linear array in the Blue Ridge Belt of the Southern Appalachian Orogen (Fig. 1) (Pratt & Lewis 1905; Hunter 1941; Hess 1955; Larrabee 1966; Misra & Keller 1978). Such linear arrays of ultramafic rocks are now recognized as lithological markers of plate sutures within orogens. In the Blue Ridge Belt, discoveries of eclogite, retrograded eclogite, and block-in-matrix structures within the ultramafic array support the interpretation that this zone marks an Ordovician (Taconic) suture (Raymond et al. 1989; Willard & Adams 1994; Adams et al. 1995; Abbott & Raymond 1997; Ryan et al. 2001). The purposes of this paper are: (1) to review the occurrences, mineralogy, petrology, and metamorphism of the ultramafic rocks; (2) to present and discuss new data and interpretations bearing on the history of the ultramafic and
associated rocks; (3) to suggest tectonic implications for the Neoproterozoic to Ordovician (Taconic) orogenic history implied by these data. The origins of alpine ultramafic rocks of the Blue Ridge Belt have been difficult to decipher, because of the fragmented nature and polymetamorphic histories of the rocks. Commonly, alpine ultramafic rocks in suture zones are parts of ophiolites that represent fragments of oceanic or suprasubduction zone (SSZ) mafic-ultramafic crust and subcrustal mantle (e.g. Moores 1970, 1981; Nicolas & Poirier 1976, chapter 11; Juteau et al. 1977; Pearce et al. 1984; Edwards 1995). Some alpine ultramafic bodies also represent subcrustal mantle slabs thrust to the surface (Davies et al 1993; Green et al 1997). In the Blue Ridge Belt, a few studies suggest that at least some of the alpine ultramafic rock masses represent ophiolite fragments (McSween & Hatcher 1985; Tenthorey et al 1996; Warner 2001). Here, we use
From'. DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 253-278. 0305-8719/037$ 15 © The Geological Society of London 2003.
254
L. A. RAYMOND ETAL.
Fig. 1. Map of the Blue Ridge Belt showing major faults, thrust blocks (and equivalent terrane names), and the locations of select ultramafic rock bodies. B, Boone; BZ, Brevard Fault Zone; FB, Fries Block; FGLF, Fries-Gossan Lead Fault; GLB, Gossan Lead Block, Toe Terrane; GMW, Grandfather Mountain Window; HFF, Hayesville—Fries Fault; MHT, Mars Hill Terrane; NC, North Carolina; GA, Georgia; SC, South Carolina; TN, Tennessee. Numbered and starred meta-ultramafic bodies, mentioned in this paper or extensively studied are: 1, Edmonds; 2, Greer Hollow; 3, Day Book; 4, Woody; 5, Webster-Addie; 6, Corundum Hill; 7, Buck Creek; 8, Hoots. •, locations of additional meta-ultramafic bodies. (Modified from Larrabee 1966; Brown et al. 1985; Hopson et al. 1989; Adams et al. 1995; Raymond 1998.)
the term 'ophiolite' in a broad sense to refer to a crudely layered sequence of mafic volcanic rocks underlain by a complex of mafic to ultramafic plutonic rocks and a basal peridotite tectonite unit. Evidence that Blue Ridge ultramafic rocks are ophiolitic in character is limited, but the available data more closely match the characteristics of ophiolitic rocks than those of other mafic-ultramafic complexes. Support for the hypothesis that these rocks are ophiolitic includes: (1) the petrotectonic association of the rocks within the regional tectonic setting; (2) petrological data; (3) geochemical data (including isotopic data) from both meta-ultramafic rocks and associated metamafic rocks (Hatcher et al. 1984; McSween & Hatcher 1985; Misra & Conte 1991; Tenthorey et al. 1996). For example, the ultramafic rocks are commonly associated with metamafic rocks interpreted to be metabasalts and metagabbros. The metabasalts (amphibolites) are tholeiitic and typically plot in ocean-floor, within-plate, or arc basalt fields on geochemical discrimination diagrams (Hatcher et al. 1984; Misra & Conte 1991). Metatroctolites associated with the metadunites of the Buck Creek body have trace element contents consistent with crystallization from a mantlederived melt in a rift setting (Tenthorey et al.
1996) and some associated amphibolites may have been gabbros (McElhaney & McSween 1983). Finally, the occurrence of dunite with harzburgite is compatible with an ophiolitic origin. None of the available evidence supports alternative interpretations that the Blue Ridge rocks represent other types of mafic-ultramafic complexes, such as layered lopolithic complexes, appinite-type complexes, Alaska-type complexes, or alkalinetype complexes. Studies conducted over the past 40 years have characterized the Blue Ridge ultramafic rocks and revealed several aspects of their histories. All of the ultramafic bodies have been metamorphosed and the bodies and rocks lack any definitive features of igneous intrusions, such as intrusive contacts, dykes and apophyses, contact metamorphism, chilled margins, or igneous textures (Swanson 1981, 1999, 2001; Abbott & Raymond 1984; Raymond 1995, p. 667ff; Raymond & Abbott 1997; Raymond & Warner 2001). The ultramafic bodies form small (1 km) masses (Fig. 2). Mineral assemblages and textures are clearly metamorphic. Up to five metamorphic recrystallization events have affected the character of the ultramafic rocks, with older events representing amphibolite-facies (and per-
CHROME-SPINEL AND BLUE RIDGE OPHIOLITES
Fig. 2. Maps and sections of selected meta-ultramafic rock bodies of the Blue Ridge Belt, (a) Day Book (map modified from Swanson 1981). (b) Grassy Creek, a moderately hydrated body (map from Brobst 1962). (c) Greer Hollow, a hydrated body, d, dunite; G, granitoid rocks; S, mica schist; UM, ultramafic rocks (mixed).
255
256
L. A. RAYMOND ETAL.
haps eclogite-facies) conditions, and younger events representing greenschist-facies conditions. Serpentinites are late-stage rocks produced in greenschist-facies conditions. Critical to unravelling the full history and tectonic significance of Blue Ridge meta-ultramafic rocks is the determination of their igneous or metamorphic protoliths. Clues to any igneous history may be hidden in mineral chemistry: Hence we examine some of that chemistry. Clarification of any igneous history that may exist also requires recognition and isolation of metamorphic features, plus resolution of a number of questions related to the later metamorphic and structural histories of the bodies. Towards that end, we provide data on and discussion of the nature and distribution of both anhydrous and almost entirely hydrated metamorphic rock bodies, the recognition of which is important to the analysis of the metamorphic history. The anhydrous bodies are composed mainly of metadunite, with local metaharzburgite. These rocks contain the highest grade mineral assemblages (olivine + Cr-spinel ± orthopyroxene) (Table 1). Commonly, the olivine, orthopyroxene, and Cr-spinel have re-equilibrated with hydrous metamorphic phases such as tremolite and chlorite (Swanson 2001). However, core compositions of some of the chrome spinels may retain a vestige of an igneous heritage. If so, as we suggest here, one possible source of these cores is Cr spinels from SSZ ophiolitic ultramafic rocks. The data however, are not unambiguous. The uncertainty means that we must assess whether the abundant metadunites are dehydrated serpentinite bodies derived from ophiolite fragments, are highgrade metamorphosed ophiolite fragments, or represent ultramafic bodies of other types. Beyond resolving the issue of the general nature of a possible protolith, we also must determine whether the anhydrous, olivine-rich bodies and hydrated bodies represent different specific protoliths (e.g. dunite v. clinopyroxenite) or the same protoliths with differences in degrees of hydration and metasomatism. The anhydrous body v. hydrated body data provide some information about
this question, but do not unequivocably resolve it. These data include the aspect ratios of the ultramafic bodies, which aid in addressing the related question: does the mode of occurrence, as subequant blocks v. highly elongated layers and lenses, reflect metamorphic or structural histories? The answers to the above questions are seemingly linked, because the petrological features control the rheological properties and strength of the rocks. The features and chemistry of some mid-ocean ridge rocks and their minerals bear some similarities to the features and chemistries of the Blue Ridge rocks and minerals. The chromium spinels from crust of the Hess Deep (a 2500 m deep fault basin at the west end of the Cocos-Nazca spreading centre, Pacific Ocean, Ocean Drilling Program (ODP) Leg 147) and the chrome spinels from Blue Ridge meta-ultramafic rocks, for example, both show enrichment in TiO2. We show that similarities in spinel chemistries reflect different petrological histories. Nevertheless, as petrogenetic indicators, the Blue Ridge spinel Al contents (and ranges of TiO2) are consistent with the hypothesis that the enclosing metadunites are SSZ rocks. Another similarity between spreading centre rocks and those of the Blue Ridge Belt, revealed by crustal gabbroic and mantle peridotite rocks recovered from six drill holes in the Hess Deep, is the presence of masses of dunite. The Hess Deep holes reveal a mid-ocean ridge-style ophiolite with dunite melt channels (Allan & Dick 1996). A comparison of features of Blue Ridge, Hess Deep, and other ophiolitic sublithospheric mantle rocks suggests the hypothesis that Blue Ridge metadunite protoliths represent sublithospheric melt channels and melt flux zones beneath a spreading centre. In this paper, we present data supportive of a melt channel-melt flux zone, SSZ, slow-spreading centre origin for Blue Ridge metadunites and discuss the Ti enrichments in Hess Deep and Blue Ridge spinels so as to assess the meaning of these data in terms of possible igneous and metamorphic histories.
Table 1. Metamorphic associations in Blue Ridge meta-ultramafic rocks Association A-l: Olivine ± Chromite ± Orthopyroxene ± Clinopyroxene ± Hornblende ± Magnetite Association A-2: Olivine ± Orthopyroxene ± Clinopyroxene ± Tremolite ± Magnesiocummingtonite ± Chlorite ± Chromite Association A-3: ± Chlorite ± Magnesiocummingtonite ± Anthophyllite ± Tremolite ± Talc ± Phlogopite ± Magnetite ± Magnesite ± Garnet Association A-4: ± Serpentine (Antigorite) ± Magnetite ± Chlorite ± Talc ± Brucite ± Tremolite Association A-5: ± Serpentine (Lizardite) ± Serpentine (Chrysotile) ± Magnetite ± Talc ± Chlorite ± Tremolite ± silica minerals (e.g. Opal) ± Magnesite ± Aragonite ± Garnierite Modified from Raymond (1995, table 31.3).
CHROME-SPINEL AND BLUE RIDGE OPHIOLITES
Methods of analysis Data for this continuing study of Blue Ridge meta-ultramafic rocks were collected both in the field and in the laboratory. In the field, mapping was conducted using traditional compass and pace methods of interpretive mapping and attitude measurement with a Brunton pocket transit. Cross-sections were constructed using both data collected in the field and data obtained from published mapping. Aspect ratios of ultramafic bodies are defined as, and were calculated from, the long axis of the meta-ultramafic bodies (usually measured from maps) divided by the maximum short axis measured perpendicular to the foliation plane of enclosing rocks. In the case of older literature, the foliation attitudes are often unreported and we used regional foliations as a guide to determine the direction in which to measure the short axis. In a few cases, the only measure possible was one of map width. In data taken from the literature, where only text descriptions were available, we used the dimensions provided. The samples used in this study were collected in the Hess Deep, a 2500 m deep fault basin at the west end of the Cocos-Nazca spreading centre in the eastern Pacific Ocean, and in the North Caroli-
257
na section of the Blue Ridge Belt of North America (Fig. 1). Hess Deep samples are from drill cores. Blue Ridge samples are grab samples selected as representative samples of the bodies being studied. Chemical analyses on minerals were conducted using electron microprobe techniques. Hess Deep samples were analysed by James Allan on a Cameca SX-50 microprobe using a focused beam, sample current of 30 nA, long counting times (for most elements), pure metal standards, and Smithsonian natural mineral standards (see Allan & Dick 1996). Blue Ridge samples were analysed by Sam Swanson, Anthony Love, Renee McCarter, and Loren Raymond using the University of Georgia, Department of Geology JEOL JXA 8600 Superprobe, with an accelerating voltage of 15 ky sample current of 15 nA, and a 2 urn beam diameter. Natural minerals were used as standards and several chromite substandards were used to check the analytical routine during each analytical run. Data were reduced online using a Bence & Albee matrix correction. Total iron was measured as FeO by the microprobe and was recalculated based on ideal stoichiometry to give the FeO and Fe2Os values. Typically, totals from analyses were in the range of 98 to >99 wt%. Low totals were probably related to assumptions involving iron
Table 2. Electron microprobe analyses of Cr-chlorites and Cr-spinels from Blue Ridge meta-ultramafic rocks Day Book DB 134 Chlorite
DB134 Cr-spinel
Frank
Henson Creek HC5 Chlorite
HC 5 Cr-spinel
0.030 32.250 0.050 32.460 SiO2 n.d. 0.050 0.050 n.d. TiO2 3.070 14.350 2.750 12.260 A1203 55.800 4.310 58.330 2.950 Cr203 0.42* 10.080 0.27* 7.770 Fe203 2.52* FeO 23.760 1.68* 21.540 NiO 0.050 0.170 0.220 0.070 n.d. 0.000 0.000 MnO n.d. 33.870 MgO 5.130 32.440 8.790 Total 97.650 83.430 97.460 86.740 Cations based on 28 oxygens (chlorite); 4 oxygens (chromite), cations normalized to 3 6.194 0.001 Si 6.387 0.001 Ti 0.001 0.001 Al 3.227 0.118 2.860 0.130 Cr 0.444 1.601 0.675 1.656 3+ 0.276 0.210 Fe 2+ _ _ 0.722 Fe 0.647 0.447 0.309 Fe2+totai 0.001 Ni 0.025 0.036 0.002 Mn 0.000 0.000 9.633 Mg 0.278 9.577 0.353
F84-5 Chlorite
F84-5 Cr-spinel
32.050 n.d. 13.720 2.250 0.32* 1.96* 0.140 n.d. 32.830 83.270
0.030 0.200 3.670 46.780 17.660 25.460 n.d. 0.000 4.250 98.050
6.322 3.191 0.351 _ 0.360 0.022 9.658
0.001 0.021 0.156 1.347 0.485 0.776 0.000 0.230
n.d., not detected. *Data obtained as FeOtotai and calculated based on stoichiometry (eight positive charges, three cations).
L. A. RAYMOND ETAL.
258
partitioning, but may be related to some unanalysed component. Mineral formulae calculated from the spinel analyses give satisfactory totals (Tables 2 and 7) and suggest the analyses are reasonable.
where a blackwall exists. No dykes have been observed extending from any of the bodies into the surrounding country rock and no zones of thermal (contact) metamorphism in surrounding rocks have been documented.
Field setting of Blue Ridge ultramafic rocks
Metamorphic character of Blue Ridge ultramafic rocks
Blue Ridge ultramafic rock bodies occur as isolated, concordant to discordant, scattered pods, lenses, slabs, and blocks in country rock of the Eastern and Central Blue Ridge belts (Table 3). These belts are composed predominantly of hornblende schist and gneiss ('amphibolite'), pelitic schist, quartzo-feldspathic schist and gneiss, or combinations of these rock types. As indicated in Figure 1, the bodies actually occur in two major thrust blocks (called terranes by some workers), the Gossan Lead Block and the Fries-Hayesville Block. Body lengths range from 1 km. Figure 2 shows maps and cross-sections of three typical Blue Ridge meta-ultramafic bodies. In general, the long axes of the bodies parallel the regional foliation and in the case of the Greer Hollow body, the meta-ultramafic rocks can be shown to have experienced folding and foliation development along with the enclosing rocks. Thus, this body (and probably others) was emplaced before deformation and some metamorphism. The contacts between country rocks and meta-ultramafic rocks are typically either sharp contacts or thin gradational zones of schist that locally form a 'blackwall' metasomatic reaction zone around the meta-ultramafic body (Sanford 1978; Swanson 1981). Contacts are relatively sharp, even in cases
The basic metamorphic character of Blue Ridge meta-ultramafic rocks has been delineated through a series of studies conducted over the past 30 years (Astwood et al. 1972; Dribus et al. 1976; Swanson & Raymond 1976; Honeycutt & Heimlich 1980; Swanson 1980, 1981, 1999, 2001; Abbott & Raymond 1984; Raymond & Abbott 1985, 1997; Raymond 1995, chapter 31; Raymond et al. 1988; Raymond et al 1999, 2001; Swanson et al. 1999; Warner 2001). Five metamorphic associations consisting of small domain, equilibrium assemblages are recognized (Table 1). These assemblages are preserved, because many of the bodies have resisted pervasive hydration and accompanying recrystallization. Thus, older assemblages have been only partly replaced by successively younger ones. The oldest assemblages are the highest grade assemblages and successively younger ones are hydrated and are of lower grade. The five groups of mineral assemblages (associations) in the meta-ultramafic rocks represent a declining P—T path of metamorphism that developed over almost 200 Ma (Abbott & Raymond 1984; see Adams et al 1995). The oldest assemblages (Association A-l) consist of olivine, chromium spinel (chromite), and, in some cases, ormopyroxene (clinopyroxene is extremely rare,
Table 3. Characteristic features of typical Blue Ridge ultramafic bodies and rock types Rock type and occurrence Hydration Dunite bodies
Low
Harzburgite bodies and lenses in dunite
Low
Orthopyroxenite bands in Low dunite and harzburgite Partially hydrated dunite Moderate and harzburgite bodies Chlorite and talc schist bodies Serpentinite bodies and veins in dunite bodies
High High
Contacts
Textures
Mineral associations
Discordant with blackwall, sheared, discordant- sharp
Equigranular, equigranular-tabular, porphyroclastic
A-l, A-2
Equigranular, equigranular-tabular, porphyroclastic Equigranular
A-l, A-2
Concordant with blackwall Discordant with blackwall, sheared, discordant-sharp Sheared, discordant-sharp Concordant to discordant, sharp with local blackwall
Equigranular, lepidoblastic to nematoblastic, felted Sharp, generally concordant Lepidoblastic to nematoblastic Sharp to gradational Lepidoblastic, meshtextured
A-l (Opx ± Ol ± Chr) A-l, A-2, A-3
A-3 A-4, A-5
CHROME-SPINEL AND BLUE RIDGE OPHIOLITES
259
Fig. 3. Photomicrographs and a photograph of meta-ultramafic rocks, (a) Photomicrograph of metadunite with Association A-l from the Day Book body, showing an equilibrium metamorphic texture with 120° grain boundary triple junctions between olivine grains, (b) Photomicrograph of metadunite of Association A-2 from Corundum Hill body showing Cr-chlorite associated with Cr-spinel. (c) Photomicrograph of porphyroblastic metaharzburgite of Association A-2 with chlorite and tremolite from the Hoots Body, (d) Photomicrograph of chlorite-anthophyllite schist of Association A-3 from the Greer Hollow body, (e) Photomicrograph of serpentinite of Association A-4 containing magnetite from the Greer Hollow body, (f) Photograph of aragonite vein (light needles) cutting metadunite of the Day Book Dunite. Coin diameter is 22 mm. Long dimension of field in (a), (b), and (d) is 6.5 mm, in (c) is 8 mm, and in (e) is 1 mm. Minerals: A, anthophyllite; Ag, aragonite, C, chrome spinel; Ch, chlorite; M, magnetite; O, olivine; Op, orthopyroxene; S, serpentine minerals; Tr, tremolite.
but does occur locally) (Table 1; Fig. 3a). These assemblages typically represent anhydrous upper amphibolite-, granulite-, or possibly eclogitefacies conditions of metamorphism. Textures in the rocks of the oldest association are clearly metamorphic (e.g. Dribus et al. 1976; Abbott & Raymond 1984; Raymond 1995, p. 667ff). Mineral assemblages of Association A-2, which partly replace those of Association A-l, are slightly hydrated assemblages containing olivine + Cr-
spinel + Cr-chlorite ± orthopyroxene + tremolite ± magnesiocummingtonite (Fig. 3b and c). Association A-2 represents amphibolite-facies conditions, but probably of lower P and T than Association A-l. Association A-3 commonly contains chlorite in abundance, plus additional hydrated phases such as anthophyllite, magnesiocummingtonite, phlogopite, and talc (Fig. 3d). This association represents low amphibolite-facies P—T conditions (Raymond 1995, chapter 31;
260
L. A. RAYMOND ETAL.
Fig. 4. Petrogenetic grid showing fields of contact metamorphism (C), greenschist facies (G), amphibolite facies (A), and eclogite facies (E), plus P-T—t metamorphic path for the Gossan Lead Block (Ashe and Alligator Back metamorphic suites of the Toe Terrane) and Fries Block (Pumpkin Patch metamorphic suite of the Cullowhee Terrane, i.e. Mars Hill Terrane) metamorphism. Metamorphic path based on Butler (1973), Abbott & Raymond (1984), Tenthorey et al. (1996), Adams & Trupe (1997), Goldberg & Dallmeyer (1997), Waters et al. (2000), Abbott & Greenwood (2001), Miller et al. (2001), and Warner (2001). Associations A-l-A-5 of Table 1.
Tenthorey et al 1996; Raymond & Abbott 1997; Warner 2001). Associations A-4 and A-5 represent hydrated conditions of lower grade (Fig. 4). Association A4 is typified by serpentine that occurs as a meshwork surrounding olivine grains, but is also represented by serpentine with magnetite in veins (Fig. 3e). The youngest mineral assemblages in many meta-ultramafic masses (Association A-5) fill veins in the metadunite or metaharzburgite and consists of serpentine and magnetite or, less commonly, minerals such as talc, aragonite, or garnierite (Fig. 3f). Veins of Association A-5 cut the serpentine meshwork of Association A-4. The serpentine-bearing assemblages probably represent P-T conditions of T +6%o are common throughout Earth history and are inconsistent with exchange and alteration by fluids strongly depleted in 18O. T, Canyon Mountain (Permian) and Bay of Islands (Ordovician) ophiolite; A, Darb Zubaydah Ophiolite, Saudi Arabia, an accreted arc terrane; •, Purtuniq ophiolite; O, Fortescue pillow lavas basement for the Hamersley Iron Formation; •, Barberton Mountain Belt; figure modified from Gregory (1991; references therein). The Purtuniq ophiolite data are from Holmden & Muehlenbachs (1993). The frequency distribution in the inset shows the results for over 100 greenstone silicate analyses from the Pilbara block, mainly from the North Pole Dome, Western Australia. DSDP, Deep Sea Drilling Project.
change estimates from sedimentology and climate modelling, and from astrophysical inferences that suggest a faint young Sun (e.g. Sagen & Mullen 1972; Hoffman et al 1998). If anything, inferences about the snowball Earth suggest that mean global temperatures may have been lower than isotopic estimates, which have them double or triple back into the past (e.g. Knauth & Epstein 1976). Third, one of the records, either the ophiolite or the carbonate record, is irrelevant to the isotopic composition of the oceans. The major contributions of ophiolite studies to the seawater oxygen isotope controversy are twofold. First, ophiolite studies confirmed inferences from dredge sample work summarized by Muehlenbachs & Clayton (1976) that hypothesized the existence of complementary reservoirs of 18Odepleted and 18O-enriched rocks in the altered oceanic crust. Second, the demonstration of the depth of penetration of seawater in a more complete geological context allows the estimation of the rate constants of exchange. Table 1 shows
rate constant calculations for Sr and O isotopes using the same assumptions of oceanic crustal production rates and continental weathering rates translated into numbers that make sense technically following the analysis of Gregory (1991; references therein) on the profiles shown in Figure 1. A global spreading rate of 3 km2 a"1 yields a characteristic age of oceanic crust of about 100 Ma assuming an oceanic crustal area of 3 X 108 km2. A chemical weathering rate of 3 km3 a"1 yields a mean age of continental crust of about 2500 Ma for a constant crustal volume of 7.5 X 109 km3 (e.g. Armstrong 1991). Tectonic rates significantly different from these rates are not geologically sustainable, otherwise the rock record would change dramatically. The age distributions on the continents and in the ocean basins would be substantially different from the current distribution with young ocean basins (5 km) seawaterhydrothermal circulation at mid-ocean ridges. Journal of Geophysical Research, 86, 2737-2755. GREGORY, R.T., GRAY, D.R. & MILLER, J.McL. 1998. Tectonics of the Arabian margin associated with the emplacement of the Oman margin along the Ibra transect: new evidence from NE Saih Hatat. Tectonics, 17, 657-760. HACKER, B.R. 1990. Simulation of the metamorphic and deformational history of the metamorphic sole of the Oman ophiolite. Journal of Geophysical Research, 95, 4895-4907. HACKER, B.R. 1991. The role of deformation in the formation of metamorphic gradients: ridge subduction beneath the Oman ophiolite. Tectonics, 10, 455-473. HACKER, B.R. 1994. Rapid emplacement of young oceanic lithosphere: argon geochronology of the Oman Ophiolite. Science, 265, 1563-1565. HACKER, B.R. & GNOS, E. 1997. The conundrum of Samail: explaining the metamorphic history. Tectonophysics, 279, 215-226. HACKER, B.R. & MOSENFELDER, J.L. 1997. Metamorphism and deformation of the sole beneath the Samail Ophiolite. Tectonics, 10, 455-473. HACKER, B.R., MOSENFELDER, J.L. & GNOS, E. 1996. Rapid emplacement of the Oman Ophiolite: thermal and geochronological constraints. Tectonics, 15, 1230-1247. HANNA, S. 1986. The Alpine deformation (Late Cretaceous and Tertiary) tectonic evolution of the Oman Mountains: a thrust tectonic approach. In: EDITOR, A. (ed.) Symposium on the Hydrocarbon Potential of Intense Thrust Zones: Organisation of Arab Petroleum Exporting Countries Conference. OAPEC, Kuwait, 2, 125-174. HANNA, S. 1990. The Alpine deformation of the Central Oman Mountains. In: ROBERTSON, A.H.F., SEARLE, M.P. & RIES, A.C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 341-359. HOPSON, C.A., COLEMAN, R.G., GREGORY, R.T., PALLISTER, J.S. & BAILEY, E.H. 1981. Geological section through the Samail Ophiolite and associated rocks along a Muscat-Ibra transect, SE Oman mountains. Journal of Geophysical Research, 86, 2527-2544. LE METOUR, J., DE GRAMONT, X. & VILLEY, M. 1986. Geological map of Masqat and Quryat, sheets NF40-4A, NF40-4D, scale 1:100000. Explanatory notes. Ministry of Petroleum and Minerals, Directorate General of Minerals, Sultanate of Oman. LE METOUR, J., RABU, D., TEGYEY, M., BECHENNEC, E, BEURRIER, M. & VILLEY, M. 1990. Subduction and obduction: two stages in the Eo-Alpine tectonometamorphic evolution of the Oman Mountains. /W.ROBERTSON, A.H.F., SEARLE, M.P. & RIES, A.C.
(eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 327-339. LE METOUR, J., MICHEL, J.C., BECHENNEC, F., PLATEL, J.P. & ROGER, J. 1995. Geology and Mineral Wealth of Oman, Ministry of Petroleum and Minerals, Muscat Oman, 285 pp. LIPPARD, SJ. 1983. Cretaceous high pressure metarnorphism in NE Oman and its relationship to subduction and ophiolite nappe emplacement. Journal of the Geological Society, London, 140, 97-104. LIPPARD, S.J., SHELTON, A.W. & GASS, I.G. 1986. The Ophiolite of Northern Oman. Geological Society of London, Memoir, 11. MANN, A. & HANNA, S.S. 1990. The tectonic evolution of pre-Permian rocks, Central and southeastern Oman Mountains. In: ROBERTSON, A.H.F., SEARLE, M.P. & RIES, A.C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 307-325. MATTAUER, M. & RITZ, J.-F. 1996. Arguments geologiques en faveur d'un modele de subduction continentale pour 1'exhumation du metamorphisme haute-pression d'Oman. Comptes Rendus de I'Academic des Sciences, Serie Ila, 322, 869-876. MlCHARD, A., BOUCHEZ, J.L. & OUAZZIANI-TOUHAMI,
O. 1984. Obduction-related planar and linear fabrics in Oman. Journal of Structural Geology, 6, 39—49. MICHARD, A., BOUDIER, F. & GOFFE, B. 1991. Obduction versus subduction and collision in the Oman case and other Tethyan settings. In: PETERS, TJ., NICOLAS, A. & COLEMAN, R. (eds) Ophiolite Genesis and Evolution of the Ocean Lithosphere. Kluwer Academic, Dordrecht, 447-467. MICHARD, A., GOFFE, B., SADDIQI, O., OBERHANSLI, R. & WENDT, A.S. 1994. Late Cretaceous exhumation of the Oman blueschists and eclogites: a two stage extensional mechanism. Terra Nova, 6, 404-413. MILLER, J.McL. 1998. The structure, metamorphism and 40 Ar/39Ar geochronology of northeast Saih Hatat, Oman: implications for the formation and preservation of high pressure rocks. Unpublished PhD thesis, Monash University, Clayton, Melbourne, Australia. MILLER, J.McL., GRAY, D.R. & GREGORY, R.T. 1998. Exhumation of high-pressure rocks in northeastern Oman. Geology, 26, 235-238. MILLER, J.McL., GRAY, D.R., GREGORY, R.T. & FOSTER, D.A. 1999. Geological and geochronological constraints on the exhumation of a high-pressure metamorphic terrane, Oman. In: RING, U., BRANDON, M.T., LISTER, G.S. & WILLET, S.D. (eds) Exhumation Processes: Normal Faulting, Ductile Flow and Erosion. Geological Society, London, Special Publications, 154, 241-260. MILLER, J.McL., GRAY, D.R. & GREGORY, R.T. 2002. Geometry and significance of internal windows and regional isoclinal folds in northeast Saih Hatat, Sultanate of Oman. Journal of Structural Geology^ 24, 359-386. MITCHELL, A.H.G. 1986. Ophiolite and associated rocks in four settings: relationships to subduction and collision. Tectonophysics, 125, 269-285.
OBDUCTION OF THE SAMAIL OPHIOLITE MONTIGNY, R., LE MER, O. & WHITECHURCH, H. 1988. K-Ar and Ar40/Ar39 study of metamorphic rocks associated with the Oman ophiolite: tectonic implications. Tectonophysics, 151, 345-362. MOORES, E.M. 1982. Origin and emplacement of ophiolites. Reviews of Geophysics, 20, 735-760. MOORES, E.M. & Twiss, R.J. 1995. Tectonics. W.H. Freeman, New York. MOUNT, V.S., CRAWFORD, R.I.S. & BERGMAN, S.C. 1998. Regional structural style of the central and southern Oman Mountains: Jabal Akhdar, Saih Hatat, and the Northern Ghaba Basin. GeoAmbia, 35 475-490. MUELLER, S. & PHILLIPS, R.J. 1991. On the initiation of subduction. Journal of Geophysical Research, 96, 651-665. NICOLAS, A. 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Dordrecht. NICOLAS, A., CEULENEER, G., BOUDIER, F. & MISSERI, M. 1988. Structural mapping in the Oman Ophiolites: mantle diapirism along an oceanic ridge. Tectonophysics, 51, 27-56. PEARCE, J.A., ALABASTER, T., SHELTON, A.W. & SEARLE, M.P. 1981. The Oman ophiolite as a Cretaceous arc-basin complex: evidence and implications. Philosophical Transactions of the Royal Society of London, Series A, 300, 299-317. ROBERTSON, A.H.F. & SEARLE, M.P. 1990. The Oman Tethyan continental margin: stratigraphy, structure, concepts and controversies. In: ROBERTSON, A.H.F., SEARLE, M.P. & RIES, A.C. (eds) Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 3-25. ROEDDER, D. 1973. Subduction and orogeny. Journal of Geophysical Research, 78, 5005-5024. ROEDDER, D. 1977. Continental convergence in the Alps. Tectonophysics, 40, 339-350. SEARLE, M.P. 1985. Sequence of thrusting and origin of culminations in the northern and central Oman Mountains. Journal of Structural Geology, 1, 129-143. SEARLE, M.P. & Cox, J. 1999. Tectonic setting, origin and obduction of the Oman ophiolite. Geological Society of America Bulletin, III, 104-122. SEARLE, M.P. & Cox, J. 2002. Subduction zone metamorphism during formation and emplacement of the Semail ophiolite in the Oman Mountains. Geological Magazine, 139, 241-255. SEARLE, M.P. & MALPAS, J. 1980. Structure and
465
metamorphism of rocks beneath the Semail ophiolite of Oman and their tectonic significance in ophiolite obduction. Transactions of the Royal Society of Edinburgh, 71, 247-262. SEARLE, M.P. & STEVENS, R.K. 1984. Obduction processes in ancient, modern and future ophiolites. In: GASS, I.G., LIPPARD, S.J. & SHELTON, A.W. (eds) Ophiolites and Oceanic Lithosphere. Geological Society Special Publications, London, 13, 303-319. SEARLE, M.P., WATERS, D.J., MARTIN, H.N. & REX, D.C. 1994. Structure and metamorphism of blueschist-eclogite facies rocks from the NE Oman Mountains. Journal of the Geological Society, London, 151, 555-576. SHELTON, A.W. & EGAN, S.S. 1991. The obduction of the northern Oman ophiolite: crustal loading and flexure. In: PETERS, TJ., NICOLAS, A. & COLEMAN, R. (eds) Ophiolite Genesis and Evolution of the Ocean Lithosphere. Kluwer Academic, Dordrecht, 469-483. TILTON, G.R., HOPSON, C.A. & WRIGHT, I.E. 1981. Uranium-lead isotopic ages of the Semail Ophiolite, Oman, with applications to Tethyan ridge tectonics. Journal of Geophysical Research, 86, 2763-2775. VILLEY, M., BECHENNEC, F., BEURRIER, M., LE METOUR, J. & RABU, D. 1986. Geological map of Yanqul, Sheet NF40-2C, Scale 1:100000. Explanatory notes. Ministry of Petroleum and Minerals, Directorate General of Minerals, Sultanate of Oman. WAKABAYASHI, J. & DILEK, Y. 2000. Spatial and temporal relationships between ophiolites and their metamorphic soles; a test of models of forearc ophiolite genesis. In: MOORES, E.M., ELTHON, D. & NICOLAS, A. (eds) Ophiolites and Oceanic Crust; New Insights from Field Studies and the Ocean Drilling. Geological Society of America, Special Papers, 349, 53-64. WARBURTON, J., BURNHILL, T.J., GRAHAM, R.H. & ISAAC, K.P. 1990. The evolution of the Oman Mountains foreland basin. In: ROBERTSON, A.H.F., SEARLE, M.P. & RIES, A.C. (eds) The Geology and Tectonics of the Oman Region. Geological Society, London, Special Publications, 49, 419-427. WENDT, A.S., D'ARCO, P., GOFFE, B. & OBERHANSLI, R. 1993. Radial cracks around a-quartz inclusions in almandine: constraints on the metamorphic history of the Oman mountains. Earth and Planetary Science Letters, 114, 449-461.
This page intentionally left blank
Subduction zone polarity in the Oman Mountains: implications for ophiolite emplacement M. P. SEARLE 1 , C. J. WARREN 1 , D. J. WATERS 1 & R. R. PARRISH 2 1
Department of Earth Sciences, Oxford University, Parks Road, Oxford OX1 3PR, UK (e-mail:
[email protected]) 2 Department of Geology, University of Leicester, and Natural Environment Research Council, Isotope Geosciences Laboratory, British Geological Survey, Keyworth, Nottingham NG12 5GG, UK Abstract: Two end-member models have been proposed to account for the structure and metamorphism of rocks beneath the Semail ophiolite in the Oman mountains. Model A involves a single, continuous NE-directed subduction away from the continental margin during the late Cretaceous. The ophiolite and underlying thrust sheets of distal to proximal oceanic sediments were emplaced a minimum of 250 km SW onto the continental margin. Subduction of Triassic—Jurassic oceanic basalts to c. lOkbar (c. 39km depth) led to the accretion of amphibolite-facies rocks to the base of the ophiolite. Thrusting propagated towards the continental margin and ended with subduction of the thinned continental crust to c. 20 kbar (c. 78 km depth), choking the subduction zone. Buoyancy forces caused the rapid exhumation of eclogites, blueschists and carpholite-grade HP rocks along the NE margin of the continental plate. During the later phase of foreland-propagating thin-skinned thrusting in the SW, NEfacing backfolding and backthrusting occurred in the hinterland, with the final exhumation of the HP rocks. Model B follows recent suggestions that a nascent SW-dipping subduction zone, dipping beneath the continental margin, existed between 130 and 95 Ma, prior to formation and emplacement of the ophiolite. A major NE-facing fold-nappe structure in the pre-Permian basement rocks of Sain Hatat is interpreted as reflecting subduction beneath the margin. Two high-pressure metamorphic events have been suggested, the first predating ophiolite emplacement, the second caused by ophiolite loading. This model is untenable, being based on a misinterpretation of the NE-facing structures in northern Saih Hatat, and on some dubious older 40 Ar/39Ar cooling ages from the eclogite-facies rocks of As Sifah. We conclude that all structures in northern Oman and all the reliable geochronology point to a single emplacement obduction event lasting from Cenomanian-Turonian time (c. 95 Ma) when amphibolites were accreted along the metamorphic sole of the ophiolite, to Campanian time, when the continental margin was subducted to the NE producing blueschists and eclogites, to the final thin-skinned emplacement of all thrust sheets, which ended before the Late Maastrichtian, at c. 68 Ma.
Since Glennie et al. (1973, 1974) published the first detailed stratigraphie and structural results of mapping the entire Oman Mountain belt, it has been universally recognized that the mountains comprise a series of relatively thin-skinned thrust sheets emplaced from NE to SW onto the stable passive continental margin of Arabia (Fig. 1). These allochthonous sheets comprise (from lower to higher units): Middle Permian to Cenomanian shelf slope carbonates (Sumeini Group), proximal to distal sediments (Hawasina complex), deepwater trench sediments and Triassic seamounts (Haybi complex; Searle & Malpas 1980; Searle 1985), and the Semail ophiolite. The Sumeini, Hawasina and Haybi thrust sheets comprise timeequivalent units of differing palaeogeographical
facies with higher units being more distal to the shelf margin. Thrusting generally propagated from NE to SW with more distal thrust sheets emplaced onto more proximal units. The Semail ophiolite crustal sequence formed during the Cenomanian, from crystallization ages of plagiogranites in the lower-crustal sequence (U-Pb zircon ages of 93.5-97.9 Ma with a mean of c. 95 Ma; Tilton et al. 1981), and from early Cenomanian to early Turonian radiolaria in cherts from the volcanic sequence (Tippit et al. 1981). Amphibolites in the metamorphic sole rocks, accreted along the base of the peridotites of the Semail ophiolite, formed at peak P-T conditions of 840-870 °C and 11.6 ± 1.6 kbar equivalent to 45-50 km depth beneath oceanic crust hanging
From: DILEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 467-480. 0305-8719/037$ 15 © The Geological Society of London 2003.
468
M. P. SEARLE ETAL.
Fig. 1. Geological map of the northern Oman Mountains.
wall (Gnos 1998; Searle & Cox 2002). 40Ar/39Ar cooling ages from hornblendes in the amphibolite sole are 95-92 Ma and record the timing at which these rocks cooled through 500 °C (Gnos & Peters 1993; Hacker 1994; Hacker et al 1996). Structural mapping, thermobarometry and geochronology clearly indicate that NE-directed oceanic subduetion of cold basaltic material to depths of 4550 km beneath the Semail ophiolite was occurring at c. 95 Ma (Searle & Malpas 1980, 1982; Lippard et al. 1986; Searle & Cox 1999, 2002). This NEdirected oceanic subduction system developed into thinner-skinned thrust tectonics with time as the Semail ophiolite, and the Haybi and Hawasina thrust sheets were progressively emplaced onto the passive margin of Arabia by a foreland-propagating 'piggy-back' style of thrusting (e.g. Searle 1985; Cooper 1988; Hanna 1990). In the southeastern part of the northern Oman mountains (Fig. 2) a high-pressure terrane is
centred around the village of As Sifah, where eclogite-facies metabasalts and metapelites crop out at the deepest structural levels exposed. The eclogites record P-T conditions of 540 ± 75 °C and c. 20kbar reflecting subduction of thinned continental crustal rocks to depths of about 7880km beneath a hanging wall that was always oceanic or ophiolitic (Searle et al. 1994; Searle & Cox 1999, 2002). Structurally overlying the eclogites are a series of blueschist- and carpholitegrade HP-LT rocks (Goffe et al 1988; El-Shazly & Coleman 1990; El-Shazly et al. 1990; El-Shazly & Liou 1991; Searle et al. 1994) showing an upwards decreasing thermal gradient. Most workers on Oman structural geology agree that subduction polarity throughout the emplacement process was NE-directed, associated with SW emplacement of oceanic units onto the continental margin (Model A, Fig. 3; e.g. Glennie et al. 1973, 1974; Searle 1985, 1988a, 1988b; Lippard et al. 1986;
SUBDUCTION ZONE POLARITY IN OMAN
469
Fig. 2. Map of the Muscat-Quriat and Saih Hatat region, after LeMetour et al. (1986) and Gray & Gregory (2000).
Searle & Cooper 1986; Bernoulli & Weissart 1987; Cooper 1988; Goffe et al. 1988; Searle et al 1990, 1994; Cawood et al 1990; Dunne et al 1990; Hanna 1990; LeMetour et al 1990; Hacker 1994; Hacker & Mosenfelder 1996; Hacker et al 1996; Searle & Cox 1999, 2002; El-Shazly et al 2001). Gregory et al (1998), Gray et al (2000) and Miller et al (2002) remapped northern Saih Hatat and proposed an entirely different tectonic model (Model B; Fig. 4). Their model involves two stages of subduction, an early, pre-ophiolite formation, 130-95 Ma SW-dipping subduction followed by a later NE-dipping subduction during emplacement, and two stages of HP metamorphism, a pre-ophiolite, Early-Mid-Cretaceous HP1,
which formed the eclogites, and an ophiolite emplacement related HP2 during the Late Cretaceous. Both these models are now critically assessed.
Model A: one NE-dipping subduction system Although some controversy still persists as to the original tectonic setting of the Semail ophiolite, either a mid-ocean ridge or suprasubduction zone setting (see Searle & Cox (1999, 2002) for a discussion), most workers agree that the emplacement of the ophiolite was related to NE-dipping, SW-verging thrusts emplacing thrust sheets of
470
M. P. SEARLE ETAL.
Fig. 3. Model A, after Searle et al. (1994). Tectonic model for the subduction-obduction evolution of the Oman ophiolite. (a) Palinspastically restored cross-section across the Arabian continental margin showing the positions of the major thrusts, the Hawasina Thrust (HT), Haybi Thrust (HyT) and Semail Thrust (ST). It should be noted that the shelf, slope (Sumeini) and basin (Hawasina) sediments are laterally time-equivalent units. The continental crust thins towards the margin in the NE. The bold-outlined boxes show the restored palaeostratigraphic positions of the protolith of the high-pressure metamorphic units along the continental margin in NE Oman. The crust beneath the distal Hawasina and Haybi units is dominantly alkali basaltic to transitional tholeiitic basaltic rocks of latest Permian—Triassic and Jurassic age. The sub-ophiolite amphibolites were formed at pressures around 7—10 kbar (28— 39 km depth) by accretion of Jurassic mid-ocean ridge basalt and Triassic Haybi-type basaltic crust to the base of the mantle sequence. Greenschists were subsequently accreted by underplating of Haybi complex sediments (pelites, quartzites and marbles) beneath the amphibolites and metamorphosed at pressures of 2-4 kbar (7-14 km depth). The Semail ophiolite plus metamorphic sole was then thrust onto the Haybi complex and thrusting propagated southwestwards with time, (b) During the later stages of the subduction-obduction process, the continental margin rocks were subducted to depths of around 80-90 km, where peak metamorphic pressures of c. 20 kbar are recorded by thermobarometry in the As Sifah unit eclogites. The sub-ophiolite thrust sheets (Haybi and Hawasina complexes) were emplaced onto the foreland, but are mainly missing in the internal (NE) parts of the Oman Mountains, where the ophiolite lies almost directly on the shelf carbonates. The subduction of continental margin material down to c. 80 km must have occurred in a phase of rapid plate motion during ophiolite obduction. Exhumation of the highpressure As Sifah unit rocks must also have occurred rapidly, back up to the same subduction zone, soon after peak metamorphism.
SUBDUCTION ZONE POLARITY IN OMAN
471
Fig. 4. Model B, after Gregory et al (1998) and Gray et al. (2000). In stages 1 and 2, a pre-Cenomanian platformdirected nascent subduction zone formed eclogites before ophiolite formation. Stage 3 is the intra-oceanic thrusting event when the ophiolite was formed and began emplacing towards the margin (95-80 Ma). During stage 4 the frontal part of the ophiolite overrode the nascent SW-dipping subduction zone. In stage 5 the present-day Semail 'thrust' formed, the folded ophiolite sheet was tectonically thinned and the HP rocks moved upward from south to north.
oceanic rocks onto the drowned continental shelf margin. Detailed structural mapping of individual culminations throughout the Oman mountains has shown that thrust sheets beneath the ophiolite have involved transport towards the Arabian continental
margin from east to west in the far north, NE to SW in the central mountains and NNE to SSW in the southeastern mountains. These include, for example, the Dibba zone in the far north (Searle 1988a, 1988b; Gnos 1998), the Sumeini Window,
472
M. P. SEARLE ETAL.
northern Oman (Searle & Malpas 1980, 1982; Searle & Cox 2002), the foreland fold-thrust belt in the Hamrat Duru range (Cooper 1988), the Asjudi and Hawasina Windows (Searle 1985; Searle & Cooper 1986), the Jebel Akhdar Window (Searle 1985; Bernoulli & Weissart 1987; Hanna 1990), the southeastern mountains (Mann & Hanna 1990) and culminations south of Saih Hatat (Cawood et al 1990). Seismic and well data beneath the foreland, both in north Oman and UAE (Dunne et al. 1990), and central Oman (Warburton et al. 1990) have confirmed the surface structural geology, i.e. that all thrust directions were towards the foreland. Detailed mapping of sub-ophiolite metamorphic sole rocks and imbricate structures beneath the Semail ophiolite confirms that thrusts and folds are related throughout to SW-directed emplacement. Imbricate faults and folds in the Hamrat Duru foreland thrust-fold belt also show that emplacement was SW-directed (Cooper 1988). Searle (1985) also described out-of-sequence thrusts, which truncate structures in the footwall on all scales from outcrop to the regional motion along the Semail 'thrust'. Many different units are in direct contact with the base of the ophiolite, and this geometry has been interpreted as late outof-sequence thrusting and/or late-stage extensional normal faulting. Although in most areas along the Oman mountains, SW-vergent folds and thrusts are evident, there are two areas where regional NE-verging thrusts and NE-facing folds have been documented, in the northern Hawasina Window (Searle & Cooper 1986) and in northern Saih Hatat (BRGM mapping; LeMetour et al. 1990; Gregory et al. 1998; Miller et al. 2002).
NE-vergent structures NE-vergent backthrusts and NE-facing folds were mapped by Searle & Cooper (1986) across the northern part of the Hawasina Window (Fig. 5). These folds are restricted to the central part of the northern Hawasina Window only. To the west in the Haybi region, and to the SE, at Jebel Akhdar, all structures show SW-directed thrusting and SWfacing folding (Searle 1985). The NE-facing folds affect the entire allochthonous tectonostratigraphy from Sumeini Group shelf edge carbonates up to the Semail ophiolite and the backthrusts cut earlier SW-directed thrusts in the footwall. NEdirected backfolding and backthrusting must therefore have formed at a late stage, after emplacement of all thrust sheets onto the margin. The internal geometry of the structures in the Hamrat Duru Group sedimentary rocks in the Hawasina Window shows an imbricate fan with SW-verging thrusts and SW-facing folds along the
southern margin of the window, and NE-vergent back thrusts and backfolds along the northern margin. These structures form a classic 'pop-up* structure related to divergent thrusts above a footwall ramp in the shelf margin. Searle and Cooper (1986) related the Hawasina Window backfolds to a local NE-facing promontory in the shelf edge along the basal decollement. These workers also described late out-of-sequence motion along the Semail 'thrust' around the Hawasina Window culmination, as the peridotites of the Semail mantle sequence rest on top of numerous different thrust sheets in the footwall. Minor extensional normal faults were also mapped, which had the effect of downthrowing the structurally higher ophiolite around the margins of the Window. In northern Saih Hatat, NE-facing folds attain very large proportions (LeMetour et al. 1986, 1990; Miller et al. 2002). These structures were noted by Glennie et al. (1974) during the initial mapping, and also by Bailey (1981) and during the BRGM mapping of the region (LeMetour et al. 1990), although their significance was not really publicized until Gregory et al. (1998), Gray et al. (2000) and Miller et al. (2002) described the structures in great detail. Our interpretation of the structures in Saih Hatat is shown in Figure 6, and the interpretation of Gregory et al. (1998), Gray et al (2000) and Gray & Gregory (2000) is shown in Figure 7. These structures form the main evidence for the model of SW-directed subduction (Model B) and the arguments against this will be laid out in the following section.
Timing of folding and thrusting The early SW-directed thrusting initiated with the accretion of HT amphibolites to the base of the ophiolite at 35-40 km depth beneath the ophiolite during the Cenomanian-Turonian (Searle & Malpas 1980, 1982; Gnos 1998). Thrusting then propagated SW with time, with successively outboard, more distal units thrust over more inboard, proximal units. As the shelf margin collapsed during the late Coniacian-early Santonian at 8884 Ma (Boote et al. 1990; Warburton et al. 1990) the Hawasina, Haybi and overlying Semail ophiolite thrust sheets were emplaced into the flexed Aruma basin along the continental margin. Final stages of the emplacement involved localized NEdirected backfolding in the northern Hawasina Window and along northern Saih Hatat. The entire process was over by the late Maastrichtian, when shallow marine rudist-bearing limestones (Simsima Formation) were deposited over all allochthonous units. Because no Tertiary cover rocks are exposed directly over the Hawasina Window, it is difficult
Fig. 5. Geological cross-sections across the Hawasina Window, after Searle and Cooper (1986), showing the late-stage localized backthrusting and backfolding to the NE. Sedimentary rocks of the Hamrat Duru Group include the Zulla Fm (Z), Guwayza sandstone (Gs) and limestone (Gl), Sidr chert (S) and Nayid limestone (N). The Sumeini group (Sm) shelf slope carbonates are the deepest structural unit and the most proximal palaeogeographically. Early SW-directed thrusts have been refolded around the fan structure and cut by later, steep NE-verging backthrusts.
474
M. P. SEARLE ETAL.
Fig. 6. Structural section across the Saih Hatat culmination and Ibra ophiolite block, showing our interpretation of the structural geometry. The HP units of As Sifah eclogites are the structurally deepest level exposed. The NE-facing backfolds of the Wadi Mayh region are upper-crustal retro-shears and backfolds, pinned to the 'autochthonous' shelf carbonates along the south side of Saih Hatat. Late-stage listric normal faults surround the Saih Hatat culmination, dropping the Ibra block ophiolite down to the south along the southern margin, and dropping the Muscat-Muttrah peridotite down to the north along the northern margin.
Fig. 7. Structural section across Saih Hatat, following the same section as shown in Figure 6, according to Gray & Gregory (2000). 1 and 2, Hawasina and Semail thrusts in the foreland belt; 3 and 4, interpreted crustal-scale shear zones; 5, Ibra dome; 6, Saih Hatat dome and regional-scale fold nappes; 7, normal faults off the coast of north Oman. Inset shows a structural profile across the Ibra block of the Semail ophiolite.
to precisely pin down the age of NE-direeted backfolding. However, it must be the youngest phase of late Cretaceous thrusting, because all earlier SW-directed thrusts and SW-facing folds have been refolded around the NE-verging backfolds. The style of folding in the PaleoceneEocene rocks of Jebel Abiad, south of the Hawasina Window, is very different, showing more open box-type folding, with upright or steep axial planes. Although shortening in the Tertiary jebels of Oman is minimal, only a few kilometres at most, uplift must have been substantial as Eocene-Oligocene limestones formed close to sea level have been uplifted to 2200 m on Jebel Abiad, SE of Quriat, south of Tiwi-Ash Shab in the
southeastern mountains (Fig. 2). Minor collapse type folding occurs along the flanks of these Tertiary mountain ranges, but Tertiary horizontal shortening is minimal.
Summary of main points (1) The NE-facing folds affect all units from deep-level eclogites up to the Permian Saiq Formation. They do not affect the highest thrust slices along the north coast from Muscat to Bandar Khuyran, that show south-directed thrusting and folding in the Triassic shelf units, Ruwi melange and ophiolite.
SUBDUCTION ZONE POLARITY IN OMAN (2) The major detachment termed the upper plate-lower plate discontinuity by Gregory et al. (1998) and Miller et al. (1998, 2002), carrying the large-scale NE-facing folds of northern Saih Hatat along the hanging wall, can be traced from surface geology down to only the upper Proterozoic basement, and not into the lower crust. (3) Metamorphic grade is similar across the upper plate-lower plate discontinuity, with carpholite-bearing metasedimentary rocks and crossite-bearing metabasalts indicating pressures of about 5-7 kbar. The major metamorphic break occurs between the Hulw unit and the As Sifah unit with a large pressure jump of around 8— 10 kbar. This contact is a major shear zone, but it is not a suture zone. (4) The timing of the NE-facing folds in Saih Hatat must have occurred prior to the deposition of the late Paleocene-early Eocene neo-autochthonous limestones, which unconformably cover all structures around the margin of northern Saih Hatat. (5) 40Ar/39Ar cooling ages of 82-79 Ma in the lower plate are associated with transposition fabrics and C' type shear bands during NE-directed shearing, and mica ages from the upper plate are 76-70 Ma (Miller et al. 1999). These cooling ages are consistent with late NE-directed backfolding associated with the final stages of Late Cretaceous culmination of Saih Hatat.
Model B: nascent SW-dipping subduction zone Gregory et al. (1998), Gray & Gregory (2000) and Gray et al. (2000) proposed a very different model involving initial subduction and high-pressure metamorphism of a small continental slab or microplate towards the Arabian continental margin, prior to ophiolite formation and emplacement (Fig. 4; stages 1 and 2). This 'nascent subduction zone' dipping SW beneath the continental margin resulted in the first stage of HP metamorphism at 130-95 Ma producing the eclogites at As Sifah. NE-directed shearing and exhumation of the eclogites occurred between 95 and 80 Ma at the same time as the ophiolite was thrusting towards the SW (Fig. 4; stage 3). A second stage of HP metamorphism of upper plate limestones and upper thrust sheets between 80 and 76 Ma resulted from structural thickening of the ophiolite (Fig. 4; stage 4), followed by gravitational collapse of the ophiolite onto the Arabian margin (Fig. 4; stage 5). Arguments against this model can be grouped into four categories, as described below.
475
Sedimentary evolution of the passive margin The Gray-Gregory model (Fig. 4) requires that the previously well-documented passive continental margin of Arabia (e.g. Glennie et al. 1973, 1974; Robertson & Searle 1990; Scott 1990) actually would have to be a destructive plate margin overlying a subduction zone, dipping SW beneath the continent. Normally, some geological evidence for this destructive margin in terms of calc-alkaline granites or volcanic rocks along an Andean-type margin would be expected. None exists throughout Oman. Indeed, the continental margin of Arabia throughout the Mid-Permian to Cenomanian was completely stable, with the sedimentation of highly fossiliferous, shallow marine carbonates, continuing through the Cretaceous up to Cenomanian times, when the Semail ophiolite crustal sequence was formed and began to emplace SW onto the margin. The sedimentarystratigraphic record suggests that it is inconceivable that the stable, shelf carbonates of the Kahmah and Wasia Groups of the Cretaceous shelf sequence were deposited anywhere other than along a passive, aseismic margin (Glennie et al. 1974; Searle et al. 1983; Rabu et al. 1990; Scott 1990). Extremely detailed biostratigraphy and chronostratigraphy of the Cretaceous shelf sequence throughout the Oman mountains shows that the Lekhwair, Kharaib, Shuaiba and Nahr Umr Formations were deposited on a stable, gently subsiding wide continental shelf (Scott 1990). The first indication of the foreland basin deposition comes in the Coniacian-Santonian (88.5-84 Ma) when the shelf edge began to collapse to accommodate the incoming thrust sheets of the Hawasina, Haybi and Semail allochthons (Warburton et al. 1990). Throughout the period 130-95 Ma, required for SW subduction beneath the margin by the GrayGregory model, there is no indication of any tectonic or magmatic activity along the shelf margin during that period whatsoever. Thick carbonate passive margins do not normally overlie active subduction zones. Size and shape of the subducting microplate The Gray-Gregory model (Fig. 4) requires a suture zone to be present along the top of their proposed 'microplate' during their stage 2 SWdirected subduction. The size of the microplate could only be the area of the HP rocks in the As Sifah unit in the lower plate, a present-day area of 2 contents (0.703.00wt%) and chondrite-normalized rare earth
element (REE) patterns, ranging from light REE (LREE) depleted (LaN/YbN 2) to LREE strongly enriched (LaN/YbN >3) (Table 1; Figs 3c, d and 4). The LREE-depleted basalts are of N-MORB character, whereas those slightly enriched in LREE are depleted in high field strength elements (HFSE) (e.g. Th > Ta; Th/Ta - 2.7-3.9; Zhu et al 1987;
OPHIOLITES IN CHINA
547
Fig. 4. Ti-Zr (Pearce & Cann 1973) and Ta-Th-Hf (Wood 1980) diagrams. Legend as in Figure 3. OFB, ocean flow basalt; LKT, low-K tholeiite; CAB, calc-alkaline basalt; WPB, within plate basalt; WPT, within plate tholeiite; VAB, volcanic arc basalt.
Xiao et al. 1992) and are more characteristic of island-arc basalts. The LREE-enriched basalts also have high Ti and P, and are probably ocean island lavas (Xiao et al. 1992). It is generally agreed that the Tangbale ophiolite formed in a forearc or back-arc oceanic basin
(Xiao et al. 1992). This part of the ocean basin opened originally in the late Cambrian-early Ordovician and closed in the mid- to late Ordovician. The emplacement age of the Tangbale ophiolite is considered to be later than early Silurian (Xiao et al. 1992).
548
Q. ZHANG ET AL.
Fig. 5. A map illustrating the general geology and distribution of ophiolites in the Palaeo-Asian tectonic belt of northern China (after Tang & Shao 1996).
East Junggar ophiolite belt The East Junggar ophiolites include Kalamaili, Aermantai, Kekesentao-Qiaoxiahala and Kuerti (Fig. 7). The Kalamaili ophiolite crops out on the eastern margin of the Junggar Basin, and consists of serpentinite (altered mainly from harzburgite), gabbro and lava. The various rock units are in fault contact so there is no continuous section exposed. Basalts in this body have moderate to high TiO2 contents (1.31-2.73%). Chondritenormalized REE patterns show both LREE depletion and slight enrichment (Li 1991, 1995). The oceanic basin from which the Kalamaili ophiolite was derived probably formed in the early Devonian and closed in the early Carboniferous (Li 1991, 1995).
Tianshan Ophiolite Belt Four ophiolite sub-belts have been recognized in the Tianshan region (Gao et al. 1998), which from north to south consist of: (1) late Palaeozoic ophiolites of the north Tianshan (represented by the Bayingou ophiolite); (2) early Palaeozoic ophiolites in the northern part of the middle Tianshan; (3) early Palaeozoic ophiolites in the
southern part of the middle Tianshan; (4) late Palaeozoic ophiolites of the south Tianshan (Fig. 8). The Changawuzi ophiolite is representative of the early Palaeozoic ophiolites in the southern part of the middle Tianshan. It has been extensively metamorphosed under greenschist-, blueschistand eclogite-facies conditions. The metabasic rocks mostly have E-MORB compositions, although a few show LREE depletion characteristic of N-MORB, suggesting formation in a backarc basin (Tang et al 1995). 40Ar/39Ar ages of phengite and glaucophane from the blueschists range from 315 to 415 Ma, suggesting formation near the end of the late Silurian (Tang et al, 1995). Basalts in the late Palaeozoic ophiolites of the south Tianshan are markedly depleted in Ti, P and LREE, and have low Zr/Y ratios (2.0-2.5). Their eNd(0 values range from +6.8 to +7.6 (where t is 334 Ma) (Gao et al. 1998), indicating formation in a suprasubduction zone setting.
Inner Mongolian ophiolite belt There are two major ophiolite sub-belts in Inner Mongolia (Fig. 9): Ondor Sum-Kedanshan-Xilamulun River (early Palaeozoic) and Solonshan-
OPHIOLITES IN CHINA
549
Fig. 6. The distribution of West Junggar ophiolites (after Bai et al. 1995).
Hegenshan (late Palaeozoic) (Bao et al. 1994; Robinson et al. 1999; Nozaka & Liu 2002). Ondor Sum ophiolites occur in the lower part of the Ondor Sum Group and consist of metabasalt and metagabbro with minor diorite locally intruded by diabase dykes. A few lenses of metaperidotite occur in the gabbro. The ophiolites are associated with high-pressure glaucophane-lawsonite schists. The volcanic rocks in Kedanshan have TiO2 contents ranging from 0.75 to 2.20 wt% and averaging 1.49wt%. The lavas have both LREE-depleted and LREE-slightly enriched patterns, similar to those of N-MORB and E-MORB, respectively. The Hegenshan ophiolite is located along the boundary of the Siberian and North China Blocks in Inner Mongolia. The ophiolite has a Sm-Nd isochron age of 403 ± 27 Ma, with eNd(7) of +8.7 ± 0.6 (Bao et al. 1994) suggesting formation in the Devonian or early Carboniferous. The Hegenshan ophiolite consists of several blocks composed chiefly of serpentinized ultramafic rocks with lesser amounts of troctolite and gabbro and sparse lavas and dykes. The ultramafic rocks consist chiefly of depleted harzburgite and minor dunite and are interpreted as mantle tectonites. In
the Hegenshan block dunite is relatively abundant and is typically associated with podiform chromitite. Several blocks have well-layered cumulate sequences of gabbro and troctolite. Sheeted dykes are absent but small mafic dykes are common in some of the ultramafic sections (Robinson et al. 1999). Basaltic rocks from Hegenshan are chiefly tholeiitic in character (Table 1) with flat REE patterns showing slight LREE depletion. Nozaka & Liu (2002) suggested the ophiolite formed at a typical mid-ocean ridge whereas Robinson et al. (1999) postulated formation in a suprasubduction zone environment, probably a back-arc basin. A few basalts are enriched in LREE and may be ocean island lavas (Robinson et al. 1999).
Qinling-Qilian-Kunlun ophiolites West Kunlun ophiolites The Kudi ophiolite is typical of the west Kunlun belt. The ophiolite consists of peridotite, minor cumulate gabbro, diabase dykes and thick-bedded basalts and basaltic andesites, associated with chert, abyssal turbidites and sea-floor olistoliths (Jiang et al. 1992; Yang et al. 1996; Wang et al.
550
Q. ZHANG ETAL.
Fig. 7. The distribution of East Junggar ophiolites (after He et al 1990).
2001b). Yang et al. (1996) suggested that the Kudi ophiolite was formed during Sinian and Cambrian time, and that the ocean closed at the end of the Silurian. A slightly younger age (OrdovicianSilurian) is suggested by recently discovered radiolaria in the Kudi melange (Zhou et al. 1998) Spinels in the Kudi peridotite are relatively high in Cr (Cr number 0.57-0.67, Yang et al 1996). Basalts are low in HFSE, and plot in the islandarc field on a Y-Cr diagram (Jiang et al. 1992; Yang et al. 1996). They have MORB-normalized spider diagrams similar to those of basalts from Tonga, confirming a suprasubduction zone origin (Figs 3c, d and 4; after Yang et al. 1996). Boninite is also present in the Kudi ophiolite and is characterized by high Mg (MgO 8.57-10.58 wt%,
Mg number 0.62-0.72) and Cr (299-516 ppm) and low HFSE and heavy REE (HREE). TiO2 is 450Ma) is underlain by the 320 Ma Renge blueschist and the Upper Permian (250 Ma) Akiyoshi accretionary complex. This spatial relationship suggests tectonic erosion or non-accretion during the intervening Siluro-Devonian time. A similar gap exists between the Lower Permian (280 Ma) Yakuno ophiolite and the underlying Jurassic Tamba accretionary complex (150 Ma). The accretionary complex is characterized by 'oceanic plate stratigraphy' composed of greenstone, chert, limestone, mudstone and sandstone in a younging order (Isozaki 1997). The basal greenstone commonly includes tholeiitic and alkaline
seamount basalt (OIB) with high Ti and Nb concentrations, but the ophiolite itself is almost free of OIB. In the present-day western Pacific, the IzuMariana and Tonga subduction zone environments are characterized by the presence of ophiolite outcrops on the trench slopes (Bloomer & Hawkins 1983; Bloomer & Fisher 1987) and blueschists (Maekawa et al. 1993, 1995), by the absence or scarcity of accretionary complexes, and by the currently active back-arc spreading. In contrast, areas off northeastern Honshu and Hokkaido are characterized by the development of vast accretionary complexes (Taira 1985) without submarine ophiolite or blueschist outcrops and without active back-arc spreading (Fig. 9). These different environments may have been repeated in any segment of the Japanese orogenic belt throughout Phanerozoic time. Periods of oceanic island arc and marginal basin development (ophiolite formation) and tectonic erosion (blueschist metamorphism) might have alternated with periods of normal subduction, during which accretionary complexes were developed. The ophiolite-blueschist association is well documented in the Japan-Primorye area, e.g. the Oeyama ophiolite-Renge blueschist (Tsujimori & Itaya 1999) and Sergeevka ophiolite-Shaiginskiy blueschist (Kovalenko & Khanchuk 1991; Zakharov et al. 1992). In the NE Japan-Sakhalin belt, the Palaeozoic Miyamori ophiolite-Motai blueschist pair (Maekawa 1988; Ozawa 1988, 1994) and the Mesozoic Horokanai ophiolite-Kamuikotan blueschist pair (Ishizuka 1985, 1987; Sakakibara & Ota 1994) are well documented. Although a major blueschist belt is absent in the Koryak Mountains, many blueschist blocks occur in the Palaeozoic and Mesozoic accretionary complexes (Stavsky et al. 1990; Dobretsov 1999). It is likely that periods of accretion and nonaccretion, as represented by the present-day Nankai Trough and Mariana Trench, respectively, have been repeated many times in different segments of the Japan-NE Russia accretionary orogenic belts in the past. Periods of ophiolite-blueschist formation and tectonic erosion at subduction zones might have been followed by periods of massive accretion. Tectonic underplating of accreted sediments beneath the mantle wedge might have facilitated the uplift of overlying ophiolite-blueschist assemblages. This idea is compatible with the geochemical signatures of ophiolitic rocks showing SSZ affinities.
Conclusions The northwestern Pacific margin extending from Japan to Russia has many ophiolites of widely
OPHIOLITES IN JAPAN AND FAR EAST RUSSIA varying ages, different petrological characteristics and distinctive tectonic histories. The following geological features suggest that these ophiolites probably formed in island arc environments in intra-oceanic settings: extremely diverse degree of melting in the residual mantle peridotite up to clinopyroxene disappearance and spinel Cr-number >0.70; the common occurrence of hydrous minerals and various metasomatic features in the mantle section; the common association with blueschist rocks; the presence of unusually thick oceanic crust. The modern Mariana and Tonga trenches, where ophiolitic rocks including highly depleted harzburgite and typical blueschist have been dredged from the sea floor, may be the modern analogues. The orogenic belts from Japan to NE Russia may have evolved through repeated stages of non-accretion, in which SSZ ophiolites and blueschists formed, and accretion, in which accretionary complexes mainly composed of clastic and volcaniclastic rocks developed. The association of highly depleted mantle harzburgite and orthopyroxene-type cumulate rocks is reinforced by reported occurrences of DH-type ophiolites from NE Russia (Shelting and Krasnaya). These ophiolites have only been reported so far from the western Pacific margins such as Hokkaido (Horokanai), Papua, and Tasmania (Adamsfield). The association of some DH-type ophiolites with boninitic volcanic rocks (Shelting, Papua, and possibly Mariana and Tonga) suggests that the depleted harzburgite is a residuum after boninitic melt production, although boninite is also reported from some ophiolites with less depleted peridotite (e.g. Robinson et al. 1983; Spadea & Scarrow 2000). Some primitive island arc tholeiite and magnesian andesite magmas could also coexist with the depleted harzburgite. The depleted harzburgite may form by either hightemperature dry melting of primary mantle or hydrous melting of previously depleted mantle. However, Os isotope studies of ophiolitic chromitites do not support much involvement of slabderived fluids in mantle melting. The Os isotope data are also inconsistent with an oceanic plateau (or LIP) origin postulated for some ophiolites with thick crustal sections. These instead may represent robust magmatic activity in SSZ environments, where they would be associated with highly depleted harzburgite massifs. We are grateful for T. Tsujimori, D. Saito, S. Miyashita, A. P. Stavsky, O. Morozov, S.A. Shcheka, A. I. Khanchuk, S. G. Byalobzhesky, W. B. Bryan and J. Hourigan for their help with fieldwork in Far East Russia. A.I. thanks S. A. Palandzhjan for valuable information on the ophiolites of the Koryak Mountains, and Y. Dilek for his encouragement in writing this paper. S. Maruyama is acknowledged for his help in arranging financial support
613
for our fieldwork. A.I. also acknowledges Grant-in-Aid for Scientific Research (C)-(2)-10640462 and -14540447 by the Ministry of Education, Japan. S. Arai is thanked for several discussions. Y. Furuhashi and D. Saito contributed the mineral analyses.
References AHMED, A.H. & ARAI, S. 2002. Unexpectedly high-PGE chromitite from the deeper mantle section of the northern Oman ophiolite and its tectonic implications. Contributions to Mineralogy and Petrology, 143, 263-278. ALABASTER, T., PEARCE, J.A. & MALPAS, J. 1982. The volcanic stratigraphy and petrogenesis of the Oman ophiolite complex. Contributions to Mineralogy and Petrology, 81, 168-183. ARAI, S. 1980. Dunite-harzburgite-chromitite complexes as refractory residue in the Sangun-Yamaguchi zone, western Japan. Journal of Petrology, 21, 141-165. ARAI, S. ET AL. 1996. Petrology of the gabbrotroctolite-peridotite complex from Hess Deep, Equatorial Pacific: implications for mantle-melt interaction within the oceanic lithosphere. In: MEVEL, C. (ed.) Proceedings of the Ocean Drilling Program, Scientific Results, 147. Ocean Drilling Program, College Station, TX, 135-155. BAZYLEV, B.A., PALANDZHJAN, S.A., GANELIN, A.V., SlLANTYEV, S.A., ISHIWATARI, A. & DMITRENKO,
G.G. 2001. Petrology of peridotites from the ophiolite melange in Cape Povorotny, Taigonos Peninsula, NE Russia: mantle processes beneath a subduction zone. Petrology, 9(2), 142-160. BEARD, J.S. 1986. Characteristic mineralogy of arcrelated cumulate gabbros: implications for the tectonic setting of gabbroic plutons and for andesite genesis. Geology, 14, 848-851. BELYI, V.F. & AKININ, V.V. 1985. Geologic Structure and Ophiolites in the Elistratova Peninsula, Volumes 1 and 2. Academy of Sciences of the USSR, Far East Science Centre, Magadan (in Russian). BLOOMER, S.H. & FISHER, R.L. 1987. Petrology and geochemistry of igneous rocks from the Tonga Trench—a non-accreting plate boundary. Journal of Geology, 95, 469-495. BLOOMER, S.H. & HAWKINS, J.W. 1983. Gabbroic and ultramafic rocks from the Mariana Trench: an island arc ophiolite. In: HAYES, D.E. (ed.) The Tectonic and Geologic Evolution of Southeast Asian Seas and Islands, Part 2. Geophysical Monograph, American Geophysical Union, 27, 294-317. BLOOMER, S.H. & HAWKINS, J.W. 1987. Petrology and geochemistry of boninite series volcanic rocks from the Mariana trench. Contributions to Mineralogy and Petrology, 97, 361-377. BOUDIER, F. & NICOLAS, A. 1985. Harzburgite and Iherzolite subtypes in ophiolitic and oceanic environments. Earth and Planetary Science Letters, 76, 84-92. BRYAN, W.B. 1991. Exploring Pacific seafloor ashore: Magadan Province, USSR. Oceanus (Woods Hole Oceanographic Institution), Summer 1991, 48-61.
614
A. ISHIWATARI ETAL
CANNAT, M., BIDEAU, D. & HEBERT, R. 1990. Plastic deformation and magmatic impregnation in serpentinized ultramafic rocks from the Garrett transform fault (East Pacific Rise). Earth and Planetary Science Letters, 101, 216-232. £INA, A., TASHKO, A. & TERSHANA, A. 1987. The Bulqiza and Gomsiqe ultrabasic massifs, ophiolites of Albanides: a geochemical comparison. Ofioliti, 12, 219-236. COFFIN, M.F. & ELDHOLM, O. 2001. Large igneous provinces: progenitors of some ophiolites? In: ERNST, R.E. & BUCHAN, K.L. (eds) Mantle Plumes: Their Identification through Time. Geological Society of America, Special Papers, 352, 59-70. COLEMAN, R.G. 1971. Plate tectonic emplacement of upper mantle peridotites along continental edges. Journal of Geophysical Research, 76, 1212-1222. COLEMAN, R.G. 1986. Ophiolites and accretion of the North American Cordillera. Bulletin de la Societe Geologique de France, 1986, 961-968. DEBARI, S.M. & COLEMAN, R.G. 1989. Examination of the deep levels of an island arc: evidence from the Tonsina ultramafic-mafic assemblage, Tonsina, Alaska. Journal of Geophysical Research, 94(B4), 4373-4391. DICK, H.J.B. 1989. Abyssal peridotites, very slow spreading ridges and ocean ridge magmatism. In: SAUNDERS, A.D. & NORRY, MJ. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications, 42, 71-105. DICK, H.J.B. & BULLEN, T.B. 1984 Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology, 86, 54-76. DILEK, Y., MOORES, E.M. & FURNES, H. 1998. Structure of modern oceanic crust and ophiolites and implications for faulting and magmatism at oceanic spreading centers. In: BUCK, W.R., DELANEY, P.T., KARSON, J.A. & LAGABRIELLE, Y. (eds) Faulting and Magmatism at Mid-Ocean Ridges. Geophysical Monograph, American Geophysical Union, 106, 219-265. DMITRENKO, G.G., MOCHALOV, A.G., PALANDZHJAN, S.A. & GORYACHEVA, E.M. 1985. Chemical Composition of the Rock-forming and Accessory Minerals of the Koryak Highland Alpine-type Ultramafics. Part 1. Rock-forming Minerals. Academy of Sciences of the USSR, Far East Science Centre, Interdisciplinary Research Institute, Magadan (in Russian). DMITRENKO, G.G., MOCHALOV, A.G. & PALANDZHJAN, S.A. 1990. Petrology and Platinum Occurrence in Lherzolite Massifs of the Koryak Mountains. Academy of Sciences of the USSR, Far East Science Centre, Interdisciplinary Research Institute, Magadan (in Russian). DOBRETSOV, N.L. 1999. Systematic periodicity of glaucophane metamorphism: an illusion or an important geologic regularity. Petrology, 7, 410-439. DOBRETSOV, N.L., WATANABE, T., NATAL'IN, B. & MIYASHITA, S. 1994. Comparison of ophiolites and blueschists of Sakhalin and Hokkaido. Ofioliti, 19, 157-176. EDWARDS, SJ. 1995. Boninitic and tholeiitic dykes in the Lewis Hills mantle section of the Bay of Islands
ophiolite: implications for magmatism adjacent to a fracture zone in a back-arc spreading environment. Canadian Journal of Earth Science, 32, 2128-2146. EDWARDS, SJ. & MALPAS, J. 1996. Melt-peridotite interactions in shallow mantle at the East Pacific Rise: evidence from OOP Site 895 (Hess Deep). Mineralogical Magazine, 60, 191-206. ENGLAND, R.N. & DA VIES, H.L. 1973. Mineralogy of ultramafic cumulates and tectonites in eastern Papua. Earth and Planetary Science Letters, 17, 416-425. FRANCIS, D. 1976. Corona-bearing pyroxene granulite xenoliths and the lower crust beneath Nunivak Island, Alaska. Canadian Mineralogist, 14, 291-298. FUJITA, K. & NEWBERRY, J.T. 1982. Tectonic evolution of northeastern Siberia and adjacent regions. Tectonophysics, 89, 337-357. FURUE, R., NIIDA, K. & KATOH, T. 1997. Origin of ophiolitic chromitites in the Kamuikotan Belt, Hokkaido. Bulletin of Hobetsu Museum, 13, 9-26 (in Japanese with English abstract). GAGGERO, L. & CORTESOGNO, L. 1997. Metamorphic evolution of oceanic gabbros: recrystallization from subsolidus to hydrothermal conditions in the MARK area (OOP Leg 153). Lithos, 40, 105-131. GREGOIRE, M., COTTIN, J.Y., GIRET, A., MATTIELLI, N. & WEIS, D. 1998. The meta-igneous granulite xenoliths from Kerguelen Archipelago: evidence of a continent nucleation in an oceanic setting. Contributions to Mineralogy and Petrology, 133, 259-283. HEBERT, R. & LAURENT, R. 1990. Mineral chemistry of the plutonic section of the Troodos ophiolite: new constraints for genesis of arc-related ophiolites. In: MALPAS, J., MOORES, E.M., PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophiolites: Oceanic Crust Analogues. Proceedings of the Troodos '87 Symposium. Geological Survey Department, Nicosia, 149-163. HERZIG, C.T., KIMBROUGH, D.L. & HAYASAKA, Y. 1997. Early Permian zircon uranium-lead ages for plagiogranites in the Yakuno ophiolite, Asago district, Southwest Japan. Island Arc, 6, 396-403. IRWIN, W.P. 1981. Tectonic accretion of the Klamath Mountains. In: ERNST, W.G. (ed.) The Geotectonic Development of California (Rubey Volume 1). Prentice Hall, Englewood Cliffs, NJ, 29-49. ISHII, T., ROBINSON, P.T., MAEKAWA, H. & FISKE, R. ET AL. 1992. Petrological studies of peridotites from diapiric serpentinite seamounts in the IzuOgasawara-Mariana forearc. In: FRYER, P., PEARCE, J. & STOKKING, L.B. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 125. Ocean Drilling Program, College Station, TX, 445-485. ISHIWATARI, A. 1985a. Igneous petrogenesis of the Yakuno ophiolite (Japan) in the context of the diversity of ophiolites. Contributions to Mineralogy and Petrology, 89, 155-167. ISHIWATARI, A. 1985b. Granulite-facies metacumulates of the Yakuno ophiolite, Japan: evidence for unusually thick oceanic crust. Journal of Petrology, 26, 1-30. ISHIWATARI, A. 1991. Time-space distribution and
OPHIOLITES IN JAPAN AND FAR EAST RUSSIA petrologic diversity of Japanese ophiolites. In: PETERS, TJ., NICOLAS, A. & COLEMAN, R.G. (eds) Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer, Dordrecht, 723-743. ISHIWATARI, A. 1994. Circum-Pacific Phanerozoic multiple ophiolite belts. In: ISHIW ATARI, A., MALPAS, J. & ISHIZUKA, H. (eds) Circum-Pacific Ophiolites (Proceedings of the 29th IGC, Kyoto, Part D). VSP, Utrecht, 7-28. ISHIW ATARI, A. & TSUJIMORI, T. 2003. Paleozoic ophiolites and blueschists in Japan and Russian Primorye in the tectonic framework of East Asia: a synthesis. Island Arc, 12, 190-206. ISHIWATARI, A., IKEDA, Y. & KOIDE, Y. 1990. The Yakuno ophiolite, Japan: fragments of Permian island arc and marginal basin crust with a hotspot. In: MALPAS, J., MOORES, E.M., PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophiolites: Oceanic Crust Analogues. Proceedings of the Troodos '87 Symposium. Geological Survey Department, Nicosia, 497-506. ISHIZUKA, H. 1985. Prograde metamorphism of the Horokanai ophiolite in the Kamuikotan Zone, Hokkaido, Japan. Journal of Petrology, 26, 391-417. ISHIZUKA, H. 1987. Igneous and metamorphic petrology of the Horokanai ophiolite in the Kamuikotan zone, Hokkaido, Japan: a synthetic thesis. Memoirs of Faculty of Science, Kochi University, Series E, 8, 1-70. ISOZAKI, Y. 1997. Contrasting two types of orogen in Permo-Triassic Japan: accretionary versus collisional. Island Arc, 6, 2-24. JAQUES, A.L. & CHAPPELL, B.W. 1980. Petrology and trace element geochemistry of the Papuan ultramafic belt. Contributions to Mineralogy and Petrology, 75, 55-70. JAQUES, A.L. & GREEN, D.H. 1980. Anhydrous melting of peridotite at 0-15 kb pressure and the genesis of tholeiitic basalts. Contributions to Mineralogy and Petrology, 73, 287-310. KHANCHUK, A.I. & PANCHENKO, I.V. 1994. Mineral assemblages of the Gankuvayam ophiolitic section in the Kuyul ophiolitic terrane of the Russia Far East. Ofioliti, 19, 257-268. KOJIMA, S. 1989. Mesozoic terrane accretion in Northeast China, Sikhote-Alin and Japan regions. Palaeogeography, Palaeoclimatology, Palaeoecology, 69, 213-232. KOVALENKO, S.V. & KHANCHUK, A.I. 1991. First find of glaucophane schist in Sikhote Alin. Doklady Akademii Nauk SSSR, 318, 692-694 (in Russian). KUROKAWA, K. 1985. Petrology of the Oeyama ophiolitic complex in the Inner Zone of Southwest Japan. Science Reports, Niigata University, Series E, Geology and Mineralogy, 6, 37-113. LACHIZE, M., LORAND, J.P. & JUTEAU, T. 1996. Calcalkaline differentiation trend in the plutonic sequence of the Wadi Haymiliyah section, Haylayn massif, Semail ophiolite, Oman. Lithos, 38, 207-232. LOOCK, G., STOSCH, H.-G. & SECK, H.A. 1990. Granulite facies lower crustal xenoliths from the Eifel, West Germany: petrological and geochemical aspects.
615
Contributions to Mineralogy and Petrology, 105, 25-41. LOPEZ, J.C. & ISHIWATARI, A. 2002. Petrogenesis of the tholeiitic basalt, calc-alkaline basaltic andesite and high magnesian andesite lava succession of the Oligo-Miocene Anamizu Formation in northeastern Noto Peninsula, central Japan. Journal of Mineralogical and Petrological Sciences, 97, 85-113. LUCHITSKAYA, M.V. 1996. Plagiogranites of the Kuyul ophiolite massif (northeastern Kamchatka, Koryak Upland). Ofioliti, 21, 131-138. MAEKAWA, H. 1988. High PIT metamorphic rocks in Northeast Japan: Motai-Matsugadaira zone. Earth Science (Chikyu Kagaku), 42, 212-219 (in Japanese with English abstract). MAEKAWA, H., SHOZUI, M., ISHII, T., FRYER, P. & PEARCE, J.A. 1993. Blueschist metamorphism in an active subduction zone. Nature, 364, 520-523. MAEKAWA, H., FRYER, P. & OZAKI, A. 1995. Incipient blueschist-facies metamorphism in the active subduction zone beneath the Mariana forearc. In: TAYLOR, B. & NATLAND, J. (eds) Active Margins and Marginal Basins of the Western Pacific. Geophysical Monograph, American Geophysical Union, 88,281-289. MATSUMOTO, I. & ARAI, S. 1997. Characterization of chromian spinel as a tool of petrological exploration for podiform chromitite. Resource Geology, 47, 189-199. MELCHER, F., GRUM, W., SIMON, G., THALHAMMER, T.V. & STUMPEL, E.F. 1997. Petrogenesis of the ophiolitic giant chromite deposits of Kempirsai, Kazakhstan: a study of solid and fluid inclusions in chromite. Journal of Petrology, 38, 1419-1458. MIYASHIRO, A. 1973. The Troodos ophiolitic complex was probably formed in an island arc. Earth and Planetary Science Letters, 19, 218-224. MIYASHITA, S. 1983. Reconstruction of the ophiolite succession in the western zone of the Hidaka Metamorphic Belt, Hokkaido. Journal of the Geological Society of Japan, 89, 69-86. MIYASHITA, S. & HASHIMOTO, S. 1975. The layered basic complex of Mt. Poroshiri, Hokkaido, Japan. Journal of Faculty of Science, Hokkaido University, Series IV, 16, 421-452. MIYASHITA, S. & YOSHIDA, A. 1988. Pre-Cretaceous and Cretaceous ophiolites in Hokkaido, Japan. Bulletin de la Societe Geologique de France, 1988, 251-260. MOORES, E.M. 1969. Petrology and Structure of the Vourinos Ophiolitic Complex of Northern Greece. Geological Society of America, Special Papers, 118. 74 p. MOORES, E.M. & VINE, F.J. 1971. The Troodos massif, Cyprus and other ophiolites as oceanic crust: evaluation and implications. Philosophical Transactions of the Royal Society of London, Series A, 268, 443-466. MORISHITA, T. & ARAI, S. 2001. Petrogenesis of corundum-bearing mafic rock in the Horoman Peridotite complex, Japan. Journal of Petrology, 42, 1279-1299. NICOLAS, A. 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer, Dordrecht.
616
A. ISHIWATARI ETAL
NICOLAS, A. & JACKSON, E.D. 1972. Repartition en deux provinces des peridotites des chaines alpines longeant la Mediterranee: implications getectoniques. Schweizerische Mineralogische und Petrographische Mitteilungen, 52, 479-495. NISHIMURA, Y. 1998. Geotectonic subdivision and areal extent of the Sangun belt, Inner Zone of Southwest Japan. Journal of Metamorphic Geology, 16, 129-140. OHARA, Y., KASUGA, S. & ISHII, T. 1996. Peridotites from the Parece Vela rift in the Philippine Sea: upper mantle material exposed in an extinct backarc basin. Proceedings of Japan Academy, Series B, 72, 118-123. OHARA, Y., STERN, J.R., ISHII, T., YURIMOTO, H. & YAMAZAKI, T. 2002. Peridotites from the Mariana Trough: first look at the mantle beneath an active back-arc basin. Contributions to Mineralogy and Petrology, 143, 1-18. OZAWA, K. 1988. Ultramafic tectonite of the Miyamori ophiolite complex in the Kitakami Mountains, Northeast Japan: hydrous upper mantle in an island arc. Contributions to Mineralogy and Petrology, 90, 159-175. OZAWA, K. 1994. Melting and melt segregation in the mantle wedge above a subduction zone: evidence from the chromite-bearing peridotites of the Miyamori ophiolite complex, northeastern Japan. Journal of Petrology, 35, 647-678. PALANDZHJAN, S.A. 1986. Ophiolite belts in the Koryak upland, northeast Asia. Tectonophysics, 127, 341-360. PEARCE, J.A., LIPPARD, S.J. & ROBERTS, S. 1984. Characteristics and tectonic significance of suprasubduction zone ophiolites. In: KOKELAAR, B.P. & HOWELLS, M.F. (eds) Marginal Basin Geology. Geological Society, London, Special Publications, 16, 79-95. PUSHCHAROVSKY, YU.M., RUZHENTSEV, S.V. & SOKO-
LOV, S.D. 1988. Tectonic thrust sheets and geologic mapping. Geotectonics, 22, 1-8 (original Russian paper published in 1969). ROBINSON, P.T., MELSON, W.G., O'HEARN, T. & SCHMINCKE, H.-U. 1983. Volcanic glass compositions of the Troodos ophiolite, Cyprus. Geology, 11, 400-404. SAITO, D., ISHIWATARI, A., TSUJIMORI, T., MIYASHITA, S. & SOKOLOV, S.D. 1999. Elistratova ophiolite in Taigonos peninsula, Far East Russia: an island arc ophiolite intruding into oceanic mantle. Memoirs of the Geological Society of Japan, 52, 303-316 (in Japanese with English abstract). SAKAKIBARA, M. & OTA, T. 1994. Metamorphic evolution of the Kamuikotan high-pressure and lowtemperature metamorphic rocks in central Hokkaido, Japan. Journal of Geophysical Research, 99, 22221-22235. SHCHEKA, S.A., ISHIWATARI, A. & VRZHOSEK, A.A. 2001. Geology and petrology of Cambrian Khanka ophiolite in Primorye (Far East Russia) with notes on its manganese-rich chromian spinel. Earth Science (Chikyu Kagaku), 55, 265-274. SHIBATA, K., IGI, S. & UCHIUMI, S. 1977. K-Ar ages of hornblendes from gabbroic rocks in Southwest
Japan. Geochemical Journal, 11, 57-64. SOBOLEV, A.V. & DANYUSHEVSKY, L.V. 1994. Petrology and geochemistry of boninites from the north termination of the Tonga Trench: constraints on the generation conditions of primary high Ca-boninite magmas. Journal of Petrology, 35, 1183-1211. SOKOLOV, S.D. 1992. Accretion Tectonics in the Koryak- Chukotka Segment of the Pacific Belt. Nauka, Moscow (in Russian). SPADEA, P. & SCARROW, J.H. 2000. Early Devonian boninites from the Magnitogorsk arc, southern Urals (Russia): implications for early development of a collisional orogen. In: DILEK, Y., MOORES, E.M., ELTHON, D. & NICOLAS, A. (eds) Ophiolites and Oceanic Crust: New Insights from Field Studies and the Ocean Drilling Program. Geological Society of America, Special Papers, 349, 461-472. STAVSKY, A.P., CHEKHOVICH, V.D., KONONOV, M.V. & ZONENSHAIN, L.P. 1990. Plate tectonic and palinspastic reconstructions of the Anadyr-Koryak region, Northeast Russia. Tectonics, 9, 81-101. STEINMANN, G. 1927. Die ophiolitischen Zonen in den Mediterranen Kettengebirgen. In: JACOB, C.H. (ed.) Geologic de la Mediterranee. Proceedings of the 14th International Geological Congress, Madrid. Graficas Reunidas, Madrid, 2, 638-667. SUHR, G. & EDWARDS, S.J. 2000. Contrasting mantle sequences exposed in the Lewis Hills massif: evidence for the early, arc-related history of the Bay of Islands ophiolite. In: DILEK, Y. ET AL. (eds) Ophiolites and Oceanic Crust: New Insights From Field Studies and the Ocean Drilling Program. Geological Society of America, Special Papers, 349, 433-442. TAIRA, A. 1985. Sedimentary evolution of Shikoku subduction zone: the Shimanto belt and Nankai trough. In: NASU, N. ET AL. (eds) Formation of Active Ocean Margins. Terra, Tokyo, 835-851. TAKAHASHI, N. 1991. Evolutional history of the uppermost mantle of an arc system: petrology of the Horoman peridotite massif, Japan. In: PETERS, TJ, NICOLAS, A. & COLEMAN, R.G. (eds) Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer, Dordrecht, 195-205. TAMURA, A., MAKITA, M. & ARAI, S. 1999. Petrogenesis of ultramafic rocks in the Kamuikotan belt, Hokkaido, northern Japan. Memoirs of the Geological Society of Japan, 52, 53-68 (in Japanese with English abstract). TAZAWA, J. 1988. Paleozoic-Mesozoic stratigraphy and tectonics of the Kitakami Mountains, northeast Japan. Earth Science (Chikyu Kagaku), 42, 165178 (in Japanese with English abstract). TILMAN, S.M. & BOGDANOV, N.A. 1992. Tectonic Map of Northeast Asia, 1:5,000,000, with Explanatory Notes. Institute of the Lithosphere, Moscow, 000. TSUJIMORI, T. & ISHIWATARI, A. 2002. Granulite facies relics in the Early Paleozoic kyanite-bearing ultrabasic metacumulate in the Oeyama belt, the Inner Zone of southwestern Japan. Gondwana Research, 5, 823-835. TSUJIMORI, T. & ITAYA, T. 1999. Blueschist-facies metamorphism during Paleozoic orogeny in southwestern Japan: phengite K-Ar ages of blueschist-
O P H I O L I T E S IN JAPAN AND FAR EAST RUSSIA facies tectonic blocks in a serpentinite melange beneath early Paleozoic Oeyama ophiolite. Island Arc, 8, 190-205. TSUJIMORI, T., SAITO, D., ISHIWATARI, A., MIYASHITA, S. & SOKOLOV, S.D. 1999. Electron microprobe element image of zoned chromian spinel with hydrous mineral inclusions in a chromitite from Elistratova ophiolite, Far East Russia. Science Reports ofKanazawa University, 43, 1-11. TSUJIMORI, T., ISHIWATARI, A. & BANNO, S. 2000. Eclogitic glaucophane schist from the Yunotani valley in Omi Town, the Renge metamorphic belt, the Inner Zone of southwestern Japan. Journal of the Geological Society of Japan, 106, 353-362. UMINO, S., YANAI, S., JAMAN, A.R., NAKAMURA, Y. & IIYAMA, J.T. 1990. The transition from spreading to subduction: evidence from the Semail ophiolite, Northern Oman Mountains. In: MALPAS, J., MOORES, E.M., PANAYIOTOU, A. & XENOPHONTOS, C. (eds) Ophiolites: Oceanic Crustal Analogues. Proceedings of Troodos '87 Symposium. Geological Survey Department, Nicosia, 375-384. VARNE, R. & BROWN, A.V. 1978. The geology and petrology of the Adamsfield ultramafic complex,
617
Tasmania. Contributions to Mineralogy and Petrology, 67, 195-207. VYSOTSKIY, S.V. 1994. High and low pressure cumulates of Paleozoic ophiolites in Primorye, eastern Russia. In: ISHIWATARI, A., MALPAS, J. & ISHIZUKA, H. (eds) Circum-Pacific Ophiolites. Proceedings of the 29th IGC, Kyoto, Vol. D. VSP, Utrecht, 145-162. VYSOTSKIY, S.V., GOVOROV, G.L, KEMKIN, I.V. & SAPIN, V.I. 2000. Boninitic ophiolite assemblage on eastern Sakhalin Island: geology and petrogenesis. Geology of Pacific Ocean, 15, 1067-1090. WALKER, R.J., PRICHARD, H.M., ISHIWATARI, A. & PIMENTEL, M. 2002. The osmium isotopic composition of convecting upper mantle deduced from ophiolite chromites. Geochimica et Cosmochimica Acta, 66, 329-345. YAMANE, M., BAMBA, M. & BAMBA, T. 1988. The first finding of orbicular chromite ore in Japan. Mining Geology, 38, 501-508. ZAKHAROV, Yu.D., PANCHENKO, I.V. & KHANCHUK, A.I. 1992. A Field Guide to the Late Paleozoic and Early Mesozoic Circum-Pacific Bio- and Geologic Events. Russian Academy of Sciences, Far East Geological Institute, Vladivostok.
This page intentionally left blank
Ophiolites in accretionary complexes along the Early Cretaceous margin of NE Asia: age, composition, and geodynamic diversity S. D. SOKOLOV 1 , M. V. LUCHITSKAYA 1 , S. A. SILANTYEV 2 , O. L. MOROZOV 1 , A. V. G A N E L I N 1 , B. A. BAZYLEV 2 , A. B. OSIPENKO 3 , S. A. P A L A N D Z H Y A N 4 & I. R. K R A V C H E N K O - B E R E Z H N O Y 1 1
Geological Institute of the Russian Academy of Sciences, 7 Pyzhevsky per., Moscow, 109017, Russia (e-mail:
[email protected]) 2
Vernadsky Institute of Geochemistry and Analytical Chemistry, 19 Kosygina St., 119991, Moscow, Russia ^Vernadsky State Geological Museum, 11/2 Mokhovaya St., Moscow, 103009, Russia
A
Institute of the Lithosphere of Marginal and Inland Seas, Russian Academy of Sciences, 22 Staromonetny per., Moscow, 109180, Russia
Abstract: The existing published data, combined with our own new field, petrographic, and geochemical observations and data show that ophiolites of the West Koryak fold system originated in a variety of tectonic environments. This fold system stretches along the boundary shared by two of NE Asia's largest tectonic units, the Verkhoyansk-Chukotka and KoryakKamchatka foldbelts. The fold system abounds in Palaeozoic and Mesozoic ophiolites and sedimentary and volcanic island-arc assemblages. The ophiolites are Palaeozoic and Mesozoic in age. The variety of geological and geochemical signatures implies ophiolite origin in diverse tectonic settings. The Early Palaeozoic ophiolites of the Ganychalan accreted terrane and Devonian(?) ophiolites of the Ust-Belaya accreted terrane are fragments of the Panthalassan oceanic lithosphere. Serpentinite melange in the Ust-Belaya terrane contains some blocks of island-arc provenance. They are probably Late Palaeozoic-Early Mesozoic in age as determined by K-Ar measurements, which require validation by other techniques. Mesozoic, chiefly Late Jurassic-Early Cretaceous ophiolites of the Beregovoi and Kuyul accreted terranes, originated in a suprasubduction-zone (SSZ) setting (ensimatic island arc and back-arc basin). Among the Mesozoic ophiolites, one finds blocks of oceanic assemblages in serpentinite melanges as well. Basalt and chert blocks of clearly oceanic derivation are viewed as detached fragments of the upper part of the oceanic lithosphere. The ophiolites have experienced a variety of accretionary scenarios. Palaeozoic ophiolites docked onto the Koni-Taigonos island arc (of Late Palaeozoic-Early Mesozoic age), probably in the Late Palaeozoic or Early Mesozoic, whereas Mesozoic ophiolites accreted onto the Uda-Murgal island arc (of Late Jurassic-Early Cretaceous age) in the terminal Early Cretaceous. Sedimentary deposits, whose base is late Albian in age, make a post-accretionary sequence. These island arcs portray the overall history of the convergent boundary between the North Asian continent and NW Pacific. Ophiolites of the Ganychalan and Ust-Belaya terranes consist of thrust sheets and, jointly with Yelistratov Peninsula ophiolites, make up the basement to the forearc of the Uda-Murgal island arc, ophiolites of Cape Povorotny and Kuyul terrane being incorporated in accretionary prisms of the same arc. Ophiolites and associated metamorphic, volcanic, and sedimentary rocks of Palaeozoic-Early Cretaceous age underwent three deformation phases, each reflecting a different stage in the evolution of the NE Asian continental margin and readily correlative with principal tectonic events in the northern Circum-Pacific region.
In NE Asia, ophiolites have been reported from the Verkhoyansk-Chukotka and Koryak-Kamchatka foldbelts (Fig. 1). The ophiolites span a broad Early Palaeozoic to Mesozoic age interval, The Verkhoyansk-Chukotka belt, except for the Kolyma loop (Fig. 1), displays a structural grain dominated by northwesterly trends resulting from
collisional processes (Pushcharovsky et al. 1992; Bogdanov & Tilman 1992; Parfenov et al. 1993). Accreted terranes found in the belt represent fragments of microcontinents, such as Chukotka, Omolon, Okhotsk, etc. Ophiolites occur sporadically within collisional piles of the Chersky Range and South Anyui suture, where they make up
From: DlLEK, Y. & ROBINSON P. T. (eds) 2003. Ophiolites in Earth History. Geological Society, London, Special Publications, 218, 619-664. 0305-8719/037$ 15 © The Geological Society of London 2003.
620
S. D. SOKOLOV ETAL.
Fig. 1. Tectonic map of NE Asia (by S. D. Sokolov and G. Ye. Bondarenko). Black areas are ophiolites: 1.1, Cape Povorotny; 1.2, Yelistratov Peninsula; 2.1, Kuyul terrane; 2.2, Ganychalan terrane; 3.1, Ust-Belaya terrane.
small, undeformed slices associated with greenschist- and amphibolite-facies metamorphic rocks (Parfenov et al. 1993; Nokleberg et al 1994; Oxmanetal. 1995). The Koryak-Kamchatka foldbelt is located east of the Okhotsk-Chukotka Volcanic Belt and stretches north-south to NE-SW (Fig. 1). It is a typical example of an accretionary continental margin formed through successive docking onto the Asian continent of outboard terranes having a variety of ages and geodynamic settings and arriving from the Pacific (Zonenshain et al. 1990; Bogdanov & Til'man 1992; Pushcharovsky et al.
1992; Sokolov 1992; Parfenov et al. 1993; Nokleberg et al. 1994). The terranes are of the following types: island arc, ophiolite, back-arc and turbidite basins, oceanic crust, and accretionary prism. Ophiolites are widespread, making up major bodies and entire terranes (Markov et al. 1982; Peyve 1984; Palandzhyan 1992; Sokolov 1992; Nokleberg et al. 1994). No consensus exists regarding the age, composition, or provenance of NE Asian ophiolites. Some workers (e.g. Fujita & Newberry 1982; Parfenov 1984; Zonenshain et al. 1990) view the ophiolites as fragments of a large oceanic basin, a
OPHIOLITES OF NORTHEAST ASIA former constituent of the Palace-Pacific. Others believe them to be relics of minor oceanic basins or rifts (Lychagin et al. 1991; Bogdanov & Til'man 1992; Oxman et al. 1995), and still others suggest that both Palace-Pacific and back-arc basin fragments come into play (Peyve 1984; Sokolov 1992; Nokleberg et al 2001). This controversy is due primarily to poor knowledge of the ophiolite assemblages; hence the critical importance of the new structural and compositional data presented here. Dating the ophiolites is crucial to unravelling their history. According to earlier workers (e.g. Coleman 1984; Ishiwatari 1994; Dilek et al. 1999; Searle & Cox 1999; Shervais 2001), ophiolites originate from a variety of tectonic environments, that obviates any palaeotectonic reconstructions or formative scenarios for the NE Asian margins without first identifying geodynamic affinities of the ophiolites. In addition, there are virtually no English-language publications on NE Asian ophiolites; hence, one more objective of this paper is to fill in this informational gap. The ophiolites discussed in this paper occur in the West Koryak fold system, which is located at the junction of the Verkhoyansk-Chukotka and Koryak-Kamchatka foldbelts (Fig. 1). In recent years, we have acquired new data on field relationships and evolution of this major tectonic unit, which incorporates accreted ophiolitic assemblages of various ages (Sokolov et al. 1999;
621
Silantyev et al. 2000). The paper focuses on new data and issues related to the tectonic setting, inner structure, composition, and age of the ophiolites.
Geological framework In NE Asia, four major tectonic units (Fig. 1) with distinctive structural grains and geological histories are recognized: (1) the Siberian craton; (2) the Verkhoyansk-Chukotka foldbelt; (3) the Koryak-Kamchatka foldbelt; (4) the OkhotskChukotka continental-margin volcanic belt (OCVB) (Fig. 1). These units represent continuous, albeit discrete, accretion onto the Siberian craton of geodynamically diverse terranes and microcontinents (Fig. 2). The West Koryak fold system lies along the boundary between the Koryak-Kamchatka and Verkhoyansk-Chukotka foldbelts. Most of the West Koryak fold system is overlain by OCVB volcanic and sedimentary rocks, and the latter also unconformably overlap the Verkhoyansk-Chukotka foldbelt. The OCVB is a Late Cretaceous Andean-type continental-margin volcanic belt. It was initiated after a major mid-Cretaceous (Aptian-Albian) phase of accretion (Fig. 2) onto the Asian continent (Sokolov 1992). The West Koryak fold system incorporates numerous island-arc assemblages and ophiolites that were brought together at the end of the Early
Fig. 2. Reconstruction showing continental growth of NE Asia (after S. D. Sokolov 1992).
622
S. D. SOKOLOV ETAL.
Cretaceous (Markov et al 1982; Parfenov 1984; Zonenshain et al 1990; Sokolov 1992; Parfenov etal 1993). The island-arc volcanic and sedimentary assemblages are of calc-alkaline affinity and range in age from Carboniferous to Early Cretaceous. Parfenov (1984) attributed these rocks to a continuous Late Palaeozoic to Mesozoic KoniMurgal island arc. However, Filatova (1988) viewed the Late Jurassic to Early Cretaceous volcanic and sedimentary sequences as part of the Uda-Murgal island arc. Zonenshain et al. (1990) identified the Koni-Murgal volcanic belt as a separate system, which they interpreted as an agglomeration of island-arc assemblages of various ages that were joined together in Mid-Cretaceous time. According to those workers, the original position of these assemblages is unknown, although they are believed to have been formed a considerable distance away from Siberia's continental margin (Zonenshain et al. 1990). These considerations drew on the pioneering palaeomagnetic data pinpointing the Omolon terrane in the Late Palaeozoic and Early Mesozoic at more southerly latitudes. These data, however, were at odds with palaeobiogeographical conclusions on the boreal faunas and Angara floras (Shapiro & Ganelin 1988). Sokolov (1992) postulated two convergent boundaries of contrasting ages in the region; one of Late Palaeozoic to Early Mesozoic age, during which the Koni-Taigonos island arc existed, and the other of Late Jurassic to Early Cretaceous age, composed of the Uda-Murgal island-arc system (Fig. 1). The volcanic and sedimentary assemblages of the Koni-Taigonos island arc are best exposed and most thoroughly studied in the Koni-Pyagina and Taigonos peninsulas (Nekrasov 1976; Zaborovskaya 1978). These areas provide a stage for reconstructing, in Permian to Mid-Jurassic times, the volcanic arc proper and the North Taigonos back-arc basin. These assemblages are also present in the Penzhina District (Khudoley & Sokolov 1998), in the Pekulnei Range, and in Chukotka (Morozov 2001). In the Penzhina segment, Carboniferous island-arc assemblages are exposed in the Kharitonya terrane and in thrust sheets within the Upupkin terrane; in this locality, they consist of coarse andesitic pyroclastic rocks and tuffaceous epiclastic rocks of Permian and Triassic ages (Khudoley & Sokolov 1998; Sokolov et al. 1999). In the Pekulnei Range and in Chukotka, the Late Palaeozoic to Early Cretaceous island-arc sequence includes metavolcanic and metasedimentary rocks, layered gabbros, and Early Mesozoic granitic rocks (Morozov 2001). Unfortunately, numerous aspects of the KoniTaigonos island arc are still unclear. These include
both the arc's polarity and basement composition, as well as the origin of its various segments. Faunal and floral data point to rock formation at high latitudes (Shapiro & Ganelin 1988; Sokolov 1992), which, in combination with structural data (Sokolov et al. 1999) and spatial position, suggest that the arc originated along a convergent boundary between the Asian continent and the NW Pacific. Various outboard terranes that arrived from the Palaeo-Pacific were accreted onto the arc (Zonenshain et al. 1990; Parfenov et al. 1993). These terranes are best exposed in the Penzhina segment, where they include the Ganychalan composite terrane and metamorphic rocks of the Upupkin terrane. Fragments of these assemblages have also been reported from the pre-arc basement of the Taigonos segment of the Uda-Murgal arc and as Ordovician deposits and ophiolites (Fig. 1; see also Fig. 15, below). Upper Jurassic to Lower Cretaceous volcanic and sedimentary rocks of the Uda-Murgal islandarc system are traceable for about 3000 km, from the Mongolia-Okhotsk foldbelt in the south along the Sea of Okhotsk coastline (via the Koni, Pyagina, and Taigonos peninsulas) as far northeastward as the Chukchi Peninsula (Fig. 1). The volcanic and sedimentary lithologies and the character and age of basement vary from place to place along the arc. In the southern segment, the only identifiable features are the volcanic portion of the island-arc system and some constituents of its associated back-arc basin. Island-arc assemblages rest on heterogeneous basement that includes fragments of the Asian continent (Siberian craton, Verkhoyansk complex, Okhotsk microcontinent), and the Koni-Taigonos Late Palaeozoic to Early Mesozoic island arc. Hence, the Late Jurassic to Early Cretaceous convergent boundary was located at an angle to the pre-existing structural grain. Throughout the study area, volcanic arc assemblages were located along the continental margin, strongly suggesting the existence of a continental-margin belt. The Taigonos and Penzhina segments provide evidence for reconstructing a lateral succession: volcanic arc-forearc-accretionary prism-trenchoceanic plate (Fig. 3). Basement to the island arc was provided by the pre-existing Koni-Taigonos arc with its accreted terranes, including the Early Palaeozoic ophiolites of the Ganychalan terrane. Within these segments, island-arc deposits were also formed in a continental margin setting. Further NE, however, the back-arc region was the locus of marine deposition, and the continentalmargin belt gave way to an ensialic arc (Sokolov etal 1999). In the Pekulnei segment, island-arc assemblages rest on heterogeneous basement that incorporates
OPHIOLITES OF NORTHEAST ASIA
623
Fig. 3. Taigonos segment of the Uda-Murgal island arc (Late Jurassic-Early Cretaceous).
fragments of both lower continental crust and oceanic lithosphere (Sokolov et al. 1999; Morozov 2001). A back-arc basin floored with oceanic crust was situated behind the arc and was probably linked to the Anyui palaeo-ocean. The northeastern Chukotka branch (Fig. 1) of the convergent boundary had heterogeneous basement that incorporated, among other things, ancient sialic crust. Consumption of oceanic crust along the Chukotka branch was not extensive, probably because of the strike-slip nature of plate interaction within this segment (Morozov 2001). Ophiolites of the West Koryak fold system (Fig. 1) occur either in forearc basement (type 1) or within accretionary prisms of the Uda-Murgal island arc (Fig. 1) (type 2). Type 1 ophiolites were accreted in the Late Palaeozoic-Early Mesozoic (Parfenov 1984; Sokolov et al. 1999) onto the Koni-Taigonos island arc. Type 2 ophiolites were accreted in Late Jurassic and, mainly, in Early Cretaceous times onto the frontal part of the UdaMurgal island-arc system (Parfenov 1984; Khanchuk et al 1990; Sokolov et al 1999). Ophiolites have been reported from the Taigonos, Penzhina, and Ust-Belaya segments of the island arc. Ophiolites are well exposed along the SE coast of the Taigonos Peninsula (Fig. 1). The largest ophiolitic outcrops occur within the accretionary pile exposed at Cape Povorotny and in pre-arc basement on the Yelistratov Peninsula (Belyi & Akinin 1985; Ishiwatari et al. 1998). In the Penzhina segment, two large ophiolitic terranes, the Ganychalan and Kuyul terranes, are well documented (Markov et al. 1982; Palandzhyan 1992; Ganelin & Peyve 2001; Nekrasov et al. 2001).
Analytical techniques Major and trace element analyses were carried out in various laboratories using a range of methods. Mineral compositions from Cape Povorotny rocks were measured in polished sections on an automated CAMEBAX-Microbeam four-channel
wavelength-dispersive electron probe at the Vernadsky Institute (GEOKHI). Whole-rock major element analyses from peridotites and gabbros were performed by X-ray fluorescence (XRF) on a Philips PW-1600 XRF automated multichannel spectrometer, and REE contents were determined by instrumental neutron activation analysis (INAA) at GEOKHI. All analytical investigations of the volcanic rocks were carried out at the Analytical Centre of the Geological Institute, Russian Academy of Sciences (GIN RAS) by INAA and inductively coupled plasma mass spectrometry (ICP-MS). Mineral chemistry of Yelistratov Peninsula ophiolitic peridotites and Ganychalan terrane plutonic rocks was analysed by N. N. Kononkova on a CAMECA CAMEBAX electron microprobe at GEOKHI at an accelerating voltage of 15 kV and beam current of 35 nA. Natural and synthetic minerals were used as standards. Rock chemistry was analysed on a PLASMA QUAD PQZ+Turbo (VG Instruments) mass spectrometer at the Institute of Mine Geology, Petrography, Mineralogy, and Geochemistry, Moscow. Routine sample preparation included dissolution in concentrated HF + HC1O4 mixture, followed by precipitation using HNOs. Analytical reproducibility was controlled using certified F, W, rare earth element (REE) + 25 ppb standard solutions and AGV-1 standard. Major, trace, and REE analyses on plagiogranites from blocks in the Main Melange unit of Cape Povorotny, Ganychalan terrane, and Kuyul terrane ophiolites, as well as ultramafic and mafic rocks of Ganychalan ophiolites were performed at the GIN RAS Analytical Centre. Major elements were measured by wet chemistry, and trace elements by XRF on a Russian-made ARF-6 quantographer in the concentration range 0.0001% to n%. REE were analysed by INAA in the range 0.000001% to n%. Ion microprobe measurements on minerals from Ust-Belaya peridotites were performed at the Northeastern Interdisciplinary Research Institute,
624
S. D. SOKOLOV ETAL.
Far East Division of the Russian Academy of Sciences (SVKNII DVO RAN), Magadan, on a CAMEBAX instrument (analysts E. M. Goryacheva and G.A. Merkulov). Major elements in ultramafic and mafic rocks were analysed at the X-ray Spectral Analysis Laboratory (SVKNII) and by gravimetric analysis at the Central Laboratory, Geological Survey, Magadan. Whole-rock K-Ar measurements were carried out by A. D. Lyuskin, at the Laboratory of Isotope Geochronology, SVKNII. REE in clinopyroxenes from the Ganychalan terrane plutonic rocks were measured on a Cameca IMS 4f ion microprobe, at the Institute of Microelectronics (IMAN), Russian Academy of Sciences, Yaroslavl. Major elements and V, Cr, Co, Ni, Cu, and Ba from Kuyul terrane peridotites were determined by wavelength-dispersive XRF using routine techniques at the Karpinsky Geological Institute (St. Petersburg, Russia). Other trace elements, including REE, were analysed by ICP-MS at the Institute of Geochemistry (Irkutsk, Russia). Mounts of 0.1 ± 0.001 g were digested with HF and HNO3 mixture in Teflon bombs for 24 h, evaporated until dry, taken up in HNO3, and once again evaporated until dry. Further HC1 was added and the product again evaporated until dry to assure quantitative removal of HF and chlorides. The samples were redissolved with deionized water. No undissolved spinels were detected in the Teflon bombs, and the fact that Cr values were comparable with those obtained by XRF suggests that all of the spinels went into solution. The samples were run on a VG Elemental Plasmaquad with long peak dwell times (320 ms per mass unit). Calibration was carried out using a set of high-Mg laboratory standards. Estimated precision is less than 10% for all of the determined elements. Microprobe analyses of minerals from Kuyul terrane peridotites were carried out at the Institute of Volcanology (Petropavlovsk-Kamchatsky, Russia) using a CAMECA CAMEBAX system equipped with a KEVEX energy-dispersive spectrometer with an accelerating voltage of 15 kV and a sample current of 15nA (counting time 100 s). Precision is estimated to be better than about 2% for all main components. Radiometric ages are taken from a number of publications, where their interpretation is provided as well.
Ophiolites in the Cape Povorotny accretionary complex Geological setting Five tectonic units are recognized in the Taigonos Peninsula (Fig. 4): (1) the Avekov terrane, which
is composed of Precambrian and Lower Palaeozoic metamorphic sequences; (2) the Pylgin suture zone, which incorporates metamorphosed Mesozoic volcanic and sedimentary rocks; (3) the Central Taigonos terrane, made up of Upper Permian-Lower Cretaceous island-arc strata; (4) the East Taigonos granite-metamorphic belt; (5) the Beregovoi terrane, which is composed of prearc complexes and the accretionary prism of the Uda-Murgal volcanic arc (Sokolov et al. 1999; Silantyev et al. 2000). A broad spectrum of igneous and metamorphic rocks are hosted by serpentinite melange in the Cape Povorotny accretionary complex and make up the following succession of tectonic units, from south to north (Fig. 5): (1) the Povorotny serpentinite melange with blocks of sheeted dykes, ultramafic rocks, and gabbro; (2) the Median serpentinite melange, with small blocks and fragments of ultramafic rock, gabbro, volcanic and terrigenous rocks, and chert; (3) the Main Melange unit, a serpentinite melange with blocks of peridotite, garnet-free and garnet-bearing amphibolites, greenschists, island-arc volcanic and sedimentary rocks, oceanic basalts and chert, and gabbro-diabase with plagiogranite veins (Fig. 6).
Petrography and geochemistry of metamorphic and igneous rocks Amphibolites occur as disrupted blocks in serpentinite melanges exposed on Cape Povorotny (Fig. 5). They consist of: (1) massive melanocratic rocks composed almost wholly of hornblende and minor plagioclase or garnet-hornblende rocks; (2) albite-hornblende schists. Judging by the characteristic mineral assemblages and mineral and bulk-rock compositions, the protoliths were made dominantly of plutonic rocks and subordinate volcanic rocks (Silantyev et al. 2000). Geochemical signatures of high-grade amphibolites have been detailed by Silantyev et al. (2000), indicating that volcanic protoliths of these rocks ranged in affinity from within-plate basalt (WPB) or enriched mid-ocean ridge basalt (E-MORB) to normal MORB (N-MORB). Mafic plutonic rocks are widespread as tectonic blocks in the Povorotny and Median melanges. Gabbro s typically occur as isolated boudins and small tectonic slices in the serpentinite matrix. Plutonic rock compositions from the Cape Povorotny ophiolite melange have a wide range, implying the existence of different gabbroic series of several geochemical types. Silantyev et al. (2000) presented chemical data indicating that gabbros chemically similar to boninite plutonic suites are abundant in the Cape Povorotny serpentinite mel-
OPHIOLITES OF NORTHEAST ASIA
625
Fig. 4. Tectonic map of the Taigonos Peninsula (after Sokolov et al. 1999).
anges. These boninite-like gabbros are low in TiC>2, their REE patterns being typical of boninites and their plutonic equivalents (low total REE contents, concave chondrite-normalized REE patterns, and considerable light REE (LREE) variations). The
Cape Povorotny boninite gabbros are associated with N-MORB and within-plate or E-MORB gabbros and dolerites. This group of plutonic rocks, including hornblende-bearing gabbro, is moderate to relatively high in TiC>2, FeO, and PiOs at
626
S. D. SOKOLOV ETAL.
Fig. 5. Map showing tectonic units of the Cape Povorotny accretionary complex (after Sokolov et al. 1999).
moderately high REE totals (analytical data have been presented by Silantyev et al. (2000)). Felsic veins in gabbro-diabase from blocks in the Main Melange unit (Fig. 6) are composed of plagiogranite and, sporadically, tonalite. The plagiogranites have magmatic textures with euhedral plagioclase crystals partly intergrown with quartz albite granophyre, suggestive of crystallization at shallow depths. The plagiogranites are composed of quartz, saussuritized plagioclase, epidote, chlorite, and magnetite. The tonalites differ from the plagiogranites in having smaller amounts of
quartz, and in that they contain light green amphibole and andesine plagioclase is andesine. The plagiogranites and tonalites have low A1203 (11.12-13.7%), K20 (0.03-0.06%) and K/ Rb ratios (0.01-0.03), and relatively high Y (3744ppm) (Table 1). Rb (7ppm) and Zr (120, 160ppm) contents of the plagiogranites (Table 1) are similar to those of the average Mid-Atlantic Ridge plagiogranites at latitude 2-3°N (Rikhter 1997). Chondrite-normalized REE patterns of the plagiogranites are nearly flat to slightly LREE en-
OPHIOLITES OF NORTHEAST ASIA
Fig. 6. Network of plagiogranite veins in gabbrodiabase from blocks in the Main Melange zone.
riched (Lan/Ybn = 1.01-1.95) and show distinct negative Eu anomalies (Eun/Eu* = 0.49-0.63, Fig. 7). The increase in total REE content from gabbro-diabase to plagiogranite and similarity of the REE patterns suggest that the rocks are cogenetic (Fig. 7). Comparison between the Cape Povorotny plagiogranites and those from the Bay of Islands ophiolites, Newfoundland (Elthon 1991), shows that both have negative Ta and Ti anomalies (Fig. 8) indicative of suprasubduction zone origin (Pearce & Norry 1979; Saunders et al 1980; Shervais 1982; Elthon 1991). Negative Ta and Nb anomalies are also shown by the Cape Povorotny plagiogranites when plotted on the ocean-ridge granite (ORG)-normalized (Pearce et al. 1984) patterns (Fig. 9). Chondrite-normalized REE patterns of plagiogranites and trondhjemites from the Maqsad ophiolite, Oman, considered by Amri et al. (1996) to have formed at a mid-ocean ridge, differ from those of Cape Povorotny plagiogranites in having lower REE totals and REE patterns with both negative and positive Eu anomalies (Fig. 10). It should be noted that Cox et al. (1999) and Searle & Cox (1999) assumed that Oman ophiolitic crust and its plagiogranite were generated above an intra-oceanic subduction zone. Basalts are the dominant volcanic rocks in the Main and Median melanges of the Cape Povorotny accretionary complex (Fig. 5). Basalts make up isolated tectonic blocks and flow units within terrigenous-tuffaceous sequences. Based on petrographic and geochemical evidence, these rocks are divided into the following groups (Sokolov et al.
627
1999; Silantyev et al. 2000): (1) boninites with low TiO2, extremely low middle REE (MREE) and heavy REE (HREE), and high MgO contents; (2) aphyric and phyric calc-alkaline basalts and low-K tholeiitic basalts; (3) pillowed and massive tholeiitic basalts with marked N- and E-MORB geochemical features. Cape Povorotny peridotites (Bazyler et al. 2000) occur in dismembered ophiolite sequences at various localities confined to NE-trending tectonic units. From SE to NW, these units crop out in the Povorotny, Median, and Main (including Greben and Beregovoi massifs) serpentinite melanges (Fig. 5). The central zone of the Greben massif is composed of spinel Iherzolite. The outer zone of the Greben massif, the Beregovoi massif, and small peridotite blocks of the Main and Median melanges are composed of harzburgite proper and Cpx-bearing harzburgite. Peridotites from the Povorotny melange are also harzburgites, but more depleted, judging from spinel compositions (Table 2). Compositions of other mineral phases have been given by Bazylev et al. (2001). The Median melange is dominated by cumulate peridotites, including pyroxene-bearing dunites, chromitites, wehrlites, and plagioclase harzburgites, which are also found in the Main Melange. Spinels from residual peridotites (Iherzolites and harzburgites) have Cr number (Cr/(Cr + Al)) ranges as wide as 0.18-0.70 (Bazylev et al. 2001), suggesting SSZ provenance for at least some of these rocks (Dick & Bullen 1984). Representative spinel compositions from the peridotites are given in Table 2. The Mg number (100Mg/(Mg + Fe)) of olivines and orthopyroxenes from the residual spinel peridotites does not correlate with the spinel Cr number (varying in the range 89.6-91.7), indicating their origin by open-system melting or melt-rock interaction, rather than by simple partial melting (Bazylev et al. 2001). Silicate mineral compositions from peridotites have been reported by Bazylev et al. (2001). Spinel compositions from the cumulate peridotites have high Cr number (0.30-0.79), elevated iron oxidation degrees, and low Ti contents, further supporting an SSZ rather than a MOR affinity for these rocks (Arai 1992). REE patterns from all the peridotite varieties including wehrlites are also LREE enriched, some of the spectra being U-shaped. They also have significant negative Nb and Zr anomalies, some samples also having negative Ti anomalies (Fig. 11). The data on rock geochemistry have been given by Bazylev et al. (2001). These features were explained by Bazylev et al. as resulting from open-system melting (Ozawa & Shimizu 1995) of mantle material accompanied by melt influx.
S. D. SOKOLOV ETAL.
628
Table 1. Major (wt%) and trace (ppm) element contents of gabbro-diabases and plagiogranites in blocks from the Main Melange zone Sample
c-2415
c-2340
c-2415/1
c-2340/2
c-2340/4
c-2340/1
Si02 TiO2 A12O3 Fe203
51.53 0.76 13.12 3.84 6.52 8.96 7.35 0.08 4.07 0.04 0.07 3.30 99.64
51.8 0.94 14.68 2.60 4.60 12.94 5.91 0.06 3.98 0.35 0.02 2.46 100.34
53.71 1.03 14.07 4.91 6.60 5.97 5.05 0.09 4.68 0.32 0.11 3.01 99.56
63.84 0.60 13.7 1.59 2.93 6.22 3.20 0.04 5.83 0.10 0.01 2.31 100.36
73.04 0.69 12.22 0.53 0.50 3.77 1.30 0.02 5.66 0.30 0.09 1.02 99.74
75.51 0.43 12.56 1.34 0.58 1.61 1.03 10wt% MgO) dykes (Elthon 1979; Elthon & Ridley 1980), identified as probable parental magmas for this mafic complex. Further to the south, the Rocas Verdes basin may have been linked to the even wider oceanic
environment of the proto-Weddell Sea (de Wit 1977; Grunow 1993a, 1993b; Mukasa & Dalziel 1996). These relations suggest that the Rocas Verdes igneous complexes, along with the associated mafic dykes and gabbros that intrude older preAndean and Andean lithologies along their flanks, are remnants of progressive stages of development of oceanic-type crust in a continental back-arc extensional tectonic environment. The Rocas Verdes formed during the initial stages of Gondwana breakup, and their origin may reflect processes associated with subduction (Bruhn et al. 1978; de Wit & Stern 1981; Alabaster & Storey 1990), reduction in plate boundary forces associated with changes in plate configuration and absolute motion (Dalziel 1981, 1986; de Wit & Stern 1981; Storey & Alabaster 1991), and/or mantle plumes involved in the opening of the South Atlantic (Cox 1978, 1988; de Wit &
ROCAS VERDES OPHIOLITES, SOUTH AMERICA
667
Fig. 2. (a) Sheeted dykes of the Sarmiento ophiolite. Geological hammer provides an approximate scale. Individual dykes are 0.5-1 m wide, and north-south-oriented sheeted dykes occur across an area 3—5 km wide in Lolos and Encuentro Fjords, southernmost Chile (Fig. 5; de Wit & Stern 1981). (b) Multiple north-south-oriented mafic dykes (dark rocks) cutting leucocratic continental basement (light rocks) along the flanks of the Sarmiento ophiolite. Field notebook in the centre of the photograph provides an approximate scale. Similar features are common along both flanks of the Rocas Verdes ophiolites and may be traced across strike into the sheeted dyke units of these ophiolites (de Wit & Stern 1981).
Fig. 3. Map showing the major lithotectonic units of southernmost South America during the Early Cretaceous. Sequential lithotectonic sections across XY during the Mid-Jurassic, Early Cretaceous and Late Cretaceous are shown in Figure 4.
Ransome 1992; Storey 1995; Storey & Kyle 1997; Dalziel et al 2000; Storey et al 2001). The Rocas Verdes and their associated rocks are remarkably well exposed (Fig. 2), and contain a wealth of information about progressive stages of continental rifting during back-arc basin formation (Dalziel et al. 1974; de Wit & Stern 1978, 1981; Saunders et al. 1979; Stern 1979, 1980), as well as petrological and metamorphic processes along MOR-type spreading centres (de Wit & Stern 1976, 1978; Stern et al. 1976; Elthon & Stern 1978; Elthon 1979; Stern 1979, 1980, 1991; Stern & Elthon 1979; Elthon & Ridley 1980; Stern & de Wit 1980; Elthon et al. 1984), and also as analogues to Archaean greenstone belts (Tarney et al. 1976; Stern & de Wit 1997); however, they have not received as much scientific attention as many other ophiolite complexes worldwide, perhaps because of their remote location and the logistic difficulties involved in their study. This
668
C. R. STERN & M. J. DE WIT
Fig. 4. Sequential lithotectonic sections (across XY in Fig. 3) during the Mid-Jurassic to Late Cretaceous, illustrating the formation and collapse of the volcano-tectonic rift zone and the Rocas Verdes back-arc basin. Figure modified after Dalziel et al. (1974), Bruhn et al (1978) and de Wit & Stern (1978, 1981).
paper reviews the major geological and petrochemical features of the Rocas Verdes ophiolite complexes and discusses some of the implications of these features for the process of back-arc basin development and the generation of oceanic-type crust in a continental margin setting.
Regional geology The formation of the Rocas Verdes basin in southernmost South America was immediately preceded by the development of a Jurassic volcano-tectonic rift zone (Fig. 4a; Bruhn et al 1978). The relatively narrow Rocas Verdes basin formed along the western edge of this much broader rift zone, which was characterized by extensive bimodal volcanism
associated with horst and graben tectonics. The voluminous silicic volcanic rocks produced by this rifting range in thickness up to >2 km. They are referred to as the Tobifera Formation in Magallanes, Chile (Bruhn et al. 1978; Hanson & Wilson 1991), and more broadly as the Chon Aike Formation in southern South America (Gust et al. 1985), that formed between 153 and 188 Ma (Riley & Knight 2001). Peraluminous granitoids in Tierra del Fuego, considered to be possible plutonic equivalents of the Tobifera Formation silicic volcanic rocks (Nelson et al 1988; Suarez et al 1990), have been dated at 157 ± 8 Ma by an Rb-Sr isochron for whole-rock samples (Herve et al 1981) and 164.1 ± 1.7 Ma by zircon U-Pb systematics (Mukasa & Dalziel 1996).
ROCAS VERDES OPHIOLITES, SOUTH AMERICA The Tobifera and Chon Aike Formations are the Patagonian portion of an even larger and longerlived Gondwana continental silicic volcanic province that produced the Upper Palaeozoic and Lower Mesozoic Choiyoi granites and rhyolites in central and northern Chile and Argentina (Kay et al. 1989; Mpodozis & Kay 1990), and Middle Jurassic rhyolites and granites of the eastern Antarctic Peninsula dated at 167-189 Ma (Dalziel & Elliot 1982; Dalziel et al. 1987; Elliot 1992; Riley & Knight 2001). This extensive silicic volcanic province, which persisted from the Late Palaeozoic into the Jurassic, or even Early Cretaceous (Kirstein et al. 2001), extended as far east as southern Africa, and south into the proto-Weddell Sea, before and during the opening of the South Atlantic (de Wit 1977; Dalziel et al. 1987; de Wit & Ransome 1992; Elliot 1992; Grunow 1993a, 1993b). The Jurassic Tobifera Formation silicic volcanic rocks unconformably overlie the eroded Palaeozoic (pre-Andean) metamorphic crystalline basement of Patagonia (Fig. 4a; Halpern 1973; Forsyth 1982; Herve et al. 1991). Rifting associated with the formation of the Tobifera volcanic rocks thinned the lithosphere below Patagonia to less than 80 km, c. 20 km thinner than its current 100 km thickness, as indicated by geotherms obtained from mantle xenoliths found within the Quaternary Patagonian plateau basalts (Skewes & Stern 1979; Stern et al 1989, 1999). The mafic igneous protoliths for granulite xenoliths, also found in the plateau basalts, probably were intruded into the base of the South American crust during the formation of this volcano-tectonic rift zone (Selverstone & Stern 1983). As Late Jurassic and Early Cretaceous rifting became focused within the Rocas Verdes basin to the SW, and the South Atlantic spreading ridge to the east, southernmost Patagonia became a stable (cratonic-like) continental platform (Figs 3 and 4b). At the same time, the southwestern margin of South America remained an active convergent plate boundary magmatic arc, as indicated by the fact that a number of I-type calc-alkaline granitic plutons of the southern Patagonian batholith, which represent the roots of this arc south and west of the Rocas Verdes belt, date between 151 and 138 Ma (Halpern 1973; Stern & Stroup 1982; Herve et al. 1984; Nelson et al. 1988; Weaver et al. 1990; Bruce et al. 1991). These plutons, which intrude the same Palaeozoic metamorphic basement underlying the Tobifera Formation volcanic rocks, are thus contemporenous with the magmatic activity that formed the Rocas Verdes mafic igneous complexes (Fig. 4). These relations are consistent with the suggestion that the Rocas Verdes basin was a back-arc
669
(marginal) basin that rifted the continental crust NE of a contemporaneous magmatic arc active along the southwestern continental margin of Gondwana, and subsequently South America, during the Gondwana continental breakup and the onset of spreading in the South Atlantic (Dalziel et al. 1974; Suarez & Pettigrew 1976; Suarez 1979; Tanner & Rex 1979; Dalziel 1981). Sediments derived from both the convergent plate boundary magmatic arc on the SW and the continental platform on the NE were deposited in the Rocas Verdes basin (Fig. 4b; Dott et al. 1977; Winn 1978; Winn & Dott 1978). Arc volcanic rocks and associated coarse volcanogenic sediments are interbedded with mafic pillow lavas along both margins of the Rocas Verdes complexes, and overlie these complexes within the centre of the basin, where they are interbedded with chert and fine-grained deep-water turbidites (Dott et al. 1977; Suarez 1979). The Rocas Verdes ophiolite complexes consist of 2-3 km thick units of predominantly mafic submarine extrusive rocks, including pillow lavas and breccias, cut by dykes and overlying a sheeted dyke complex (Figs 5 and 6). Below the 300500 m thick sheeted dyke complex (Fig. 2a), the lower contact of which may be either sharp or gradational (Fig. 6), c. 1 km of massive diabases and coarse-grained gabbros are exposed. North-tosouth variations in eruptive products (de Wit & Stern 1978), ophiolite pseudostratigraphy (de Wit & Stern 1978, 1981), and petrochemistry (Stern 1979, 1980; Saunders et al. 1979) have been shown to result from smaller degrees of extension in the northern part and greater degrees of extension in the southern part of the Rocas Verdes basin (Fig. 7). Mafic dykes and gabbros of similar composition cut older lithologies on the flanks of the Rocas Verdes ophiolites (Fig. 2b). Here we describe first the mixed mafic-felsic outcrops that flank the Rocas Verdes ophiolites, followed by the Sarmiento ophiolite complex at the northern, narrow end of the Rocas Verdes belt, and then the Tortuga complex at the southern, wider end of this belt (Fig. 1). These igneous complexes are interpreted as representing remnants of the progressive stages of development of the igneous floor of the extensional back-arc Rocas Verdes basin, which ranged from intermediate between continental and oceanic to typically oceanic in character (Fig. 7).
Mixed mafic-felsic terranes flanking the ophiolites Flanking the belt of Rocas Verdes ophiolites are mixed mafic-felsic outcrops in which basaltic
670
C. R. STERN & M. J. DE WIT
Fig. 5. Geological maps of the regions around the Sarmiento and Tortuga ophiolites, at the northern and southern extremes, respectively, of the Rocas Verdes belt, southern Chile.
dyke swarms (Fig. 2b) and large diabase and gabbro sills and stocks intrude remnants of the pre-Cretaceous continental crust, including the Palaeozoic metamorphic basement and Jurassic silicic volcanic and plutonic rocks. Similar mafic intrusions seldom occur within the Lower Cretaceous basin sedimentary rocks, which overlie both the ophiolite complexes and these mixed mafic felsic rock units. The dyke swarms within these mixed mafic-felsic blocks have regular orientations, which parallel both the strike of the ophiolite complexes and the orientation of sheeted dykes within these complexes. Volumetrically, these mafic intrusions gradually increase towards and merge with the intrusive components of the ophiolite complexes. The basaltic igneous rocks in these mixed mafic-felsic rock units are greenstones or spilites, altered by hydrothermal metamorphism without formation of schistosity, similar to the metamorphic overprint observed within the Rocas Verdes ophiolite complexes (Stern et al. 1976; Elthon & Stern 1978; Stern & Elthon 1979). The major and immobile trace element (Ti, Zr, Y and rare earth elements (REE)) compositions of the metamorphosed mafic dykes, sills and stocks in these mixed mafic-felsic terranes flanking the
Rocas Verdes ophiolites are comparable with those of basaltic dykes and lavas from within the ophiolites (Fig. 8; Stern 1979, 1980; de Wit & Stern 1981). In particular, the REE contents of these dykes and sills are similar to those of dykes and lavas from the Sarmiento ophiolite complex, and both have similar normalized light REE (LREE; La) to heavy REE (HREE; Yb) ratios, with (La/Yb)N >1 (Fig. 9; Stern 1980). Both the field and geochemical data are consistent with the suggestion that the basaltic dykes and sills in the mixed mafic-felsic terranes flanking the ophiolites are a phase of the same igneous activity that formed the Rocas Verdes ophiolite complexes. The sheeted dyke complexes within the ophiolites imply 100% extension along oceanic type spreading centres, and the mixed maficfelsic terranes that flank the ophiolites also imply extension during intrusion of mafic magmas, but less than 100%. The boundary between the mafic ophiolite complexes and the flanking mixed mafic-felsic blocks is diffuse. The variation from 100% pre-Cretaceous continental rocks to 100% mafic rocks may occur across a narrow or relatively wide distance of several tens of kilometres. The pre-Cretaceous continental rocks are progressively more disrupted with the volumetric increase
ROCAS VERDES OPHIOLITES, SOUTH AMERICA
671
Fig. 6. Schematic sections across the Sarmiento and Tortuga ophiolite complexes, modified after de Wit & Stern (1978, 1981). (Note the abrupt igneous contact between the sheeted dykes and plagiogranites in the Sarmeinto complex and the continuous transition between dykes and massive diabase in the Tortuga complex, where plagiogranites are absent.) Acid (trondhjemite) crustal xenoliths also occur in the Sarmiento complex, but are absent in the Tortuga complex.
of mafic rocks above 60-70%. When mafic rocks dominate by >75%, they cause melting and remobilization of the host country rocks, leading to the reconstitution of continental crust through the formation of hybrid rock types (de Wit & Stern 1981). Field relations representing different stages of this process are best preserved along the flanks of the Sarmiento ophiolite complex, in the narrow northern part of the original basin (Fig. 5). Along the western flank of the Sarmiento complex, on Young Island, basaltic dyke swarms and gabbros intrude horizontally bedded silicic volcanic rocks of the Jurassic Tobifera Formation (de Wit & Stern 1981). In situ brecciation of the silicic rocks occurs near large mafic bodies and in areas of high concentrations of mafic dykes, creating raggedly outlined blocks of silicic lavas set in a new pale brown-green hybrid igneous matrix. This hybrid matrix is clearly a mixture of mafic and felsic components, although the exact physical process of mixing, whether by partial melting and/ or mechanical brecciation, remains obscure. However, the hybrid matrix is observed to reintrude as an independent magma, on a large scale, the entire
rock sequence of silicic rocks and mafic dykes and stocks (de Wit & Stern 1981).
Sarmiento ophiolite complex A vertical section 1-3 km thick of the Sarmiento ophiolite complex, the northernmost in the Rocas Verdes belt, is exposed in the Lolos and Encuentro fjords (Fig. 5; Dalziel et al 1974; de Wit & Stern 1978, 1981) of southern Chile. In these two fjords, coarse-grained gabbros and ferro-gabbros cut by basaltic dykes occur at sea level (Fig. 6). Dyke concentration increases vertically, as do the volumetric proportions of finer-grained diabases and leucocratic rocks. A sheeted dyke complex, consisting of 100% dykes (Fig. 2a), occurs at c. 1 km above sea level. The base of the sheeted dyke complex in the Sarmiento complex is an abrupt igneous contact marked by the intrusion of leucocratic rocks into the sheeted dykes by magmatic stoping. Approximately 300-500 m above this contact the first screen of extrusive pillow lavas and breccias is observed within the dyke complex. The volume per cent of these screens increase upwards, and the dyke complex is overlain by an
Fig. 7. Schematic cross-sections through three parts of the Rocas Verdes basin in the Early Cretaceous, modified after de Wit & Stern (1981): (a) in the northernmost part (north of 50°S) of the basin, where the igneous floor of the basin was continental in character; (b) in the area of the Sarmiento complex, where the igneous floor of the basin was intermediate between continental and oceanic in character; (c) in the southern part of the basin, in the area now preserved as the Tortuga complex, where the igneous floor of the basin was oceanic in character.
ROCAS VERDES OPHIOLITES, SOUTH AMERICA
Fig. 8. Titanium (Ti) v. zirconium (Zr) concentrations for the major Mesozoic igneous rock suites from southernmost South America, modified after de Wit & Stern (1981). Mafic dykes and sills (A; Stern 1980) intruding pre-Cretaceous continental crust flanking the Rocas Verdes ophiolite complexes (Fig. 2) plot along the same fractionation trend as do basalts from the ophiolites (bold line; Stern 1979), and also within the field of ocean-floor basalts (OFB; Pearce & Cann 1973). Silicic country rocks from Young Island (•; de Wit & Stern 1981) plot in the field of Jurassic Tobifera silicic volcanic rocks (Bruhn et al. 1978). Trondhjemites from within the Sarmiento complex (•; de Wit & Stern 1981) plot within the fields of both Tobifera silicic volcanic rocks and silicic plutons from the Patagonian batholith (Bruhn et al. 1978; Stern & Stroup 1982). In contrast, plagiogranites from the Sarmiento complex have significantly higher Zr contents (Stern 1979).
estimated 2 km of extrusive rocks, themselves cut by numerous mafic dykes (de Wit & Stern 1978). Two chemically and petrologically distinct types of leucocratic rocks may be distinguished in the Sarmiento complex (Saunders et al. 1979; Stern 1979; de Wit & Stern 1981; Elthon et al. 1984; Stern et al. 1992). One type, termed plagiogranites (Figs 6 and 8), is fine grained, occurs as both the massive body that intrudes the base of the sheeted dyke complex and also as dykes within this complex, and has relatively high incompatible trace element (Zr, Y, REE) concentrations compared with the mafic rocks of the complex (Figs 8 and 9), but with similar (La/Yb)N >1 and large negative Eu anomalies (Fig. 9). These plagiogranites have been shown to be derived from the mafic rocks of the complex by a process of closed-system crystal-liquid fractionation involving separation of mineral phases found in ferrogabbros (>20 wt% FeO; Stern 1979), which occur within the gabbro units of the Sarmiento complex. Ferro-basalts (>15wt% FeO and >2 wt% TiO2) and FeO- and TiO2-rich (>8 wt% FeO and >lwt% TiO2) intermediate (55-65 wt% SiO2) icelandites, which are products of less extensive
673
Fig. 9. Rare earth element (REE) concentrations, normalized to the REE content of a chondrite meteorite, of dykes and sills flanking the Sarmiento (•) and Tortuga (A) ophiolite complexes, as well as dykes and lavas from these ophiolites (Stern 1980).
fractionation, also occur as dykes in the sheeted dyke complex and cutting both gabbros and extrusive rocks. Approximately 30% of the dykes in the sheeted dyke complex of the Sarmiento complex consist of these evolved compositions. The second type of leucocratic rocks found within the Sarmiento complex is coarse grained and occurs as blocks, ranging from several centimetres to tens of metres across, within the gabbro unit of the complex ('Acid xenoliths' in Fig. 6). These rocks are referred to as trondhjemites (Fig. 8). They have distinctly lower Zr (Fig. 8), Y and HREE contents, and thus higher (La/Yb)N (de Wit & Stern 1981, p. 251, fig. 12), than plagiogranites in the Sarmiento complex. The trondhjemites are chemically similar to both the silicic volcanic rocks of the Jurassic Tobifera volcanic rocks and granitic plutons from the Patagonian batholith (Fig. 8), and they are interpreted as xenoliths of country rocks within the ophiolite. Trondhjemite xenoliths have been cut by dykes and are in places brecciated and remelted, producing hybrid magmas that have reintruded the gabbros of the Sarmiento complex. These relations resemble those in the remobilized country rocks flanking the ophiolite complexes.
674
C. R. STERN & M. J. DE WIT
The Sarmiento complex underwent hydrothermal 'ocean-floor' metamorphism, characterized by the growth of secondary minerals without the development of schistosity, prior to its uplift and exposure in the Andes (de Wit & Stern 1976; Stern et al, 1976; Elthon & Stern 1978; Stern & Elthon 1979; Elthon et al. 1984). The metamorphic overprint of this metamorphism on the pseudostratigraphy of the ophiolite exhibits a steep vertical gradient, passing from zeolite to amphibolite facies in 2 km (Fig. 10), and with equally abrupt variations in the extent of metamorphic replacement of igneous minerals and textures. Metamorphic facies boundaries are irregular and disequilibrium retrograde effects are common. The extent of metamorphic replacement is most intense just above and within the sheeted dyke complex, but decreases markedly within the gabbro unit of the ophiolite, probably because of restricted access of circulating sea water in the deeper plutonic levels of this ophiolite complex. Oxygen isotopes confirm that the extrusive rocks are enriched in 18O relative to fresh basalts as a result of interaction with sea water at relatively low temperatures (500 °C) interactions with sea water. Water-rock
ratios involved in this metamorphism decrease from between 15 and 90 in the extrusive section of the Sarmiento complex, to < 1 in the plutonic portion of the complex (Elthon et al. 1984). However, in detail, the effects of metamorphism are locally controlled by variations in permeability related to fault zones and the complex thermal history associated with the generation of the ophiolites. Metamorphism resulted in large-scale migration of KaO, Na21 (Fig. 9; Saunders et al. 1979; Stern 1980), similar to mafic magmas produced during the early stages of evolution of the Larson Harbour complex, a southern extension of the Rocas Verdes on South Georgia (Alabaster & Storey 1990; Storey & Alabaster 1991). The observed chemical variations within the Sarmiento complex, which includes ferro-basalt and intermediate icelandite dykes, as well as plagiogranites, are best explained by limited open-system magma chamber behaviour, followed by closed-system crystalliquid igneous fractionation (Stern 1979).
Fig. 10. Histograms illustrating the distribution of metabasalts and metagabbros, from the various stratigraphic units of the Sarmiento complex, among the metamorphic facies defined in the inset in the upper right of the figure (Elthon & Stern 1978; Stern & Elthon 1979). Filled bars represent relative abundance of rocks without relict higher temperature facies minerals, and diagonally ruled bars are rocks with such relict minerals. The left side of the diagram summarizes the observed petrographic variations in the extent of metamorphic replacement of original igneous minerals and textures.
ROCAS VERDES OPHIOLITES, SOUTH AMERICA Tortuga ophiolite complex The Tortuga ophiolite complex, the southernmost in the Rocas Verdes belt, is best exposed on Navarino and Milne Edwards islands (Fig. 5; de Wit & Stern 1978, 1981; Elthon & Ridley 1980). In the Tortuga complex, neither silicic plagiogranites nor trondhjemites have been observed, and the lower contact of the sheeted dyke complex is gradational. Dykes grade downwards into medium-grained diabase cut by later dykes (Fig. 6) and finally cumulate gabbros, which include both plagioclase-rich and olivine-bearing varieties. The igneous rocks of the Tortuga complex exhibit a more restricted chemical range than those of the Sarmiento complex, and no ferrobasalts, icelandites or silicic plagiogranites occur in the Tortuga complex. However, high-MgO (>10wt% MgO) dykes occur cutting the deeper gabbro and massive diabase level of this complex, below the sheeted dyke complex. These highMgO dykes are similar in composition to liquids derived by high degrees (25-30%) of partial melting of mantle Iherzolite at 20 kbar pressure, and they contain phenocrysts of highly aluminous picotite spinel consistent with high-pressure formation (Elthon 1979). They are interpreted as the parental mantle-derived magmas from which the more evolved basalts in the Tortuga complex formed. The observed chemical variations among the mafic dykes and lavas of the Tortuga complex may be modelled by open-system magmatic differentiation (Stern 1979). This model involves the periodic input of mantle-derived high-MgO basaltic magmas into the base of a magma chamber within which gabbros are forming along the floor and walls and from which differentiated (lower MgO) basalts are being erupted through the roof (Stern 1979; Elthon & Ridley 1980; Stern & de Wit 1980). Based on the determination of liquids in equilibrium with the cumulate minerals in the gabbros of the Tortuga complex, Elthon & Ridley (1980) concluded that the lack of systematic trends in FeO/MgO in vertical transects through the gabbros is consistent with open-system differentiation within a periodically refilled magma chamber. The relatively high FeO/MgO of the exposed gabbros implies that a significant volume of more magnesian cumulates must be associated with the Tortuga complex, but are not yet exposed. The tholeiitic basalts of the Tortuga complex have (La/Yb)N 780 Ma (Kemp et al 1980). Pallister et al. (1988) interpreted the Jabal Al Wask massif as a serpentinite-matrix melange. U-Pb zircon ages of gabbro (dyke in a peridotite) and trondhjemite samples from the Jabal Al Wask massif range from 770 to 740 Ma (Pallister et al 1988). Ophiolites of the Bir Umq suture zone The Bir Umq suture zone runs subparallel to the Yanbu suture and separates the Hijaz terrane to the north from the Jeddah terrane to the south (Fig. 2). Near the suture zone, the Jeddah terrane is composed of dioritic to tonalitic plutons of the Taif arc complex that were probably intruded into an older arc basement (Stoeser & Camp 1985). Magmatism in the Taif arc might have been waning by 715 Ma when the initial suturing between the Hijaz and Jeddah terranes was in progress (Nasseef & Gass 1977). The Bir Umq suture zone continues farther SW across the Red Sea into Sudan and connects with the Amur-Nakasib Suture, which separates the Gebeit terrane in the north from the Haya terrane in the south within the Nubian Shield (Nassief et al. 1984; Reischmann 2000). The
Ingessana ophiolite constitutes the main ophiolite complex in the Amur-Nakasib Suture in Sudan (Fig. 1). The Bir Umq and Jabal Thurwah massifs are the two main ophiolite occurrences in the Bir Umq suture zone (Table 1). Bir Umq ophiolite. The Bir Umq ophiolite occurs in an ENE-trending large thrust sheet at the eastern end of the suture zone (number 8 in Fig. 2) and consists of peridotite, gabbro, volcanic rocks, and overlying chert and tuff; no sheeted dyke complex has been observed in this ophiolite (Ahmed & Hariri 2001). The peridotites consist of altered harzburgite, dunite and pyroxenite. Crspinels in the harzburgite and dunite are TiC>2poor ( 11 km thick, underlain by a dunite-dominated ultramafic section of 2 km thickness. Ophiolitic exposures also occur as roof pendants in a younger granitic batholith and as slivers along the Najd system faults (Quick & Gregory 1995). At Darb Zubaydah, the ophiolite forms an east-dipping homocline. The orogenic granite-diorite-monzogranite intrusions are 720-640 Ma old, and a younger group of smaller granitic intrusions are 640-570 Ma in age (Quick 1990). Ultramafic rocks consist mainly of dunite cumulates, strongly altered to serpentine, talc, chlorite and magnetite. Chromian spinel grains are not foliated, as is commonly seen in tectonites, and possess very thick ferritchromite and magnetite rims with Cr numbers ranging from 0.75 to 0.85 (Quick 1990). Such high Cr numbers are typical of modern forearc peridotites, and are distinctly different from those of mid-ocean ridge or fracture zone peridotites (Dick & Bullen 1984). The existence of harzburgite is suggested by the presence of sparse poikilitic bastite pseudomorphs. The dominance of dunite in the ultramafic section, high Cr numbers of chromian spinel, and the depleted nature of the peridotites suggest high degrees of partial melting, and/or the percolation of large volumes of transient melt through a more stagnant upper mantle. Gabbroic and diabasic dyke intrusions are widespread in the ultramafic rocks. No sheeted dyke complex has been observed in the Darb Zubaydah ophiolite. The crustal section of the ophiolite contains interbedded basalt-andesite-rhyolite flows in addition to tuffs, lahar deposits, and turbidites formed in a submarine environment within, or on the flanks of, a volcanic arc. Pelagic sedimentary rocks are absent. Basalts and andesites have calcalkaline compositions with high-alumina (16.919.9%) and low-Ti (