Precambrian Ophiolites and Related Rocks, Volume 13 (Developments in Precambrian Geology)

This page intentionally left blank

Developments in Precambrian Geology, 13

PRECAMBRIAN OPHIOLITES AND RELATED ROCKS

DEVELOPMENTS IN PRECAMBRIAN GEOLOGY Advisory Editor Kent Condie Further titles in this series 1. B.F. WINDLEY and S.M. NAQVI (Editors) Archaean Geochemistry 2. D.R. HUNTER (Editor) Precambrian of the Southern Hemisphere 3. K.C. CONDIE Archean Greenstone Belts 4. A. KRÖNER (Editor) Precambrian Plate Tectonics 5. Y.P. MEL’NIK Precambrian Banded Iron-formations. Physicochemical Conditions of Formation 6. A.F. TRENDALL and R.C. MORRIS (Editors) Iron-Formation: Facts and Problems 7. B. NAGY, R. WEBER, J.C. GUERRERO and M. SCHIDLOWSKI (Editors) Developments and Interactions of the Precambrian Atmosphere, Lithosphere and Biosphere 8. S.M. NAQVI (Editor) Precambrian Continental Crust and Its Economic Resources 9. D.V. RUNDQVIST and F.P. MITROFANOV (Editors) Precambrian Geology of the USSR 10. K.C. CONDIE (Editor) Proterozoic Crustal Evolution 11. K.C. CONDIE (Editor) Archean Crustal Evolution 12. P.G. ERIKSSON, W. ALTERMANN, D.R. NELSON, W.U. MUELLER and O. CATUNEANU (Editors) The Precambrian Earth: Tempos and Events

Developments in Precambrian Geology, 13

PRECAMBRIAN OPHIOLITES AND RELATED ROCKS Edited by

T.M. KUSKY Department of Earth and Atmospheric Sciences Saint Louis University St. Louis, MO 63103, USA

2004

Amsterdam - Boston - Heidelberg - London - New York - Oxford Paris - San Diego - San Francisco - Singapore - Sydney - Tokyo

ELSEVIER B.V. Sara Burgerhartstraat 25 P.O. Box 211 1000 AE Amsterdam The Netherlands

ELSEVIER Inc. 525 B Street Suite 1900 San Diego CA 92101-4495 USA

ELSEVIER Ltd The Boulevard Langford Lane Kidlington Oxford OX5 1GB UK

ELSEVIER Ltd 84 Theobalds Road London WC1X 8RR UK

© 2004 Elsevier B.V. All rights reserved. This work is protected under copyright by Elsevier B.V., and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier’s Rights Department in Oxford, UK; phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. Requests may also be completed on-line via the Elsevier homepage (http://www.elsevier.com/locate/permissions). In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555; fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier’s Rights Department, at the fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

First edition 2004 ISBN: 0-444-50923-2 (hardbound) ISNN: 0166-2635 (Series)

∞ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992
(Permanence of Paper).

Printed in Italy.

v

CONTRIBUTING AUTHORS

A.M. AL-SALEH Geology Department, King Saud University, PO Box 2009, Riyadh, Saudi Arabia I.I. BABARINA Geological Institute of the RAS, Pyzhevsky per. 7, Moscow 109017, Russia E.V. BIBIKOVA Vernadsky Institute of Geochemistry and Analytical Chemistry of RAS, Kosygin st. 19, Moscow 17975, Russia M.M. BOGINA Institute of Ore Deposits, Petrography, Geochemistry and Mineralogy of RAS, Staromonetny per. 35, Moscow 109017, Russia R.W. CARLSON Department of Terrestrial Magnetism, Carnegie Institute of Washington, Washington, DC, USA P.L. CORCORAN Department of Geology, University of Western Ontario, Canada J. DANN 90 Old Stow Road, Concord, MA 01742, USA ([email protected]) M. DE WIT CIGCES, Department of Geological Sciences, University of Cape Town, Rondebosch 7701, South Africa ([email protected], http://www.uct.ac.za/depts/cigces) J. ENCARNACIÓN Department of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63103, USA (jpe@eas. slu.edu, http://www.eas.slu.edu/people/jpencarnacion/jpeslu.html) F. JUN Department of Geology, Peking University, Beijing 100871, China R. GANLEY Department of Earth and Atmospheric Sciences, Saint Louis University, 3507 Laclede Ave., St. Louis, MO 63103, USA A. GLASS Earth Sciences Department, Cardiff University, PO Box 914, Main Building, Park Place, Cardiff, South Glamorgan CF1434E, UK ([email protected]) T.L. GROVE Department of Earth, Atmospheric, and Planetary Sciences, MIT, Building 54-1224, Cambridge, MA 02139, USA ([email protected]) A. HOFMANN School of Geosciences, University of the Witwatersrand, P. Bag 3, Wits 2050, South Africa

vi

Contributing authors

X.N. HUANG Department of Geology, Peking University, Beijing 100871, China I.M. HUSSEIN Institut für Geowissenschaften, Universität Mainz, 55099 Mainz, Germany R. HUSON Department of Earth and Atmospheric Sciences, Saint Louis University, 3507 Laclede Ave., St. Louis, MO 63103, USA ([email protected], http://www.eas.slu.edu/people/students/rlhuson/index.html) P.R. JOHNSON Saudi Geological Survey, PO Box 54141, Jiddah 21514, Saudi Arabia ([email protected]) F.H. KATTAN Saudi Geological Survey, PO Box 54141, Jiddah 21514, Saudi Arabia R. KERRICH Department of Geological Sciences, University of Saskatchewan, SK, Canada S7N 5E2 (robert.kerrich@ usask.ca) A.N. KONILOV Geological Institute of the RAS, Pyzhevsky per. 7, Moscow 109017, Russia A. KONTINEN Geological Survey of Finland, PO Box 1237, FIN-70211 Kuopio, Finland A. KRÖNER Institut für Geowissenschaften, Universität Mainz, 55099 Mainz, Germany ([email protected], http://www.uni-mainz.de/∼kroener/) K.A. KRYLOV Department of Geological and Environmental Sciences, Stanford University, CA 94305-2115, USA (kirka@geo. tv-sign.ru) T.M. KUSKY P.C. Reinert Endowed Chair of Natural Sciences, Department of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63103, USA ([email protected], http://www.eas.slu.edu/programs/geograd.html# kusky) J.H. LI Department of Geology, Peking University, Beijing 100871, China ([email protected]) J. LYTWYN Department of Geosciences, University of Houston, Houston, TX 77058, USA W.U. MUELLER Department of Geology, University of Quebec at Chicoutimi, Canada B.A. NATAL’IN ˙ Maden Fakültesi, Jeoloji Bölümü, Ayazaˇga 34469, Istanbul, ˙ ITÜ Turkey ([email protected]) X.L. NIU Department of Geology, Peking University, Beijing 100871, China S.W. PARMAN Department of Earth, Atmospheric, and Planetary Sciences, MIT, Building 54-1224, Cambridge, MA 02139, USA ([email protected])

Contributing authors

vii

P. PELTONEN Geological Survey of Finland, PO Box 96, FIN-02151 Espoo, Finland ([email protected]) J. PFÄNDER Institut für Mineralogie, Universität Münster, Corrensstr. 24, D-48149 Münster, Germany (pfaender@ mpch-mainz.mpg.de) A. POLAT Department of Earth Sciences, University of Windsor, Windsor, Ontario, Canada N9B 3P4 ([email protected]) I.S. PUCHTEL Department of the Geophysical Sciences, The University of Chicago, 5734 South Ellis Avenue, Chicago, IL 60637, USA ([email protected], http://geosci.uchicago.edu/∼ ipuchtel/) T. RAHARIMAHEFA Department of Earth and Atmospheric Sciences, Saint Louis University, 3507 Laclede Ave., St. Louis, MO 63103, USA T. REISCHMANN Institut für Geowissenschaften, Universität Mainz, 55099 Mainz, Germany ([email protected]) A.V. SAMSONOV Institute of Ore Deposits, Petrography, Geochemistry and Mineralogy of RAS, Staromonetny per. 35, Moscow 109017, Russia A.M.C. SENGÖR ¸ ˙ Maden Fakültesi, Jeoloji Bölümü, Ayazaˇga 34469, Istanbul, ˙ ITÜ Turkey ([email protected]) A.A. SHCHIPANSKY Geological Institute of the RAS, Pyzhevsky per. 7, Moscow 109017, Russia ([email protected]) A.I. SLABUNOV Karelian Research Center of RAS, Institute of Geology, Pushkinskaya st. 11, Petrozavodsk 185610, Karelia, Russia R.J. STERN Geosciences Department, University of Texas at Dallas, Box 830688, 2601 N. Floyd Rd., Richardson, TX 75083-0688, USA (http://www.utdallas.edu/dept/geoscience/) B. YIBAS Pulles Howard and de Lange, Environmental and Water Quality Management, PO Box 861, Auckland Park 2006, South Africa ([email protected]) C. ZHENG School of Earth and Space Sciences, Peking University, Beijing 100871, China

This page intentionally left blank

ix

CONTENTS

Contributing authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . T.M. Kusky

1

PROTEROZOIC OPHIOLITES AND RELATED ROCKS Chapter 1.

The Jormua Ophiolite: A Mafic-Ultramafic Complex from an Ancient Ocean-Continent Transition Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . P. Peltonen and A. Kontinen

35

Chapter 2.

The 1.73 Ga Payson Ophiolite, Arizona, USA . . . . . . . . . . . . . . . . . . . . . . . . . . J.C. Dann

73

Chapter 3.

Neoproterozoic Ophiolites of the Arabian-Nubian Shield . . . . . . . . . . . . . . . . . . . . Robert J. Stern, Peter R. Johnson, Alfred Kröner and Bisrat Yibas

95

Chapter 4.

Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure . . . . . . . 129 Peter R. Johnson, Fayek H. Kattan and Ahmed M. Al-Saleh

Chapter 5.

The Wadi Onib Mafic-Ultramafic Complex: A Neoproterozoic Supra-Subduction Zone Ophiolite in the Northern Red Sea Hills of the Sudan . . . . . . . . . . . . . . . . . . . . . . 163 I.M. Hussein, A. Kröner and T. Reischmann

Chapter 6.

Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite in Tuva, Central Asia: Crustal Growth by Island Arc Accretion . . . . . . . . . . . . . . . . . 207 J.A. Pfänder and A. Kröner

ARCHEAN OPHIOLITES AND RELATED ROCKS Chapter 7.

Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt, North China Craton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Timothy M. Kusky, Jainghai Li, Adam Glass and X.N. Huang

Chapter 8.

Re-Os Isotope Chemistry and Geochronology of Chromite from Mantle Podiform Chromites from the Zunhua Ophiolitic Mélange Belt, N. China: Correlation with the Dongwanzi Ophiolite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 T.M. Kusky, J.H. Li, T. Raharimahefa and R.W. Carlson

Chapter 9.

Geochemical and Petrographic Characteristics of the Central Belt of the Archean Dongwanzi Ophiolite Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 R. Huson, T.M. Kusky and J.H. Li

x

Contents

Chapter 10. Microstructures of the Zunhua 2.50 Ga Podiform Chromite, North China Craton and Implications for the Deformation and Rheology of the Archean Oceanic Lithospheric Mantle 321 Xiongnan Huang, Jianghai Li, T.M. Kusky and Zheng Chen Chapter 11. Neoarchean Massive Sulfide of Wutai Mountain, North China: A Black Smoker Chimney and Mound Complex Within 2.50 Ga-Old Oceanic Crust . . . . . . . . . . . . . . . . . . . . 339 Jianghai Li, Tim Kusky, Xianglong Niu, Feng Jun and Ali Polat Chapter 12. Inferred Ophiolites in the Archean Slave Craton . . . . . . . . . . . . . . . . . . . . . . . . . 363 P.L. Corcoran, W.U. Mueller and T.M. Kusky Chapter 13. 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia . . . . . . . . . . . . . . . . . 405 Igor S. Puchtel Chapter 14. 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences from the North Karelian Greenstone Belt, NE Baltic Shield, Russia . . . . . . . . . . . . . . . . . . . . . . . 425 A.A. Shchipansky, A.V. Samsonov, E.V. Bibikova, I.I. Babarina, A.N. Konilov, K.A. Krylov, A.I. Slabunov and M.M. Bogina Chapter 15. The Belingwe Greenstone Belt: Ensialic or Oceanic? . . . . . . . . . . . . . . . . . . . . . . 487 Axel Hofmann and Tim Kusky MODELS FOR THE EVOLUTION OF OCEANIC CRUST WITH TIME Chapter 16. Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites: Evidence of Archean Fore-Arc Magmatism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539 S.W. Parman and T.L. Grove Chapter 17. Precambrian Arc Associations: Boninites, Adakites, Magnesian Andesites, and Nb-Enriched Basalts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567 A. Polat and R. Kerrich Chapter 18. Archean Greenstone Belts Do Contain Fragments of Ophiolites . . . . . . . . . . . . . . . . 599 Maarten J. de Wit ANALOGS TO PRECAMBRIAN OPHIOLITES Chapter 19. Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites? . . . . . . . 615 John Encarnación Chapter 20. The Resurrection Peninsula Ophiolite, Mélange and Accreted Flysch Belts of Southern Alaska as an Analog for Trench-Forearc Systems in Precambrian Orogens . . . . . . . . . . 627 Timothy M. Kusky, Rose Ganley, Jennifer Lytwyn and Ali Polat Chapter 21. Phanerozoic Analogues of Archaean Oceanic Basement Fragments: Altaid Ophiolites and Ophirags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675 A.M.C. S¸ engör and B.A. Natal’in Chapter 22. Epilogue: What if Anything Have We Learned About Precambrian Ophiolites and Early Earth Processes? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 727 Timothy M. Kusky Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 739

xi

PREFACE

Recent developments have shown that many full and partial ophiolites are preserved in Precambrian cratons. This book provides a comprehensive description and discussion of the field aspects, geochemistry, geochronology, and structure of the best of these ophiolites. The book also presents syntheses of the characteristics of ophiolites of different ages, and presents an analysis of what the characteristics of these ophiolites mean for the thermal and chemical evolution of the earth. This book emphasizes new studies of Precambrian Geology that have documented ophiolites, ophiolitic fragments, and ophiolitic melanges in many Precambrian terranes. Each chapter focuses on individual Precambrian ophiolites or regions with numerous Precambrian ophiolites, and covers field aspects, petrology, geochemistry, geochronology, and other descriptive aspects of these ophiolites, and also delves into more theoretical and speculative aspects about the interpretation of the significance of these ancient ophiolites. The first section of the book focuses on Precambrian ophiolites and associated rock suites in Proterozoic terranes. Each chapter emphasizes field relationships, structural associations, geochronology, and chemistry to present a well-rounded perspective on some of the best documented Proterozoic ophiolites in the world. These descriptions serve as a useful standard of the characteristics of the preserved Proterozoic oceanic realm that can serve also for comparison with younger and older ophiolitic sequences. The second part of the book focuses on Archean ophiolites. The first several chapters examine different aspects of the Dongwanzi ophiolite and Zunhua ophiolitic melange in the North China Craton, and correlative ophiolitic terranes in the ca. 2.5 Ga Central Orogenic Belt of the craton. The next chapter examines the controversy surrounding interpretation of several greenstone belts in northern Canada’s Slave Province as ophiolites. The following chapters document Archean ophiolites from other terranes around the world, including Middle and Early Archean belts in Asia and Africa. The final section examines several younger orogens, including the Altaids, southern Alaska convergent margin, and the Philippines, as possible modern analogs to Precambrian, and especially Archean ophiolites. A synopsis of the main points of the papers in the book is presented at the end of the Introduction, and discussed again in a concluding chapter to highlight what has been learned about Precambrian oceanic spreading systems from this compilation of descriptions of Precambrian ophiolites. A table of some of the diagnostic, characteristic, typical, and rare aspects of ophiolites of all ages is presented in order to help determine if tectonically deformed and metamorphosed sequences in ancient mountain belts may be considered ophiolites. This comparative approach is important in that it enables users to more realistically characterize an allochthonous mafic/ultramafic rock sequences as ophiolitic than some other arbitrary classification schemes that have proposed requiring three or four

xii

Preface

of the Penrose-style ophiolitic units to be present in Precambrian sequences for a specific rock sequence to be considered ophiolitic. Once these tectonic fragments are recognized as remnants of ancient ocean floor, great progress may be made in understanding early Earth history.

ACKNOWLEDGEMENTS First and foremost I thank my wife Carolyn, and children Shoshana and Daniel, for their patience and understanding while writing and compiling this volume. I also extend my warm thanks to the students who have helped me with editing and formatting the text, particularly Rose Ganley, Angie Bond, Rusunoko Made, and Elisabet Head. Special thanks also go to Series Editor Kent Condie for inviting me to write and compile this volume, and to the editorial staff at Elsevier including Patricia Masser and Friso Venstra. I also wish to thank the many people who have helped shape my opinions about the Precambrian and those who have reviewed chapters in this volume. Outstanding among these groups are especially Kevin Burke, Bill Kidd, John Dewey, Paul Hoffman, Celail Sengör, Win Means, Akiho Miyashiro, Declan De Paor, Bruce Marsh, Dwight Bradley, Li Jianghai, Brian Windley, Alfred Kröner, Bob Stern, Bob Tucker, Maarten de Wit, Sam Bowring, Rich Goldfarb, and Peter Hudleston. Finally, I thank the many colleagues who have worked in the field with me, both on greenstone belts of Canada, USA, Africa, Australia, China, and Arabia, and on younger ophiolites of Newfoundland, Alaska, Oman, and the Appalachians. Timothy M. Kusky

xiii

DEDICATION

This book is dedicated to Kevin Burke, whose unfaltering commitment to understanding processes of plate tectonics on the Earth through time has inspired many geologists to critically examine some of the planet’s oldest rock sequences, searching for records of ancient plate interactions. Without Kevin’s knowledge, teaching, humor, and inspiration, many careers would have taken different routes and scientific discoveries and models such as those presented in this book may never have materialized.

This page intentionally left blank

1

INTRODUCTION T.M. KUSKY

Understanding the early history of the Earth and how the planet developed from a planetary nebula to its present, life-sustaining state is one of the most fundamental problems in Earth Sciences, and one that has occupied the thoughts of scholars and sages for centuries. To approach these questions it is necessary to synergize data from a variety of sources to estimate how plate tectonics, which is presently the surface expression of planetary heat loss, has evolved from a period of higher heat flow from the early Earth to its present state. This knowledge leads to a better understanding of how the surface and interior of the planet have evolved with time, and how previous interactions of the lithosphere, atmosphere, hydrosphere, and biosphere have been driven by the fundamental heat loss from the interior of the earth. These broad geodynamic goals require the integration of data from many different fields, including structural geology and tectonics, geophysics, petrology and geochemistry, sedimentology, and biology. It also involves several different scales of observation, from global geophysical data sets, to regional synthesis, through the outcrop scale, to the microscopic and molecular levels, which yield clues about the boundary conditions of formation and deformation. Understanding the early history of the Earth is thus a multi-disciplinary, multi-scale problem. To make headway in understanding how the early Earth operated, and how it may have differed or been similar to today’s Earth, it is necessary to compare studies of important Precambrian geological provinces with possible modern analogs, as well as other planets. Thus, in this book there are a number of studies that make comparisons between Precambrian and younger geological settings, and other chapters that describe active and Phanerozoic tectonic settings as potential analogs to Archean terranes, greenstone belts, and ophiolites.

1. GENERAL CHARACTERISTICS OF OPHIOLITES Ophiolites are a distinctive association of rocks interpreted to form in a variety of plate tectonic settings, including oceanic spreading centers, back-arc basins, forearcs, arcs, and other extensional magmatic settings including those in association with plumes (Moores, 1982, 2002; Gass et al., 1984; Lippard et al., 1986; Nicolas, 1989; Parson et al., 1992; Peters et al., 1991; de Wit and Ashwal, 1997; Dilek et al., 2000). A complete ophiolite grades downward from pelagic sediments into a mafic volcanic complex comprised mostly of pillow basalts, underlain by a sheeted dike complex. These are underlain by

2

Introduction

gabbros exhibiting cumulus textures, then tectonized peridotite, resting above a thrust fault that marks the contact with underlying rock sequences (Dewey and Bird, 1971; Dewey, 1977). The term “ophiolite” refers to this distinctive rock association (e.g., Sylvester et al., 1997), although many workers interpret the term to mean structurally emplaced oceanic lithosphere rocks formed exclusively at mid-ocean ridges. Many ophiolites are altered to serpentinite, chlorite, albite, and epidote rich rocks, possibly by hydrothermal sea floor metamorphism. In the 1960s and 1970s much research was aimed at defining a type ophiolite succession, which became known as the Penrose-type of ophiolite (Anonymous, 1972). More recent research has revealed that the variations between individual ophiolites are as significant as any broad similarities between them (e.g., Casey et al., 1981; Moores, 2002; Karson, 2001; Dilek et al., 2000; Dilek and Newcomb, 2003). A classic Penrose-type ophiolite is typically 5–15 km thick, and if complete, consists of the following sequence from base to top, with a fault marking the base of the ophiolite. The base of most ophiolites consists of harzburgite, consisting of olivine + orthopyroxene (± chromite), often forming strongly deformed or transposed compositional layering, forming harzburgite tectonite. The lowest unit in some ophiolites is lherzolite, consisting of olivine + clinopyroxene + orthopyroxene, generally interpreted to be fertile, undepleted mantle. In some ophiolites, harzburgite overlies lherzolite. The harzburgite is generally interpreted to be the depleted mantle from which overlying mafic rocks were derived, and the deformation is related to the overlying lithospheric sequence flowing away from the ridge along a shear zone within the harzburgite. The harzburgite sequence may be more than 10 km thick in some ophiolites, such as the Semail ophiolite in Oman, and the Bay of Island ophiolite in Newfoundland. See Fig. 1. Resting above the harzburgite is a group of rocks that were crystallized from magma derived by partial melting of the harzburgite. The lowest unit of these crustal rocks includes crystal cumulates of pyroxene and olivine, forming distinctive layers of pyroxenite, dunite, and other olivine + clinopyroxene + orthopyroxene peridotites including wehrlite, websterite, and pods of chromite + olivine. The boundary between these rocks (derived by partial melting and crystal fractionation) and those below from which melts were extracted is one of the most fundamental boundaries in oceanic lithosphere and defines the petrologic Moho, or base of the crust. In this case, the Moho is a chemical boundary, without a sharp seismic discontinuity. A seismic discontinuity occurs about half a kilometer higher than the chemical Moho in ophiolites. The layered ultramafic cumulates grade upwards into a transition zone of interlayered pyroxenite and plagioclase-rich cumulates, then into an approximately 1 km thick unit of strongly layered gabbro. Individual layers within this thin unit may include gabbro, pyroxenite, and anorthosite. The layered gabbro is succeeded upward by several to 5 km of isotropic gabbro, which is generally structureless but may have a faint layering. The layers within the isotropic gabbro in some ophiolites define a curving trajectory, interpreted to represent crystallization along the walls of a paleomagma chamber. The upper part of the gabbro may contain many xenoliths of diabase, pods of trondhjemite (plagioclase plus quartz), and may be cut by diabase dikes.

1. General Characteristics of Ophiolites

3

Fig. 1. Schematic columnar sections of representative oceanic crust and corresponding ophiolite types (modified after Moores, 2002). (A) Complete ophiolite sequence according to the “Penrose Conference definition” (Anonymous, 1972), characteristic of a magma-rich (generally fast spreading) ridge. (B) Faulted, incomplete sequence characteristic of a magma-starved spreading where tectonic processes dominate, designated a “Hess-type”ophiolite from Hess’s (1962) characterization of oceanic crust as serpentinized peridotite. (C) Complex composite section of oceanic island-arc sequences developed on oceanic crust. Designated a “Smartville-type” for the Smartville Complex in the northwest Sierra Nevada, California (e.g., Dilek et al., 1991). (D) Possible hotspot (oceanic plateau) section of oceanic crust. (E) A new, ”transitional type” of ophiolite, recognized where oceanic crust forms over extended continental crust, with an example from the Tihama Asir area of Saudi Arabia on the Red Sea. See text for discussion.

4

Introduction

The next highest unit in a complete, Penrose-style ophiolite is typically a sheeted dike complex, consisting of a 0.5–2 km thick complex of diabasic, gabbroic, to silicic dikes that show mutually intrusive relationships with the underlying gabbro. In ideal cases, each diabase dike intrudes into the center of the previously intruded dike, forming a sequence of dikes that have chilled margins developed only on one side. These dikes are said to exhibit one-way chilling. In most ophiolites that are not severely deformed or metamorphosed, examples of one way chilling may be found, but statistically the one-way chilling may only show directional preference in 50–60% of cases. The sheeted dikes represent magma conduits that fed basaltic flows at the surface. These flows are typically pillowed, with lobes and tubes of basalt forming bulbous shapes distinctive of underwater basaltic volcanism. The pillow basalt section is typically 0.5–1 km thick. Interstices between the pillows may be filled with chert, and sulfide minerals are common. Many ophiolites are overlain by deep-sea sediments, including chert, red clay, in some cases carbonates, or sulfide layers. Many variations are possible, depending on tectonic setting (e.g., conglomerates may form in some settings) and age (e.g., siliceous biogenic oozes and limestones would not form in Archean ophiolites, before the life forms that contribute their bodies developed on the Earth).

2. VARIATIONS BETWEEN OPHIOLITES AND OCEANIC CRUST FORMED UNDER DIFFERENT CONDITIONS Research in ophiolite studies and in situ oceanic crust in the past decade has revealed a much greater diversity in the abundance and sequence of rock types present than originally known when the Penrose-definition of an ophiolite was published (e.g., Moores, 2002; Karson, 2001; Dilek et al., 2000; Dilek and Newcomb, 2003). For instance, it is now thought that Penrose-style ophiolites form at fast spreading ridges such as the East Pacific Rise, whereas slow-spreading ridges like the Atlantic may have massive serpentinite overlying harzburgite tectonite, in turn faulted against pillows, dikes and gabbros. In some of these slow-spreading systems, extension is proceeding faster than magma can upwell to replace the lost volume of oceanic crust (e.g., Dick et al., 2003). Ophiolites produced in arc environments, either the main arc, forearc, or back arc, typically have thick mafic cumulate sections, interstratified volcanoclastic, pelagic, or hemipelagic sediments, and thick lava sequences. Many arc-type ophiolites are intruded by plutons with mafic to silicic compositions, that fed upper parts of the arc sequence. Oceanic crust produced at hot spots has particular relevance to the Archean, since some models for the Archean suggest that the mantle may have been slightly hotter in earlier times, and possibly analogous to hot spots. However, hot spots are not much hotter than surrounding mantle but, but simply produce greater amounts of magma than surrounding mantle. Places like Iceland, where a hot spot is superimposed on a mid-ocean ridge may be particularly analogous to Precambrian oceanic spreading environments, where both upwelling and enhanced melting processes are occurring. Oceanic plateaus, with thick crustal sections, may have been the norm for oceanic crust in the Archean, so it is difficult, when

2. Variations between Ophiolites and Oceanic Crust Formed under Different Conditions

5

assessing oceanic crust formed in early times, to differentiate between so-called normal oceanic crust and plateau-type crust. Oceanic plateaus have crustal sections that are similar to Penrose-style ophiolites, but may reach or exceed 10–15 km in thickness. It may be best, during our present early stages of studies of ancient oceanic crust, to use the standard of the present when describing Archean sequences. In this way, any similarities or differences between the present and Precambrian can be better-assessed than approaches that use a moving target of reference, such as the “alternative Earth” model of Hamilton (2003), or the ad-hoc models of Bickle et al. (1994). Several of the ophiolites described in this volume appear to have formed within the transition from rifted continental margins to ocean spreading centers during early stages of ocean opening, then were structurally detached and/or deformed and incorporated into convergent margins during ocean closure. These ophiolites are distinctive from classical Penrose-style ophiolites and others formed in forearc and back arc environments, and we coin the term “transitional ophiolites” for this new class of ophiolite. During early stages of ocean formation, continental crust and lherzolite of the subcontinental mantle is extended forming graben on the surface, and ductile mylonites at depth. Sedimentary basins may form in the graben, and as the extension continues magmatism sometimes affects the rifted margin, either forming volcanic rifted margins, or migrating to a spreading center forming an oceanic spreading center. New asthenospheric mantle upwells along the new ridge, and may intrude beneath the extended continental crust. In some cases, wedges of extended mid-to-lower continental crust overlying mylonitic lherzolitic sub-continental mantle become intruded by numerous dikes and magmas from this new asthenospheric mantle. These dikes then feed a crustal gabbroic magma chamber closer to the surface, which in turn may feed a dike complex and basaltic pillow/massive lava section. If preserved, this unusual sequence forms a “transitional ophiolite”, grading down from subaquatic sediments, to pillow lavas, dikes, sheeted dikes, layered gabbro, dunite and pyroxenite cumulates, then remarkably into stretched, typically mylonitic granitic mylonites, underlain by lherzolite. The lherzolite tectonic may be underlain by harzburgite tectonite or harzburgite. (See Fig. 1.) Examples of this type of transitional ophiolite are found in the Proterozoic Jourma complex, and in some of the Slave Province ophiolites (see papers by Peltonen and Kontinen, 2004, and Corcoran et al., 2004). Modern analogs for such transitional ophiolites are found around the Red Sea, including at Tihama Asir, Saudi Arabia, where a 5–10 Ma old transitional ophiolite has a dike complex overlying layered gabbro, which in turn overlies continental crust. Also, on Egypt’s Zabargad Island, oceanic mantle is exposed, and it is likely that the crustal structure near this region preserves transitional ophiolites as well. The main lesson here is that ophiolites may form in many tectonic settings, from extended continental crust, to mid ocean ridges, to forearcs, arcs, back arcs, to triple junctions along convergent margins. Once we identify a sequence as ophiolitic, then we need to identify the tectonic environment in which it formed before we can make realistic comparisons to younger environments to determine how plate tectonic style has changed with time.

6

Introduction

3. PROCESSES OF OPHIOLITE AND OCEANIC CRUST FORMATION The sequence of rock types described above are a product of specific processes that occurred within the oceanic spreading centers along which the ophiolites formed (see reviews by Anonymous, 1972; Moores, 1982; Nicolas, 1989; Parson et al., 1992; Dilek et al., 2000). Variations in the rock sequence, mineralogy, chemistry, or structure of ophiolites with time may be related to variations in the processes that produced the ophiolites. With higher heat production in the early Earth, it is important to document these variations to determine how the Earth may have lost heat in early times of high heat production. It is not clear if the early mantle responded to the higher heat production by a significant increase in temperature, a change in viscosity and greater ease in convective overturn, or some other process (e.g., Abbott and Mencke, 1990; Arndt et al., 1997; Condie, 1981, 1997b; Grove et al., 1994; Helmstaedt and Shulze, 1989; Martin, 1986, 1993; McKenzie and Bickle, 1988; Nisbet et al., 1993; Parman et al., 1997; Turcotte and Schubert, 2002; Parman and Grove, 2004). Studies of Precambrian ophiolites and related rocks have great potential to unravel the secrets of early heat loss from the planet. As the mantle convects and the asthenosphere upwells beneath mid ocean ridges, mantle pyrolites, harzburgites and lherzolites undergo partial melting of 10–15% in response to the decreasing pressure. The percentage of partial melt may have been different in early times if mantle temperatures were significantly higher. So far, estimates of partial melt fractions estimated from Precambrian ophiolites have not determined whether or not the Precambrian melt fractions were on average greater than, less than, or similar to those of the younger record. The melts derived from the harzburgites rise to form a magma chamber beneath the ridge, forming the crustal section of the oceanic crust. As the magma crystallizes the densest crystals gravitationally settle to the bottom of the magma chamber, forming layers of ultramafic and higher mafic cumulate rocks. Above the cumulates, a gabbroic fossil magma chamber forms, typically with layers defined by varying amounts of pyroxene and feldspar crystals. In many examples the layering in ophiolites has been shown to be parallel to the fossil margins of the magma chamber. An interesting aspect of the magma chamber is that periodically, new magma is injected into the chamber, changing the chemical and physical dynamics. These new magmas are injected during extension of the crust so the magma chamber may effectively expand infinitely if the magma supply is continuous, as in fast spreading ridges. In slow spreading ridges the magma chamber may completely crystallize before new batches of melt are injected. Studies of ancient ophiolites therefore have the potential to estimate relative rates of extension and magma supply, a line of research that has not yet matured in Precambrian ophiolite studies. As extension occurs in the oceanic crust, dikes of magma shoot out of the gabbroic magma chamber, forming a diabasic (fine-grained rapidly cooled magma with the same composition as gabbro) to gabbroic sheeted dike complex. The dikes have a tendency to intrude along the weakest, least crystallized part of the previous dike, which is usually in the center of the last dike to intrude. In this way each dike intrudes the center of the previous dike, forming a sheeted dike complex characterized by dikes that have only one chill margin, most of which face in the same direction. In some Phanerozoic ophiolites, varia-

4. Historical Recognition of Archean Ophiolites

7

tions in the thickness and character of the dikes with depth have been related to temperature changes with depth. The dike complex thus represents a potential indicator of geothermal gradients in ancient ophiolites, but extracting such information from Precambrian ophiolites may be difficult due to the paucity of Precambrian dike complexes, plus the effects of deformation and metamorphism have obscured many original relationships. Many of the dikes reach the surface of the sea floor, where they feed basaltic lava flows. Basaltic lava flows on the sea floor are typically in the form of bulbous pillows that stretch out of magma tubes, forming the distinctive pillow lava section of ophiolites. The top of the pillow lava section is typically quite altered by sea floor metamorphism including mineralized veins that culminate in deposits of black smoker-type hydrothermal vents (e.g., Harper, 1999; Li et al., 2004). Early life forms probably flourished around these deep sea hydrothermal vents, but few studies have focused on searching for early life forms around Precambrian sea floor black smokers (but see Huang et al., 2004). Studies of lava vesicularity, and the nature of interpillow sediments, have the potential to yield clues about the water depth and hence crustal thickness of ancient ophiolites, but few studies have yet explored these potentially critical relationships. The pillow lavas are overlain by sediments deposited on the sea floor. In the Phanerozoic oceans, if the oceanic crust formed above the calcium carbonate compensation depth, the lowermost sediments may be calcareous. These would be succeeded by siliceous oozes, pelagic shales, and other deep water sediments as the sea floor cools, subsides, and moves away from the mid ocean ridge. There has been no comprehensive analysis of the types of sediments expected to be deposited on Precambrian oceanic crust as it moved away from spreading centers, nor how this sequence may have changed systematically with time. A third sequence of sediments may be found on the ophiolites. These would include sediments shed during detachment of the ophiolite from the sea floor basement, and its thrusting (obduction) onto the continental margin. The type of sediments deposited on ophiolites should have been very different in some of the oldest ophiolites that formed in the Precambrian. For instance, in the Proterozoic and especially the Archean, organisms that produce the carbonate and siliceous oozes were not present, as the organisms that produced these sediments had not yet evolved. Thus, study of the sedimentary sequences deposited on Precambrian ophiolites may yield important information about Precambrian sea water and atmospheric conditions, sedimentation processes, and about the development and evolution of life in the early oceans.

4. HISTORICAL RECOGNITION OF ARCHEAN OPHIOLITES Very few complete Phanerozoic-like ophiolite sequences have been recognized in Archean greenstone belts, leading some workers to the conclusion that no Archean ophiolites or oceanic crustal fragments are preserved (e.g., Bickle et al., 1994). However, as emphasized by Sylvester et al. (1997), the original definition of ophiolites (Anonymous, 1972) includes “dismembered”, “partial”, and “metamorphosed” varieties, and there is no justification for new arbitrary definitions that attempt to exclude portions of Archean

8

Introduction

greenstone belts that contain two or more parts of the full ophiolite sequence, especially in structurally complex settings such as found in greenstone belts (e.g., Harper, 1985; de Wit et al., 1987; Kusky 1990, 1991). Similarly, many Proterozoic ophiolites are dismembered, or partial sequences (Kröner, 1985; Berhe, 1990; Dann, 1991). Archean oceanic crust was possibly thicker than Proterozoic and Phanerozoic counterparts, resulting in accretion predominantly of the upper section (basaltic) of oceanic crust (Burke et al., 1976; Hoffman and Ranalli, 1988; Moores, 1986; Burke, 1995; Kusky and Vearncombe, 1997). The crustal thickness of Archean oceanic crust may in fact have resembled modern oceanic plateaux (e.g., Sleep and Windley, 1982; Kusky and Kidd, 1992). If this were the case, complete Phanerozoic-like MORB-type ophiolite sequences would have been very unlikely to be accreted or obducted during Archean orogenies. In contrast, only the upper, pillow lava-dominated sections would likely be accreted (Kusky and Polat, 1999). Portions of several Archean greenstone belts have been interpreted to contain dismembered or partial ophiolites. Accretion of MORB-type ophiolites has been proposed as a mechanism of continental growth in a number of Archean, Proterozoic, and Phanerozoic orogens (Kusky and Polat, 1999). It is worthwhile to investigate these claims to better understand the crustal structure and tectonic setting in which these Archean ophiolites formed. Several suspected Archean ophiolites have been particularly well-documented. One of the most disputed is the circa 3.5 Ga Jamestown ophiolite in the Barberton greenstone belt of the Kaapvaal craton (Brandl and de Wit, 1997). de Wit et al. (1987) describe a 3 km thick tectonomagmatic sequence including a basal peridotite tectonite unit with chemical and textural affinities to Alpine-type peridotites, overlain by an intrusiveextrusive igneous sequence, and capped by a chert-shale sequence. This partial ophiolite is pervasively hydrothermally altered and shows chemical evidence for interaction with sea water with high heat and fluid fluxes (de Wit et al., 1990). SiO2 and MgO metasomatism and black-smoker like mineralization is common, with some possible hydrothermal vents traceable into banded iron formations, and subaerial mudpool structures. These features led de Wit et al. (1982, 1992) to suggest that this ophiolite formed in a shallow sea, and was locally subaerial, analogous to the Reykjanges ridge of Iceland. In this sense, Archean oceanic lithosphere may have looked very much like younger oceanic plateaux lithosphere. Maarten de Wit (2004) presents an updated description and interpretation of the Barberton belt and Jamestown ophiolite in this volume, incorporating years of additional mapping. Several partial or dismembered ophiolites have been described from the Slave Province, although these too have been disputed (e.g., King and Helmstaedt, 1997). From the Point Lake greenstone belt in the central Slave Province, Kusky (1991) described a fault-bounded sequence grading downwards from shales and chemical sediments (umbers) into several kilometers of pillow lavas intruded by dikes and sills, locally into multiple dike/sill complexes, then into isotropic and cumulate-textured layered gabbro. The base of this partial Archean ophiolite is marked by a 1 km thick shear zone composed predominantly of mafic and ultramafic mylonites, with less-deformed domains including dunite, websterite, wehrlite, serpentinite, and anorthosite. By using down-plunge projections and sectionbalancing techniques, Kusky (1991) estimated that the shear zone at the base of this ophi-

4. Historical Recognition of Archean Ophiolites

9

olite accommodated a minimum of 69 km of slip. Although this still allows the ophiolite to have formed at or near extended older continental crust that forms the Anton terrane to the west (Kusky, 1989), the actual amount of transport was probably much greater. Synorogenic conglomerates and sandstones were deposited in several small foredeep basins, and are interbedded with mugearitic lavas (and associated dikes), all deposited/intruded in a foreland basin setting. Kusky (1990) suggested that portions of the Cameron and Beaulieu River greenstone belts of the southern Slave Province contain ophiolitic components. The belts are cut by numerous layer-parallel shear zones, but some sections are composed mostly of tholeiitic pillow basalts, others contain approximately equal quantities of pillows and dikes, and a few sections consist of nearly 100% mafic dikes. The bases of these greenstone belts are marked by up to 500 m thick shear zones (locally containing mélanges), with tectonic blocks of gabbro, mafic volcanics, peridotite, and slivers of the underlying quartzofeldspathic gneiss (with extensive mafic dike complexes) and its autochthonous cover. Original relationships between dikes in the basement complex and dikes in the basal parts of the greenstone belts have not been established, but older-generation mafic dikes do not cut intervening sedimentary sequences nor the shear zone that separates the greenstone belt from the basement. Helmstaedt et al. (1986) describe a pillow lava sequence that grades down into sheeted dikes and gabbro from the Yellowknife greenstone belt, but interpreted the basal contact of the belt as an unconformity on a banded iron formation, an interpretation questioned by Kusky (1987). The dikes and pillow lavas are geochemically similar to MORB (MacLaughlin and Helmstaedt, 1995), although Isachsen et al. (1991), and Isachsen and Bowring (1997) have shown that the Yellowknife greenstone belt contains several different, and probably unrelated volcanic and sedimentary sequences, separated by as much as 50 Ma and spanning an age interval of 200 Ma. In this volume, Patricia Corcoran, Wulff Mueller, and Tim Kusky present a review of the current status of the ophiolitic interpretation of some of the greenstone belts in the Slave Province, synthesizing fifteen years of debate on the interpretation. Harper (1985) and Wilks and Harper (1997) describe rocks of the South Pass area in the Wind River Range, Wyoming, as containing a dismembered metamorphosed Archean ophiolite. This ophiolite contains all of the units of a complete ophiolite except the basal peridotite tectonite, and contacts between all units are shear zones. Cumulate textures in ultramafic rocks and gabbros are present, as are small exposures of a sheeted dike complex. Pillow lavas are associated with metapelites and banded iron formation. It has been argued (Bickle et al., 1994) that the paucity of well-developed sheeted dike complexes known from Archean greenstone belts indicates that they are not ophiolites. But sheeted dikes are not well-preserved in many Phanerozoic ophiolites, especially when they are metamorphosed and deformed to the extent that most Archean greenstone belts are. Abbott (1996) argues that sheeted dikes are not necessarily formed in every ocean floor sequence. Despite this, sheeted dike complexes have been discovered in several of the ophiolitic greenstone belts described above. Well-developed sheeted dike complexes have also been mapped in several locations in the Kalgoorlie terrane of the Yilgarn craton (Fripp and Jones, 1997). Multiple cooling units of dolerite, high-Mg mafic rocks, and ser-

10

Introduction

pentinite are truncated at an angle between 35◦ –80◦ by an unconformably overlying mafic volcanic breccia, pillow breccia, and lenses of pillow lava that strike parallel to bedding in overlying sedimentary rocks in the Kanowna Lake area. Fripp and Jones (1997) interpret this unit as a sheeted dike complex overlain by a volcanic carapace. At the Cowan Lake Six Islands locality, Fripp and Jones (1997) describe lherzolite and dunite that grade up into websterite and gabbro with pyroxenite layers. These rocks are overlain by high-Mg mafic and picritic basalts that occur in multiple tabular cooling units, interpreted as sheeted dikes that exhibit both one-way and two-way chill margins. These are overlain by chert, silicified mudstone, shale and graywacke turbidites, which locally occur as partially assimilated xenoliths (containing zircons) within the intrusive rocks. Fripp and Jones (1997) interpret the Lake Cowan greenstone locality to include the peridotitic lower plutonic sequence that marks the transition zone between mantle and crust in ophiolite suites. This transition zone sequence is overlain by a sheeted dike complex, but the extrusive magmatic carapace is omitted by faulting at this locality. Fripp and Jones (1997) note the many similarities between the Kalgoorlie ophiolites and Phanerozoic ophiolites such as the Samail, Troodos, and Bay of Islands massifs. Kimura et al. (1993) interpret parts of the Larder Lake and Beardmore-Geraldton greenstone belts in the Abitibi and Wabigoon subprovinces of the Superior Province to include ophiolitic fragments accreted in arc environments, in a manner analogous to the setting of the basalt/chert slivers of the Schreiber-Hemlo belt (Kusky and Polat, 1999). The Larder Lake belt occurs in the southern part of the Abitibi greenstone belt, and consists of pillow basalts and banded iron formation (BIF) tectonically stacked with terrigeneous turbidites. The pillow basalts and BIF are interpreted to be the upper part of an oceanic plate stratigraphy, offscraped and interdigitated with trench turbidites in an accretionary wedge setting similar to Alaska (Kusky et al., 1997a, 1997b, 2004a; Kusky, 2004) or Japan (Isozaki et al., 1990). Kimura et al. (1993) and Williams et al. (1991) also suggest a similar exotic origin for basalts and iron formation tectonically interleaved with terrigeneous turbidites in the Beardmore-Geraldton area in the southern part of the Wabigoon subprovince. In both of these examples, the accreted trench turbidites and ophiolitic slivers are intruded and overprinted by arc-related plutons and lavas, formed when the trench stepped back and intruded its own accretionary wedge. A similar accretionary wedge setting and oceanic crustal origin for slivers of basalt in greenstone belts of the Pilbara craton has been proposed by Isozaki et al. (1992). Evidence for the creation and obduction of oceanic crust in the Archean is not limited to field relationships as described above. Jacob et al. (1994) report that the geochemistry of diamondiferous eclogites from the Udachnaya Mine, Siberia (Puchtel et al., 1997), are most consistent with derivation from subducted slabs of Archean oceanic crust that were extensively hydrothermally altered prior to subduction. Similarly, many eclogite samples from South African kimberlites are also interpreted as remnants of subducted Archean oceanic crust (e.g., Jagoutz et al., 1984; MacGregor and Manton, 1986; Carlson et al., 2000). In summary, dismembered ophiolites are a widespread component of Archean greenstone belts, and many of these apparently formed as the upper parts of Archean oceanic crust. Most of these appear to have been accreted within forearc and intra-arc tectonic set-

5. Historical Recognition of Proterozoic Ophiolites

11

tings. The observation that Archean greenstone belts have such an abundance of accreted ophiolitic fragments compared to Phanerozoic orogens suggests that thick, relatively buoyant, young Archean oceanic lithosphere may have had a rheological structure favoring delamination of the uppermost parts during subduction and collisional events (Hoffman and Ranalli, 1988; Kusky and Polat, 1999). 5. HISTORICAL RECOGNITION OF PROTEROZOIC OPHIOLITES In contrast to the Archean, a number of ophiolites have been recognized and generally accepted in Proterozoic terranes for a number of years. The Late Proterozoic ArabianNubian Shield hosts a number of ophiolite-decorated sutures, and boasts one of the highest ophiolite densities known for a Proterozoic terrane on the planet. The Arabian-Nubian Shield is part of the East African Orogen that stretches from the Arabian-Nubian Shield in the north, through parts of India, east Africa and Madagascar, and has uncertain links with Neoproterozoic orogens in Antarctica and elsewhere around the globe. The East African Orogen has a complex history including a record of the break-up of Rodinia at circa 900– 800 Ma, and the evolution of numerous arc systems, oceanic plateaux, oceanic crust, and sedimentary basins. Neoproterozoic closure of the Mozambique Ocean sutured east and west Gondwana along the length of the East African Orogen (Stern, 1994; Kusky et al., 2003a, 2003b). An accretionary collage of arc, ophiolitic, and microcontinental terranes formed during closure of the Mozambique ocean is now preserved in the Arabian-Nubian shield. Some of the arc terranes appear to represent juvenile additions to the continental crust during this time period, whereas others may have been built on older continental basement on the margins of the Mozambique Ocean. Many of the ophiolites in the East African Orogen and Arabian-Nubian shield record different aspects of this complex history. Careful field mapping, geochronology, geochemistry, and structural analyses of some of these ophiolites has led to significant improvements in understanding of heat loss from the Neoproterozoic Earth, continental growth, and preservation of juvenile crustal terranes. Paleo and Meso-Proterozoic ophiolites are less abundant than the Neoproterozoic ophiolites so well preserved in the Arabian-Nubian Shield. A few Mesoproterozoic ophiolitic terranes have been described from the Karelian Shield, Cape Smith Belt, West Africa, and the SW USA (e.g., Abouchami et al., 1990; Boher et al., 1992; Dann, 2004; Scott et al., 1991, 1992; St-Onge et al., 1989). However, until recently, no Mesoproterozoic ophiolites were known. The first section of this book presents descriptions of several recently recognized Proterozoic ophiolite sequences, as well as reviews of some of the more classic Proterozoic ophiolitic terrains. 6. THE ROLE OF OCEANIC LITHOSPHERE IN CONTINENTAL GROWTH Although the rate of continental growth is a matter of geological debate (Fyfe, 1978; Dewey and Windley, 1981; Armstrong, 1991; McCulloch and Bennett, 1994; Taylor and McLennan, 1995; Rudnick, 1995; Artemieva and Mooney, 2001), most geological data

12

Introduction

indicates that the continental crust has grown by accretionary and magmatic processes taking place at convergent plate boundaries since the early Archean (Burke et al., 1976; Sleep and Windley, 1982; Taylor and McLennan, 1995; Friend et al., 1988; Card, 1990; Condie, 1994, 1997a; Sengör and Natal’in, 1996; Windley et al., 1996). Arc-like trace element characteristics of continental crust suggest that subduction zone magmatism has played an important role in the generation of the continental crust (Taylor and McLennan, 1995; Hofmann, 1988; Tarney and Jones, 1994; Rudnick, 1995); a corollary of this observation is that oceanic crust and lithosphere must have been created in order to be subducted. Convergent margin accretionary processes that contribute to the growth of the continental crust include oceanic plateau accretion, oceanic island arc accretion, normal ocean crust (mid-ocean ridge) accretion/ophiolite obduction, back-arc basin accretion, and arc-trench migration/Turkic-type orogeny accretion (Ben-Avraham et al., 1981; Kusky and Polat, 1999). These early accretionary processes are typically followed by intrusion of late stage anatectic granites, late gravitational collapse, and late strike slip faulting (Kusky, 1993; 1998). Together, these processes release volatiles from the lower crust and mantle and help to stabilize young accreted crust and form stable continents (see also Abbott, 1991; Abbott and Mooney, 1995; Abbott et al., 1997), and also alter in various ways the original chemical and physical relationships that make the primary tectonic setting of the rock suites difficult to determine. Thus, in order to understand the earliest physical and chemical conditions in potential Precambrian ophiolite suites it is typically necessary to first unravel a complex suite of later superimposed structures and mineral assemblages. Sengör et al. (1993) and Sengör and Natal’in (1996, and see also this volume) proposed a new type of orogeny, so-called “Turkic or accretionary-type orogeny”, for continental growth. These orogenic belts possess very large sutures (up to several hundred km wide) characterized by subduction-accretion complexes and arc-derived granitoid intrusions, similar to the Circum-Pacific accreted terranes (e.g., Alaska, Japan). These subductionaccretion complexes are composed of tectonically juxtaposed fragments of island arcs, back-arc basins, ocean islands/plateaux, trench turbidites, and micro-continents (Sengör, 1993; Sengör and Natal’in, 1996). Another important feature of these orogens is the common occurrence of orogen parallel strike-slip fault systems, resulting in lateral stacking and bifurcating lithological domains (Sengör and Natal’in, 1996). In these respects, the accretionary-type orogeny may be considered as unified accretionary model for the growth of the continental crust. In the concluding section of this book, Sengör and Natal’in (2004) present an analysis of the Altaids orogen and compare ophiolitic fragments therein with similar ophiolitic fragments in Precambrian terranes, especially Archean granitegreenstone terranes. Their similarity is remarkably, suggesting a common origin.

7. CONVERGENT MARGIN ANALOGS TO ARCHEAN GREENSTONE BELTS The southern Alaska convergent margin consists of belts of accreted PaleozoicCenozoic flysch, mélange, and ophiolites. Paleogene near-trench plutons intruded the margin diachronously from 61 Ma in the west to 50 Ma in the east during migration of a

7. Convergent Margin Analogs to Archean Greenstone Belts

13

trench-ridge-trench triple junction. Recognizing these plutons as a product of ridge subduction has several implications for forearc evolution and interpretation of linear belts of plutons, flysch, ophiolites, and mélange in Precambrian orogens. Forearcs are not necessarily places characterized exclusively by high-P, low-T, metamorphic series and a lack of plutonism, but may contain high-T, low-P metamorphism in association with belts of plutonic rocks if the forearc was affected by ridge subduction. Similarly, belts of magmatic rocks in Precambrian orogens (e.g., Wylie et al., 1997; Rapp, 1997) may not necessarily represent individual arc terranes, but could be a paired arc/forearc system that experienced ridge subduction. The record of ridge subduction events varies considerably depending on plate geometry and rates of triple junction and slab window migration. However, some of the hallmark signatures of ridge subduction in forearcs include along strike diachronous intrusion of TTG (tonalitic, trondhjemitic, granodiorite), to granitic plutons, high-T metamorphism, emplacement of ophiolites in the forearc, diachronous gold mineralization, and belts of anomalous complex faulting. Structural, thermal and magmatic aspects of the Chugach terrane are similar to the geology of Archean granite-greenstone terranes. In both, deformation is locally melt-dominated, and plutons follow a low-K series from diorite to trondhjemite. Metamorphism is of a high-T, low-P series. Most Archean granite-greenstone terranes acquired their first-order structural and metamorphic characteristics at convergent plate margins, where large accretionary wedges similar in aspect to the Chugach, Makran, and Altaids grew through offscraping and accretion of oceanic plateaux, oceanic crustal fragments, juvenile island arcs, rifted continental margins, and pelagic and terrigeneous sediments. Some suites of TTG in these terrains appear to have been generated during ridge subduction events, suggesting that ridge subduction is an important process in continental growth. Ridge subduction was likely an important process in the Archean, when the total number of plates was higher, and the number of ridge-trench encounters was greater. The southern Alaska margin serves as a relatively modern example of processes important in Archean forearc evolution and continental growth. 7.1. Near Trench Magmatism Associated with Ridge Subduction Near-trench magmatic rocks in southern Alaska related to subduction of the Kula-Farallon ridge include muscovite-biotite granodiorites, leucotonalites, trondhjemites, and dikes with basaltic to rhyolitic compositions. The ages of these plutons are diachronous along strike (Bradley et al., 1994, 2003; Kusky et al. 1997a, 1997b, 2003a, 2003b), and track the migration of the Kula-Farallon-North America triple junction as it migrated along the Cordilleran margin. The plutons can be used as time-markers to help construct a model for the structural evolution of the wedge during ridge subduction (Kusky et al. 1997a, 1997b, 2003a, 2003b). 7.2. Generation and Emplacement of Ophiolites at the Triple Junction A geochemical study and petrologic model for the generation of the Resurrection and Knight Island ophiolites (Lytwyn et al., 1997; Kusky et al., 2004b, 2004c, 2004d), formed

14

Introduction

at the triple junction and soon after incorporated into the accretionary wedge concluded that the T-MORB’s show geochemical variations related to derivation from and mixing of multiple parental magmas, derived through near-fractional melting of variably depleted mantle sources in a subaxial melting column. Geochemical evidence for contamination of the magmas by sedimentary material similar in composition to nearby flysch, supports the interpretation that these ophiolites formed in a near-trench triple junction environment. Kusky and Young (1999) relate emplacement of the ophiolite to ridge subduction, and show that the ophiolite was formed at a triple junction, incorporated into the accretionary prism, and intruded by near-trench magmas, all within 3.6 Ma. Magmas of the Aialik pluton intruded along a shear zone formed during incorporation of the ophiolite into the accretionary prism. The sedimentary geochemistry of the shale/turbidite sequence overlying the ophiolite shows a progressively increasing terrigenous source up-section, as the future Resurrection ophiolite approached the Chugach accretionary prism. 7.3. Comparison to Crustal Growth Mechanisms in the Precambrian Comparison of crustal growth processes in accretionary prisms affected by ridge subduction with crustal growth as seen in Archean shield areas offers insight about processes in Archean subduction zones. From this, it should be possible to eventually estimate if ridge subduction was more common in Precambrian terranes, when plate boundary lengths were greater (and total number of plates greater), and subducting lithosphere was on average younger. Ridge subduction related processes should have been more important in the older geologic record, as the number of triple junctions increases nearly twice as fast as the number of plates. With the probable larger number of plates accommodating greater heat production in the Archean, the number of triple junctions would increase, and the role of ridge subduction is thus expected to have been more important in the growth of the continental crust in Precambrian times.

8. SYNOPSIS OF THIS VOLUME The first section of this book contains several chapters that present descriptions and interpretations of Proterozoic ophiolites, including reviews of previously described ophiolites, and some descriptions of sequences interpreted as ophiolites for the first time. 8.1. Proterozoic Ophiolites and Related Rocks Until the recent discovery of the 2.5 Ga Dongwanzi ophiolite, many students of Precambrian geology regarded the circa 1.96–2.0 Ga Jourma ophiolite of Finland and the 1.998 Ga Purtuniq ophiolite from the Cape Smith Belt as the world’s oldest, nearly complete well-preserved ophiolites. The Jourma ophiolite is a remarkable polyphase ophiolite that preserves evidence for formation at an ocean-continent transition. In this volume, Petri Peltonen and A. Kontinen present a detailed description and review of recent work on the

8. Synopsis of This Volume

15

Jourma ophiolite by the Finnish Geological Survey and other teams. They conclude that the Jormua Ophiolite is an allochtonous mafic-ultramafic rock complex, thrusted onto the Karelian Craton margin, that formed within a passive margin environment ∼ 100 km southwest from its present position. This complex consists of two distinct units: (1) fragments of ancient subcontinental lithospheric mantle that became exposed beneath the Archean craton by detachment faulting following the final break-up of the craton, and (2) alkaline and tholeiitic igneous suites that were emplaced within and through the lithospheric mantle at ∼ 2080 Ma and 1950 Ma, respectively. The mantle peridotites had yielded melt already before they were intruded by the oldest suite of dikes at > 2800 Ma. These old dikes are “dry” clinopyroxene cumulates being products of an Archean magmatic episode. Later, during the initial stages of continental break-up at ∼ 2080 Ma, this same piece of mantle became extensively intruded by hydrous alkaline magmas that resulted in formation of high-pressure hornblendite-garnetite cumulates deep in the ophiolite stratigraphy and fine grained OIB-type dikes at more shallow levels. Simultaneously, the residual peridotites became metasomatized due to porous flow of the melt in the peridotite matrix. Alkaline magmatism was soon followed by lithospheric detachment faulting that exposed the subcrustal peridotites at the seafloor, where they became covered by tholeiitic (EMORB) pillow and massive lavas and intruded by coeval dikes and gabbros. Since transitional contacts between all main ophiolite units can be demonstrated, the Jormua Ophiolite Complex is interpreted to represent a practically unbroken sample of seafloor from an ancient oceancontinent transition (OCT) zone, strikingly similar to that reported from the Cretaceous West Iberia non-volcanic continental margin. From Payson Arizona, Jesse Dann (2004) presents maps, photographs, and data on the 1.73 Ga Payson Ophiolite. This well-exposed but unusual ophiolite is disposed as a shallow-dipping, layered sequence of coeval gabbro, sheeted dikes, and submarine volcanic rocks, partly disjointed by later intrusion and deformation. A sheeted dike complex is spectacularly exposed on cliffs and in stream sections in shallow canyons. The dike complex is rooted in underlying gabbro, as shown by gabbro-dike mingling and mutual intrusion. An unusual continuous zone of intense alteration marks the transition from sheeted dikes to submarine volcanics. The Payson ophiolite has many arc-like characteristics. A tonalite/dacite magmatic suite occurs as rare lavas and as dikes and hypabyssal plutons mutually intrusive with the basaltic sheeted dikes and gabbro. An older basement complex occurs as roof pendants in gabbro and screens in the sheeted dike complex. Dann suggests a model for the Payson ophiolite in which an intra-arc basin formed by seafloor spreading along an arc-parallel strike-slip fault system, and relates its emplacement within the arc and accretion to North America to events associated with the ca. 1.70 Ga Yavapai Orogeny. Proterozoic ophiolites were first widely recognized from the Arabian-Nubian Shield. In this volume, Robert Stern, Peter Johnson, Alfred Kröner, and Bisrat Yibas present a review of the Neoproterozoic ophiolites of the Arabian-Nubian Shield, whereas Peter Johnson, Fayak Kattan, and Ahmed Al-Saleh describe the field characteristics of some of the better-exposed ophiolites in more detail. Ophiolites of the Arabian-Nubian Shield range in age from 690 to 890 Ma and in the northern part of the shield, occur as nappe complexes marking suture zones between terranes. Although dismembered and altered, all of the diag-

16

Introduction

nostic components of ophiolites can be found, including harzburgite, cumulate ultramafics, layered and higher level gabbro and plagiogranite, sheeted dikes, and pillow basalt lavas. Allochthonous mafic-ultramafic complexes in the southern part of the shield in Ethiopia and Eritrea are interpreted as ophiolites, but are more deformed and metamorphosed than those in the north. Reconstructed ophiolitic successions have crustal thicknesses of 2.5 to 5 km. The Arabian-Nubian shield ophiolitic mantle was mostly harzburgitic, and exhibits chemical compositions comparable to modern forearcs and distinctly different from midocean ridges and backarc basin peridotites. Arabian-Nubian shield ophiolites are often associated with a thick (1–3 km) sequence of cumulate ultramafic rocks, which define a transition zone between the seismic and petrologic Mohos. These cumulates are dominated by dunite, with subordinate pyroxene-rich lithologies. Cumulate ultramafics grade upwards into layered gabbro. Both tholeiitic and calc-alkaline affinities are present, and a significant, although subordinate, amount of boninites have been identified. The ArabianNubian shield ophiolitic lavas include both LREE-depleted and LREE-enriched varieties, but as a group are slightly LREE-enriched. On a variety of discrimination diagrams, the lavas plot in fields for MORB, BABB, arc tholeiite, and boninite. Nd-isotopic compositions indicate derivation from a long-depleted mantle source. Mineral and lava compositions are consistent with the hypothesis that most Arabian-Nubian shield ophiolites formed in ‘suprasubduction zone’ (SSZ) settings, and the high Cr# of Arabian-Nubian shield ophiolitic harzburgites suggests a forearc environment. Stern et al. (2004) conclude that studies of deep water sediments deposited on Arabian-Nubian shield ophiolites are needed to better characterize and understand the Neoproterozoic ocean where the ophiolites formed. In a chapter that is complimentary to the overview by Stern and others, Peter Johnson, Fayek Kattan, and Ahmed Al-Saleh describe field relationships from a number of the Arabian Shield ophiolites. Where most complete, they consist of serpentinized peridotite, gabbro, dike complex, basalt, and pelagic rocks. However, because of folding and shearing, the majority of the ophiolites lack one or more of these diagnostic lithologies. Nonetheless, the incomplete assemblages are identified as ophiolites because they minimally include peridotite and gabbro, in many cases are associated with basalt, and in all cases show evidence of emplacement by thrusting and shearing rather than intrusion. The ophiolites range in age from ∼ 870 Ma to ∼ 695 Ma, documenting a 200-million year period of oceanic magmatism in the Arabian shield, and are caught up in ∼ 780 Ma to ∼ 680 Ma suture zones that reflect a 100-million year period of terrane convergence. All the ophiolites are strongly deformed, metamorphosed, and altered by silicification and carbonatization. Low-grade greenschist facies metamorphism predominates, but in places the rocks reach amphibolite grade. Alteration resulted in the development of listwaenite, particularly in shear zones, and locally the only evidence that mafic-ultramafic rocks underlie a given area is the presence of upstanding ridges of listwaenite that are resistant to erosion. Kinematic indicators in shear zones indicate that the ophiolites were affected by both strike-slip and vertical displacements. Variations in senses of shear observed along and across strike demonstrate considerable strain partitioning during deformation. However, prevailing senses of shear can be discerned for several of the ophiolites that, in conjunction with other structural

8. Synopsis of This Volume

17

observations, indicate the main shear trajectories of the shear zones containing the ophiolites. Jabal Ess, Jabal Tharwah, and Bi’r Umq ophiolites were emplaced during periods of dextral transpression on the Yanbu and Bi’r Umq sutures, respectively. The Bi’r Tuluhah ophiolite was emplaced during sinistral transpression of the Hulayfah-Ad Dafinah-Ruwah suture, and the Halaban ophiolite was emplaced during west-directed convergence on the Halaban suture. In a final paper on the Arabian-Nubian Shield, Ibrahim Hussein, Alfred Kröner, and Thomas Reischmann describe aspects of the 808 ± 14 Ma Wadi Onib mafic-ultramafic complex, located within the Onib-Sol Hamed suture in the northern Red Sea Hills of the Sudan, and relate these features to suprasubduction zone processes. The Wadi Onib ophiolite consists, from bottom to top, of a basal peridotite unit, an exceptionally thick (2–3 km) transitional zone of interlayered cumulates, isotropic gabbroic with plagiogranite bodies, a sheeted basic dike complex, and pillowed basaltic lavas containing fragmentary lenses of ribbon chert and/or graphitic to shaly carbonates. Whereas the basal unit is strongly serpentinized and/or carbonatized, the transitional zone comprises abundant and well preserved pyroxenites. The transition zone also shows a polycyclic cumulate arrangement that possibly originated from multiple magma pulses rather than from tectonic interslicing. Moreover, mineral grading, gravity stratification and a spectrum of folds with varying geometrical dispositions and amplitudes within discrete layers as well as a vertical metamorphic zonation (suggesting seafloor hydrothermal processes) are evident within the Onib ophiolitic sequence. In particular, the volcanic component is Ti-rich, has a transitional IAT/MORB character and is indistinguishable from anomalous MORB and/or marginal basin basalts. Thus, the Onib is envisaged to be of arc/back-arc (marginal) basin affiliation, and it classifies as a supra-subduction zone (SSZ) rather than normal MORB-type ophiolite. The ophiolitic sequence probably resulted from parental magma(s) generated through multi-stage partial fusion of mantle peridotite. In a paper summarizing the youngest ophiolite described in this volume, Joerg Pfänder and Alfred Kröner (2004) present field, geochronologic, and geochemical data on the tectono-magmatic evolution, age and emplacement of the 570 million year old Agardagh Tes-Chem ophiolite in Tuva, Central Asia. This ophiolite is located in the Palaeozoic Central Asian Mobile Belt which formed during subduction-accretion processes lasting from the early Neoproterozoic to the late Palaeozoic. The ophiolite was obducted onto the TuvaMongolian Massif (microcontinent?) in the early Palaeozoic towards the SE along N- and NW-dipping faults and is embedded within a tectonic mélange and thus is part of an accretionary wedge. Dating of three small zircon fractions from a plagiogranite by the evaporation technique yielded a 207 Pb/206Pb age of 569.6 ± 1.7 Ma, which reflects the crystallization age of the plutonic section of the ophiolite. Geochemical data reveal an island arcrelated origin for the ophiolite, where typical island arc volcanic rocks predominate over MORB-like pillow lavas. In contrast to the highly incompatible element-enriched volcanic rocks, all plutonic rocks of the ophiolite are depleted, and mineral compositions of ultramafic cumulates indicate the presence of boninitic parental melts. The ophiolite therefore consists of an association of island arc and back-arc related sequences that have been amalgamated during subduction-accretion and collisional obduction. Isotopic and trace element

18

Introduction

data reveal the existence of a depleted and refractory mantle source beneath Central Asia, from which the volcanic and plutonic rocks of the ophiolite were formed. However, source contamination took place by sediment subduction, before the parental melts of the island arc volcanic rocks were formed. 8.2. Archean Ophiolites and Related Rocks The second main section of the book presents descriptions of a number of well-exposed Archean mafic-ultramafic sequences that have been suggested to be possible ophiolites. Recognizing ophiolites in the Archean record has been more controversial than calling similar sequences in the Proterozoic record ophiolites, and many of the papers in this section discuss this bias of some workers against calling Archean ophiolite-like sequences ophiolites. In a series of four chapters on the Dongwanzi ophiolite, Timothy Kusky, Jianghai Li, and their co-workers describe various aspects of a remarkable complete but dismembered ophiolite sequence discovered in the North China craton in 2001. In the first overview paper, Timothy Kusky, Jianghai Li, Adam Glass, and Xiongnan Huang describe the general field characteristics of the 2.5 Ga Dongwanzi ophiolite, and discuss its regional tectonic setting. Banded iron formation structurally overlies several tens of meters of variably deformed pillow lavas and mafic flows. These are in structural contact with a 2 km thick mixed gabbro and sheeted dike complex with gabbro screens, exposed discontinuously along strike for more than 20 km. The dikes consist of metamorphosed diabase, basalt, hb-cpx-gabbro, and pyroxenite. Many have chilled margins developed on their NE sides, indicating one-way chilling. The dike/gabbro complex is underlain by several kilometers of mixed isotropic and foliated gabbro, which develop compositional layering approximately two kilometers below the sheeted dikes, and then over several hundred meters merge into strongly compositionally layered gabbro and olivine-gabbro. The layered gabbro becomes mixed with layered pyroxenite/gabbro marking a transition zone into cumulate ultramafic rocks including serpentinized dunite, pyroxenite and wehrlite, and finally into strongly deformed and serpentinized olivine and orthopyroxene-bearing ultramafic rocks interpreted as depleted mantle harzburgite tectonites. A U/Pb zircon age of 2.505 Ga from gabbro of the Dongwanzi ophiolite makes it the world’s oldest recognized, laterally-extensive complete ophiolite sequence. Characteristics of this remarkable ophiolite may provide the best constraints yet on the nature of the Archean oceanic crust and mantle, and offer insights to the style of Archean plate tectonics and global heat loss mechanisms. In a companion paper, Rachael Huson, Timothy Kusky, and Jianghai Li (2004) compare major and trace element concentrations of rocks from the Dongwanzi ophiolite with known concentrations from well-studied ophiolites and rocks from tectonic settings to determine tectonic environment of formation. Major element analysis shows samples are subalkalic (in particular, calc-alkaline) to alkalic. Trace element analysis shows enrichment of large ion lithophile elements as well as depletion of high field strength elements relative to mid-ocean ridge basalts. Calc-alkaline geochemical characteristics of oceanic rocks have predominantly been identified in suprasubduction zone settings and their occurrence in the

8. Synopsis of This Volume

19

Donwangzi ophiolite suggests a similar tectonic setting. Trace element signatures are also similar to suprasubduction zone ophiolites indicating formation above a subduction zone. The Dongwanzi ophiolite is but one of the largest well-preserved greenstone belts in the Central Orogenic belt that divides the North China craton into eastern and western blocks. More than 1,000 other fragments of gabbro, pillow lava, sheeted dikes, harzburgite, and podiform-chromite bearing dunite occur as tectonic blocks (tens to hundreds of meters long) in a biotite-gneiss and BIF matrix, intruded by tonalite and granodiorite, in the Zunhua structural belt. Blocks in this metamorphosed Archean ophiolitic mélange preserve deeper levels of oceanic mantle than the Dongwanzi ophiolite. The ophioliterelated mélange marks a suture zone across the North China Craton, traced for more than 1,600 km along the Central orogenic belt. Many of the chromitite bodies are localized in dunite envelopes within harzburgite tectonite, and have characteristic nodular and orbicular chromite textures, known elsewhere only from ophiolites. The chromites have variable but high chrome numbers (Cr/Cr + Al = 0.74–0.93) and elevated P, also characteristic of suprasubduction zone ophiolites. The high chrome numbers, coupled with TiO2 wt% < 0.2 and V2 O5 wt% < 0.1 indicate high degrees of partial melting from a very depleted mantle source and primitive melt for the chromite. As reported in the chapter by Kusky, Li, Raharimahefa, and Carlson (2004d), a Re-Os model age from the chromites indicates an age of 2547 ± 10 Ma, showing that they are the same age as the Dongwanzi ophiolite. The range in initial Os isotopic compositions in the chromites in these ophiolitic blocks is small and well within the range seen in modern ophiolites. The chondritic to sub-chondritic initial ratio also is interesting in that it shows more similarity to the values found for abyssal peridotites than OIB’s, pointing to an ocean-ridge rather than plume setting for the initial formation of these peridotites. The ultramafic and ophiolitic blocks in the Zunhua mélange are therefore interpreted as dismembered and strongly deformed parts of the Dongwanzi ophiolite. Xiongnan Huang, Jianghai Li, Timothy Kusky, and Zheng Chen describe a remarkably well-preserved suite of microstructures from the Zunhua podiform chromite, and discuss implications for the deformation and rheology of the Archean oceanic lithospheric mantle. The Zunhua podiform chromite preserves typical magmatic fabrics including nodular and orbicular textures, and magmatic flow structures. The magmatic textures indicate that the Zunhua podiform chromite was formed through five-stages of evolution, with the following time sequence: disseminated chromite, net-like veins, antinodular, orbicular and nodular textures. The evolution of the texture series can be interpreted to result from fast flowing magmatic flowing systems. They result from the vertical accretion of the oceanic mantle. The podiform chromite ores show strong deformation with development of pull-apart structures, banding, folds, and mylonitic foliation. These structures were formed at high temperature in the oceanic mantle during the oceanic ridge spreading as the ores were caught up by plastic flow and sheared transversely. The Zunhua podiform chromite bodies result from active magmatic accretion and strong high-temperature plastic flow, therefore a fast spreading oceanic ridge is suggested for its formation. Silicate mineral inclusions within the chromium spinel and geochemical characteristics of the Zunhua ophiolite support a geological setting in a suprasubduction belt.

20

Introduction

Several hundred kilometers to the southwest of the Dongwanzi ophiolite, Jianghai Li, Timothy Kusky, Niu Xianglong, and Feng Jun describe the textures and mineralogy of a Neoarchean massive sulfide deposit in the Wutai Mountains, recognizing a black smoker chimney and mound complex within 2.50 Ga old oceanic crust. The Wutai VMS is one of largest sediment-hosted sulfide deposits in China. It forms small lenses, thin sheets, and tabular bodies of massive to layered sulfide, disseminated through a forearc mélange belt. Although they are reworked by late deformation, sulfide deposits formed at different crustal levels still can be identified, including relicts of chimneys, pyritic siliceous exhalite, massive crystallized sulfides, talus of massive sulfides and stockwork zones. The country rock of the Wutai VMS ores show intense silicification and chloritization. Epidosites have been identified within mafic rocks. Under microscopic observations, porous sulfides show a mineralogical zonation around micro-conduits. The colloform textures developed delicate banding and concentric textures. The vuggy cavities are commonly lined by concentric layers consisting of idiomorphic pyrite and silica. The Wutai VMS are spatially associated with convergent plate boundaries, formed in the upper sequence of a former Neoarchean oceanic basin. They have been overthrust by foreland-thrust belts following closure of an oceanic basin. The preliminary studies reveal the presence of black smoker chimneys preserved in the Wutai Mountains, which suggest that seafloor black smoker activity at about 2.50 Ga plays an important role for generation and accumulation of Wutai VMS. In addition, the Wutai VMS is quite similar to Besshi-type deposits, it is inferred to be generated in the setting of forearc, later tectonically transported in mélange belts during continental collision. The possibility that some mafic greenstone belts in the Slave Craton of northern Canada and Nunavut may be ophiolites has been a contentious issue with debate spanning much of the late 1980’s, 1990’s, and early 2000’s. In the sixth chapter in this section, three authors (Patricia Corcoran, Wulff Mueller, and Timothy Kusky) with different views on this subject have co-written a paper that attempts to lay-out the most important observations, and reconcile possible interpretations with these observations. This chapter reviews three distinct areas in the Slave craton and assesses their potential of containing ophiolite sequences. These include the (1) Yellowknife, (2) Point Lake, and (3) Beaulieu and Cameron River volcanic belts. Since in all these case, the mafic-ultramafic sequences that have some ophiolitic characteristics rest structurally over continental crust, the best modern analogy for Slave Province ophiolites may be Tethyan-type ophiolites. The 6 km thick Chan Formation of the Yellowknife volcanic belt resembles modern ophiolites with tholeiitic massive to pillowed flows, abundant gabbro dikes and sills, interflow sedimentary rocks, and a mafic sheeted dike swarm. The base of this crustal-floored sequence is sheared and locally stitched by late-tectonic plutons and the dunite-peridotite-gabbro segment is lacking, so if it is an ophiolite, it only contains the upper parts of the sequence. The inferred base of the Point Lake volcanic belt is composed of mafic mylonite with low-strain domains of gabbro, pyroxenite, dunite, and peridotite. The mafic mylonite is overlain by gabbro, layered gabbro, minor mafic dikes, pillowed flows, massive flows, hyaloclastite, and local chert. A well-defined sheeted dike swarm is absent although mafic dikes are locally preserved in the crustal sequence, and other dikes cut underlying granite. The Beaulieu and

8. Synopsis of This Volume

21

Cameron River volcanic belts are spatially associated with mafic dike swarms that intrude the Sleepy Dragon basement complex. However, the currently juxtaposed dike swarm and mafic volcanic belt are not necessarily directly related, since they are everywhere separated by a major shear zone with significant displacements. Mafic massive and pillowed flows and sub-volcanic sills are predominant above the sheared basement contact. In the strict sense, these belts or belt segments do not fit the definition of a complete ophiolite, but do meet the general requirements in that they are allochthonous mafic sequences consisting of submarine volcanics and intrusives. Ophiolites form in numerous tectonic settings and complete preservation from tectonized mantle to surficial ocean floor products is highly unlikely, especially for Archean rocks. Therefore, the nature of the basement contacts is particularly significant. If the contacts are tectonic, then parts of the ophiolitic sequences may have been sheared off, which is commonly the case for the mafic-ultramafic intrusive component. Recent models have compared parts of certain Slave Province greenstone belts with supra-subduction zone settings including arcs and back-arcs, extensional settings such as mid-ocean ridges. Some of the ophiolitic Slave Province greenstone belts have characteristics that suggest they formed along ocean-continent transition zones, similar to the Jourma ophiolite of Finland and the Cretaceous west Iberian continental margin. One of the important points that Corcoran, Mueller, and Kusky develop is that identifying ophiolite sequences based solely on geochemistry is overly simplistic, and regional geological context, structure, and stratigraphy is required (e.g., Pearce, 1987; Wood et al., 1979). Ophiolites are generally considered a distinct suite of obducted ocean floor rocks with a highly varied geochemical affinity depending on tectonic setting (Sylvester et al., 1997; Dilek et al., 2000). As pointed out by Eldridge Moores (2002) the “ophiolite conundrum” marks a discrepancy between structural and stratigraphic setting, and geochemical characteristics. Moores argues that the mantle is heterogeneous at all scales and geodynamic settings, and that a distinct geochemical signature for ophiolites is lacking. This has ramifications especially for the Archean, in which volcano-sedimentary sequences are generally incomplete, structures are complex, Fe-tholeiites and komatiites are abundant (rare to non-existent in modern ophiolites), and mantle compositions and temperatures were possibly different. Identifying the tectonic setting of a dismembered mafic (-ultramafic) volcanic sequence thus becomes enigmatic. Must there be a complete stratigraphic sequence (i.e., dunite-peridotite, tectonite, gabbro, sheeted dykes, pillows, pelagic sedimentary rocks) to qualify a specific sequence as an ophiolite? What portion of the succession is necessary in order to be called an ophiolite? Is there a distinction between Phanerozoic and Archean ophiolites? Further north in the Aldan Shield in eastern Siberia, Igor Puchtel (2004) describes a 3.0 billion year old partial ophiolitic sequence from the Olondo greenstone belt. The Olondo greenstone belt is distinguished from the other greenstone belts in the Aldan Shield by an abundance and a great facies diversity of mafic-ultramafic rocks. The rocks are relatively well preserved both geologically and geochemically compared to other Archean ophiolitelike sequences worldwide, and thus can be regarded as valuable witnesses of the early history of the Earth. The Olondo greenstone belt contains one of the oldest ophiolite-like sequences on the planet. The age of the Olondo greenstone belt at 3.0 Ga is intermediate

22

Introduction

between the two most commonly cited periods of global crust-forming activity, namely, 2.7 and 3.4 Ga (Condie 1995, 1998). Thus, the study of this belt can help fill the gap in our understanding the significance of the tectonothermal and chemical evolution of the Earth during the time period between the early and late Archean. Andrey Shchipansky and others (2004) describe Neoarchean subduction-related assemblages of the North Karelian greenstone belt, in the northeast part of the Baltic Shield, Russia. This belt contains some of the world’s oldest known boninite series rocks, occurring in at least in two areas of the belt. The first area, referred to here as the Khizovaara structure, shows evidence of a late Archean ocean-island volcanic arc collage formed during two tectonic episodes nearly 2.8 Ga ago. The second area, named the Iringora structure, preserves distinctive features of an ophiolite pseudostratigraphy, including not only gabbro and lava units, but also remnants of a sheeted dike complex. The major and trace element chemistry of the Iringora ophiolitic gabbro, dike and lava units suggests a comagmatic series with a continuous compositional variation from more primitive mafic to strictly boninitic melts. In terms of major and trace element abundance, the boninite series of the North Karelian greenstone belt is practically indistinguishable from the Group I and II of the Troodos upper pillow lavas. These occurrences strongly suggest that Neoarchean subduction-related processes including boninite-hosting supra-subduction zone ophiolites have not changed substantially over the past 2.8 Ga. The Belingwe belt in Zimbabwe is probably the best-known well-preserved late Archaean greenstone belt in the world. Despite the presence of well exposed rocks of very low metamorphic grade and low strain, the tectonic evolution of the Belingwe belt has been a matter of much controversy, with debates focusing on whether the parts of the greenstone succession are oceanic in nature, or whether they were erupted through underlying continental crust. Axel Hoffman and Tim Kusky (2004) present a synthesis of the geology of the Belingwe belt, and assess various tectonic models for the belt’s origin. The Belingwe greenstone belt comprises two distinct greenstone successions. The lower, 2.9–2.8 Ga old Mtshingwe Group consists of four stratigraphic units, an intermediate to felsic volcanic and volcanoclastic unit, an ultramafic to mafic lava plain sequence, a conglomerate-shale sedimentary sequence, and a unit of tectonically imbricated sedimentary and volcanic rocks. Although geochronological, geochemical and lithological characteristics are broadly known, the tectonic evolution of the Mtshingwe Group remains a matter of speculation. Controversy surrounds the intensely studied, 2.7 Ga old Ngezi Group, which consists of a thin basal sedimentary sequence, a thick ultramafic to mafic volcanic sequence, and an upper sedimentary succession. The basal unit rests unconformably on up to 3.5 Ga old granitoid gneisses and Mtshingwe Group rocks, consists of fluvial to shallow-marine sedimentary rocks, and is similar to cratonic cover successions. The structurally overlying volcanic unit is a submarine lava plain sequence of massive and pillow basalts with komatiites near the base and andesites near the top. The upper sedimentary unit represents a foreland basin sequence and consists of karstified carbonate ramp limestones overlain by deeper-water turbidite deposits. Autochthonous versus allochthonous models have been proposed for the tectonic evolution of the Ngezi Group. Proponents of the autochthonous model regard the Ngezi Group as a conformable sequence that formed in an ensialic rift

8. Synopsis of This Volume

23

setting above a mantle plume. Other workers regard the volcanic sequence as an allochthonous unit of oceanic crust that was obducted onto continental basement. A great number of arguments have been proposed in recent years from structural, sedimentological, and geochemical studies for and against the different models. A critical reappraisal of the various arguments indicate the lack of convincing evidence for an ensialic and autochthonous origin. Arguments for an allochthonous origin are strong, whereas an oceanic origin can only be inferred by assuming that modernistic plate tectonic processes were operating in the late Archean. 8.3. Models for the Evolution of Oceanic Crust with Time Chapters in the third section of the book focus on melting and petrological processes in the Archean mantle, sub-ridge, and sub-arc environments, and on models for the evolution of oceanic crust with time. Authors of papers in this section have synthesized data from several different belts, and place important constraints on the nature of the Archean mantle. Archean greenstone belts are known for their hallmark deposits of komatiites, magnesium rich lavas that many petrologists have suggested indicate significantly higher temperatures for the Archean mantle. These estimated temperatures were in turn used by many workers to derive unusual non-uniformitarian and non-actualistic models for tectonics on the early Earth. Steve Parman and Tim Grove show field and petrologic data from the Barberton Greenstone belt that suggests an alternative interpretation, that the Archean mantle may not have been so different from that of today. The paper by Parman and Grove is therefore very significant in that it removes any reason for assuming that plate tectonics should have been drastically different from today. The Barberton Greenstone Belt is one of several mid- to late-Archean greenstone belts that lie along the eastern margin of the Kaapvaal craton (Brandl and de Wit, 1997). With an age of 3.49–3.46 Ga (Lopezmartinez et al., 1992), the BGB is among the oldest of the Kaapvaal Craton’s greenstone belts and is part of the nucleus around which the Late Archean greenstone belts to the north (e.g., Murchison and Giyani) and to the south (e.g., Nondweni and Commondale) were attached. Parman and Groves focus their discussion on the komatiites and related basaltic komatiites from the Komati and Hoogenoeg formations. These two formations form a continuous stratigraphic section and have been the main focus of their research, though reference is also made to komatiites in the Barberton Greenstone Belt’s smaller and less well preserved komatiite-bearing Sandspruit, Theespruit, Mendon and Weltevreden sequences. In the end Parman and Groves put the Barberton data in the context of the global komatiite data set, showing that komatiites do not require exceptionally high mantle temperatures to form. Ali Polat and Robert Kerrich (2004) synthesize data on known occurrences of boninites, adakites, magnesian andesites, and Nb-enriched basalts, and related these to Precambrian arc associations. Boninitic lavas have recently been reported from several Precambrian terranes, including the ∼ 3.8 Ga Isua terrane of West Greenland; 2.8 Ga Opatica and the 2.7 Ga Abitibi terranes of the Superior Province; the 2.8 Ga North Karelian terrane of the Baltic Shield; and the 1.9 Ga Flin Flon terrane in the Trans-Hudson orogen. In the Isua belt, boninitic flows coexist with pillow basalts and picrites. Boninitic lavas, and low-Ti

24

Introduction

tholeiitic basalts, outcrop over a 300 km corridor in the Abitibi volcanic-plutonic subprovince. They are intercalated with a stratigraphically lower ocean plateau association of komatiites and basalts, and an upper volcanic arc association of tholeiitic to calc-alkaline arc basalts; accordingly there was contemporaneous eruption of neighboring plume and arc magmas. The 2.8 Ga Opatica boninitic lavas are spatially and temporally associated with arc-type volcanic rocks. The 2.8 Ga Baltic Shield boninitic rocks are related to a supra-subduction ophiolite complex. All of these Precambrian boninitic lavas share the low-TiO2, high Al2 O3 /TiO2 ratios, U-shaped REE patterns, and negative Nb but positive Zr anomalies of Phanerozoic counterparts; however, SiO2 contents are variable. Boninites of Phanerozoic age occur in ophiolites or intra-oceanic island arcs, such as the Izu-Bonin-Mariana arc system. These primary liquids are interpreted as second-stage high-temperature, low-pressure melting of a depleted refractory mantle wedge fertilized by fluids and/or melts, above a subduction zone. Precambrian boninitic lavas are likely products of the same conjunction of processes. Low-Ti tholeiites lack the LREE enrichment coupled with negative Nb anomalies of the boninites. They had a similar depleted wedge source, but without a subduction zone component. An association of adakites, magnesian andesites (MA), and Nb-enriched basalts (NEB) with “normal” tholeiitic to calc-alkaline basalts and andesites has recently been described from the 2.7 Ga Wawa and Confederation volcanic-plutonic terranes of the Superior Province. Cenozoic adakites are considered to form by slab melting; MA the product of hybridization of adakite liquids with the peridotitic mantle wedge; and NEB melting of the residue of the MA wedge source. This volcanic association is found in Cenozoic arcs characterized by shallow subduction of young, hot oceanic lithosphere. Archean equivalents likely formed under comparable tectonic settings. U-shaped REE patterns in conjunction with positive Zr anomalies of Archean and Phanerozoic boninites can be modeled by a depleted peridotitic wedge fertilized by adakite liquids and/or hydrous fluids in a convergent margin. Consequently, Phanerozoic type arcs were operating in Archean convergent margins. Imbrication of komatiite-basalt ocean plateau volcanic sequences with arcs solves the apparent Mg#, Ni deficit of some models for Archean upper continental crust. Higher geothermal gradients in Archean subduction zones may have played an important role for the growth of continental crust. In a final, concluding chapter for the this section, Maarten de Wit synthesizes data from Archean greenstone belts around the world and concludes that “Archean greenstone belts do contain fragments of ophiolites” (a direct pun on an earlier, and flawed paper by Bickle et al., 1994). Maarten de Wit notes that most Archean greenstone belts are so severely tectonized so that reconstruction of their rock assemblages revealing original autochthonous relationships is a daunting task (de Wit and Ashwal, 1997; Kusky and Vearncombe, 1997). There are about 260 individual Archean greenstone belts worldwide. Few of these have been studied in sufficient detail to provide relatively reliable information about pre-2.5 Ga geological processes (de Wit and Ashwal, 1997). Greenstone belts represent some of the earliest records of Earth history, but they are not restricted to the Archean. For example, the large Neoproterozoic Arabian-Nubian shield has an Archean-like cratonic crust with at least 7 major greenstone belts, most of which

8. Synopsis of This Volume

25

comprise island arc-like successions and associated (but often dismembered) ophioliteassemblages (Berhe, 1997; see Stern et al., 2004; Johnson et al., 2004; and Hussein et al., this volume). Similarly, the Baltic shield contains greenstone belt sequences ranging in age from > 3.1 Ga (Mesoarchean) to 1.9 Ga (Mesoproterozoic). Some of the Mesoproterozoic greenstone belts share characteristics of many Archean greenstone belts (e.g., abundant komatiites), whilst others share characteristics of Phanerozoic ophiolites (Sorjonen-Ward et al., 1997, and this volume). A wide spectrum of tectonic environments is preserved within Archean greenstone belts, and many individual belts are mixtures of components from different tectonic environments and in particular from island arc terrains (de Wit and Ashwal, 1997; Kusky and Vearncombe, 1997). It is claimed nevertheless by some that oceanic crustforming environments are not preserved amongst this mixture of tectonic regimes because in their views no rocks assemblages in Archean greenstone belt sequences exhibit sufficient features to warrant definitive classification as an ophiolite (Bickle et al., 1994; Hamilton, 1998, 2003). The difficulty in recognizing and even defining ophiolites has been acknowledged widely and is not addressed here (Anonymous, 1972; and this volume). In his short contribution de Wit outlines some probable and some possible ophiolite sequences that have been reported from a number of Archean greenstone belts around the world. He also comments on the likely tectonic implications of these examples to better resolve Archean processes. 8.4. Analogs to Precambrian Ophiolites The final section of the book encompasses three important Phanerozoic analogs to processes that are currently producing rocks assemblages and structures that resemble Archean ophiolitic greenstone belts. John Encarnación (2004) describes the northern Philippines as a possible modern analogue for some Precambrian greenstone belts. It has a ∼ 150 Myr history of multiple and overlapping periods of oceanic crust generation, arc volcanism, sedimentation, and deformation dominated by wrench tectonics. At least five ophiolite complexes of distinct age make up most of the basement—all having a distinct suprasubduction zone signature, relationships reminiscent of the Yellowknife Belt in the Slave Province. Arc plutons are predominantly of the diorite-tonalite series with minor alkali-feldspar bearing rocks. Sedimentary basins probably floored by oceanic crust are dominated by immature sediments and volcaniclastics and are locally up to ∼ 10 km thick. The whole arc and ophiolitic complex is in the process of being accreted to Eurasia, where it may be preserved in a broad “suture zone” between Eurasia and Australia and/or the Americas. Southern Alaska’s Mesozoic-Cenozoic Chugach-Prince William terrane is an unusual forearc in that it contains belts of graywacke-dominated flysch, mélange, and ophiolitic fragments all intruded by a suite of tonalite-trondhjemite-granodiorite plutons, and large parts of the accretionary prism are metamorphosed to the greenschist, amphibolite, or granulite facies. In the second chapter of this section, Tim Kusky, Rose Ganley, Jennifer Lytwyn, and Ali Polat (2004c) describe the overall structural geometry, abundance and

26

Introduction

types of rocks and rock suites present, the petrogenetic relationships between rock suites, and the metamorphic style are all strongly reminiscent of Archean granite-greenstone terranes. As such, the southern Alaska forearc represents one of the world’s best modern analogs to early stages in the evolution of Archean granite-greenstone terranes. Belts of flysch, mélange and accreted ophiolites are described, and particular attention is paid to details of the geology of the 57 ± 1 Ma Resurrection Peninsula ophiolite as a remarkable analog to some Archean greenstone belts. The Resurrection ophiolite formed in a neartrench environment as the Kula-Farallon ridge was being subducted beneath North America. The magmatic sequence includes pillow lavas, sheeted dikes, gabbros, trondhjemites, and a poorly-exposed ultramafic section. The lavas show mid-ocean ridge basalt and arclike geochemical signatures, interpreted to reflect compositionally diverse melts derived from near-fractional melting of a variably depleted mantle source, mixed with variable amounts of assimilated continentally-derived flysch. A sedimentary sequence overlying the ophiolite preserves a continuous record of turbidite sedimentation deposited on the ophiolite as it was transported to North America and emplaced in the Chugach accretionary prism. The top of the sedimentary section is truncated by the Fox Island shear zone, a 1 km thick, greenschist-facies, west-over-east thrust related to the emplacement of the ophiolite into the accretionary wedge. The Fox Island shear zone is intruded by a 53.4 ± 0.9 Ma granite, showing that the ophiolite formed, was transported to the North American continent, overthrust by a major accretionary prism-related thrust, and intruded by granite all within 3.6 ± 1.4 Ma. Geological relationships in the southern Alaska forearc are instructive, in that if similar relationships were found in an Archean granite-greenstone terrane, they would probably currently be interpreted to reflect calc-alkaline mafic-felsic volcanic-plutonic complexes intruded and erupted through a complex metasedimentary sequence. Many Precambrian forearc ophiolites and accretionary prisms may have gone unrecognized because the processes of forearc ophiolite emplacement and intrusion by near-trench magmas at triple junctions has been poorly documented. In the final chapter of the book, A.M. Celail Sengör and Boris Natal’in synthesize Phanerozoic analogues of Archean oceanic basement fragments in the Altaids, distinguishing between nearly complete ophiolites, and severely dismembered ophiolitic bodies they term ophiorags. Sengör and Natal’in note that in the minds of most geologists, orogenic belts are linear/arcuate, long and narrow zones of intense deformation. That is why, irregularly shaped areas of widespread “orogenic deformation” interspersed with abundant fragments of the members of the ophiolite association in the Precambrian, but especially in the Archaean, have been thought as products of processes no longer operative. However, the geology of the Altaid orogenic system in Asia greatly resembles in its overall map aspects, lithological content, structural characteristics, and in the distribution and types of fragments of floors of former oceans to the Archaean granite-greenstone terrains. In the Altaids, ophiolites are now encountered in three main settings: (1) Ophiolites that occur as basement of ensimatic arcs, (2) ophiolites and ophirags that occur in former forearcs now entrapped within transform sutures: (a) Ophiolites as backstop to accretionary wedges, (b) Ophirags within accretionary wedges, and (3) Ophiolites and ophirags in collisional

References

27

suture zones that have usually evolved from members of the second category. Ophiolites and ophirags have a widespread distribution within the orogenic edifice. This distribution was brought about by processes that shaped the Altaid edifice, namely, generation of suprasubduction zone forearc basements created by pre-arc spreading, back-arc basin opening, subduction-accretion, trench-linked strike-slip faulting including arc slicing and arc shaving faults and associated ocean floor spreading processes, and collision of buoyant pieces along suture zones. Without appreciating the nature and sequence of these processes and their superimposition, it is impossible to understand the rules that govern the distribution of oceanic basement fragments in the Altaids. These processes have led to a tremendous degree of structural shuffling of previously distant environments and a large degree of dismembering of formerly more complete ocean floor fragments. The preservation is highly selective and favors upstanding and buoyant segments of ocean floors. Such pieces are embedded most commonly in metapelitic/metapsammitic or, more rarely, in serpentinitic matrices in mélange/wildflysch complexes. Sengör and Natal’in contend that the same rules apply to the greenstone belts of the Precambrian and greatly hinder their deciphering in the absence of biostratigraphic control.

REFERENCES Abbott, D., 1991. The case for accretion of the tectosphere by buoyant subduction. Geophysical Research Letters April, 1–4. Abbott, D., 1996. Plumes and hotspots as sources of greenstone belts. Lithos 37, 113–127. Abbott, D.H., Drury, R., Mooney, W.D., 1997. Continents as lithological icebergs: the importance of buoyant lithospheric roots. Earth Planet. Sci. Lett. 149, 15–27. Abbott, D.H., Mencke, W.H., 1990. The length of the global plate boundary at 2.4 Ga. Geology 18, 58–61. Abbott, D., Mooney, W., 1995. The structural and geochemical evolution of the continental crust: support for the oceanic plateau model of continental growth. Rev. Geophys. Suppl. Uni. Geodesy Geophys. 1991–94, 231–242. Abouchami, W., Boher, M., Michard, A., Albarede, F., 1990. A major 2.1 Ga event of mafic magmatism in west Africa: An early stage of crustal accretion. J. Geophys. Res. 95, 17605–17629. Anonymous, 1972. Ophiolites. Geotimes 17, 24–25. Armstrong, R.L., 1991. The persistent myth of crustal growth. Austral. J. Earth Sci. 38, 613–630. Arndt, N.T., Albarede, F., Nisbet, E.G., 1997. Mafic and ultramafic magmatism. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 233–254. Artemieva, I.M., Mooney, W.D., 2001. Thermal thickness and evolution of Precambrian lithosphere—A global study. Journal of Geophysical Research 106, 16,387–16,414. Ben-Avraham, Z., Nur, A., Jones, D., Cox, A., 1981. Continental accretion: from oceanic plateaus to allochtonous terranes. Science 213, 47–54. Berhe, S.M., 1990. Ophiolites in north and east Africa: implications for Proterozoic crustal growth. J. Geol. Soc. London 147, 41–57. Berhe, S.M., 1997. The Arabian-Nubian Shield. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 761–771.

28

Introduction

Bickle, M.J., Nisbet, E.G., Martin, 1994. Archean greenstone belts are not oceanic crust. J. Geology 102, 121–138. Boher, M., Abouchami, W., Michard, A., Albarede, A., Arndt, N.T., 1992. Crustal growth in west Africa at 2.1 Ga. J. Geophys. Res. 97, 345–369. Bradley, D.C., Haeussler, P.J., Kusky, T.M., 1994. Timing of Early Tertiary ridge subduction in southern Alaska. In: Till, A. (Ed.), Geologic Studies in Alaska by the U.S. Geological Survey during 1992. U.S. Geol. Survey Bull. 2068, 163–177. Bradley, D.C., Kusky, T.M., Haeussler, P., Rowley, D.C., Goldfarb, R., Nelson, S., 2003. Geologic signature of early ridge subduction in the accretionary wedge, forearc basin, and magmatic arc of south-central Alaska. In: Sisson, V.B., Roeske, S., Pavlis, T.L. (Eds.), Geology of a Transpressional Orogen Developed During a Ridge-Trench Interaction Along the North Pacific Margin. Geological Society of America Special Paper 371, 19–50. Brandl, G., de Wit, M.J., 1997. The Kaapvaal Craton. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 581–607. Burke, K., 1995. Did plate tectonics operate in the Archean? In: Precambrian 95, Program and Abstracts, p. 277. Burke, K., Dewey, J.F., Kidd, W.S.F., 1976. Dominance of horizontal movements, arc and microcontinental collisions during the later permobile regime. In: Windely, B.F. (Ed.), The Early History of the Earth. Wiley, pp. 113–129. Card, K.D., 1990. A review of the Superior Province of the Canadian Shield, a product of Archean accretion. Precambrian Res. 48, 99–156. Carlson, R.W., Boyd, F.R., Shirey, S.B., Janney, P.E., Grove, T.L., Bowring, S.A., Schmitz, M.D., Bell, D.R., Gurney, J.J., Richardson, S.H., Pearson, D.G., Hart, R.J., Wilson, A.H., Moser, D., 2000. Continental Growth, Preservation, and Modification in Southern Africa. GSA Today 10, 1–7. Casey, J.F., Dewey, J.F., Fox, P.J., Karson, J.A., Rosencrantz, E., 1981. Heterogeneous nature of oceanic crust and upper mantle: A perspective form the Bay of Islands ophiolite complex. In: Emiliani, C. (Ed.), The Oceanic Lithosphere. In: The Sea, vol. VII. Wiley, New York, pp. 305– 338. Condie, K.C., 1981. Archean Greenstone Belts. In: Developments in Precambrian Geology, vol. 3. Elsevier, New York, p. 434. Condie, K.C., 1994. Greenstones through time. In: Condie, K.C. (Ed.), Archean Crustal Evolution. Elsevier, Amsterdam, pp. 85–120. Condie, K.C., 1995. Episodic ages of greenstones; a key to mantle dynamics? Geophysical Research Letters 22, 2215–2218. Condie, K.C., 1997a. Contrasting sources for upper and lower continental crust: the Greenstone Connection. J. Geology 105, 729–736. Condie, K.C., 1997b. Plate Tectonics and Crustal Evolution. Butterworth–Heinemann, Bath. Condie, K.C., 1998. Episodic continental growth and supercontinents; a mantle avalanche connection? Earth and Planetary Science Letters 163, 97–108. Corcoran, P.L., Mueller, W.U., Kusky, T.M., 2004. Inferred Ophiolites in the Archean Slave Province. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 363–404. Dann, J.C., 1991. Early Proterozoic ophiolite, central Arizona. Geology 19, 590–593. Dann, J.C., 2004. The 1.73 Ga Payson Ophiolite, Arizona, USA. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 73–93.

References

29

de Wit, M.J., 2004. Archean Greenstone Belts do contain fragments of ophiolites. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 599–613. de Wit, M.J., Ashwal, L.D. (Eds.), 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, p. 809. de Wit, M.J., Hart, R.A., Martin, A., Abbott, P., 1982. Archean abiogenic and probable biogenic structures associated with mineralized hydrothermal vent systems and regional metasomatism, with implications for greenstone belt studies. Econ. Geol. 77, 1783–1802. de Wit, M.J., Hart, R.A., Hart, R.J., 1987. The Jamestown ophiolite complex, Barberton mountain belt: a section through 3.5 Ga oceanic crust. J. African Earth Sci. 6, 681–730. de Wit, M.J., Tredoux, M., Hart, R.A., 1990. EOS 71, 969–970. de Wit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., de Ronde, C.E.J., Green, R.W.E., Tredoux, M., Peberdy, E., Hart, R.A., 1992. Formation of an Archean continent. Nature 357, 553–562. Dewey, J.J., 1977. Suture zone complexities: A review. Tectonophysics 40, 53–67. Dewey, J.F., Bird, J., 1971. Origin and emplacement of the ophiolite suite: Appalachian ophiolites in Newfoundland. In: Bird, J.M. (Ed.), Plate Tectonics. J. Geophys. Res. 76, 3179–3206. Dewey, J.F., Windley, B.F., 1981. Growth and differentiation of the continental crust. Philos. Trans. Roy. Soc. London A 301, 189–206. Dick, H.J.B., Lin, J., Schouten, H., 2003. An ultraslow-spreading class of oceanic ridge. Nature 426, 405–412. Dilek, Y., Thy, P., Moores, E.M., 1991. Episodic dike intrusions in the northwestern Sierra Nevada, California—implications for multistage evolution of a Jurassic arc terrane. Geology 19, 180–184. Dilek, Y., Moores, E., Elthon, D., Nicolas, A., 2000. Ophiolites and Oceanic Crust. Geological Society of America Special Paper 349, 552. Dilek, Y., Newcomb, S., 2003. Ophiolite concept and the evolution of geological thought. Geological Society of America, Companion to Special Paper 349, 470. Encarnación, J.P., 2004. Northern Philippines ophiolites: Modern analogues to Precambrian ophiolites? In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 615–626. Friend, C.R.L., Nutman, A.P., McGregor, V.R., 1988. Late Archean terrane accretion in the Godthab region, southern west Greenland. Nature 335, 535–538. Fripp, R.E.P., Jones, M.G., 1997. Sheeted intrusions and peridotite-gabbro assemblages in the Yilgarn Craton, western Australia: elements of Archean ophiolites. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 422–437. Fyfe, W.S., 1978. Evolution of the Earth’s crust: modern plate tectonics to ancient hotspot tectonics. Chem. Geol. 23, 89–114. Gass, I.G., Lippard, S.J., Shelton, A.W., 1984. Ophiolites and Oceanic Lithosphere. In: The Geological Society Special Publications, vol. 13. Blackwell Scientific, Oxford, p. 413. Grove, T.L., Gaetani, G.J., de Wit, M.J., 1994. Spinifex textures in 3.49 Ga Barberton Mountain belt komatiites: evidence for crystallization pf water-bearing, cool magmas in the Archean. EOS Transactions, American Geophys. Union 75 (16), 354. Hamilton, W.B., 1998. Archean tectonics and magmatism. International Geology Reviews 40, 1–39. Hamilton, W.B., 2003. An alternative Earth. GSA Today 13, 4–13. Harper, G.D., 1985. A dismembered Archean ophiolite, Wind River Mountains, Wyoming (USA). Ophioliti 10, 297–306. Harper, G.D., 1999. Structural styles of hydrothermal discharge in ophiolite/sea floor systems. Reviews in Economic Geology 8, 53–73.

30

Introduction

Helmstaedt, H., Shulze, D.J., 1989. Southern African kimberlites and their mantle sample: implications for Archean tectonics and lithosphere evolution. In: Kimberlites and Related Rocks. Blackwell, Carleton, pp. 358–368. Helmstaedt, H., Padgham, W.A., Brophy, J.A., 1986. Multiple dikes in the lower Kam Group, Yellowknife greenstone belt: Evidence for sea floor spreading? Geology 14, 562–566. Hofmann, A.W., 1988. Chemical differentiation of the Earth: the relationships between, mantle, continental crust, and oceanic crust. Earth Planet. Sci. Lett. 90, 297–314. Hofmann, A., Kusky, T.M., 2004. The Belingwe greenstone belt: ensialic or oceanic? In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 487–537. Hoffman, P.F., Ranalli, G., 1988. Archean oceanic flake tectonics. Geophys. Res. Lett. 15, 1077– 1080. Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithospheric mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337. Huson, R., Kusky, T.M., Li, J.H., 2004. Geochemical and petrographic characteristics of the central belt of the Archean Dongwanzi ophiolite complex. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 283–320. Hussein, I., Kröner, A., Reischmann, T., this volume. The Wadi Onib mafic-ultramafic complex, a Neoproterozoic suprasubduction zone ophiolite in the northern Red Sea Hills, Sudan. Isachsen, C.I., Bowring, S.A., 1997. The Bell Lake Group and Anton Complex—a basement cover sequence beneath the Yellowknife greenstone belt revealed and implicated in greenstone belt formation. Can. J. Earth Sci. 34, 169–189. Isachsen, C.I., Bowring, S.A., Padgham, W.A., 1991. U-Pb zircon geochronology of the Yellowknife volcanic belt, N.W.T.: New constraints on the timing and duration of greenstone belt magmatism. J. Geol. 99, 55–67. Isozaki, Y., Maruyama, S., Furuoka, F., 1990. Accreted oceanic materials in Japan. Tectonophysics 181, 179–205. Isozaki, Y., Endo, H., Ohta, I., Suzuki, S., Maruyama, S., Kimura, G., 1992. Middle Archean (3.3 Ga) Cleaverville accretionary complex in northwestern Pilbara block W. Australia. In: Maruyama, S. (Ed.), Evolving Earth Symposium, Okazaki, Abstracts, pp. 128–131. Jacob, D., Jagoutz, E., Lowry, D., Mattey, D., Kudrjavtseva, G., 1994. Diamoniferous eclogites from Siberia: Remnants of Archean oceanic crust. Geochimica et Cosmochimica Acta 58, 5191–5207. Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004. Neoproterozoic ophiolites in the Arabian Shield: Field relations and structure. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162. Jagoutz, E., Dawson, J.B., Hoernes, S., Spettel, B., Wanke, H., 1984. Anorthositic oceanic crust in the Archean Earth. 15th Lunar and Planet Sci. Conf., 395–396. Karson, J.A., 2001. Oceanic crust when Earth was young. Science 292, 1076–1078. Kimura, G., Ludden, J.N., Desrochers, J.-P., Hori, R., 1993. A model of ocean crust accretion for the Superior province, Canada. Lithos 30, 337–355. King, J.E., Helmstaedt, H., 1997. The Slave Province, North-West Territories, Canada. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 459–479.

References

31

Kröner, A., 1985. Ophiolites and the evolution of tectonic boundaries in the late Proterozoic ArabianNubian Shield of north east Africa and Arabia. Precambrian Res. 27, 277–300. Kusky, T.M., 1987. Comment on “Multiple dikes in the lower Kam Group, Yellowknife Greenstone Belt: Evidence for Archean sea-floor spreading?”. Geology 15, 280–282. Kusky, T.M., 1989. Accretion of the Archean Slave Province. Geology 17, 63–67. Kusky, T.M., 1990. Evidence for Archean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10, 820–841. Kusky, T.M., 1993. Collapse of Archean orogens and the generation of late- to post-kinematic granitoids. Geology 21, 925–928. Kusky, T.M., 1998. Tectonic setting and terrane accretion of the Archean Zimbabwe craton. Geology 26, 163–166. Kusky, T.M., 2004. The Southern Alaska Accretionary Orogen as an Analog for Precambrian TrenchForearc Systems. In: Asia-Oceanic Geological Society Joint Meeting, Singapore, July 5–9, Abstract Volume SE 14. Kusky, T.M., Bradley, D.C., Haeussler, P., Karl, S., 1997a. Controls on accretion of flysch and mélange belts at convergent margins: Evidence from The Chugach Bay thrust and Iceworm mélange, Chugach Terrane, Alaska. Tectonics 16, 855–878. Kusky, T.M., Bradley, D.C., Haeussler, P., 1997b. Progressive deformation of the Chugach accretionary complex, Alaska, during a Paleogene ridge-trench encounter. J. Str. Geol. 19, 139–157. Kusky, T.M., Bradley, D.C., Ganley, R., 2004a. Melange. Ophiolite, and Flysch Belts of the Southern Alaska Accretionary Orogen as an Analog for Trench-Forearc Systems in Precambrian Orogens. In: International Workshop on Accretionary Orogens and Continental Growth, Taipei, Taiwan, May 18–23, Proceedings Volume. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004b. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Kusky, T.M., Ganley, R., Lytwyn, J., Polat, A., 2004c. The Resurrection Pininsula ophiolite, mélange, and accreted flysch belts of southern Alaska as an analog for trench-forearc systems in Precambrian orogens. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 627–674. Kusky, T.M., Li, J.H., Raharimahefa, T., Carlson, R.W., 2004d. Re-Os isotope chemistry and geochronology of chromite from mantle podiform chromites from the Zunhua ophiolitic mélange belt, N. China: Correlation with the Dongwanzi ophiolite. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 275–282. Kusky, T.M., Abdelsalam, M., Stern, R., Tucker, R., 2003a. Preface to Special Issue Precambrian Research on The East African and Related Orogens, and the Assembly of Gondwana. Precambrian Research 123 (2–4), 81–85. Kusky, T.M., Bradley, D.C., Donley, D.T., Rowley, D., Haeussler, P., 2003b. Controls on intrusion of near-trench magmas of the Sanak-Baranof belt, Alaska, during Paleogene ridge subduction, and consequences for forearc evolution. In: Sisson, V.B., Roeske, S., Pavlis, T.L. (Eds.), Geology of a Transpressional Orogen Developed During a Ridge-Trench Interaction Along the North Pacific Margin. Geological Society of America Special Paper 371, 269–292.

32

Introduction

Kusky, T.M., Kidd, W.S.F., 1992. Remnants of an Archean oceanic plateau, Belingwe greenstone belt, Zimbabwe. Geology 20, 43–46. Kusky, T.M., Vearncombe, J.R., 1997. Structural aspects. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 91–124. Kusky, T.M., Polat, A., 1999. Growth of Granite-Greenstone Terranes at Convergent Margins and Stabilization of Archean Cratons. In: Marshak, S., van der Pluijm, B. (Eds.), Special Issue on Tectonics of Continental Interiors. Tectonophysics 305, 43–73. Kusky, T., Young, C., 1999. Emplacement of the Resurrection Peninsula ophiolite in the southern Alaska Forearc During a Ridge-Trench Encounter. Journal of Geophysical Research 104 (B12), 29,025–29,054. Li, J.H., Kusky, T.M., Niu, X.L., Feng, J., Polat, A., 2004. Neoarchean massive sulfide of Wutai Mountain, North China: A black smoker chimney and mound complex within 2.50 Ga-old oceanic crust. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 339–361. Lippard, S.J., Shelton, A.W., Gass, I.G., 1986. The Ophiolite of Northern Oman. In: The Geological Society Memoirs, vol. 11. Blackwell Scientific, Oxford, p. 178. Lopezmartinez, M., York, D., Hanes, J.A., 1992. A Ar40 /Ar39 Geochronological Study of Komatiites and Komatiitic Basalts from the Lower Onverwacht Volcanics-Barberton Mountain Land, SouthAfrica. Precambrian Research 57 (1–2), 91–119. Lytwyn, J.N., Casey, J.F., Gilbert, S., Kusky, T.M., 1997. Arc-like mid-ocean ridge basalt formed seaward of a trench-forearc system just prior to ridge subduction: An example from subaccreted ophiolites in southern Alaska. Journal of Geophysical Research 102 (B5), 10,225–10,243. MacGregor, I.D., Manton, W.I., 1986. Roberts Victor eclogites: ancient oceanic crust. J. Geophys. Res. 91, 14063–14079. MacLaughlin, K., Helmstaedt, H., 1995. Geology and geochemistry of an Archean mafic dike complex in the Chan Formation: basis of a revised plate tectonic model of the Yellowknife greenstone belt. Can. J. Earth Sci. 32, 614–630. Martin, H., 1986. Effect of steeper Archean geothermal gradient on geochemistry of subduction zone magmas. Geology 14, 753–756. Martin, H., 1993. The mechanisms of petrogenesis of the Archean continental crust-comparison with modern processes. Lithos 30, 373–388. McCulloch, M.T., Bennett, V.C., 1994. Progressive growth of the Earth’s continental crust and depleted mantle: Geochemical Constraints. Geochim. Cosmochim. Acta 58, 4717–4738. McKenzie, D., Bickle, M.J., 1988. The volume and composition of melts generated by extension of lithosphere. J. Petrol. 29, 624–679. Moores, E.M., 1982. Origin and emplacement of ophiolites. Rev. Geophys. 20, 735–750. Moores, E., 1986. The Proterozoic ophiolite problem, continental emergence, and the Venus connection. Science 234, 65–68. Moores, E.M., 2002. Pre-1 Ga (Pre-Rodinian) ophiolites: Their tectonic and environmental implications. Geol. Soc. Amer. Bull. 114, 80–95. Nicolas, A., 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, p. 367. Nisbet, E.G., Cheadle, M.J., Arndt, N.T., Bickle, M.J., 1993. Constraining the potential temperature of the Archean mantle: A review of the evidence from komatiites. Lithos 30, 291–307. Parman, S.W., Dann, J.C., Grove, T.L., de Wit, M.J., 1997. Emplacement conditions of komatiite magmas from the 3.49 Ga Komati formation, Barberton greenstone belt, south Africa. Earth Planet. Sci. Lett. 150, 303–323.

References

33

Parman, S.W., Grove, T.L., 2004. Petrology and geochemistry of Barberton komatiites and basaltic komatiites: Evidence of Archean fore-arc magmatism. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 539–565. Parson, L.M., Murton, B.J., Browning, P. (Eds.), 1992. Ophiolites and Their Modern Oceanic Analogs. Geological Society Special Publication 60, 330. Pearce, J., 1987. An expert system for the tectonic characterization of ancient volcanic rocks. J. Volcanol. Geotherm. Res. 32, 51–65. Peltonen, P., Kontinen, A., 2004. The Jormua ophiolite: A mafic-ultramafic complex from an ancient ocean-continent transition zone. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 35–71. Peters, Tj., Nicolas, A., Coleman, R.G., 1991. Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Boston, p. 903. Pfänder, J.A., Kröner, A., 2004. Tectono-magmatic evolution, age, and emplacement of the Agardagh Tes-Chem ophiolite in Tuva, Central Asia: Crustal growth by island arc accretion. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 207–221. Polat, A., Kerrich, R., 2004. Precambrian arc associations: Boninites, Adakites, Magnesian Andesites, and Nb-enriched Basalts. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 567–597. Puchtel, I., 2004. 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 405–424. Puchtel, I.S., Shchipansky, A.A., Samsonov, A.V., Zhuralev, D.Z., 1997. The Karelian granitegreenstone in Russia. In: de Wit, B., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 699–706. Rapp, R.P., 1997. Heterogeneous source regions for Archean granitoids: experimental and geochemical evidence. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs On Geology and Geophysics, vol. 35, pp. 267–279. Rudnick, R.L., 1995. Making continental crust. Nature 378, 571–578. Scott, D.J., St Onge, M.R., Lucas, S.B., Helmstaedt, H., 1991. Geology and chemistry of the Early Proterozoic Purtuniq ophiolite, Cape Smith Belt, Northern Quebec, Canada. In: Peters, Tj., et al. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Ministry of Petroleum and Minerals, Sultanate of Oman, pp. 817–849. Scott, D.J., Helmstaedt, H., Bickle, M.J., 1992. Purtuniq ophiolite, Cape Smith belt, northern Quebec, Canada: A reconstructed section of Early Proterozoic oceanic crust. Geology 20, 173–176. Sengör, A.M.C., 1993. Turkic-type orogeny in the Altaids: implications for the evolution of continental crust and methodology of regional tectonics. Transactions of the Leicester Literary and Philos. Soc. 87, 37–51. Sengör, A.M.C., Natal’in, B.A., Burtman, V.S., 1993. Evolution of the tectonic collages and Palaeozoic crustal growth in Eurasia. Nature 364, 299–307. Sengör, A.M.C., Natal’in, B.A., 1996. Turkic-type orogeny and its role in the making of the continental crust. Annual Reviews of Earth and Planetary Sciences 24, 263–337. Sengör, A.M.C., Natal’in, B.A., 2004. Phanerozoic analogues of Archean oceanic basement fragments: Altaid ophiolites and ophiorags. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 675– 725.

34

Introduction

Shchipansky, A.A., et al., 2004. 2.8 Ga Boninite-hosting partial suprasubduction zone ophiolite sequences from the North Karelian greenstone belt, NE Baltic Shield, Russia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 425–486. Sleep, N., Windley, B.F., 1982. Archean Tectonics: constraints and inferences. J. Geol. 90, 363–379. Sorjonen-Ward, P., Nironen, M., Luukkonen, E., 1997. Greenstone associations in Finland. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35. Clarendon Press, Oxford, pp. 677–698. Stern, R.J., 1994. Arc Assembly and continental collision in the Neoproterozoic East African Orogen: Implications for the assembly of Gondwanaland. Annual Reviews of Earth and Planetary Sciences 22, 319–351. Stern, R.J., Johnson, P., Kröner, A., Yibas, B., 2004. Neoproterozoic Ophiolites of the ArabianNubian Shield. St-Onge, M.R., Lucas, S.B., Scott, D.J., Begin, N.J., 1989. Evidence for the development of oceanic crust and for continental rifting in tectonostratigraphy of the early Proterozoic Cape Smith belt. Geoscience Canada 16, 119–122. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.C., 1997. Volcanic Aspects. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 55–90. Tarney, J., Jones, C.E., 1994. Trace element geochemistry of orogenic igneous rocks and crustal growth models. J. Geol. Soc. London 151, 855–868. Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust. Rev. Geophys. 33, 241–265. Turcotte, D.L., Schubert, G., 2002. Geodynamics. Cambridge Univ. Press, Cambridge, p. 456. Wilks, M.E., Harper, G.D., 1997. Wind River Range, Wyoming Craton. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 508– 516. Williams, H.R., Stott, G.M., Heather, K.B., Muir, T.L., Sage, R.P., 1991. Wawa Subprovince. In: Thurston, P.C., Williams, H.R., Sutclife, H.R., Stott, G.M. (Eds.), Geology of Ontario. Ont. Geol. Survey Spec. Vol. 4 (1), 485–539. Windley, B.F., Whitehouse, J., Ba-Bttat, M.A.O., 1996. Early Precambrian gneisses and Pan-African island arcs in Yemen: crustal accretion of eastern Arabian shield. Geology 24, 131–134. Wood, D.A., Joron, J.L., Treuil, M., 1979. A re-appraisal of the use of trace elements to classify between magma series erupted in different tectonic settings. Earth Planet. Sci. Lett. 45, 326–336. Wyllie, P.J., Wolf, M.B., van der Laan, S.R., 1997. Conditions for formation of tonalites and trondhjemites: magmatic sources and products. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 91–124.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

35

Chapter 1

THE JORMUA OPHIOLITE: A MAFIC-ULTRAMAFIC COMPLEX FROM AN ANCIENT OCEAN-CONTINENT TRANSITION ZONE P. PELTONENa AND A. KONTINENb a Geological Survey b Geological Survey

of Finland, P.O. Box 96, FIN-02151, Espoo, Finland of Finland, P.O. Box 1237, FIN-70211, Kuopio, Finland

The Jormua Ophiolite is an allochtonous mafic-ultramafic rock complex, thrusted onto the Karelian Craton margin, that formed within a passive margin environment ∼ 100 km southwest from its present position. This complex consists of two distinct units: (1) fragments of Archean subcontinental lithospheric mantle that became exposed from beneath the Karelian craton by detachment faulting following the final break-up of the craton, and (2) alkaline and tholeiitic igneous suites that were emplaced within and through the lithospheric mantle at ∼ 2.1 Ga and 1.95 Ga, respectively. At the prerift stage of continental breakup (c. 2.1 Ga), residual lithospheric peridotites became intruded by alkaline melts that formed “dry” clinopyroxene cumulate dikes. Slightly later, this same piece of mantle became extensively intruded by hydrous alkaline magmas that resulted in formation of high-pressure hornblendite-garnetite cumulates deep in the ophiolite stratigraphy and fine grained OIB-type dikes at more shallow levels. Simultaneously, the residual peridotites became metasomatized due to porous flow of the melt in the peridotite matrix. Alkaline magmatism was soon followed by lithospheric detachment faulting that exposed the subcrustal peridotites at the seafloor, where they at 1.95 Ga became covered by tholeiitic (EMORB) pillow and massive lavas and intruded by coeval dikes and gabbros. Since transitional contacts between all main ophiolite units can be demonstrated, the Jormua Ophiolite Complex is interpreted to represent a practically unbroken sample of seafloor from an ancient ocean-continent transition (OCT) zone, strikingly similar to that reported from younger similar tectonic settings, such as the Cretaceous West Iberia nonvolcanic continental margin.

1. INTRODUCTION Though included only as a connotation in the current de facto ophiolite definition (Anonymous, 1972), it is generally accepted that ophiolite complexes represent fragments of oceanic lithosphere that formed in a number plate tectonic settings (oceanic spreading ridges, island arcs, back arc basins, leaky transforms, nascent ocean basins, etc.). The DOI: 10.1016/S0166-2635(04)13001-6

36

Chapter 1: The Jormua Ophiolite

special importance of Precambrian ophiolites is that their presence or absence in the rock record is usually considered one of the critical evidences for or against operation of modern type plate tectonism in Precambrian. In addition, the chemical composition of associated basalts provide an uncontaminated window into the Precambrian convective mantle, and in rare cases—such as Jormua—the petrology of the mantle rocks can be studied in situ from excellent outcrops. The 1.95 Ga Jormua complex was the first early Proterozoic ophiolite ever reported (Kontinen, 1987). Since then a few early Proterozoic and even Archean ophiolites have been discovered and described but still their number is notably low (Scott et al., 1992; Dann, 1991; Kusky et al., 2001). Being in many respects similar to the Northern Apennines ophiolites (e.g., Rampone and Piccardo, 2001) and sharing several of the salient features of the modern oceancontinent transition zones (Louden and Lau, 2001) the Jormua Ophiolite Complex has been interpreted as a break-up-related “passive margin ophiolite” (Kontinen, 1987; Peltonen et al., 1996, 1998). It is made up of two components of distinct origin: (a) subcontinental (> 2.8 Ga) lithospheric mantle (SCLM) component, and (b) asthenospheric component that consists of remnants of the 1.95 Ga mantle diapir and various types of igneous rocks emplaced within and through the mantle tectonites ∼ 1.95–2.1 Ga. Thus, the main importance of Jormua for understanding the evolution of Precambrian plate tectonic processes is the evidence it provides of continental rifting and break-up related tectonic and magmatic processes. The other ophiolitic rocks in Finland—the Outokumpu and Nuttio complexes (Fig. 1)—which appear to share the same large-scale tectonic setting along the Karelian Craton margin, provide us samples from more mature oceanic basin and island-arc setting, respectively (Koistinen, 1981; Vuollo and Piirainen, 1989; Hanski, 1997). Together, in a synthesis that remains to be compiled, these occurrences provide a tantalizing opportunity to interpret the break-up of the Karelian Continent, and the igneous processes that took place in the Svecofennian ocean from its birth until its closure. This contribution is intended to be a compact overview of the geology of the Jormua Ophiolite Complex and a state-of-art summary of what is known and inferred about its genesis. Emphasis is placed on the field description of the main ophiolite units. Chemical and isotopic compositions of both the crustal and mantle rocks have extensively been recently described elsewhere (Peltonen et al., 1996, 1998) and only some salient points shall be touched in this context. For the same reason no analytical data is included. There are still many lines of research that have not yet been applied to this rare piece of ancient oceanic and continental lithosphere. We encourage more research on Jormua so that more of its messages from the geological past will be revealed. In the meantime, we believe that the tectonic framework upon which the future work will bounce is already well established.

2. THE REGIONAL SETTING OF THE JORMUA OPHIOLITE COMPLEX The Jormua Ophiolite Complex is the northernmost and the most completely preserved example of the ophiolite fragments within the Paleoproterozoic North Karelia Schist Belt

2. The Regional Setting of the Jormua Ophiolite Complex

37

Fig. 1. Geological map of the eastern part of the Fennoscandian Shield emphasizing the location of the Jormua Ophiolite Complex and other early Proterozoic ophiolite fragments. Archaean blocks: PC = Pudasjärvi Complex; IC = Iisalmi Complex; EFC = Eastern Finland Complex. Modified from Koistinen et al. (2001) and Hanski (1997).

38

Chapter 1: The Jormua Ophiolite

and the Kainuu Schist Belt in the central part of the Fennoscandian Shield (Fig. 1). These two Paleoproterozoic schist belts, which consist mainly of 2.3–1.90 Ga metasediments (grouped under the term Karelian in the Finnish literature), are located 0–100 km to the east of the suture between their Archaean basement structure (Karelian Craton) and the 1.93–1.80 Ga Svecofennian island arc domain (Fig. 1). Approximately 1.9 Ga ago the southwest margin of the Karelian Craton was covered by thrusted allochthonous complexes and deformed by the related compressional and subsequent wrench/thrust tectonics. The thrusting took place before the D1 deformation as a response to the collision between the Karelian Craton and the cratonized Svecofennian arc collage (Gaál and Gorbatschev, 1987; Kärki and Laajoki, 1995; Kohonen, 1995; Korsman et al., 1999). The thrust belt contains a 200 km long chain of ophiolite fragments whose distribution is related to the early thrusting, further modified by the later multistage regional deformation (Koistinen, 1981; Kontinen, 1987; Kärki and Laajoki, 1995). The Kainuu Schist Belt that encloses the Jormua Ophiolite occupies a structural depression between two rigid blocks of the basement structure: the Eastern Finland Complex and Pudasjärvi-Iisalmi Complexes (Fig. 1). The belt is up to 30 km wide but nowhere more than 2–3 km thick on the basis of structural and gravimetric data. It thus represents a relatively thin veneer of faulted and folded autochthonous and allochthonous supracrustal rocks and ophiolite fragments on the thick Archaean basement structure. The belt is dissected by a folded N-S trending strike-slip fault zone along which the Archean basement blocks have moved both vertically and horizontally (Kärki and Laajoki, 1995). The central part of the Kainuu Schist Belt around the Jormua Ophiolite comprises three major unconformity or thrust-bound lithofacies: (1) the autochthonous, cratonic to epicratonic Jatuli (2.3–2.1 Ga) sequence, consisting predominantly of quartzites derived from fluvial to shallow marine, mature quartz-rich sands; (2) the 2.1–1.95 Ga rift-phase related “lower Kaleva” assemblage, characterized by metaturbiditic conglomerates, quartz wackes, graywackes and shales, as well as turbidite-hosted P-Mn-C-rich silicate-carbonate iron formations and abundant, metal-rich (Cu, Ni, Zn) graphitic black schists; and (3) the allochthonous “upper Kaleva” sequences, dominated by deep marine metaturbiditic graywacke-shale deposits whose depositional age is bracketed by the age of the Jormua (∼ 1.95 Ga; Peltonen et al., 1998) and the youngest detrital zircon SHRIMP ages of 1.92 ± 0.12 Ga (Claesson et al., 1993). Many of the detrital zircons in these “upper Kalevian” metagraywackes are thus younger than the mafic sequence of the Jormua Ophiolite. The Jatulian mature arenites, and the “lower Kalevian” sequence, which are presumably rift-related deposits contain mainly recycled Archean detritus, whereas the younger, flyschlike “upper Kalevian” metagraywackes and shales contain also abundant Proterozoic material. An additional important geological unit, probably closely related to the origin of the Jormua Ophiolite Complex, is the ∼ 1.96 Ga Otanmäki gneissic, peralkaline-alkaline Atype granite to the SSW of Jormua (Fig. 1). These granites have intruded Archaean gneisses and Jatuli-type cover sediments and are present as a narrow strongly foliated tectonic slice. At their southern margin the Otanmäki gneissic granites are in a faulted contact with the rocks of the Kainuu Schist Belt and the Archaean rocks of the Iisalmi Complex. We are

3. Principal Features of the Main Blocks

39

tempted to believe that the Otanmäki gneissic alkaline granite is an allochthonous unit not belonging to the basement structure of the Kainuu Schist Belt.

3. PRINCIPAL FEATURES OF THE MAIN BLOCKS The earliest stages of the Svecofennian deformation, related to the early tectonic processes that contributed to the detachment of the Jormua Ophiolite Complex from oceanic environment and its subsequent thrusting across the foreland, involved tectonic disruption of the original ophiolite assemblage. Consequently, the Jormua Ophiolite now consists of four major fault-bounded blocks (Fig. 2), which represent diverse parts through the ancient ocean-continent transition zone (Peltonen et al., 1996, 1998). Extensive shearing along the Kainuu Schist Belt in the latest stages of the Svecofennian tectonism (Kärki and Laajoki, 1995) and associated parasitic faulting and folding further disrupted and deformed the ophiolitic blocks, some of which now have forms of shear-controlled “megaaugens”. The salient characteristics of these blocks are summarized in Table 1. The eastern block consists of several fault-bounded slices of serpentinized mantle tectonites (mainly harzburgites) and minor dunitic pods, some of which enclose small nodularand orbicular-textured podiform chromitite lenses. Mantle peridotites have been intruded by gabbro stocks and basaltic sheeted dike complexes. The eastern block is practically the only block that is associated with the extrusive seafloor sequence. Reconstructed stratigraphy of the ophiolite suggests that the dike complex and locally also mantle peridotites are directly overlain by metabasaltic pillow lavas intercalated with some massive lavas and pillow breccias. According to diamond drilling data the pillow lavas at the western margin of the eastern block are overlain by basic tuffs intercalated with sedimentary carbonates. These metatuffs are—at least in a tectonostratigraphic sense—overlain by the typical upper Kaleva metaturbiditic black shales and graywackes. Thus, in terms of igneous stratigraphy, the eastern blocks expose the most complete ophiolite within the Jormua igneous complex. The northern block is scantily exposed and hence the least studied of the Jormua blocks (Fig. 2). Nevertheless, the presently available data suggests that this block is fairly similar with the eastern block. The main exposed components are mantle tectonites, gabbroic feeder dikes, and more extensive gabbro-diorite intrusions and some massive metabasaltic amphibolites, probably of sheeted dike origin. Highly altered, epidote and iron sulfide-rich gabbros are known from one locality. The central block resembles the eastern block in some respects but several fundamental differences are evident (Table 1). Both extrusive rocks and gabbroic pods and dikes are uncommon within the central block. Instead, the mantle tectonites have been intruded by abundant fine grained EMORB dikes. The abundance of these dikes progressively increases from NW to SE (Fig. 2). In the northwest single anastomosing dikes intrude peridotites. In the following, these dikes are referred to as “deep dikes” because of their apparent location deep in the ophiolite stratigraphy. Towards southeast “deep dikes” gradually coalescence into thicker units finally forming a spectacular sheeted dike complex with only rare interdike screens of mantle peridotites or gabbro. The mantle peridotites (serpentinites) of the

40

Chapter 1: The Jormua Ophiolite

Fig. 2. Geological map of the Jormua Ophiolite Complex. Modified after Kontinen (1998).

3. Principal Features of the Main Blocks

41

Table 1. Characteristics of the Jormua Ophiolite complex Rock type

Western block

Central block

Eastern block

Northern block

Crustal units Extrusive unit Sheeted dike complex Gabbros/plagiogranites Ultramafic cumulates

− − − −

− ++ − −

++ ++ ++ ?

+ + + −

Intrusive to mantle tectonites EMORB-dikes Gabbroic feeder dikes Chromitite pods OIB-type dikes Clinopyroxenite mantle dikes Hornblendite mantle dikes Garnetite veins Carbonatitic veins

? − − ? ++ ++ + +

++ + − + + − − −

++ ++ + − − − − −

+ + − − − − − −

Mantle tectonites Lherzolites (> 3 wt% Al2 O3 ) Depleted lherzolites (1
Al2 O3
3 wt%) Harzburgites and dunites (< 1 wt% Al2 O3 )

+ ++ −

− ++ +

+ ++ +

? + +

+, present; ++, abundant; −, absent.

central block are intruded by an additional generation of ultramafic-mafic dikes which are absent in the eastern block. These dikes, which have OIB-type geochemical characteristics and resemble ultramafic lamprophyres, are intruded by the more voluminous EMORB dikes and are therefore labelled as “early dikes”. While the EMORB dikes intersect the mantle tectonite foliation at high angles the general trends of the OIB-type dikes are subparallel. This clearly implies that EMORB and OIB magmas were emplaced during distinct episodes, and that between the emplacement of OIB and EMORB dikes, the tectonite foliation was rotated from subvertical to almost horizontal. The contact between the western block and central blocks is unexposed. Thus, the possibility remains that they form a single long continuous sheet, but their significant internal differences suggest that this is probably not the case (Fig. 2). First, the western block is not associated with “upper Kaleva” graywackes, but instead is bounded by slices of the Archaean basement and rift-related “lower Kaleva” sediments. Second, the mantle peridotites are less depleted in their basaltic constituents and harzburgitic and dunitic residues are uncommon. Furthermore, gabbros, basaltic dikes and pillow lavas are absent from the western block. Instead, mantle peridotites are extensively intruded by clinopyroxenite and hornblendite dikes and pods representing igneous cumulates that crystallized in melt channels and pathways within the upper mantle. Isotopic data suggest that these cumulates are broadly coeval and closely related to the OIB-type dikes found in the central block

42

Chapter 1: The Jormua Ophiolite

(Peltonen et al., 1998). Overall, the western block shares more common features with orogenic lherzolite massifs (i.e., subcontinental lithospheric mantle) than with true ophiolites.

4. RECONSTRUCTED STRUCTURE OF THE COMPLEX Although the Jormua Ophiolite Complex has been tectonically disrupted into several distinct blocks, the reconstruction of the original igneous stratigraphy can be made with reasonable confidence. This is because transitional contacts between the main ophiolitic units, i.e., extrusive rocks, sheeted dikes complex, gabbros and mantle peridotites can be demonstrated in the field. A revised stratigraphic reconstruction for the Jormua Ophiolite Complex is presented in Fig. 3. The upper part of the stratigraphic column is based on the geology of the eastern and central blocks of the Jormua Ophiolite Complex. The extrusive unit consists of pillow and massive lavas capped by some basic tuffite and sedimentary carbonates. The extrusive unit is relatively thin (0–400 m) and one may argue that locally lavas were deposited directly onto mantle tectonites exposed at the seafloor. In modern seafloor, mid-ocean ridges with mantle peridotite outcrops are characterized by slow spreading rates and deep axial valleys that are expected to form at magma-poor ridge regions where a substantial fraction of the oceanic lithosphere is made of tectonically uplifted mantle material (Cannat, 1993). Downwards in the stratigraphic section, pillow lavas occur as interdike screens within the uppermost sheeted dikes complex, and some dikes are present within the lava unit implying coeval formation of lavas and dikes. Thick mafic-ultramafic cumulate layers are absent from the Jormua Ophiolite Complex. This is, however, not because of incomplete preservation of the original sequence, but is a characteristic feature of slow spreading-ridge ophiolites where thick axial lithosphere prevents magma pooling at crustal depths. In Jormua, the sheeted dike complex is not rooted in the cumulates but instead is rooted in the mantle tectonites (Figs. 2 and 3). Going downwards the coherent sheeted dike complex gradually gives way to dike swarms with abundant mantle tectonite interdike screens and finally into sparse anastomosing basaltic dikes (“deep dikes”) intruding mantle tectonites. Furthermore, indirect evidence for the absence of large magma chambers in Jormua is provided by the geochemistry of the basalts; the analysed lava and sheeted dike samples cannot be related to each other by fractional crystallization processes but represent rather unmodified melt fractions which did not undergo fractionation at an intermediate magma storage (Peltonen et al., 1996). Isotropic gabbro (+ plagiogranite) pods characterize the middle section of the stratigraphic column (Fig. 3). Importantly, these gabbros are frequently cross-cut by sheeted dikes, but are never observed to intrude the dike complex. This suggests that gabbros and seafloor magmatism represent distinct magmatic episodes and that the U-Pb zircon ages provided by the gabbros give a maximum age for the seafloor volcanism. However, the absolute time difference is believed to be small, with the gabbro stocks being related to the initial stages of the continental rifting, while lavas and dikes record a slightly more advanced stage of the oceanization. Distinct types of gabbros are found beneath the large high-level gabbro-plagiogranite pods. These occur in the form subvertical dikes that

4. Reconstructed Structure of the Complex

43

Fig. 3. Igneous stratigraphy of the Jormua Ophiolite Complex. Age data from Huhma (1986), Kontinen (1987), Peltonen et al. (1998), and Peltonen et al. (2003, unpublished).

intrude the mantle tectonites (hereafter: gabbroic feeder dikes). Their internal fractionation is indicative of them being feeder dikes for upper-level magmas. They have yielded equal crystallization age (1953 ± 2 Ma; Peltonen et al., 1998) with the gabbro stocks (1960 ± 12 Ma) and plagiogranites (1954 ± 11 Ma; Kontinen, 1987) and it is not evident whether they are comagmatic with high-level gabbro stocks or with sheeted dikes or lavas. The lower part of the stratigraphic column refers to the western block of the complex. Since the western block is lithologically distinct from other blocks and the contact between the western and central block is unexposed a continuous stratigraphic column was not drawn. This part of the Jormua Ophiolite Complex does not resemble a true ophiolite but bears striking similarities with fragments of subcontinental lithospheric mantle (SCLM), such as orogenic lherzolite bodies of the French Pyrenees. Thus, the Jormua Ophiolite consists of two distinct parts; one similar to slow-spreading type ophiolites (oceanic lithosphere) and another that is similar to orogenic lherzolites (subcontinental lithospheric mantle). Importantly, juxtaposition of these parts is not coincidental since sim-

44

Chapter 1: The Jormua Ophiolite

ilar c. 1.95 Ga old dikes facies are present in both fragments. This suggests that these two distinct parts share a common history and led Peltonen et al. (1998) to suggest that as a whole the Jormua Ophiolite Complex records an almost continuous sequence across an ocean-continent transition (OCT) zone.

5. ALTERATION AND METAMORPHISM Seafloor metamorphism, alteration during obduction and tectonic transport, and finally the Svecofennian regional metamorphism, have together resulted in extensive destruction of the primary mineralogy of the mafic and ultramafic lithologies of the Jormua Ophiolite. Fortunately, however, chemical changes with respect to most elements are far less pronounced. The metamorphic history of the ophiolite commenced already at the seafloor stage—evidence of which has mostly been lost by later imprints. Hydrothermal circulation on the seafloor resulted in alteration of basalts and veining of the basaltic dikes by albite-filled fractures (Fig. 4f). Lavas, in turn, are anomalously depleted in Fe, a feature that has been related to the seafloor weathering (Peltonen et al., 1996). It is not clear anymore to what extent the mantle peridotites were serpentinized at this stage. However, since the basaltic lid is thin in Jormua and field evidence suggests that some mantle tectonites were exposed at the seafloor, it is likely that mantle peridotites became at least partially serpentinized before obduction. The alteration that took place during the obduction and tectonic transport is only poorly characterized. The serpentinization of the mantle tectonites continued due to their interaction with meteoric waters. This alteration proceeded at relatively low temperature conditions resulting in extensive replacement of olivine and pyroxenes by pseudomorphic lizardite and local carbonatization and silicification of the margins of the tectonite massifs. Basaltic and gabbroic dikes in contact with peridotites became rodingitized due to their interaction with serpentinizing fluids. This resulted in considerable loss of silica and alkalies from the dikes (Fig. 8) and replacement of the primary igneous mineral parageneses of the gabbro dikes by hydro-grossular garnet, diopside, epidote, and chlorite. Much of the present mineral parageneses of the Jormua mafic and ultramafic rocks, however, represent the metamorphic equilibria attained during the Svecofennian regional metamorphism. Within the Kainuu Schist Belt metamorphism culminated under low-P high-Ttype conditions in the amphibolite facies between 1.87 and 1.85 Ga. The cooling was a prolonged event and it was not until ∼ 1.80 Ga that temperatures fell below 500 ◦ C. Thus, the metamorphism significantly outlasted the deformation that was essentially over by 1.86 Ga (Tuisku, 1997). The metamorphic mineral paragenesis in the metabasalts typically is sodic plagioclase + actinolitic hornblende ± epidote ± chlorite. Due to the activity of CO2 -rich metamorphic fluids some lava samples became slightly depleted in LREE. This is evident in the Sm-Nd isochron diagram where basalt samples form an isochron with a slope corresponding to an age of 1.72 ± 0.12 Ga (Peltonen et al., 1996). The slope of this isochron is strongly controlled by the three pillow-lava samples having the most LREE-depleted patterns. Since this “errorchron” age is significantly less than the U-Pb zircon age (∼ 1.95 Ga)

5. Alteration and Metamorphism

45

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 4. Outcrops of (a) pillow lava, (b) pillow breccia, (c) “deep dikes”, i.e., EMORB dikes intruding mantle tectonites at the root zone of the sheeted dikes complex. Note the anastomosing shape of the dikes and the dark alteration selvages at their margins due to interaction with adjacent serpentinites (former residual mantle peridotites), width of the photo ∼ 2 m, (d) sheeted dikes complex with 100% dike-in-dike sets, width of the photo ∼ 3 m, (e) sheeted dikes with gabbro interdike screens, (f) “deep dike” with seafloor alteration-related albite veins overprinted by late alteration selvage between dike and enclosing mantle tectonite at left, width of the photo ∼ 30 cm. (f) Reprinted with the permission from Journal of Petrology, vol. 37, Oxford Univ. Press.

46

Chapter 1: The Jormua Ophiolite

of the Jormua Ophiolite, or the age of the obduction (∼ 1.90–1.87 Ga), it is clear that the LREE depletion of these three samples cannot be of igneous origin. In mantle tectonites, the regional metamorphism resulted in dehydration and replacement of the earlier low-T pseudomorphic lizardite serpentine by non-pseudomorphic antigorite serpentine. Present stable metamorphic parageneses of serpentinites are dependent on the bulk rock composition and include antigorite+magnetite and antigorite+tremolite+magnetite in the eastern and central block serpentinites. Olivine is added to the parageneses in the western block. On the basis of published mineral stability curves (e.g., Will et al., 1990) we estimate that these parageneses imply peak-metamorphic temperatures of approximately 480 and 530 ◦ C for pressures of 2 and 5 kb, respectively, in the west and slightly lower temperatures in the east. As a result of nearly complete serpentinization, the analysed H2 O(tot)-contents of peridotite samples now range between 9.4 and 12.0 wt%. However, since SiO2 /MgO ratios of Jormua serpentinites are still very similar to fresh or only slightly serpentinized mantle peridotites elsewhere, it is highly probable that serpentinization largely conserved both SiO2 and MgO, in which case the volume of the peridotites must have increased (O’Hanley, 1996). Within the Jormua Ophiolite Complex talc-carbonate rocks occur as altered marginal variants of serpentinite massifs. Their mineralogy is dominated by carbonate and talc in approximately equal proportions, together with some magnetite and sulfides. Extensive talc-carbonate alteration is restricted to narrow marginal zones of serpentinites. Talccarbonate alteration may be a post-metamorphic process (Eckstrand, 1975) or, alternatively, antigorite-carbonate-talc and carbonate-talc assemblages could have been stabilized under prograde conditions but at significantly higher XCO2 than the carbonate-free mineral assemblages. Similar metamorphic zonation in serpentinite bodies has been described from metakomatiites (Gole et al., 1987). By analogy, the concentric metamorphic mineralogy of serpentinite massifs could have been created when the breakdown of premetamorphic serpentines (XH2 O = 1) and the infiltration of serpentinite margins by CO2 rich fluid (originating from adjacent calcareous metasediments and pre-metamorphic alteration zones undergoing decarbonatization) generated gradients in the composition of metamorphic fluid. Carbonate-free assemblages represent equilibration beyond the limit of CO2 -infiltration. Peltonen et al. (1998) discussed in length the mobility of REE during serpentinization or deserpentinization reactions. They came to the conclusion that in the absence of suitable ligands (such as sulfate or carbonate) in the fluids, the alteration— although severe—did not seriously affect the REE characteristics of the peridotites.

6. THE CRUSTAL UNIT 6.1. Lavas and Hyaloclastites The extrusive crustal part of the Jormua Ophiolite is rather thin, approximately averaging only 100–400 m (Fig. 3). The thickness of the whole basaltic lid of the ophiolite that includes lavas, sheeted dikes and high-level gabbro stocks was variable (< 500 to > 1.5 km).

6. The Crustal Unit

47

Locally, it is apparent that the basaltic flows were deposited directly onto mantle peridotites. Most of the extrusive metavolcanic rocks of the Jormua Ophiolite Complex occur within the eastern block (Fig. 2; Table 1) as tectonic slices up to 400 m thick and kilometers in length, and typically in faulted contact with adjacent lithologies. Typically, lavas are bordered by serpentinites and talc-carbonate rocks or gabbros-plagiogranites in the structural footwall (to the west), and by upper Kalevian black schists and metagraywackes in the structural hanging wall (to the east). Most of the extrusive part of the Jormua consists of pillow lavas (∼ 50%; Fig. 4a) together with substantial amounts of pillow breccias (Fig. 4b) and hyaloclastites that make up to 25% of the extrusive unit. The remaining 25% consists of massive lava flows or flow parts. Some dikes and gabbro intrusions, one > 15 m thick differentiated gabbro-pyroxenite sill, for example, are present within the extrusive sequence. Importantly, interstitial or intercalated terrigenous sedimentary material is completely absent. Based on the presence of hyaloclastite interpillow matrix, minor pillow breccias, and the vesicle-rich nature of some of the pillowed flows Kontinen (1987) argued that the lavas erupted in a relatively shallow-water environment. Relict porphyritic and glomeroporphyritic textures are commonly recognizable in lavas which at present consist essentially of recrystallized plagioclase and nematoblastic calcic amphibole. The chemical composition of the lavas, such as their high Mg# [Mg/(Mg+Fe2+ tot ) = 0.59 to 0.73] and high Cr and Ni abundances suggest that lavas are not strongly fractionated. On the Cr vs. Y diagram (Fig. 5), for example, lava samples plot along or only slightly below the partial melting trend suggesting that the primary magma has undergone only minor fractionation (Fig. 5). Since chromite saturates early in MORB (Fisk and Bence, 1980) fractionation should rapidly deplete the residual melt in chromium. Most lava samples cannot be related to each other (or to the sheeted dikes) by fractional crystallization. Instead, the chemical composition of the lavas is consistent with them representing individual melt fractions produced through variable degrees of partial melting from which only minor amounts of chromite, olivine, and plagioclase have been segregated. Such chemical characteristics are in perfect agreement with the igneous stratigraphy inferred from field observations that demonstrated the absence of large magma chambers where pre-eruptive fractionation could have taken place. The trace element composition of Jormua basalts has been extensively discussed by Kontinen (1987) and Peltonen et al. (1996). Lavas (and most of the dikes) form a rather homogeneous group that has a composition closely similar to that of enriched mid-ocean ridge basalts (EMORB) of the modern seafloor. They are characterized by flat REE (Fig. 6) and, e.g., Nb, Ta, and Th abundances in excess of that typical for NMORB. On the basis of trace element and Nd isotope data Peltonen et al. (1996) concluded that the chemical composition of Jormua lavas and sheeted dikes result from mixing of NMORB and OIB mantle sources. Furthermore, on the Th/Yb vs. Ta/Yb diagram (Pearce, 1983) lava (and dike) samples form a coherent group of analyses that plot along the diagonal mixing array between depleted and enriched mantle sources. This implies that they all contain broadly uniform amounts of enriched endmember, and are free of any geochemical crustal (arctype) signature.

48

Chapter 1: The Jormua Ophiolite

Fig. 5. Cr vs.Y plot for lavas and basaltic dikes of the Jormua Ophiolite. Mineral vector calculations after Peltonen et al. (1996). The subhorizontal line is the partial melting trend of Pearce (1982).

6.2. Sheeted Dike Complex The sheeted dike complex, with its > 1 km thickness and ∼ 10 km2 areal coverage, is the major crustal unit of the Jormua Ophiolite. The central block (Fig. 2) emphasizes one of the most spectacular phenomena of Jormua: the “rooting” of the sheeted dike complex in the uppermost mantle. Stratigraphically the lowest part of the complex consists of individual anastomosing basaltic dikes (“deep dikes”) that intrude the mantle tectonite (Fig. 4c). Margins of thick “deep dikes” are characterized by 10–30 cm wide, dark green, chlorite rich alteration selvages while thin “deep dikes” may be thoroughly altered. Compared to dike interiors the alteration margins are strongly depleted in, e.g., Si, Ca, and alkalies, and enriched in Mg, Cr, Ni, and loss of ignition reflecting their pervasive hydration and double diffusive interaction with mantle peridotite (Peltonen et al., 1996). This interaction took place either during serpentinization of the peridotite host (rodingitization), or later during regional metamorphism, or both. Moving upwards in the stratigraphy “deep dikes” coalescence into groups of multiple dikes with progressively smaller mantle tectonite and gabbro interdike screens demonstrating the consanguinity of “deep dikes” to the coherent sheeted dikes complex. Presence of lava, gabbro and mantle tectonite as interdike screens

6. The Crustal Unit

49

Fig. 6. Chondrite-normalized (Boynton, 1984) rare earth element plots. (a) Lavas and basaltic dikes. Two distinct basalt suites are present: EMORB-type basalts with flat REE patterns and “early suite” OIB-type lavas with fractionated HREE and high LaN/YbN. (b) Gabbros and plagiogranites of the Jormua Ophiolite. The gabbros include samples from both from high-level gabbro stocks and gabbroic feeder dikes which have similar patterns and equal range of absolute concentrations.

(septa) in the Jormua sheeted dikes complex demonstrate that the contacts between the main ophiolite units are transitional, as required for ophiolite recognition by, e.g., Bickle et al. (1994).

50

Chapter 1: The Jormua Ophiolite

At higher crustal levels the sheeted dike complex consist of 100% of subparallel metadolerite and metabasalt dikes (Fig. 4d). Dikes in the sheeted complex are generally 20– 120 cm thick, aphyric or plagioclase-phyric with sharp chilled mutual contacts. The older dikes, which tend to be thick and coarse grained, were intruded by thinner and finer grained younger dikes, and became transformed to interdike screens of “half”, “one-way chilled”, and “marginless” dikes. Branching and apophyses along dike margins are common. Some of the dikes contain abundant plagioclase concentrated by magmatic flow in the dike interiors. No other microtextures have survived the recrystallization during regional metamorphism. The occasional presence of widely separated serpentinite and gabbro septa attest in a striking way the magnitude of extension during the formation of the mafic lid of the Jormua Ophiolite Complex (Fig. 4e). Although the “deep dikes”, sheeted dikes and overlying basalts are clearly coeval magmatic rocks with broadly similar chemical and isotopic compositions (Kontinen, 1987; Peltonen et al., 1996), minor differences emphasize the intricacy of this magmatism. The Cr (ppm) vs. Y (ppm) diagram (Fig. 5) emphasizes that while the lavas represent almost unmodified melt fractions, most of the “deep dikes” and sheeted dikes are significantly more fractionated and cannot represent feeders for the lavas. Probably this is related to the rate of the magma flow in the dike. Those dikes that acted as feeders for the lavas were characterized by high flow rates which prevented extensive fractional crystallization of the melt and resulted in eruption of lavas with primitive composition. Most of the dikes are, however, “blind” and never fed any basalts but instead underwent fractionation during ascent. 6.3. Gabbros A thick layered gabbro unit, characteristic of many classic young ophiolites is distinctly absent in the Jormua Ophiolite Complex. Instead, all gabbro occurrences in Jormua, even the largest ones (0.5 × 1.5 km in areal extent), represent stocks and dikes intrusive into the uppermost mantle. Most gabbro outcrops are made of isotropic or “varied-textured”, coarse to pegmatoid gabbro (Fig. 7a) and any clear modal layering or banding is absent even in the largest occurrences. Gabbro bodies in the central part of the eastern block are frequently intruded by fine grained to aphanitic, usually aphyric, typically 0.2 to 1.2 m wide basaltic dikes. These are separated by cm to meter wide gabbro screens (septa) that comprise 30–60% of many of the gabbro outcrops (Fig. 4e). Locally subparallel dikes form several meter wide zones of dike-in-dike sets. Many of the dikes show clear chilled margins against the gabbro screens and older dikes. Two types of gabbros have been distinguished in the field: grayish green Mg-gabbros, and dark green Fe-gabbros characterized by relatively low MgO, low SiO2 and distinctly high Fe and Ti reflected in abundance of ilmenite. Mineral assemblages in gabbros are usually thoroughly metamorphic: of the primary phases only An-rich plagioclase is sporadically preserved in the Mg-gabbros and coarse ilmenite in the Fe-gabbros. The most voluminous Fe-gabbros are present in the central part of the eastern block. Transitions from adjacent Mg-gabbros take place within a few meters by abrupt increase in ilmenite

6. The Crustal Unit

51

(a)

(b) Fig. 7. Outcrops of (a) varied-textured ilmenite-bearing high-level gabbro stock, and (b) leucotonalite (plagiogranite) dikes and veins in fine grained high-level gabbro.

content. Grain-size is characteristically variable ranging from fine to coarse and even pegmatoid over short distances. Irregular fine grained dike-like parts in some ferrogabbro outcrops suggest that coeval Fe-basalt dikes are rooting in the Fe-gabbro stocks. However, no Fe-rich lavas or dikes in the sheeted dike complex have been recognized so far. On the AFM ternary diagram gabbro samples follow a tholeiitic fractionation trend showing pronounced Fe (+ Ti) enrichment in the most evolved samples of the igneous suite (Fig. 8). Incompatible element abundances (such as Ti and Zr) are significantly lower in all gabbros compared to lavas or sheeted dikes of the complex (Kontinen, 1987). This implies that the gabbros are cumulates with only small amounts of intercumulus liquid remaining. Either substantial post-cumulus growth has taken place or, maybe more likely, the gabbroic cumulates have been depleted in residual liquid by filter pressing due to their crystallization in a dynamic environment. Furthermore, the compatible elements Cr and

52

Chapter 1: The Jormua Ophiolite

Fig. 8. AFM triangular plot for gabbros and plagiogranites of the Jormua Ophiolite Complex emphasizing the alkali depletion of gabbroic feeder dikes due to their rodingitization and, on the other hand, the extreme alkali (sodium) enrichment of the plagiogranites. Discrimination line after Irvine and Baragar (1971).

Ni are also present in significantly lower concentrations than is common for the lavas and dikes with the same Mg level. This suggests that the parental magma for the gabbros was already relatively evolved and implies that they probably are not coeval with lavas or dikes. 6.4. Plagiogranites Plagiogranites (i.e., leucotonalites) are well exposed only at one locality in the eastern block where they are closely associated with gabbros, microgabbros and diorites. Two main types of occurrences are present. First, some plagiogranite is present as a few meters wide and at least several tens of meters long zones of multiple, successive, and apparently syn-magmatically deformed dike injections within high-temperature ductile shear zones in Mg-gabbros. Plagiogranite in these zones is present as networks of irregular, cm to dm

7. The Mantle Section

53

thick, branching and cross-cutting dikes. Outcrop features suggest coeval emplacement of microgabbroic-basaltic dikes with the plagiogranite dikes and mingling of these magmas. Second, plagiogranite is also found as dike networks within microgabbro-diorite that occurs as marginal facies of large ferrogabbro bodies (Fig. 7b). Outcrop features suggest emplacement of the microgabbros-diorites and plagiogranites as a multistage progressive process involving mingling of the various pulses of magma. The leucotonalitic to trondhjemitic segregations (plagiogranites) of Jormua Ophiolite Complex have the chemical characters of ocean ridge granites being, e.g., very high in Na2 O and low in K2 O (Fig. 8). Chondrite normalized REE patterns of the leucotonalite segregations are somewhat fractionated at a relatively high level of REE abundances and have negative Eu minimas (Fig. 6b). The high Y and Nb concentrations of the dioritesleucotonalites places them in the within-plate granite field in the Nb vs. Y discrimination diagram of Pearce et al. (1984), which is in line with the overall EMORB character of the Jormua mafic rocks. Zircons from one leucotonalite segregation yielded a somewhat unprecise U-Pb age of 1954 ± 12 Ma. In addition, εNd (1.95 Ga) for this sample is +1.9, which is close to the +2 average for the lavas and dikes implying that plagiogranite origin is intimately related to the oceanic crust-forming magmatism.

7. THE MANTLE SECTION 7.1. Mantle Peridotites (Metaserpentinites) Mantle peridotites comprise approximately 55% of the area of the Jormua Ophiolite Complex. At present, the peridotites are mainly antigorite metaserpentinites whose mineralogical composition is controlled by bulk rock compositions and metamorphic grade. The eastern and central block peridotites are still characterized by outcrop textures typical for residual mantle tectonites, and by the absence of any well-defined textural or compositional layering. Serpentine pseudomorphs (bastite recrystallized to antigorite) after residual orthopyroxene occur as high relief lumps in the weathered outcrops (Fig. 9a). They are embedded within a “groundmass” consisting of somewhat darker serpentine + magnetite dust that is replacing mantle olivine. Locally, pyroxene pseudomorphs form mm to cm thick bands and strained chromite grains form schlierens thus defining the tectonite foliation (Fig. 9b). This foliation is cross-cut by 1950 Ma old gabbroic feeder dikes implying that the foliation is of mantle origin and not due to ∼ 1880 Ma regional deformation. Chromite is the only primary mineral that is preserved in the mantle peridotites (Liipo et al., 1995). Most of the disseminated chromite grains in the serpentinites have thoroughly been altered to ferri-chromite and chromian magnetite. Only occasionally do the serpentinites contain disseminated grains with unaltered chromite cores. In these grains the mutual boundary between the chromite core and ferri-chromite outer shell is sharp both optically and chemically. Ferri-chromite in turn gradually grades to chromian magnetite towards the grain margin. Electron microprobe analyses imply that at the scale of hand samples, there is only minor intra and inter grain variation in Cr# [Cr/(Cr + Al)] of

54

Chapter 1: The Jormua Ophiolite

(a)

(b)

(c)

(d)

Fig. 9. (a) “Knobby”-textured serpentinite (former residual harzburgite) where bastite pseudomorphs after orthopyroxene remain as high-relief knots in the antigorite serpentine matrix after mantle olivine, (b) chromite (now largely chromian magnetite) striations define mantle foliation, (c) residual mantle tectonite infiltrated by basaltic melt. Such samples are responsible for the relatively high REE abundances in some peridotite samples and their flat chondrite-normalized patterns (Fig. 11a), (d) small podiform chromitite body enclosed by talc-carbonate altered eastern block mantle tectonite. (a) Reprinted with the permission from Journal of Petrology, vol. 39, Oxford Univ. Press.

the chromite cores. Mg# [Mg/(Mg + Fe2+ )], in turn, varies greatly being higher in the interiors of the large chromite cores and lower in interiors of the small ones and gradually decreases from the chromite cores towards their margins (Kontinen and Peltonen, 1998). This implies that Mg# have been strongly modified from the original mantle values by alteration and metamorphism. However, the apparent immobility of Cr and Al in the chromite cores suggests that the variation in the Cr# (45–70) may still closely reflect the primary mantle melting-controlled values. Relatively high and variable Cr# suggests that the eastern and central block peridotites represent residues after variable but generally high degrees of partial melting. Chemical composition of the metaserpentinites imply similar origin as was deduced from the compositions of chromite. On the Pd/Ir vs. Ni/Cu metal ratio diagram of Barnes et al. (1988) talc- and carbonate-free serpentinite samples form a tight cluster close to the

7. The Mantle Section

55

mantle field (Fig. 10a). Two talc-carbonate altered samples are displaced from the mantle field: a serpentinite-talc schist contact zone sample (0.73 wt% CO2 ) contains 170 ppm Cu (> 5 times that of primitive mantle) and one of the massive serpentinites (1.72 wt% CO2 ) is strongly depleted in Pd. Uniform and low Pd/Ir is inconsistent with a cumulate origin for any of these samples. Due to similar partition coefficients between melt and residual sulfides during partial melting the Pd/Ir is not sensitive to variations in the degree of mantle melting. This is, instead, illustrated by Cr/Al vs. Ni/Al plot (Fig. 10b). In this diagram— where both ratios increase as a function of mantle melting—Jormua serpentinites record a wide range of ratios indicating variable degrees of melt extraction from peridotites. Incompatible element abundances bear evidence for a complex post-melting history of the Jormua mantle peridotites. Chondrite-normalized REE patterns bring out major differences between peridotites from different blocks of the Jormua Ophiolite Complex (Fig. 11). Serpentinites from the eastern block yielded two distinct types of patterns. First, several samples are characterized by U-shaped chondrite-normalized patterns—a form that is typical for dunites and harzburgites from the basal sections of ophiolites elsewhere (e.g., McDonough and Frey, 1989). Such patterns indicate that peridotites have first been depleted in LREE and MREE during mantle melting and later enriched in LREE. Second, some eastern block samples contain much more REE and yield flat patterns, similar to those of Jormua EMORB, but at a lower level. Some of these samples may represent small dunitic cumulate pods but some are characterized by typical residual mantle outcrop textures and more likely represent residual peridotites with substantial amounts of infiltrated and trapped basaltic melt (Fig. 9c). Western block peridotites yield truly distinct forms of chondrite-normalized patterns. They are characterized by relatively steep patterns between HREE and MREE but somewhat depleted LREE resulting in upward-concave patterns. These enriched peridotites yield similar initial 143 Nd/144Nd as the OIB-type “early dikes” and hornblenditic mantle dikes that intrude the western block peridotites. This was explained by Peltonen et al. (1998) by coeval flow of alkaline melt in dikes (conduit flow) and in the residual peridotite matrix (porous flow) that was at least partly driven by filter pressing of the melt from the dikes. 7.2. Chromitites Massive chromitite bodies and peridotites rich in disseminated chromite are known only from the eastern block where they occur within a 700 m long talc-carbonate altered peridotite slice between two pillow lava slices (Fig. 2). Most of the chromitite-bearing lithology in this relatively poorly exposed zone comprises brecciated serpentinized dunites with scattered small (cm to dm size) broken fragments of massive, nodular, or orbicular chromitite. Only two larger than one meter-size chromitite bodies are currently known (Fig. 9d). The largest of the presently known pods, which is about 0.8 m wide and at least 5 m long is located in heavily carbonated serpentinite fringed by talc-carbonate rocks. Margins of this folded and boudinaged chromitite lens comprise strongly fractured and altered chromite, the fractures being filled with carbonate and chlorite, whereas the core part of the body contains surprisingly fresh coarse grained chromite (Fig. 12). This chromite has a moder-

56

Chapter 1: The Jormua Ophiolite

Fig. 10. (a) Pd/Ir vs. Ni/Cu plot of Barnes et al. (1988) for serpentinite samples from the Jormua Ophiolite. With the exception of some talc-carbonate altered samples they all plot within the mantle field consistent with their residual origin. Low Pd/Ir is inconsistent with cumulate origin for any of the samples. (b) Cr/Al vs. Ni/Al diagram illustrating the compositional variability of the Jormua mantle tectonites due to extraction of variable melt fractions. The trend from the Lizard (least depleted) to Vourinous peridotites (most depleted) illustrates that produced by increasing degree of partial melting of mantle peridotite. Reference fields after Roberts and Neary (1993). Symbols in (a) and (b): open circle (eastern block peridotites), filled circle (central block peridotites), and black triangle (western block peridotites).

7. The Mantle Section

57

(a)

(b) Fig. 11. Primitive mantle (McDonough and Sun, 1995) normalized REE patterns for (a) eastern block serpentinites and (b) western block serpentinites. Note the large range in REE abundances for the eastern block samples indicating that some peridotites contain substantial amounts of infiltrated basaltic liquid (see also Fig. 9c) while some have U-shaped patterns more typical for oceanic peridotites. The pattern shapes for the western block peridotites are distinct and mimic those of alkaline (hornblenditic) dikes of the western block (Fig. 15b) suggestive of them being metasomatized by corresponding melt or fluid.

58

Chapter 1: The Jormua Ophiolite

Fig. 12. Photomicrograph of podiform chromitite from the eastern block of the Jormua Ophiolite Complex. Transparent light, width of the field ∼ 3 cm, photo by J. Väätäinen.

ately aluminous composition with average Mg# and Cr# of 76 and 55, respectively, and low TiO2 content of ∼ 0.24 wt%. Small silicate inclusions were present in many of the chromite grains but are now commonly replaced by secondary minerals. The preserved ones comprise relatively sodic tsermakitic hornblende having compositions strikingly similar with the amphibole inclusions in chromitites of some oceanic and ophiolitic peridotites (e.g., McElduff and Stumpfl, 1991). Interestingly, the chromitites yielded initial γOs values of +0.8 ± 0.5 and 3.0 ± 0.1 consistent with their derivation from a convective MORB-like oceanic mantle at the time of the ophiolite formation ∼ 1.95 Ga (Tsuru et al., 2000). 7.3. Gabbroic Feeder Dikes The eastern block mantle peridotites are intruded by a suite of gabbroic dikes. Typically these dikes are couple of meters wide and show symmetric texture indicative of them representing subvertical channels. They range from subparallel to discordant relative to the mantle foliation and branches from the main dikes cross-cut the foliation at high angle (Fig. 13). Locally, the veining of the host peridotite has resulted in detachment of peridotite xenoliths from the dike wall, leaving them ”floating” in the gabbro. Typically, the dike margins are composed of coarse grained (up to 5 cm long) often plastically deformed clinopyroxene crystals enclosed by fine grained dark green intercumulus material, whose primary composition remains obscure because of its pervasive alteration to chlorite. Locally, the clinopyroxene crystals are aligned parallel to the dike margins indicative of the

7. The Mantle Section

Fig. 13. A sketch of an outcrop of gabbroic feeder dike. Note that the dike is broadly parallel to the mantle foliation, but in the meantime brecciates the peridotite—features that are indicative of semibrittle environment for gabbro emplacement. 59

60

Chapter 1: The Jormua Ophiolite

magma flow in the dike. Towards their core parts the dikes comprise increasingly coarse grained or pegmatoidal clinopyroxene + plagioclase + ilmenite cumulates. In many places the plagioclase has been thoroughly replaced by grossular garnet and epidote, and ilmenite by sphene and rutile as a result of rodingitization and subsequent regional metamorphism. The irregular contacts of the gabbros suggest that the mantle tectonite was undergoing deformation at the time of the emplacement, and the coarse grain size of these rather thin dikes imply their slow cooling at the high ambient temperatures of the host peridotite. Importantly, too, the alteration history of the gabbroic feeder dikes and basaltic “deep dikes” are distinct. The former are often pervasively rodingitized with abundant grossular garnet while the latter are altered (epidotized) but not rodingitized sensu stricto (never garnet). This is believed to be indicative that gabbros intruded fresh (i.e., nonserpentinized) peridotites that later became serpentinized resulting in coeval rodingitization of the enclosed gabbro dikes. In contrast, the absence of rodingitization reactions in the “deep dikes” suggests that they intruded already extensively serpentinized peridotites. Rare earth element patterns of the gabbroic feeders are identical to those of the highlevel gabbros (Fig. 6b). Large ranges in the elemental abundances testify to the strong across-dike fractionation that took place in the dikes and that is broadly similar with the fractionation of the high level gabbros. Importantly, samples from both the gabbro dikes and high level stocks yield similar initial 143 Nd/144Nd values and fit along a well-defined Sm-Nd isochron. The slope of the isochron corresponds to an age (1936 ± 43 Ma) that is equal to the U-Pb zircon ages of both the gabbro dikes and high-level stocks (Peltonen et al., 1998). As a summary, we believe that the internal structure, alteration, and chemical and isotopic composition of these gabbroic feeder dikes imply that they represent feeders for the high-level gabbro bodies. However, because gabbros are frequently cross-cut by EMORB dikes, but are never observed to intrude the basaltic dikes, the gabbros are not directly related to the extrusive rocks in Jormua. More likely, their emplacement preceded (probably by a few million years) that of the sea floor volcanism. This is consistent with the evidence of the rheological properties of mantle at the time of the emplacement of the gabbro dikes (above) which suggests that they intruded the peridotites at relatively high ambient temperatures, perhaps before the mantle was exposed beneath the continental crust by lithospheric detachment faulting. 7.4. “Early” OIB-Type Dikes A distinct suite of fine grained dikes with chemical characteristics similar to oceanic island basalts (OIB) have been observed in the central block only (Figs. 3 and 6a). They occur as subvertical, NNE-trending, < 10–200 cm wide dikes that run subparallel with chromite foliations of the enclosing mantle tectonite. Their mineralogy and LILE element abundances have been strongly modified due to rodingitization. At present they mainly consist of chlorite, ilmenite, magnetite, sphene, apatite, carbonate and trace sulfides. Immobile elements, such as REE, Zr, Nb, Y, and Al can still be applied to determine their origin and led Peltonen et al. (1996) to argue that they represent primitive alkaline melts

7. The Mantle Section

61

(a)

(b)

(c)

(d)

Fig. 14. (a) An outcrop illustrating the cross-cutting relationship of “early” OIB-type dikes and deep dikes (EMORB) within the central block mantle tectonites, (b) clinopyroxenitic cumulate dike intruding mantle tectonite, (c) thick hornblenditic mantle dike (gray) with garnetite and garnet rich veins (mottled), (d) a close-up of garnetite vein with > 50 vol% garnet crystals (white; now pseudomorphosed) growing inside from the conduit wall defining comb-layering. (c) Reprinted with the permission from Journal of Petrology, vol. 39, Oxford Univ. Press.

(ultramafic lamprophyres) from a ∼ 2.3 Ga [Nd(TDM)] plume source. Also “early dikes” are completely devoid of any subduction-related geochemical component. Since abundant EMORB dikes cross-cut the OIB-dikes at high angles, it is evident that they represent distinct episodes of magmatism (Fig. 14a). Preliminary U-Pb zircon ages determined by SIMS from one OIB-dike are consistent with field evidence that emplacement of OIB-dikes indeed preceded that of tholeiitic gabbro magmatism and ocean floor volcanism by some tens of millions of years (Peltonen et al., in preparation). Most likely their emplacement was related to the initial stages of continental rifting. These dikes are among the oldest Precambrian alkaline rocks described in the literature (Blichert-Toft et al., 1996).

62

Chapter 1: The Jormua Ophiolite

7.5. Clinopyroxenite Feeder Dikes Cumulus-textured clinopyroxenite dikes are common within the western block of the Jormua Ophiolite (Figs. 3 and 14b). Such dikes are present but rare in the central block and are completely absent from the eastern block. These dikes are typically from a few dm up to one meter wide having relatively straight and linear sharp contacts against the enclosing mantle tectonites. Locally they occur in thick dike-in-dike swarms or may form small cumulate pods within the mantle tectonite. They are subparallel with the mantle foliation but in detail are discordant. These dikes consist of clinopyroxene ortho- or mesocumulates with only minor intercumulus material and are collectively called “dry” clinopyroxenites to distinguish them from hydrous hornblendite-garnetite veins (next section). Most of the primary clinopyroxene crystals, however, have been retrogressively replaced by fibrous actinolite. The high abundances of, e.g., Mg, Cr and Ni, convex-upward primitive mantle normalized REE patterns (Fig. 15a) and low abundances of incompatible elements all reflect the accumulation of calcic pyroxene and low modal amount of intercumulus melt in these dikes. Particularly close analogies for such mantle dikes can be found from the orogenic lherzolite massifs of the French Pyrenees (e.g., Conquéré, 1971; Bodinier et al., 1987a). Two clinopyroxene cumulate dikes from central and western blocks yielded magmatic, growth-zoned zircons with relatively large spread of 207 Pb/208Pb ages between 3106 and 2718 Ma. Sm-Nd isotope data, however, clearly suggests that the crystallization age of the dikes is Proterozoic. This implies that Archean zircon grains in these dikes are xenocrysts inherited from deeper sources of the continental mantle (Peltonen et al., 2003). Therefore, central and western block mantle tectonites (which are intruded by such dikes) must represent ancient subcontinental lithospheric mantle (SCLM) that became exposed from underneath the Karelian craton during the 1.95 Ga rifting event. Since the clinopyroxenitic mantle dikes bear no evidence of having gone through melting after their formation, the mantle peridotites exposed in the central and western blocks cannot be the source for the Jormua gabbros, dikes or lavas. 7.6. Hornblendite Mantle Dikes and Garnetite Veins In addition to “dry” clinopyroxenites the western block tectonites are characterized by abundant hydrous intrusives. Frequently, the hydrous veins are subparallel with the “dry” dikes, but in detail cross-cut them. They form a somewhat heterogeneous suite of dikes and veins with significant modal and grain size variations within individual dikes. Hydrous mantle dikes include at least the following rock types (in approximate order of abundance): pure medium-grained hornblendites (Fig. 14c), garnetite veins (Fig. 14d), pegmatitic hornblendites, magnetite-ilmenite-zirconolite-rich cumulates and carbonatitic segregations. By their chemical composition the hydrous dikes are more evolved than the clinopyroxenites. Igneous mineralogy is not well-preserved but primitive mantle normalized REE patterns are particularly informative for the origin of hydrous dikes (Figs. 15b, c). These patterns

7. The Mantle Section

63

Fig. 15. Primitive mantle (McDonough and Sun, 1995) normalized REE patterns for cumulustextured alkaline mantle dikes from the Jormua Ophiolite: (a) clinopyroxene cumulate dikes, (b) hornblendites, and (c) garnet-rich veins.

64

Chapter 1: The Jormua Ophiolite

largely reflect the mineralogical composition of the dikes. Hornblendites are characterized by steep fractionated patterns with convex-upward LREE part of the spectrum, while garnetite vein patterns clearly reflect the accumulation of garnet that strongly partitions HREE. These types of dikes, especially the garnet-bearing ones that imply high crystallization pressures, do not occur in the mantle sections of true ophiolites. They are, however, common in orogenic lherzolite massifs (SCLM), such as Lherz and Freychinéde, French Pyrenees (Conquéré, 1971; Bodinier et al., 1987b; Fabriés et al., 1991; Woodland et al., 1996) where they represent high-pressure cumulates of the continental magmatism that passed through the uppermost subcontinental lithospheric mantle. In Jormua, the hydrous dikes are intimately associated also with sporadic occurrences of carbonatite-like vein material. In fact, petrographic observations suggest that there exists a complete sequence from hornblendite veins with intercumulus carbonatite to veins consisting of more than 50% carbonates. This may be indicative that the carbonatitic material represents an extreme differentiation product of the hornblenditeproducing alkaline magma. Carbon isotope values for these carbonatites are typical for upper mantle (unpublished) and clearly distinct from secondary (metamorphic) carbonates of the talc-carbonate rocks. At present, carbonatitic segregations consist of carbonates (> 50 vol%), amphiboles, phlogopite, chlorite, magnetite, apatite, ilmenite, zircon, zirconolite and baddeleyite. Microscopic vugs consisting of carbonatitic material have been described from mantle xenoliths and mantle xenocrysts (e.g., Ionov et al., 1996; Zhang and Liou, 1994) but, to our knowledge, the carbonatitic veins within the Jormua mantle tectonites are the first mesoscopic carbonatite occurrence described from mantle samples. Zircons from one carbonatite vein have been dated by SIMS (Peltonen et al., in preparation). This sample yielded a bimodal distribution of ages. Most of grains record ages close to 1950 Ma that equals the age of the EMORB magmatism in Jormua. Some grains, however, are significantly older being ∼ 2.1 Ga. The results indicate that the igneous age of the carbonatites (and by corollary hydrous dikes in general) is close to 2.1 Ga but that most of the grains were recrystallized, as indicated by their morphology and lack of color, in the 1950 Ma thermal event. Such a time sequence is supported by field observations that imply that the only EMORB dike which is known from the western block cross-cuts not only the “dry” clinopyroxenites but also hydrous dikes. Importantly, the age of this carbonatite vein equals that of the central block OIB-dikes upper in the ophiolite stratigraphy, which indicates that hydrous cumulate dikes of the western block represent deep-seated equivalents of the OIB-type alkaline dikes (Fig. 3).

8. GEODYNAMIC SETTING OF THE JORMUA OPHIOLITE COMPLEX AND EVOLUTION OF THE KARELIAN CONTINENTAL MARGIN The internal stratigraphy of the Jormua Ophiolite Complex (e.g., absence of thick cumulate layers; thin basaltic lid), together with the presence of subcontinental lithospheric mantle imply that Jormua formed at the final stages of continental rifting representing

8. Geodynamic Setting of the Jormua Ophiolite Complex and Evolution of the Karelian Continental Margin 65

the first sea floor to be generated. On the basis of the structure of the Jormua alone, it cannot be judged whether this break-up led to development of a major ocean. Therefore, two alternative paleogeographic settings of origin are possible: (a) either the Jormua Ophiolite Complex formed between the Eastern Finland and Pudasjärvi-Iisalmi Archean complexes (Fig. 1) within a continental rift zone that never developed into major ocean, or (b) Jormua formed within the westernmost passive margin of the craton and has been tectonically transported over the Pudasjärvi-Iisalmi complexes (Fig. 1). The rift zone model (a) seriously contradicts the lithofacies of the associated (“upper Kaleva”) metasediments. Instead, we prefer the passive margin model (b) because there is no evidence of rift sedimentation around 1.95 Ga within the Kainuu Schist Belt, and because the Jormua Ophiolite (as Karelidic ophiolite fragments in general) has been obducted together with deep water slope-rise graywackes (Koistinen, 1981; Kontinen and Sorjonen-Ward, 1991) that were deposited less than 1920 ± 10 Ma ago (Huhma, 1986; Claesson et al., 1993). The monotonous deep water turbiditic lithofacies, complete absence of any type of volcanic interbeds or synsedimentary intrusions, and the presence of 1.92–1.97 old sedimentary source component from a remote unknown terrain (Claesson et al., 1993) clearly exclude the origin of “upper Kaleva” as a continental rift fill. In the passive margin model the Jormua Ophiolite can be envisaged as screens of oceanic lithosphere within a recently developed (∼ 1950 Ma) continental margin which, following the thermal subsidence, became covered by the “upper Kaleva” slope-rise turbidites (Fig. 16). This model implies that both the Jormua Ophiolite and the somewhat older (early rifting) Otanmäki alkaline gneisses are allochthonous and that their present appearance as mega-boudinaged slices within basement and cover sediment slices is due to imbrication and shearing. The deposition of the upper Kaleva metasediments as well as ophiolite obduction took place somewhere between 1920 ± 20 Ma and 1871 ± 5 Ma. The lower limit is provided by the youngest detrital zircons in these metasediments (Claesson et al., 1993), and the upper limit by the age of the oldest granite intruding the “upper Kaleva” schists (Huhma, 1986). Evidence of the flat-lying nappes that were responsible for the thrusting of the Jormua klippe onto the craton some 1.9 Ga ago has mostly been destroyed by the subsequent regional deformation. Though there is evidence of a 2 Ga mantle plume in the south-east are Fennoscandian Shield (e.g., Puchtel et al., 1998), the limited amount of 1950 Ma volcanism at the westernmost margin of the Karelian Craton suggests that it developed as a non-volcanic continental margin. The presence of both exposed continental lithospheric mantle and asthenosphere-derived igneous rocks in the Jormua Ophiolite Complex suggest a heterogeneous and asymmetric stretching and rifting process, which resulted in delamination and exposure of the subcontinental lithospheric mantle in an early stage of the ocean opening (Fig. 16; Whitmarsh et al., 2001). Phanerozoic ophiolites similar to the Jormua include the Ligurian/western Alps ophiolites that also have been related to continental break-up by asymmetric passive rifting (e.g., Lemoine et al., 1987; Rampone and Piccardo, 2001, and references therein). Modern environments that may be analogous to the provenance of the Jormua include the West Iberian non-volcanic margin where the continent-ocean transition zone consists of partially serpentinized continental mantle tectonites veined by

66

Chapter 1: The Jormua Ophiolite

Fig. 16. Proposed tectonic setting for the protolith of the Jormua Ophiolite Complex within an ocean-continent transition zone. Modified after Whitmarsh et al. (2001) emphasizing the situation at the present West Iberia passive margin.

EMORB dikes, gabbros and pyroxenites akin to those in Jormua (Chian et al., 1999; Cornen et al., 1999). Sedimentary deposits along this ∼ 1.95 Ga non-volcanic continental margin that could be related to the break-up/ocean opening are rare. One of the reasons for the obvious sediment starved nature is the peneplain nature of the Karelian continent at that time and thus lack of higher mountainous terrains along the break-up zone to supply large quantities of detritus. Furthermore, there is evidence that suggests that the whole Karelian Craton may have been submerged 1980 Ma ago shortly before the opening (Puchtel et al., 1998). The absence or very thin Jatuli-type sedimentary cover along the continental margin indicates that the break up zone probably rose above sea level and was subjected to some erosion before its rifting. Despite the obvious sediment starved nature of the opening, there still is puzzlingly little evidence of the break-up along the craton margin, and most of this evidence comes from thrusted units. One explanation could be that the break-up zone and inferred thinned margin is buried below the Svecofennian terrane in the west of the suture (Fig. 1). This scenario is however not supported by isotope data from syn- and postcollisional granites to the west of the exposed suture (Huhma, 1986; Lahtinen and Huhma, 1997; Rämö et al., 2001). Therefore, perhaps the Svecofennia-Archaean suture represents a major strike-slip fault

9. Epilogue

67

along which much of the thinned continental margin and underlying mantle was removed already before the amalgamation of the Svecofennian terrane. In this case the thrusting of the eastern Finland ophiolites and related units may have been a tectonic episode that significantly preceded and was unrelated to the actual Svecofennian collision. The present tectonic setting of the Jormua Ophiolite Complex (within klippes transported far from their inferred tectonic root west of the suture) provides few clues for unraveling the possible causes of detachment of the Jormua from the seafloor and its subsequent obduction on to the continent. One and purely speculative suggestion would be that maybe there was a shortlived event of attempted subduction beneath the continental margin preceding the obduction. This may have resulted in uplifting of the thin transitional crust that in turn facilitated the bulldozing of ultramafic seafloor and “upper Kaleva” turbidites from somewhat more distant oceanic realm onto the continent. Continental mantle in rift zones lies open to serpentinization (e.g., Perez-Gussinye and Reston, 2001) that could have made the provenance of Jormua boyonant and facilitated the obduction.

9. EPILOGUE The Jormua Ophiolite is a truly unique mafic-ultramafic rock complex that consists of two distinct components: (1) Archean subcontinental lithospheric mantle that became exposed on the seafloor due to detachment faulting, and (2) a suite of convective mantlederived alkaline and tholeiitic igneous rocks that intruded the SCLM before and during the continental rifting between 2.1 and 1.95 Ga. The mantle exposed in Jormua thus mainly represents the uppermost sub-crustal lithosphere of the Archean Karelian Craton. Intriguingly, mantle xenoliths representing this same continental mantle have been recently recovered from ∼ 500 Ma kimberlite pipes that intrude the craton margin 100 km SSE of Jormua. The mantle xenolith suite consists garnet peridotites and mantle eclogites derived from a depth range of ∼ 110–240 km (Peltonen et al., 1999; Kukkonen and Peltonen, 1999). Combined, an astonishing window to understand the evolution of the Karelian mantle for over ∼ 3 Ga period and over its entire vertical depth seems to be at hand. Unfortunately, the high degree of alteration and metamorphism of the Jormua mantle samples decreases their value; primary minerals are generally no longer available and the multistage alteration hampers study of the most subtle geochemical aspects. Nevertheless, relatively good exposure of glacially polished outcrops over an area of 50 km2 will provide many new insights to the structure and composition of Precambrian mantle.

REFERENCES Anonymous, 1972. Penrose field conference on ophiolites. Geotimes 17, 24–25. Barnes, S.-J., Boyd, R., Korneliussen, A., Nillsson, L.-P., Often, M., Pedersen, R.B., Robins, B., 1988. The use of mantle normalization and metal ratios in discriminating between the effects of partial melting, crystal fractionation and sulphide segregation on platinum-group elements, gold,

68

Chapter 1: The Jormua Ophiolite

nickel and copper: examples from Norway. In: Prichard, H.M. (Ed.), Geo-Platinum. Elsevier, pp. 113–143. Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Blichert-Toft, J., Arndt, N.T., Ludden, J.N., 1996. Precambrian alkaline magmatism. Lithos 37, 97– 111. Bodinier, J.-L., Guiraud, M., Fabriés, J., Dostal, J., Dupuy, C., 1987a. Petrogenesis of layered pyroxenites from Lherz, Freychinéde and Prades ultramafic bodies (Ariége, French Pyrenees). Geochimica et Cosmochimica Acta 51, 279–290. Bodinier, J.-L., Fabriès, J., Lorand, J.-P., Dostal, J., Dupuy, C., 1987b. Geochemistry of amphibole pyroxenite veins from the Lherz and Freychinède ultramafic bodies (Ariège, French Pyrenees). Bulletin de Mineralogie 110, 345–358. Boynton, W.V., 1984. Geochemistry of rare earth elements: meteorite studies. In: Henderson, P. (Ed.), Rare Earth Element Geochemistry. Elsevier, pp. 63–114. Cannat, M., 1993. Emplacement of mantle rocks in the seafloor at mid-ocean ridges. Journal of Geophysical Research 98, 4163–4172. Chian, D., Louden, K.E., Minshull, T.A., Whitmarsh, R.B., 1999. Deep structure of the oceancontinent transition in the southern Iberia Abyssal Plain from seismic refraction profiles: Ocean Drilling Program (Legs 149 and 173) transect. Journal of Geophysical Research 104, 7443–7462. Claesson, S., Huhma, H., Kinny, P.D., Williams, I.S., 1993. Svecofennian detrital zircon ages— implications for the Precambrian evolution of the Baltic Shield. Precambrian Research 64, 109– 130. Conquéré, F., 1971. Les pyroxénolites à amphibole et les amphibolites associées aux lherzolites du gisement de Lherz (Ariège, France): un example du rôle de l’eau au cours de la cristallisation fractionnée des liquides issus de la fusion partielle de lherzolites. Contributions to Mineralogy and Petrology 33, 32–61. Cornen, G., Girardeau, J., Monnier, C., 1999. Basalts, underplated gabbros and pyroxenites record the rifting process of the West Iberian margin. Mineralogy and Petrology 67, 111–142. Dann, J.C., 1991. Early Proterozoic ophiolite, central Arizona. Geology 19, 590–593. Eckstrand, O.R., 1975. The Dumont serpentinite; a model for control of nickeliferous opaque mineral assemblages by alteration reactions in ultramafic rocks. Economic Geology 70, 183–201. Fabriés, J., Lorand, J.-P., Bodinier, J.-L., Dupuy, C., 1991. Evolution of the upper mantle beneath the Pyrenees: evidence from the orogenic spinel lherzolite massifs. Journal of Petrology, Special Lherzolites Issue, 55–76. Fisk, M.R., Bence, A.E., 1980. Experimental crystallization of chrome spinel in FAMOUS basalt 527-1-1. Earth and Planetary Science Letters 48, 111–123. Gaál, G., Gorbatschev, R., 1987. An outline of the Precambrian development of the Baltic Shield. Precambrian Research 35, 15–52. Gole, M.J., Barnes, S.J., Hill, R.E.T., 1987. The role of fluids in the metamorphism of komatiites, Agnew nickel deposit, Western Australia. Contributions to Mineralogy and Petrology 96, 151– 162. Hanski, E.S., 1997. The Nuttio serpentinite belt, central Lapland: An example of Palaeoproterozoic ophiolitic mantle rocks in Finland. Ofioliti 22, 35–46. Huhma, H., 1986. Sm-Nd, U-Pb and Pb-Pb isotopic evidence for the origin Early Proterozoic Svecofennian crust in Finland. Geological Survey of Finland Bulletin 337, 47.

References

69

Ionov, D.A., O’Reilly, S.Y., Genshaft, M.G., Kopylova, Y.S., 1996. Carbonate-bearing mantle peridotite xenoliths from Spitsbergen: phase relations, mineral compositions and trace element residence. Contributions to Mineralogy and Petrology 125, 375–392. Irvine, T.N., Baragar, W.R.A., 1971. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences 8, 523–548. Kärki, A., Laajoki, K., 1995. An interlinked system of folds and ductile shear zones—late stage Svecokarelian deformation in the central Fennoscandian Shield, Finland. Journal of Structural Geology 17, 1233–1247. Kohonen, J., 1995. From continental rifting to collisional crustal shortening—Paleoproterozoic Kaleva metasediments of the Höytiäinen area in North Karelia, Finland. Geological Survey of Finland Bulletin 380, 79. Koistinen, T., 1981. Structural evolution of an early Proterozoic sratabound Cu-Co-Zn deposit, Outokumpu, Finland. Transactions of the Royal Society of Edinburgh, Earth Sciences 72, 115–158. Koistinen, T.J., Stephens, M.B., Bogatchev, V., Nogulen, Ø., Wennerström, M., Korhonen, J., Geological map of the Fennoscandian Shield, scale 1:2 000 000. Geological Surveys of Finland, Norway and Sweden, and the North-West Department of Natural Resources of Russia. Kontinen, 1987. An Early Proterozoic ophiolite—the Jormua mafic-ultramafic complex, northeastern Finland. Precambrian Research 35, 313–341. Kontinen, A., Sorjonen-Ward, P., 1991. Geochemistry of metagraywackes and metapelites from the Palaeoproterozoic Nuasjärvi group, Kainuu schist belt and the Savo Province, north Karelia: implications for provenance, lithostratigraphic correlation and depositional setting. Geological Survey of Finland Special Paper 12, 21–22. Kontinen, A., 1998. Geological map of the Jormua area 1:50 000. Geological Survey of Finland Special Paper 26, Appendix. Kontinen, A., Peltonen, P., 1998. Excursion to the Jormua ophiolite complex. In: Hanski, E., Vuollo, J. (Eds.), International Ophiolite Symposium and Field Excursion: Generation and Emplacement of Ophiolites through Time, August 10–15, 1998, Oulu, Finland. Geological Survey of Finland Special Paper 26, 69–89. Korsman, K., Korja, T., Pajunen, M., Virransalo, P., 1999. The GGT/SVEKA transect: structure and evolution of the continental crust in the Paleoproterozoic Svecofennian orogen in Finland. International Geology Review 41, 287–333. Kukkonen, I.T., Peltonen, P., 1999. Xenolith-controlled geotherm for the central Fennoscandian Shield—implications for the lithosphere-asthenosphere relation. Tectonophysics 304, 301–315. Kusky, T.M., Li, J.H., Tucker, R., 2001. The Archean Dongwanzi ophiolite complex, North China Craton: 2.505-billion-year-old oceanic crust and mantle. Science 292, 1142–1145. Lahtinen, R., Huhma, H., 1997. Isotopic and geochemical constraints on the evolution of the 1.93– 1.79 Ga Svecofennian crust and mantle in Finland. Precambrian Research 82, 13–34. Lemoine, M., Tricart, P., Boillot, G., 1987. Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): In search of a genetic model. Geology 15, 622–625. Louden, K., Lau, H., 2001. Insights from scientific drilling on rifted continental margins. Geoscience Canada 28, 187–195. Liipo, J., Vuollo, J., Nykänen, V., Piirainen, T., Pekkarinen, L., Tuokko, I., 1995. Chromites from the early Proterozoic Outokumpu-Jormua Ophiolite Belt: a comparison with chromites from Mesozoic ophiolites. Lithos 36, 15–27. McDonough, W.F., Frey, F.A., 1989. Rare earth elements in upper mantle rocks. In: Lipin, B.R., McKay, G.A. (Eds.), Geochemistry and Mineralogy of Rare Earth Elements. In: Reviews in Mineralogy, vol. 21, pp. 99–145.

70

Chapter 1: The Jormua Ophiolite

McDonough, W.F., Sun, S.-s., 1995. The composition of the Earth. Chemical Geology 120, 223–253. McElduff, B., Stumpfl, E.F., 1991. The chromitite deposits of the Troodos Complex, Cyprus— evidence for the role of a fluid phase accompanying chromite formation. Mineralium Deposita 26, 307–318. O’Hanley, D.S., 1996. Serpentinites. Records of Tectonic and Petrological History. Oxford Univ. Press, p. 277. Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plate boundaries. In: Thorpe, R.S. (Ed.), Andesites; Orogenic Andesites and Related Rocks. John Wiley & Sons, pp. 525–548. Pearce, J.A., 1983. Role of subcontinental lithosphere in magma genesis at active continental margins. In: Hawkesworth, C.J., Norry, M.J. (Eds.), Continental Basalts and Mantle Xenoliths. Shiva Publishing, UK, pp. 230–249. Pearce, J.A., Harris, N.B.W., Tindle, A.G., 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology 25, 956–983. Peltonen, P., Kontinen, A., Huhma, H., 1996. Petrology and geochemistry of metabasalts from the 1.95 Ga Jormua Ophiolite, northeastern Finland. Journal of Petrology 37, 1359–1383. Peltonen, P., Kontinen, A., Huhma, H., 1998. Petrogenesis of the mantle sequence of the Jormua Ophiolite (Finland): Melt migration in the upper mantle during Palaeoproterozoic continental break-up. Journal of Petrology 39, 297–329. Peltonen, P., Huhma, H., Tyni, M., Shimizu, N., 1999. Garnet-peridotite xenoliths from kimberlites of Finland: nature of the continental mantle at Archaean craton-Proterozoic mobile belt transition. In: Gurney, J.J., Gurney, J.L., Pascoe, M.D., Richardson, S.H. (Eds.), Proceedings of the 7th International Kimberlite Conference, vol. 2, pp. 664–676. Peltonen, P., Mänttäri, I., Huhma, H., Kontinen, A., 2003. Archean zircons from the mantle: the Jormua Ophiolite revisited. Geology 31, 645–648. Perez-Gussinye, M., Reston, T.J., 2001. Rheological evolution during extension at nonvolcanic rifted margins: Onset of serpentinisation and development of detachments leading to continental breakup. Journal of Geophysical Research 106, 3961–3976. Puchtel, L.S., Arndt, N.T., Hofmann, A.W., Haase, K.M., Kröner, A., Kulikov, V.S., Kulikova, V.V., Garbe-Schönberg, C.-D., Nemchin, A.A., 1998. Petrology of mafic lavas within the Onega plateau, central Karelia: evidence for 2.0 Ga plume-related continental crustal growth in the Baltic Shield. Contributions to Mineralogy and Petrology 130, 134–153. Rampone, E., Piccardo, G.B., 2001. The ophiolite-oceanic lithosphere analogue: new insights from the Northern Apennines (Italy). In: Dilek, Y., Moores, E.M., Elthon, D., Nicholas, A. (Eds.), Ophiolites and Oceanic Crust: New Insights from Field Studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 21–34. Rämö, O.T., Vaasjoki, M., Mänttäri, I., Elliot, B.A., Nironen, M., 2001. Petrogenesis of the postkinematic magmatism of the Central Finland Granitoid Complex. Journal of Petrology 42, 1971– 1993. Roberts, S., Neary, C., 1993. Petrogenesis of ophiolitic chromitite. In: Prichard, H.M., Alabaster, T., Harris, N.B.W., Neary, C.R. (Eds.), Magmatic Processes and Plate Tectonics. Geological Society Special Publication 76, 257–272. Scott, D.J., Helmstaedt, H., Bickle, M.J., 1992. Purtuniq ophiolite, Cape Smith Belt, northern Quebec, Canada: a reconstructed section of Early Proterozoic oceanic crust. Geology 20, 173–176. Tsuru, A., Walker, R.J., Kontinen, A., Peltonen, P., Hanski, E., 2000. Re-Os isotopic systematics of the 1.95 Ga Jormua Ophiolite Complex, northeastern Finland. Chemical Geology 164, 123–141.

References

71

Tuisku, 1997. An introduction to Palaeoproterozoic (Svecofennian) regional metamorphism in Kainuu and Lapland, Finland. In: Evins, P., Laajoki, K. (Eds.), Archaean and Early Proterozoic (Karelian) Evolution of the Kainuu-Peräpohja Area, Northern Finland. Res Terrae, Ser. A 13, 27–31. Vuollo, J., Piirainen, T., 1989. Mineralogical evidence for an ophiolite from the Outokumpu serpentinites in North Karelia, Finland. Bulletin of the Geological Society of Finland 61, 95–112. Whitmarsh, R.B., Manatschai, G., Minshull, T.A., 2001. Evolution of magma-poor continental margins from rifting to seafloor spreading. Nature 413, 150–154. Will, T.M., Powell, R., Holland, T.J.B., 1990. A calculated petrogenetic grid for ultramafic rocks in the system CaO-FeO-MgO-Al2 O3 -SiO2 -CO2 -H2 O at low pressures. Contributions to Mineralogy and Petrology 105, 347–358. Woodland, A.B., Kornprobst, J., McPherson, E., Bodinier, J.-L., Menzies, M.A., 1996. Metasomatic interactions in the lithospheric mantle: petrologic evidence from the Lherz massif, French Pyrenees. Chemical Geology 134, 83–112. Zhang, R.Y., Liou, J.G., 1994. Significance of magnesite paragenesis in ultrahigh-pressure metamorphic rocks. American Mineralogy 79, 397–400.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Published by Elsevier B.V.

73

Chapter 2

THE 1.73 GA PAYSON OPHIOLITE, ARIZONA, USA J.C. DANN 90 Old Stow Road, Concord, MA 01742, USA

The 1.73 Ga Payson Ophiolite is a shallow-dipping, layered sequence of coeval gabbro, sheeted dikes, and submarine volcanic rocks, partly disjointed by later intrusion and deformation. The sheeted dike complex is spectacularly exposed as cliffs and water-polished outcrop in many shallow canyons. Gabbro-dike mingling and mutual intrusion attest to rooting of the sheeted dike complex in the underlying gabbro. A stratigraphically continuous zone of intense alteration marks the transition from sheeted dikes to submarine volcanics. A tonalite/dacite suite occurs as rare lavas and as dikes and hypabyssal plutons, mutually intrusive with the basaltic sheeted dikes and gabbro. An older basement complex occurs as roof pendants in gabbro and screens in the sheeted dike complex. An actualistic tectonic model of an intra-arc basin formed by seafloor spreading along an arc-parallel strike-slip fault system explains the origin of the Payson Ophiolite, its emplacement within the arc, and accretion to North America during the ca. 1.70 Ga Yavapai Orogeny.

1. INTRODUCTION The 1.73 Ga Payson Ophiolite (Dann, 1991, 1992, 1997a, 1997b) is about 90 km northeast of Phoenix within the Yavapai-Mazatzal orogenic belt of central Arizona (Fig. 1A). It is one of only a few Early Proterozoic ophiolites known worldwide with a well developed sheeted dike complex and the horizontally layered structure characteristic of Phanerozoic ophiolites and modern oceanic crust (see Moores, 2002, for review). The sheeted dike complex and transitions to overlying volcanic and underlying gabbroic rocks are well exposed as water-polished outcrops in easily accessed canyons (stop one in field trip guide, Karlstrom et al., 1990). Most of the ophiolite remains gently dipping. Fold-and-thrust deformation is localized, and no penetrative fabric occurs in the ophiolite. Even the overlying turbidites have only a locally developed, spaced fracture cleavage. Although low greenschist-grade metamorphism affected the ophiolite, large areas of the gabbro are 95% igneous minerals. The quality of exposure and preservation of primary features facilitated detailed structural mapping and petrological, geochemical, and geochronological analyses of the Payson Ophiolite (PO). The purpose of this paper is summarize the work done and discuss its contribution to our understanding of the mechanics of seafloor spreading, plate tectonics, and continental assembly during the Early Proterozoic. DOI: 10.1016/S0166-2635(04)13002-8

74

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Fig. 1. (A) Geologic map showing the location of the Payson Ophiolite within the Mazatzal crustal block in the Early Proterozoic of central Arizona (modified from Karlstrom et al., 1990, and Anderson, 1989; ‘T’ is Tonto Creek). (B) Map showing the location of central Arizona within the 1.6–1.8 Ga orogenic belt of North America (from Hoffman, 1989). (C) Tectonostratigraphic columns comparing the Ash Creek and Mazatzal crustal blocks divided by the Moore Gulch shear zone.

2. Regional Setting

75

First, we need to feel confident that the complicated map pattern (Fig. 2) actually conceals the layered structure that distinguishes an ophiolite from other crustal sections. Ophiolites are defined by their pseudostratigraphy or horizontally layered structure of mantle tectonite, gabbro, sheeted dikes, and submarine volcanic and sedimentary rocks (Moores, 1982). Importantly, the transitional zones between lithologic layers indicate that the layers were forming simultaneously (unlike real stratigraphy). In addition, the distribution of tonalites, hydrothermal alteration, and chemical sediments record important processes ongoing during development of the crust. In most greenstone belts, deformation and intrusion of late granitoids has dislocated submarine volcanic sequences from their hypabyssal equivalents. What makes the PO special is that this connection is intact. In addition, we need to distinguish between (1) intrusion of a dike swarm into an older plutonic complex and (2) rooting of sheeted dikes in coeval gabbro. These questions were addressed by detailed structural mapping and analysis of the field relations (Dann, 1992). Second, the story of the origin and emplacement of the ophiolite is recorded in its relationship to the rest of the orogenic belt (Dann and Bowring, 1997). Ophiolites commonly occur within terranes or shear zone-bound crustal blocks that record tectonostratigraphic histories and original tectonic settings that are incompatible with neighboring terranes (Fig. 1C). From the comparative tectonostratigraphic histories, the geochemistry of the ophiolitic magma, the geometry of extensional and convergent tectonism, and with reference to modern examples, an actualistic tectonic model can be developed that provides a useful predictive framework for understanding the creation and assembly of Proterozoic crust in the southwestern United States.

2. REGIONAL SETTING The Payson Ophiolite is in the Mazatzal crustal block bound by the Moore Gulch shear zone on the northwest and a belt of post-assembly granites to the southeast (Fig. 1A). This 50–60 km wide crustal block occurs at the eastern end of a 600 km transect of the Proterozoic orogenic belt, exposed in the Transition Zone between the Colorado Plateau and the Basin and Range Province. In central Arizona this orogenic belt consists of submarine volcanic and volcaniclastic rocks that host massive sulfide deposits and are intruded by granitoid plutons. These lithologic associations are interpreted by most workers to represent magmatic arcs (e.g., Anderson, 1989). The magmatic arcs formed over a 40 m.y. period from ca. 1.75 to 1.71 Ga (Bowring et al., 1991), locally involving older crust. From the high estimated rate of crustal growth, the predominance of juvenile magmatic arc rocks, and the juxtaposition of distinct terranes, Karlstrom and Bowring (1988) proposed that the orogenic belt formed by accretion of arc terranes along a convergent plate boundary. Most of the deformation along the block boundaries reflects post-assembly crustal shortening and differential uplift (Bowring and Karlstrom, 1990). Consequently, the role of the block boundaries during the assembly of terranes remains speculative, but it may be particularly important in the origin and emplacement of the Payson Ophiolite.

76

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Fig. 2. Geologic Map and cross sections of the Payson Ophiolite (contacts outside the ophiolite from Wrucke and Conway, 1987). Sheeted dikes are best exposed in American Gulch (AG), Rattlesnake Canyon, (RC), St. John’s Creek (SJ), and along the east flank of the Mazatzal Mountains (EF). The Larson Spring Formation of the basement complex, intruded and overlain by the ophiolite, are best exposed at the East Verde River (a), Crackerjack Mine (b), Larson Spring (c), Center Creek (d), and Gisela (e).

3. Payson Ophiolite

77

3. PAYSON OPHIOLITE The distribution and orientation of dikes reflects the structure of the ophiolite. Sheeted dikes underlie submarine volcanic rocks on the west side of the map (EF, Fig. 2) and overlie gabbro east of the Tertiary valley (AG, RC, Fig. 2). The sheeted dikes dip about 75 degrees to the northeast where gabbro devoid of dikes occurs (‘rv’, Fig. 2). The northeastto-southwest sequence of gabbro, sheeted dikes, and volcanic rocks, combined with the 75 degree northeast dip of the dikes, indicates that the ophiolite pseudostratigraphy dips about 15 degrees to the southwest (cross section, Fig. 2). This reconstruction, based on a perpendicular relationship between dikes and the pseudostratigraphy, is justified by the angular relationships between bedded basaltic volcanic rocks and (1) underlying sheeted dikes (EF, Fig. 2) and (2) bedded felsic volcaniclastic rocks in the basement complex (Dann, 1997a). Parallel to the ophiolite pseudo-stratigraphy, the ca. 1.70 Ga Payson granite intruded its gabbroic roof as a sheet dipping 15–25 degrees to the southwest (Conway et al., 1987). Due to the shallow dip of both the Payson Granite and the ophiolite, the gabbro-norite of Round Valley (‘rv’, Fig. 2) is the deepest level of the ophiolite exposed. The mantle section of the PO remains hidden beneath the Payson Granite. Tertiary normal faults created the sediment-filled valley (Fig. 2) and the complicated map pattern by juxtaposing different levels of the ophiolite pseudostratigraphy. Description of the ophiolite proceeds from gabbro to volcanic rocks with particular emphasis on the transitions that establish the contemporaneity of the dikes with both the gabbro and volcanic section. 3.1. Gabbro Most of the exposed ophiolite is gabbro that forms a pseudostratigraphic layer of plutons beneath the sheeted dike complex. The Round Valley gabbro (‘rv’, Fig. 2) in the northeastern part of the map area is coarse-grained, completely devoid of mafic dikes, and the least altered (1–5%) of all mafic rocks in the ophiolite. The NE-SW cross section shows that the RV solidified about 1.5–2 km below the sheeted dike complex (inset, Fig. 2). Hornblende gabbro mantles the RV, and toward the northwest-trending transition to the sheeted dike complex, it becomes increasingly fractured, altered, and intruded by mafic dikes. Just below the sheeted dike complex, distinct plutons are resolved, based on texture, flow fabrics, and mineralogy, especially the isotropic quartz diorite (‘d’, Fig. 2) that yielded a U/Pb zircon age of 1.73 Ga (Bowring et al., 1991). Gabbro also intruded the sheeted dike complex and the volcanic section as sill-like bodies (e.g., ‘sm’, Fig. 2). All the gabbroic rocks contain hornblende, locally with coarse-grained, ophitic and poikolitic textures (< 4 cm). Elongate plagioclase crystals and folded modal layering define a lineation and/or foliation in the gabbros below the sheeted dike complex. The lineation is primarily orthogonal to the dikes, indicating flow of crystal-rich magma parallel to the direction of extension (Dann, 1997a, Figs. 5, 6). However, the orientation of the layering varies and is locally parallel to the dikes. Structural analysis, based on the orientation of bedding in the basement complex, reveals that the plutonic core of the ophiolite was not tectonically rotated and that magmatic layering with locally steep dips is a primary feature (Dann, 1997a, Fig. 13).

78

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Rothery (1983) and Greene (1989) reached similar conclusions for the Semail and Lokken ophiolites, respectively. In general, the magmatic layering is less reliable for structural reconstructions than the orientation of sheeted dikes and distribution of the pseudostratigraphy. 3.2. Sheeted Dike Complex The sheeted dike complex is spectacularly exposed in four major canyons as continuous water-polished outcrop, cliffs of parallel slabs of dikes, and hillsides of dikes standing in relief (Fig. 3A). As a result of the overlapping and nearly parallel intrusion, most dikes have well defined chilled margins against other dikes (Figs. 3B, C), and dike splitting and low-angle cross cutting displaced segments of early dikes over tens of meters. Over 600 m of measured section permit an estimate of the proportions of mafic and felsic dikes and screens of gabbro and granitic basement and other features that distinguish one area of sheeted dikes from another (see Dann, 1997a, Fig. 8, and Dann, 1997b, Fig. 3, for examples of sections). The most pronounced difference between the top and bottom of the sheeted dike complex is the 3-fold increase in the average width of dikes (e.g., compare Fig. 4A and Fig. 5E). The depth-thickness relationship inspired a new model for the vertical development of sheeted dike complexes (Dann, 1997b). No complete section from top to bottom of the sheeted dike complex is exposed. The base of the sheeted dike complex and transition to the underlying gabbro is best exposed in the Rattlesnake Canyon area (RC, Fig. 2). The top of the sheeted dike complex and transition to submarine volcanic rocks is only exposed on the west side of the Tertiary valley (EF, Fig. 2). American Gulch (AG, Fig. 2) provides the best display of the intrusive sequence of mafic and dacitic dikes into the basement screens (Figs. 3B–D). Analysis of dike widths places AG at an intermediate level within the sheeted dike complex. These three areas of sheeted dikes form the corners of a triangle, giving the appearance that the sheeted dike complex is not a continuous layer. However, the granitic pluton in the middle of the triangle (SJ, Fig. 2) contains a large roof pendant of 100% sheeted dikes that testify to the original continuity of the layer of sheeted dikes that was eroded off the top of the pluton. 3.3. Transition from Gabbro to Sheeted Dikes Best exposed in the Rattlesnake Canyon area (RC, Fig. 2), the transition from gabbro to sheeted dikes trends northwest, parallel to the trend of the dikes. Over about 100 m, dikes increase from < 50% in gabbro to > 90% with gabbroic screens. The dikes are generally thick and coarse-grained (Figs. 4A, B) with tonalitic dikes up to 15 m thick. Early thick dikes have weakly chilled contacts that locally mingle with gabbroic screens (Fig. 4B). The transition zone contains small dioritic intrusions and locally developed intrusion breccias. Some screens contain a mixture of coarse-grained gabbroic material and finer-grained dike material that is interlayered parallel to a flow foliation/lineation (Fig. 4C). Xenoliths of porphyritic basalt aligned in the flow foliation are identical to dikes cutting the gabbro

3. Payson Ophiolite

79

Fig. 3. Sheeted dikes of American Gulch (AG, Fig. 2). (A) View looking northwest down into the gulch. Paler dike (V) is a 3 m thick dacitic dike that is also on outcrop map (D). (B) Water-polished outcrop of sheeted dikes. Basaltic dikes (b) intrude a dacitic dike (x) that intrudes a granitic basement screen (g). (C) One-way chilling (arrows) of basaltic dikes against porphyritic dike (scale bar is 1 cm). (D) Outcrop map of sheeted dikes showing 3 dacitic dikes and 18 basaltic dikes.

80

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

(Fig. 4D). In the gabbro below the sheeted dikes, mafic dikes are pulled apart parallel to the flow foliation, forming trains of enclaves with deformed shapes and mingled contacts. In addition, the gabbro contains conjugate shear bands locally cut by dikes, indicating that the gabbro underwent hot sub-solidus extension consistent with the dikes (see Dann, 1997a, Fig. 6). Gabbro-dike mingling, mutual intrusion, and stretching of xenoliths parallel to the flow foliation in the gabbro indicate that the dikes are rooted in coeval gabbro (see Dann, 1997b, Fig. 6, for summary figure). Flowing gabbroic crystal mush cannibalized the base of the dikes, forming trains of deformed enclaves and mixed dike-gabbro layers along the direction of flow. The transition represents a fluctuating rheological boundary between brittle fracture with dike intrusion and fluid flow of crystal-rich gabbroic magma. This process caused a stepwise decrease in the abundance of dikes with depth, such that the sheeted dike complex bottoms out and dikes are completely absent 1–1.5 km below the sheeted dike complex. Similar features that attest to the dynamic process of dikes rooting in the underlying gabbro occur in many ophiolites (Pederson, 1986; Furnes et al., 1988; Nicholas and Boudier, 1991; Skjerlie and Furnes, 1996). 3.4. Volcanic Section The volcanic section is a thin 400–500 m thick sequence that lies between overlying turbidites (Fig. 5A) and underlying sheeted dikes (Fig. 5E) along the west side of the map area (EF, section Y–Y , Fig. 2). The section consists mostly of basaltic sheet flows with rare outcrops of pillowed flows. The pillows have a 2–3 cm thick band of vesicles inside well-defined selvages. Flow tops are readily recognizable by large amygdules, breccia, jasper infillings, and interflow sediments (Fig. 5B). Besides basaltic flows, the volcanic section contains an auto-brecciated dacitic flow and associated debris-flow deposits. Clastic interflow sediments include volcanic conglomerates, greywacke, and tuffs. Graded beds and scour-and-fill structures indicate consistent northwest younging (Fig. 5B). Chemical interflow sediments include thick lenses of magnetite-rich banded iron formation, jasper, and chert. Sediments are locally deformed by the weight of overlying flows (Fig. 5B). The base of the volcanic section is intruded by thin basaltic sills and dikes that are distinguished by their cross-cutting relations or vesicle layering parallel to chilled margins. Although the paucity of pillowed flows is unusual for a submarine volcanic section, the thickness of the volcanic section (400–500 m) is typical of many ophiolites (usually < 1 km;

Fig. 4. Base of sheeted dike complex. (A) Two thick mafic dikes (arrows, 4 m wide; map folder at top for scale) in the sheeted dike complex of Rattlesnake Canyon (RC, Fig. 2). (B) Mingled and poorly chilled contact (arrow) between mafic dike and gabbroic screen (gb; pen for scale). (C) Gabbroic screen (gb) in sheeted dikes (arrows point to chills), showing early dike material (dark and fine-grained, between arrows) mingled with, and drawn out along the flow foliation in, the gabbro (1 m across field of view). (D) Porphyritic dike xenoliths elongate parallel to the flow foliation of the gabbro are identical to porphyritic dikes cutting the gabbro (pen for scale).

3. Payson Ophiolite

81

82

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Fig. 5. (A) Graded beds of the turbidite sequence overlying the ophiolite (arrow points in younging direction). (B) Interflow clastic sediment, graded from course sand above the amygdaloidal flow top (arrow base) to pale tuff that penetrates crack in overlying flow (arrow tip). (C) Silicified mafic rock of the altered transition from sheeted dikes to volcanic flows (coin for scale). Silica veining (‘s’) outlines pseudo-breccia texture (‘p’). (D) Schematic column showing increasing alteration (white) at the top of the sheeted dike complex. (E) Relatively unaltered sheeted dikes, several 100 m below the volcanic section, with an average width of about 80 cm.

3. Payson Ophiolite

83

Moores, 1982) and sections of oceanic crust exposed on the ocean floor (< 500 m; Francheteau et al., 1992). 3.5. Transition from Sheeted Dikes to Submarine Basalt The transition from sheeted dikes to volcanic rocks is marked by a cliff-forming, stratigraphically continuous, siliceous zone. This transitional zone and the overlying volcanic rocks are steeply dipping to overturned along the east flank of the Mazatzal Mountains (EF, section Y–Y , Fig. 2), where canyons provide cross-sectional exposures. The volcanic section has a sharp lower contact with the massive siliceous zone. In contrast, the sheeted dike complex has a gradational upper contact with the siliceous zone (Fig. 5D). With increasing degrees of alteration, the sheeted dikes merge upwards with the siliceous zone where all primary features are obliterated and the rock takes on a pseudo-breccia texture (Fig. 5C). A few dikes cut the base of the siliceous zone and some bedded, tuffaceous, cherty sediments occur near the top, indicating that this transition zone consists of a mixture of dikes and flows. Although the siliceous transition is internally disjointed by conjugate fractures and shear bands and is the only ophiolitic unit with a crude, nonpenetrative cleavage, the magmatic transition from sheeted dikes to volcanic rocks is intact. The magmatic connection between the dikes and overlying volcanic flows is also indicated by their co-varying geochemistry and phenocryst types (e.g., plag-phyric flows overlie areas of plag-phyric dikes, etc). Although some sills and dikes occur within the volcanic section, the transition from 100% dikes to the volcanic section occurs within about 100 m. The abrupt transition from sheeted dikes to volcanic rocks is characteristic of many well-studied ophiolites (Pallister, 1981; Moores, 1982; Rosencrantz, 1983; Harper, 1984; Baragar et al., 1987) and distinguishes ophiolites from other types of volcanic centers. 3.6. Tonalites and Dacites A suite of tonalitic and dacitic rocks occurs in all three exposed layers of the ophiolitegabbro, sheeted dikes, and volcanic rocks. Rare dacitic flows, breccias, and tuffs are interbedded with the basaltic flows. Tonalitic and dacitic dikes make up about 10% of the sheeted dike complex and locally up to 38%. They are mutually intrusive with, and parallel to, the mafic dikes (Figs. 3B, D). The PO is unusual for the high proportion of dacitic dikes in the sheeted dike complex. These dikes are white to pink in color, and the contrast with the dark green and gray basaltic dikes shows off the cross cutting contacts and makes it easier to reconstruct the intrusive sequence. Besides the mutually crosscutting relations in the sheeted dike complex, composite dikes in the gabbro have co-mingled dacitic and basaltic phases that testify to the coeval intrusion of these two distinct magma types (see Dann, 1997a, Fig. 13). In addition, the tonalite/dacite suite includes small sub-spherical mafic enclaves that indicate mingling and incomplete mixing of basalt in the tonalitic magma. Locally, granitic screens melted and intruded the mafic dikes, suggesting that the abundance of dacitic magma may have been generated from felsic basement within the ophiolite. On

84

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

the other hand, the geochemistry of the upper gabbro and tonalite are complimentary, suggesting that the tonalite represents a residual liquid, filter pressed from the gabbro (Dann, 1992), similar to tonalites in the Karmoy Ophiolite (Pederson and Malpas, 1984). A more detailed petrological and geochemical study is needed to determine the petrogenesis of the variety of tonalites within the PO.

4. BASEMENT COMPLEX The basement complex consists of (1) coarse-grained granitoids, and (2) hypabyssal granite overlain by (3) submarine felsic volcaniclastic rocks of the Larson Spring Formation (Fig. 2). The basement complex is intruded by gabbro and cut by mafic dikes. Few ophiolites contain such a well-defined suite of older felsic rocks, but most of the ophiolite is devoid of roof pendants or screens. The basement complex indicates that the ophiolite developed from extension of older arc crust. It provides a valuable reference frame within the ophiolite, recording a tectonic event that preceded development of the ophiolite as well as the degree of magmatic extension. A northeast-trending belt of coarse-grained, foliated, tonalite and quartz monzodiorite (Fig. 2) occurs as roof pendants in gabbro and is intruded by a swarm of basaltic dikes and small plutons of isotropic porphyritic diorite. The presence of alkali feldspar, low An content of the plagioclase, darker hornblende, and more quartz and biotite distinguishes this suite of rocks from the gabbro and quartz diorite of the ophiolite. Abundant zircons yield a U/Pb age of 1.75 Ga (Dann et al., 1989), making it 20 m.y. older than the ophiolite and one of the oldest rocks in central Arizona. What this pluton intruded is not exposed. Screens in the sheeted dike complex and roof pendants in gabbro define a northeasttrending belt of felsic volcanic rocks, the Larson Spring Formation, and underlying, finegrained, isotropic granitoids (‘a–d’, Fig. 2). Near Larson Spring (‘c’, Fig. 2), the most intact felsic section underlain by hypabyssal granite sits as a block in gabbro, intruded by basaltic dikes (< 20%). Felsic volcaniclastic breccia, beds graded from breccia to porcelainite, and plagio-arenites with scour-and-fill structures indicate that the bedding dips consistently to the northwest. Massive, aphyric, felsic flows, mafic sediments, and chert also occur locally. How much older than the ophiolite these rocks are is not known.

5. HYDROTHERMAL ALTERATION Extensional tectonism generates hydrothermal circulation by facilitating the shallow emplacement of hot magma (dikes) and increasing the fracture porosity. Seafloor hydrothermal alteration and later burial metamorphism produce similar greenschist assemblages in mafic rocks. The effects of seafloor alteration can be distinguished by locating alteration or hydrothermal products that are unambiguously associated with intrusion or eruption of magma. In addition, the location of the most severe alteration is diagnostic of a narrow axial zone of intrusion, the hallmark of seafloor spreading.

6. Geochemistry

85

The sheeted dike complex of the PO displays patterns of alteration that are unambiguously associated with intrusion of the dikes. First, the interiors of some dikes with identifiable chilled margins are completely replaced by epidote and quartz. These green epidosites are locally intruded by grey greenschist dikes that retain their igneous textures and whole rock compositions. Second, dikes with a reddish-brown, hematitic alteration near the top of the sheeted dike complex are split by dikes with the more usual greenschist alteration. Third, veins of quartz (with or without sulfides), especially common along the margins of dikes, are also cut by late dikes. Based on the cross cutting relations, the epidosite, hematitic alteration, and quartz veins are temporally associated with intrusion of the dikes. The epidosites are typically found in sheeted dike complexes of ophiolites and are interpreted to require strongly localized, high temperature, fluid fluxes (Gillis and Banerjee, 2000). The most intensely altered unit in the Payson Ophiolite is the siliceous zone that marks the sheeted dike-volcanic transition. Except a few patches of sediment near the top and late dikes near the base of this zone, all original outcrop-scale features are obliterated by quartz-chlorite alteration that locally renders a pseudo-breccia texture (Fig. 5C). Disseminated sulfides are common, and weathering of small patches of gossan creates orange iron straining on some cliff exposures. Silicification and mineralization are concentrated at the transition from sheeted dikes to volcanic rocks in both modern oceanic crust (Alt et al., 1986) and in other ophiolites (e.g., Josephine Ophiolite; Harper et al., 1988). The transition from sheeted dikes to volcanic rocks is the site of eruption on the seafloor. Therefore, the alteration represented by the siliceous zone occurred around vents at the axial zone of seafloor spreading, where fractures open as fissures and permeability is highest. What distinguishes alteration of the PO from other ophiolites is the high degree of silicification to form an erosion-resistant, stratigraphically continuous zone at the dike-volcanic transition. Chemical sediments are interlayered with the volcanic flows and represent the exhalative products of a hydrothermal system that was active during seafloor spreading. A Cu-Pb massive sulfide deposit occurs just above the transition from sheeted dikes to submarine basalts in the southernmost exposure of the PO (Wessels and Karlstrom, 1991) in Tonto Creek (‘T’ on Fig. 1A). Massive sulfide deposits are common components of ophiolites (Gillis and Banerjee, 2000).

6. GEOCHEMISTRY All components of the 1.73 Ga Payson Ophiolite—submarine basalts, sheeted dikes, gabbro, and tonalite—as well as the 1.75 Ga basement complex have geochemical signatures of magmatic arc rocks (Dann, 1991, 1992). These include light rare earth element (LREE) and large-ion lithophile element (LIL) enrichment and relative high-field strength element (HFSE) depletion, typical of arc rocks (Pearce et al., 1984). The basaltic rocks plot in the ‘arc’ or ‘suprasubduction zone’ field in all tectonic discrimination diagrams (Th-HfTa, Ti-Cr, Cr-Y, etc.). Analyses of mafic dikes define a tholeiitic fractionation trend of

86

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

increasing FeO, TiO2 , P2 O3 , and V with decreasing MgO, typical of arc tholeiites. Hydrous igneous minerals (hornblende, biotite) in the gabbros and clinopyroxene-controlled fractionation of the dikes indicate that the parental magma was hydrous, consistent with generation above a subduction zone. Nd isotopic analyses indicate the influence of an older LREE-enriched component (Dann et al., 1993). The ophiolite has the geochemical and isotopic features of the high Ce/Yb suite of arc magmas (cf. Hawkesworth et al., 1993). Sheeted dikes may form during rifting of arcs, rifting of continental crust, or rifting of volcanic islands (e.g., Hawaii (Walker, 1987) or Canary Islands (Stillman, 1987)). However, the geochemistry alone indicates that the PO formed during an extensional phase in the evolution of a magmatic arc.

7. DEFORMATION OF THE OPHIOLITE Four stages of deformation are recorded by structures of the PO. First, rotation of the basement complex occurred prior to development of the ophiolite. Then, crustal extension guided the intrusion of basaltic magma, culminating in seafloor spreading. Structures that formed during intrusion of the basaltic magma provide important clues about the tectonic setting. Third, the ophiolite and overlying rocks are affected by two Early Proterozoic episodes of coaxial convergent deformation, the ca. 1.70 Ga Yavapai orogeny (D2 ) and the ca. 1.67 Ga Mazatzal Orogeny (D3 ). An earlier period of deformation (D1 ) only affected terranes west of the Moore Gulch Fault (Figs. 1A, C). Finally, Tertiary extensional faulting created the sediment-filled valley that cuts through the middle of the map area (Fig. 2). The bedded rocks of the Larson Spring Formation occur as narrow screens in the sheeted dike complex in Center Creek (‘d’, Fig. 2). The orientation of bedding defines an angular unconformity beneath the volcanic section of the ophiolite, which exposed granite on the pre-ophiolite paleosurface (see Dann, 1997a, Fig. 12). This angular relationship, the consistent orientation of bedding in the roof pendants, and repetition of the same lithologies from one roof pendant to another suggests that blocks of the basement complex rotated along listric normal faults prior to development of the ophiolite. Fault-bound domains of sheeted dikes with dips diverging 30 degrees from adjacent domains indicate block rotations along normal faults in the RC area. Outcrops of the overlying Cambrian Tapeats Formation are only rotated 5–10 degrees by Tertiary normal faults. As a result, the larger block rotations may reflect normal faults active during development of the ophiolite. Normal faults and rotated blocks occur along the mid-ocean ridges and in well-exposed ophiolites. Rotated dikes define grabens in the Troodos Ophiolite (Varga and Moores, 1985), and in the Josephine Ophiolite, entire crustal sections were rotated as much as 50 degrees relative to the overlying sediments and underlying Moho (Harper, 1984). Proving their syn-ophiolite origin, vertical dikes cut the rotated sections of sheeted dikes in these examples. This relationship has not yet been found in the PO. The most prominent fold of the sheeted dike complex occurs in RC area (‘s’, Fig. 2). The dikes and foliation in gabbroic screens rotate about 75 degrees clockwise at they approach, and end at, a poorly exposed, northeast-trending boundary with a younger granite. On

8. Tectonostratigraphic Analysis

87

an outcrop scale, dextral shear zones and faults are intruded by dikes. Consequently, this boundary may have been a dextral strike-slip shear zone active during development of the ophiolite. A system of northeast-trending strike-slip shear zones, including the boundaries of the Mazatzal block, is consistent with arc-parallel strike-slip faults that commonly occur within island arcs. Fold-and-thrust structures in the Mazatzal Mountains east of the Tertiary valley (Fig. 2) record two episodes of convergent deformation, D2 and D3 , separated by the unconformity at the base of the Mazatzal Group (section Y–Y , Fig. 2). Overall NE-trending structures, in addition to the northeasterly trend of Yavapai-Mazatzal orogenic belts across North America (Fig. 1C), indicate the presence of a NE-trending convergent margin, subduction zones, and island arcs during the main phase of crustal assembly. In the Mazatzal block, the ca. 1.70 Ga Yavapai Orogeny produced the northeast-trending syncline in the turbidites overlying the ophiolite (‘f’ and section Y–Y , Fig. 2). Crustal thickening drove uplift, subaerial exposure, and erosion, marked by an unconformity that truncated the fold in the turbidites and, to the south, cut down into the sheeted dikes. The unconformity is closely associated with eruption of the ca. 1.70 Ga, subaerial, Red Rock Rhyolite and intrusion of hypabyssal equivalents and large sheets of granite (i.e., Payson Granite, Fig. 2). Only subvolcanic facies are preserved in the map area (Fig. 2). Siliciclastic sedimentation of the Mazatzal Group covered the unconformity. Deposition of these rocks in a foreland setting suggests that the earlier Yavapai Orogeny involved accretion of the Mazatzal block to North America. Coaxial with the Yavapai structures, the Mazatzal Orogeny produced a foreland system of thrust faults and related folds in the Mazatzal Group, defining the D3 -phase of convergent deformation. Despite the deformation of the bedded sequences east of the Tertiary valley, the plutonic core of the exposed ophiolite was not folded.

8. TECTONOSTRATIGRAPHIC ANALYSIS The tectonostratigraphic history of the Mazatzal block is recorded in three distinct units: (1) the 1.75 Ga basement complex, (2) the 1.73 Ga PO and overlying basin-filling sedimentary and submarine volcanic sequences (1.73–1.71 Ga), and (3) 1.70 Ga subaerial rhyolite and related granite and later fluvial to shallow-marine siliciclastic sediments (Figs. 1C, 2). The 1.75 Ga granitoids of the basement complex and the 1.73 Ga PO are incompatible with the tectonostratigraphic history (Fig. 1C) of the adjacent Ash Creek block and other blocks to the northwest (Fig. 1A). The Ash Creek block records submarine arc volcanism and northwest-trending D1 deformation prior to intrusion of the 1.735 Ga Cherry Batholith (Fig. 1C; Karlstrom and Bowring, 1991). Northwest-trending dikes or other evidence of rifting while the PO was forming are lacking. Likewise, the basement complex shows no indication of the northwest-trending D1 deformation. Juvenile Nd isotopic signatures of the Ash Creek block contrast with evidence for a LREE-enriched component in all rocks of the Mazatzal block (Dann et al., 1993). This period of incompatibility between adjacent blocks suggests that movement along Moore Gulch shear zone juxtaposed the Ash Creek and Mazatzal crustal blocks. The lack of evidence for a subduction zone or low angle fault

88

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

that juxtaposed the two crustal blocks is best explained by a strike-slip boundary. Late dip-slip movement along Moore Gulch shear zone juxtaposes rocks from different crustal depths and obscures early fabrics. The PO is overlain by a volcano-sedimentary sequence, characterized by a lack of stratigraphic continuity across the Mazatzal block due to both facies changes and structural imbrication. In the map area, the basalts of the PO are directly overlain by dacitic volcaniclastic breccia with lenses of jasper and then a thick sequence of turbidites (Fig. 5A). Ash beds within the turbidites have U/Pb zircons ages of ca. 1.72 Ga (Dann et al., 1989). West of the map area (‘H’, Fig. 2), andesitic flows and coarse volcaniclastic breccias are interbedded with, and overlain by, turbidites and pelitic sediments with ash beds. These relations indicate that andesitic arc volcanoes erupted before and during turbidite deposition (Anderson, 1989) and on, or adjacent to, the PO. Finally, three granodiorite plutons (Fig. 2) intruded the PO at ca. 1.71 Ga (Conway et al., 1987). Anderson (1989) mapped a sequence of slates across the Moore Gulch shear zone, suggesting an overlap and contiguity of the two crustal blocks. In addition, ca. 1.70 Ga Yavapai deformation is recorded on both sides of the shear zone. Consequently, the Ash Creek and Mazatzal blocks must have been juxtaposed prior to the main phase of convergent deformation. After Yavapai deformation, the Ash Creek and Mazatzal blocks underwent differential uplift, probably reflecting different crustal profiles established early in their development. This difference is best appreciated by noting that the Mazatzal crustal block is unique, not only for the PO but also for preserving at least 4 unconformities and 3 transitions from plutonic to coeval volcanic rocks. Apparently, the Mazatzal block was uplifted less than adjacent blocks, suggesting less over-thickening during convergent deformation. The unique character of the crustal section probably began with the origin and emplacement of the ophiolite.

9. ORIGIN AND EMPLACEMENT OF THE PAYSON OPHIOLITE Ophiolites commonly originate in marginal basins above subduction zones because this tectonic setting predisposes them to be incorporated into continental crust during accretion of arcs or collision of continents. The formative tectonic setting and mechanics of emplacement are closely related, a theme that is important to any tectonic model of ophiolites. Seafloor spreading produced the horizontal layered structure of the PO as indicated by (1) the laterally extensive sheeted dike complex, (2) rooting of the sheeted dikes in coeval gabbro, (3) the abrupt transition to submarine lavas, and (4) the distribution of hydrothermal alteration and its exhalative products. The supra-subduction zone signature of the mafic rocks and older and younger arc lithologies implies that seafloor spreading took place within a magmatic arc. Within a 40 m.y. period, the basement complex developed within an arc, extensional tectonics rifted the arc and opened an intra-arc basin by seafloor spreading, arc volcanics and derived turbidites filled the basin, and arc granitoids intruded the basin floor. In modern arc settings, seafloor spreading creates both intra-arc and backarc submarine basins.

9. Origin and Emplacement of the Payson Ophiolite

89

The D2 and D3 convergent deformations affecting the Mazatzal block reflect a northeast-trending convergent margin that controlled the assembly of arc terranes and/or collisional orogenesis. Northwest-trending dikes of the PO formed by northeast-southwest extension, parallel to the convergent boundary prior to continental assembly. The schematic block diagram in Fig. 6A shows the Mazatzal and Ash Creek blocks prior to the ca. 1.70 Ga D2 Yavapai orogeny. In a Cretaceous back-arc basin represented by the Rocas Verdes ophiolite (de Wit and Stern, 1981), dikes are parallel to, and extension was perpendicular to, the convergent boundary. So, if the PO formed above a northeast-trending subduction zone, a simple back-arc model does not fit. Alternatively, arc-parallel extension occurs in modern arcs along arc-parallel strike-slip faults. For example, the Marinduque intra-arc basin in the Philippines developed from a pull-apart structure to seafloor spreading along the arc-parallel, strike-slip Philippine fault zone (Sarewitz and Lewis, 1991). This small basin, about the size of the Mazatzal block, contains a fossil axial-spreading center orthogonal to the strike-slip faults (Fig. 6B). Volcanoes rise from the basin floor. Turbidites are pouring in and interfingering with the volcanic debris. Strike-slip faults were active during seafloor spreading. When the basin crust is finally uplifted and exposed, we probably would see gabbro, sheeted dikes, and submarine volcanics overlain by turbidites and, like the PO, screens and roof pendants of older arc crust. In addition, we might see evidence for strike-slip shear zones within the ophiolite. Since the ophiolite remains above the subduction zone after its formation, we might expect to see arc plutons intruding the basin crust. As the basin is transported along the fault, the ophiolitic crust will be juxtaposed with arc crust that formed 100’s of km away in the same arc (Sarewitz and Lewis, 1991), probably with a contrasting tectonostratigraphic history. The Philippine arc is a collage of terranes including fragments of older deformed crust and several ophiolites, juxtaposed along major faults with up to 1000 km of displacement. An important feature of this model is that this juxtaposition occurs prior to the main phase of convergent deformation that accretes the arc to the continent. The Payson Ophiolite may be the oldest example of this mode of ophiolite generation and emplacement, attesting to the complex evolution of Early Proterozoic magmatic arcs leading up to the assembly of continental crust. High estimated crustal growth rates, indicated by large areas of juvenile crust like Yavapai-Mazatzal orogenic belt, suggest that continental assembly is episodic. Patchett and Chase (2002) estimate a 16% probability for margin-parallel strike-slip movement > 400 km that could effectively concentrate juvenile crust in small regions, giving the impression of higher than actual crustal growth rates. Direct evidence for early strike-slip movement along terrane boundaries that are reactivated during convergent deformation and post-assembly differential uplift is inherently difficult to recover. As a result, only by piecing together the tectonostratigraphic histories of terranes can the role of strike-slip tectonics in the assembly of continental crust be appreciated. Further analysis of the Payson Ophiolite and associated rocks is needed to better understand plate tectonics and the timing and mechanics of continental assembly during the Early Proterozoic.

90

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Fig. 6. (A) Schematic model of Mazatzal and Ash Creek crustal blocks at the time of formation of the Payson Ophiolite (ca. 1.73 Ga) and prior to D2 deformation of the ca. 1.70 Ga Yavapai Orogeny. The contrast in tectonostratigraphies requires an active boundary to juxtapose the distinct terranes. Arc-parallel extension in a step-over zone within a system of arc-parallel strike-slip faults rifted older arc crust and culminated in seafloor spreading and formation of an intra-arc basin. The basin was a locus of deposition within the arc. Transported along the arc by strike-slip movement, the basin was juxtaposed with the Ash Creek block prior to convergent deformation. (B) Similar in size to the Mazatzal block, the Marinduque intra-arc basin formed by seafloor spreading along the Philippine strike-slip fault zone in the Philippine arc (modified from Sarewitz and Lewis, 1991).

References

91

REFERENCES Alt, J.C., Honnerez, J., Laverne, C., Emmermann, R., 1986. Hydrothermal alteration of a 1 km section through the upper oceanic crust, Deep Sea Drilling Project Hole 504B: Mineralogy, chemistry, evolution of seawater-basalt interactions. Journal of Geophysical Research 91, 10309–10335. Anderson, P., 1989. Proterozoic plate tectonic evolution of Arizona. In: Jenny, J.P., Reynolds, S.P. (Eds.), Geologic Evolution of Arizona. Arizona Geological Society Digest 17, 17–148. Baragar, W.R.A., Lambert, M.B., Baglow, N., Gibson, I., 1987. Sheeted dikes of the Troodos ophiolite, Cyprus. In: Halls, H.C., Fahrig, W.F. (Eds.), Mafic Dike Swarms. Geological Association of Canada Special Paper 34, 257–272. Bowring, S.A., Karlstrom, K.E., 1990. Growth, stabilization, and reactivation of Proterozoic lithosphere in the southwestern United States. Geology 18, 1203–1206. Bowring, S.A., Karlstrom, K., Chamberlain, K.R., Dann, J.C., 1991. The tectonic evolution of central Arizona. Geologic Society of America Abstracts with Programs 23 (4), A7. Conway, C.M., Karlstrom, K.E., Silver, L.T., Wrucke, C.T., 1987. Tectonic and magmatic contrasts across a two-province Proterozoic boundary in central Arizona. In: Davis, G.H., VandenDolder, E.M. (Eds.), Geologic Diversity of Arizona and Its Margins: Excursions to Choice Areas. Arizona Bureau of Geology Special Paper 5, 158–175. Dann, J.C., 1991. Early Proterozoic ophiolite, central Arizona. Geology 19, 590–593. Dann, J.C., 1992. Origin and emplacement of the Early Proterozoic Payson ophiolite, central Arizona. Ph.D. dissertation. Washington University, St. Louis, MO, p. 340. Dann, J.C., 1997a. Pseudostratigraphy and origin of the Early Proterozoic Payson Ophiolite, central Arizona. Bulletin of the Geological Society of America 109, 347–365. Dann, J.C., 1997b. Branching sheeted dikes and seafloor spreading within an Early Proterozoic intraarc basin. Journal of Geophysical Research 102 (24), 917–924, 929. Dann, J.C., Bowring, S.A., Coleman, D.S., 1993. Geochronologic, isotopic, and structural constraints on the evolution of an Early Proterozoic intra-arc basin, central Arizona. Geological Society of America Abstracts with Programs 25 (5), 237. Dann, J.C., Chamberlain, K.R., Bowring, S.A., 1989. An Early Proterozoic ophiolite, central Arizona: U-Pb constraints on its structure and development. Geological Society of America Abstracts with Programs 21 (6), A23. Dann, J.C., Bowring, S.A., 1997. The Early Proterozoic Payson ophiolite and the Yavapai-Mazatzal orogenic belt, central Arizona. In: de Wit, M.J., Ashwall, L. (Eds.), Tectonic Evolution of Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 781–790. de Wit, M.J., Stern, C.R., 1981. Variations in the degree of crustal extension during formation of a back-arc basin. Tectonophysics 72, 229–260. Francheteau, J., Armijo, R., Cheminee, J.L., Hekinian, R., Lonsdale, P., Blum, N., 1992. Dyke complex of the East Pacific Rise exposed in the walls of the Hess Deep and the structure of the upper oceanic crust. Earth and Planetary Science Letters 111, 109–121. Furnes, H., Pederson, R.B., Stillman, C.J., 1988. The Leka Ophiolite Complex, central Norwegian Caledonides: field characteristics and geotectonic significance. Journal of the Geological Society of London 145, 401–412. Gillis, K.M., Banerjee, N.R., 2000. Hydrothermal patterns in supra-subduction zone ophiolites. In: Dilek, Y., Moores, E., Elthon, D., Nicolas, A. (Eds.), Ophiolites and Oceanic Crust: New Insights from Field Studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 283–297.

92

Chapter 2: The 1.73 Ga Payson Ophiolite, Arizona, USA

Greene, T., 1989. Magmatic evolution of the Lokken SSZ ophiolite, Norwegian Caledonides: relationships between anomalous lavas and high-level intrusions. Geological Journal 24, 251–274. Harper, G.D., 1984. The Josephine ophiolite, northwestern California. Geological Society of America Bulletin 95, 1009–1026. Harper, G.D., Bowman, J.R., Kuhns, R., 1988. A field, chemical, and stable isotope study of subseafloor metamorphism of the Josephine ophiolite, California-Oregon. Journal of Geophysical Research 93, 4625–4656. Hawkesworth, C.J., Gallagher, K., Hergt, J.M., McDermott, F., 1993. Mantle and slab contributions in arc magmas. Annual Review of Earth and Planetary Sciences 21, 175–204. Hoffman, P.E., 1989. Precambrian geology and tectonic history of North America. In: Bally, A.W., Palmer, A.R. (Eds.), The Geology of North America—An Overview. Geological Society of America, The Geology of North America A, pp. 447–512. Karlstrom, K., Bowring, S.A., 1988. Early Proterozoic assembly of tectonostratigraphic terranes in southwestern North America. Journal of Geology 96, 561–576. Karlstrom, K.E., Bowring, S.A., 1991. Styles and timing of Early Proterozoic deformation in Arizona: constraints on tectonic models. In: Karlstrom, K.E. (Ed.), Proterozoic Geology and Ore Deposits of Arizona. Arizona Geological Society Digest 19, 1–10. Karlstrom, K.E., Doe, M.F., Wessels, R.L., Bowring, S.A., Dann, J.C., Williams, M.L., 1990. Juxtaposition of Proterozoic crustal blocks: 1.65–1.60 Ga Mazatzal orogeny. In: Gehrels, G.E., Spencer, J.E. (Eds.), Geologic Excursions through the Sonoran Desert Region, Arizona, and Sonora. Arizona Geological Survey Special Paper 7. Moores, E.M., 1982. Origin and emplacement of ophiolites. Reviews of Geophysics and Space Physics 20, 735–760. Moores, E.M., 2002. Pre-1 Ga (pre-Rodinian) Ophiolites: Their tectonic and environmental implications. Bulletin of the Geological Society of America 114, 80–95. Nicholas, A., Boudier, F., 1991. Rooting of the sheeted dike complex in the Oman ophiolite. In: Peters, T., Nicholas, A., Coleman, R.G. (Eds.), Ophiolite Genesis and Evolution of Oceanic Lithosphere. Kluwer Academic, Dordrecht, The Netherlands, pp. 39–54. Pallister, J.S., 1981. Structure of the sheeted dike complex of the Samail ophiolite near Ibra, Oman. Journal of Geophysical Research 86, 2661–2672. Patchett, P.J., Chase, C.G., 2002. Role of transform continental margins in major crustal growth episodes. Geology 30, 39–42. Pearce, J.L., Lippard, S., Roberts, S., 1984. Characteristics and tectonic significance of suprasubduction zone ophiolites. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basin Geology: Volcanic and Associated Sedimentary and Tectonic Processes in Modern and Ancient Marginal Basin. Blackwell Scientific Publications, Oxford, England, pp. 77–94. Pederson, R.B., 1986. The nature and significance of magma chamber margins in ophiolites: examples from the Norwegian Caledonides. Earth and Planetary Science Letters 77, 100–112. Pederson, R.B., Malpas, C.A., 1984. The origin of oceanic plagiogranites from the Karmoy ophiolite, Western Norway. Contributions to Mineralogy and Petrology 88, 36–52. Rosencrantz, E., 1983. The structure of sheeted dikes and associated rocks in the North Arm massif, Bay of Islands ophiolite complex, and the intrusive process at oceanic spreading centers. Canadian Journal of Earth Sciences 20, 787–801. Rothery, D.A., 1983. The base of a sheeted dyke complex, Oman ophiolite: implications for magma chambers at oceanic spreading axes. Journal of the Geological Society of London 140, 287–296.

References

93

Sarewitz, D.R., Lewis, S.D., 1991. The Marinduque intra-arc basin, Philippines: basin genesis and in situ ophiolite development in a strike-slip setting. Geological Society of America Bulletin 103, 597–614. Skjerlie, K.P., Furnes, H., 1996. The gabbro-dyke transition zone demonstrated on Tviberg, SolundStavfjord ophiolite complex. Geological Magazine 133, 573–582. Stillman, C.J., 1987. A Canary island dyke swarm: Implications for the formation of oceanic islands by extensional fissural volcanism. In: Halls, H.C., Fahrig, W.F. (Eds.), Mafic Dike Swarms. Geological Association of Canada Special Paper 34, 243–255. Varga, R.J., Moores, E.M., 1985. Spreading structure of the Troodos ophiolite, Cyprus. Geology 13, 846–850. Walker, G.P.L., 1987. The dike complex of Koolau volcano, Oahu: Internal structure of a Hawaiian rift zone. U.S. Geological Survey Professional Paper 1350, 961–993. Wessels, R.L., Karlstrom, K.E., 1991. Evaluation of the tectonic significance of the Proterozoic Slate Creek shear zone in the Tonto Basin area. In: Karlstrom, K.E. (Ed.), Proterozoic Geology and Ore Deposits of Arizona. Arizona Geological Society Digest 19, 193–210. Wrucke, C.T., Conway, C.M., 1987. Geologic map of the Mazatzal Wilderness and contiguous roadless areas, Gila, Maricopa, and Yavapai counties, Arizona. United States Geological Survey OpenFile Report 87-664, scale 1:48,000.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

95

Chapter 3

NEOPROTEROZOIC OPHIOLITES OF THE ARABIAN-NUBIAN SHIELD ROBERT J. STERNa , PETER R. JOHNSONb , ALFRED KRÖNERc AND BISRAT YIBASd a Geosciences

Department, University of Texas at Dallas, Box 830688, Richardson, TX 57083-0688, USA b Saudi Geological Survey, P.O. Box 54141, Jiddah 21514, Saudi Arabia c Institut für Geowissenschaften, Universität Mainz, 55099 Mainz, Germany d Pulles Howard and de Lange, Environmental and Water Quality Management, P.O. Box 861, Auckland Park 2006, South Africa

Ophiolites of mid-Neoproterozoic age are abundant in the Arabian-Nubian Shield (ANS) of NE Africa and Arabia. ANS ophiolites range in age from 690 to 890 Ma and litter a region that is 3000 km N-S and > 1000 km E-W. In the northern ANS, ophiolites occur as nappe complexes marking suture zones between terranes. Although dismembered and altered, all of the diagnostic components of ophiolites can be found: harzburgite, cumulate ultramafics, layered as well as higher level gabbro and plagiogranite, sheeted dikes, and pillowed basalt. Allochthonous mafic-ultramafic complexes in the southern ANS, in Ethiopia and Eritrea, are interpreted as ophiolites, but are more deformed and metamorphosed than those in the north. Reconstructed ophiolitic successions have crustal thicknesses of 2.5 to 5 km. The ANS ophiolitic mantle was mostly harzburgitic, containing magnesian olivines and spinels that have compositions consistent with extensive melting. Cr# for spinels in ANS harzburgites are mostly > 60, comparable to spinels from modern forearcs and distinctly higher than spinels from mid-ocean ridges and backarc basin peridotites. ANS ophiolites are often associated with a thick (1–3 km) sequence of cumulate ultramafic rocks, which define a transition zone between seismic and petrologic Mohos. These cumulates are dominated by dunite, with subordinate pyroxene-rich lithologies. Cumulate ultramafics transition upwards into layered gabbro. Several crystallization sequences are inferred from ANS transition zones and cumulate gabbro sections. In all samples studied, olivine and spinel crystallized first, followed (in order of decreasing abundance) by cpx-plag, cpxopx-plag, and opx-cpx-plag. ANS ophiolitic lavas mostly define a subalkaline suite characterized by low K and moderate Ti contents, that has both tholeiitic and calc-alkaline affinities and includes a significant, although subordinate, amount of boninites. The lavas are fractionated (mean Mg# = 55) but have higher abundances of Cr (mean = 380 ppm) and Ni (mean = 135 ppm) than would be expected for such a low Mg#. The ANS ophiolitic lavas include both LREE-depleted and LREE-enriched varieties, but as a group are slightly DOI: 10.1016/S0166-2635(04)13003-X

96

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

LREE-enriched: mean (Ce/Yb)n ∼ 2.2 and Sm/Nd ∼ 0.30. On a variety of discrimination diagrams, the lavas plot in fields for MORB, BABB, arc tholeiite, and boninite. ANS lavas cluster around Ti/Zr = 97, indicating that Ti-bearing phases did not precipitate early. Nd isotopic compositions indicate derivation from a long-depleted mantle source, with εNd(t) ∼ +5 to +8. Mineral and lava compositions are consistent with the hypothesis that most ANS ophiolites formed in ‘suprasubduction zone’ (SSZ) settings, and the high Cr# of ANS ophiolitic harzburgites suggests a forearc environment. Geochemical studies of deep water sediments deposited on ANS ophiolites are needed to better characterize and understand the Neoproterozoic ocean where ANS ophiolites formed.

1. INTRODUCTION The Arabian-Nubian Shield (ANS) in NE Africa and W. Arabia is the largest tract of juvenile continental crust of Neoproterozoic age on Earth (Patchett and Chase, 2002). This crust was generated when smaller terranes of arc and back arc crust were generated within and around the margins of a large oceanic tract known as the Mozambique Ocean, which formed in association with the breakup of Rodinia ∼ 800–900 Ma (Stern, 1994). Oceanic plateaus may also have been accreted (Stein and Goldstein, 1996). These crustal fragments collided as the Mozambique Ocean closed, forming arc-arc sutures, composite terranes, the Arabian-Nubian Shield (ANS; Fig. 1), and the larger collisional belt known as the East African Orogen (Stern, 1994; Kusky et al., 2003). The Arabian-Nubian Shield was caught between fragments of East and West Gondwanaland as these collided at about 600 Ma (Meert, 2003) to form a supercontinent variously referred to as Greater Gondwanaland (Stern, 1994), Pannotia (Dalziel, 1997) or just Gondwanaland. The ANS was subsequently buried by Phanerozoic sediments but has been exposed by uplift and erosion on the flanks of the Red Sea in Oligocene and younger times. Several lines of evidence support the idea that the ANS is juvenile Neoproterozoic crust, including non-radiogenic initial Sr and radiogenic initial Nd isotopic compositions for a

Fig. 1. Location of the Arabian-Nubian Shield and location of ophiolites and related rocks within it. (A) Political and modern geographic features of the region. B = Bahrain, Dj = Djibouti, Is = Israel, Jrdn = Jordan, K = Kuwait. Location of Figs. 2A and B shown in dashed rectangles. (B) Location of Precambrian basement exposures, crustal types, and ophiolites and ophiolitic rocks (shown in black). Abbreviations for some of the better studied ophiolites follow. Saudi Arabia: H = Halaban, JT = Jebel Tays; JU = Jebel Uwayjah; JE = Jebel Ess; AA = Al ‘Ays (Wask); BT = Bi’r Tuluhah; A = Arjah; DZ = Darb Zubaydah; BU = Bi’r Umq; Th = Thurwah; JN = Jebel Nabitah; T = Tathlith. Egypt: F = Fawkhir; Br = Barramiya; Gh = Ghadir; AH = Allaqi-Heiani; G = Gerf. Sudan: OSH = Onib-Sol Hamed; Hs = Hamisana; AD = Atmur-Delgo; K = Keraf; R = Rahib; M = Meritri; Os = Oshib; Kb = Kabus (Nuba Mts); I = Ingessana; Kk = Kurmuk. Eritrea: Hg = Hagar Terrane. Ethiopia: Zg = Zager Belt; DT = Daro Tekli Belt; Bd = Baruda; TD = Tulu Dimtu; Y = Yubdo; A = Adola; My = Moyale. Kenya: S = Sekerr, B = Baragoi. The Bi’r Umq-Nakasib suture is defined by the Bi’r Umq-Thurwah-Meritri-Oshib ophiolite belt.

97

(A)

(B)

1. Introduction

98

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

wide range of igneous rocks. The ANS can be isotopically defined as the region in NE Africa and Arabia where Nd-model ages approximate crystallization ages (Stern, 2002). These indications that abundant juvenile continental crust and mantle lithosphere were generated during Neoproterozoic time in the region has been confirmed by Nd and Sr isotopic studies of samples of mafic lower crust and mantle lithosphere brought up as xenoliths in Tertiary lavas from Saudi Arabia, which also indicate that the lower crust and lithospheric mantle of the region formed during Neoproterozoic time (Henjes-Kunst et al., 1994; McGuire and Stern, 1993). Ophiolites and ophiolitic rocks are remarkably abundant in the Arabian-Nubian Shield. They are scattered across most of the ANS, over a distance of ∼ 3000 km from the farthest north (Jebel Ess) almost to the equator, and from Rahib in the west to Jabal Uwayjah (45 ◦ E) in the east, encompassing an area of about two million square kilometers (Fig. 1). Ophiolites are particularly well studied in Arabia (see companion paper by Johnson et al. (2004). If ophiolites are the remains of oceanic lithosphere, then the ANS is a massive graveyard of Neoproterozoic oceanic lithosphere. The abundance of ophiolites is a further indication that ANS crust and lithosphere were produced by processes similar to those of modern plate tectonics. Their abundance has made it difficult to define the orientation of sutures solely from the distribution of ophiolitic rocks (Church, 1988; Stern et al., 1990). This is further complicated by the fact that not all ANS mafic-ultramafic complexes formed in a seafloor spreading environment—some appear to be roots of island arcs, such as Darb Zubaydah in Arabia (Quick and Bosch, 1989)—and others are autochthonous layered intrusions, such as Dahanib in Egypt (Dixon, 1981). Care must be exerted to avoid misidentification, but even the most conservative estimates indicate that there is a remarkable abundance of ophiolites in the ANS. Ophiolites were first recognized in the region by Rittmann (1958), but were otherwise ignored until the pioneering study of Bakor et al. (1976). This was followed by a flurry of field, petrological, and geochemical studies throughout the 1980’s. This level of activity has decreased significantly through the 1990’s and into the 21st century. This review has three objectives. First, it is intended to summarize the most important observations of this first phase of studying ANS ophiolites. Second, because the ophiolites of the Arabian-Nubian Shield are so common and so well-exposed, it is hoped that this example will provide a basis of what is expected to be preserved when a major episode of juvenile crust formation associated with modern-style plate tectonics occurs. Finally, it is hoped that this overview will stimulate a resurgence of ANS ophiolite studies. Note that the Arabic word for mountain is variously spelled ‘Jebel’ (Egypt), ‘Gebel’ (Sudan), or ‘Jabal’ (Arabia). For simplicity, we use the ‘Jebel’ spelling throughout this contribution.

2. OUTCROP PATTERNS We define the ANS as the northern, juvenile part of the Neoproterozoic EAO. More restrictive definitions have the shield ending in the south with the southernmost contiguous

3. Crustal Structure

99

outcrops of basement exposed around the Red Sea. The Arabian-Nubian Shield as defined here includes basement outliers well to the south in Ethiopia and Kenya. For the purposes of this presentation, the ANS is that part of the East African Orogen characterized by juvenile Neoproterozoic crust and is also where ophiolites and ophiolitic rocks are encountered (Fig. 1). The ANS can be usefully subdivided into northern and southern halves. North of the Bi’r Umq-Nakasib suture, which extends NE from the Oshib and Meritri ophiolites in Sudan and continues across the Red Sea in Arabia through the Thurwah and Bi’r Umq ophiolites, the ophiolite belts trend approximately E-W. It is relatively easy to identify structures developed during ophiolite obduction in this region. Greenschist-facies metamorphism is characteristic for these ophiolites, and diagnostic features, such as pillowed basalts, are well preserved. This is shown in Table 1, which lists ophiolites with a Penrosetype succession; these are from the northern ANS or from the Bi’r Umq-Nakasib suture. Such excellent preservation largely reflects the fact that these northern ophiolites are relatively undisturbed by steep N-S structures developed during terminal collision between E and W Gondwanaland, which are pervasive farther south (Fig. 2A). Many ophiolites in the northern ANS are, however, disrupted by NW-trending strike-slip faults and shear zones of the Najd fault system (Sultan et al., 1988). Remote sensing has proven to be an effective way to map the distribution of spectrally distinct lithologies, especially serpentinites and amphibole-bearing mafic rocks, providing a quantitative, if indirect, assessment of the distribution and abundance of disrupted ophiolites in the basement of Egypt (Sultan et al., 1986). A more detailed presentation of the field relations and structure of Arabian Shield ophiolites is presented by Johnson et al. (2004). It is a significantly greater challenge to identify ophiolites to the south of the Bi’r Umq-Nakasib suture. This area was closer to and thus more intensely affected by the endNeoproterozoic terminal collision, such that structures related to ophiolite obduction are transposed or obliterated (Abdelsalam and Stern, 1996). Basement structures dip steeply (Fig. 2B), units are intensely deformed and shuffled by high-angle thrusting and subhorizontal shearing, and metamorphism is typically amphibolite-facies (Yihunie, 2002). Diagnostic features and emplacement fabrics for units that might originally have been ophiolites are not common. Purists would hesitate to identify the linear mafic-ultramafic complexes of the southern ANS as ophiolites, but the association of harzburgitic ultramafics in association with MORB-like and even boninitic mafic units as well as the regional association of southern ANS mafic-ultramafic complexes to the abundant and unequivocal ophiolites of the northern ANS makes it likely that these mostly represent ophiolites in different stages of preservation.

3. CRUSTAL STRUCTURE The best preserved ophiolites in the northern ANS contain all or most of the components of complete ‘Penrose’ ophiolites (Table 1), including pillowed basalts, gabbros,

100

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Table 1. ANS ophiolites with a “Penrose-type” assemblage Name Darb Zubaydah Bi’r Tuluhah Wadi Khadra Halaban Ess Al ‘Ays (Wask) Tharwah Bi’r Umq Tathlith Oshib-Ariab belt Arbaat Atmur-Delgo

Country Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Sudan Sudan Sudan

Sol Hamed Wadi Onib Gebel Gerf Wadi Ghadir Fawkhir

Sudan Sudan Egypt Egypt Egypt

Reference Quick (1990) Pallister et al. (1988) Quick (1991) Al-Saleh et al. (1998) Pallister et al. (1988) Bakor et al. (1976) Nassief et al. (1984) Shanti (1983) Pallister et al. (1988) Abdel-Rahman (1993) Abdelsalam and Stern (1993) Harms et al. (1994), Schandelmeier et al. (1994) Fitches et al. (1983) Hussein et al. (1983) Zimmer et al. (1995) El-Bayoumi (1983) El-Sayed et al. (1999)

and tectonized harzburgites. Several ANS ophiolites have sheeted dike complexes (such as Ghadir, Onib, and Ess) but these are not always reported. Even the well-preserved ophiolites are faulted, folded, and otherwise disrupted, so that reconstructing a complete ophiolite pseudostratigraphy is difficult and equivocal. Nevertheless, three such reconstructions of ANS ophiolite crustal structure are shown on Fig. 3. These reconstructions differ in the relative abundances of volcanics, pillowed lavas, sheeted dikes, and gabbro but all suggest that the oceanic crust represented by ANS ophiolites was generally in the range of 2.5 to 6 km thick. As discussed in the next section, ANS ophiolites were generated and em-

Fig. 2. Remote sensing images of ANS ophiolites, showing the different outcrop patterns of ophiolites in the northern (A) and southern (B) parts of the Arabian-Nubian Shield. Locations shown in Fig. 1. (A) Allaqi-Heiani Suture along the Egypt-Sudan border. Image is approximately 300 km across and N is towards the top of image. Dashed line approximates trace of Allaqi-Heiani Suture. Note that the general E-W structure of the ophiolite belt, which formed during suturing of the SE Desert and Gabgaba terranes is only disrupted by younger N-S structures (developed during terminal collision between East and West Gondwanaland) of the N-S Hamisana Shear Zone. This outcrop pattern indicates the ophiolites and associated accretionary structures are subhorizontal. NASA astronaut photograph (S32-74-100). (B) Landsat TM image of basement units in N. Eritrea and E. Sudan, showing dominance of complex deformation related to terminal collision between East and West Gondwanaland, resulting in ∼ N-S structures. Terrane names are modified after Drury and Filho (1998). Scene is about 90 km across.

3. Crustal Structure

101

102

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Fig. 3. Reconstructed crustal sections of ANS ophiolites. Al ‘Ays (Al Wask), Saudi Arabia (Bakor et al., 1976); Bi’r Umq, Saudi Arabia (Al-Rehaili and Warden, 1980); Onib, Sudan (Kröner et al., 1987).

placed relatively early in the history of the ANS and EAO. Fragments from dismembered ophiolites are common in the ANS, and it becomes more difficult to interpret these as once being allochthonous pieces of oceanic crust as these fragments become more deformed and metamorphosed. Suffice it to say that not all ultramafic rocks in the shield are—or were— parts of ophiolites. There are many layered igneous intrusions containing non-ophiolitic ultramafics and gabbros and there are many examples of non-ophiolitic pillowed lavas. Nevertheless, the association of harzburgitic ultramafics and low-K tholeiitic metabasalt argues strongly that these once belonged to a coherent ophiolite.

4. Age

103

Fig. 4. Ages of ophiolites in the Arabian-Nubian Shield, binned at 25 million year intervals. Neoproterozoic time (544 to 1000 Ma) is subdivided into Tonian (1000–850 Ma), Cryogenic (850 to ∼ 600 Ma) and Neoproterozoic III (∼ 600–544 Ma), after Knoll (2000). Age range when Arabian-Nubian Shield was tectonically and magmatically active (870 Ma to end of Neoproterozoic) is also given. Ophiolite ages are U-Pb and Pb-Pb zircon ages and Sm-Nd ages from (Stacey et al., 1984; Claesson et al., 1984; Pallister et al., 1988; Kröner et al., 1992; Zimmer et al., 1995; Worku, 1996). Mean age for ophiolites ±1 standard deviation is given.

4. AGE ANS ophiolites have been reliably dated using U-Pb zircon techniques (Stacey et al., 1984; Pallister et al., 1988) and Pb-Pb zircon evaporation techniques (Kröner et al., 1992; Zimmer et al., 1995) on zircons separated from gabbros and plagiogranites. Other ages have been generated using Sm-Nd mineral and whole-rock techniques (Claesson et al., 1984; Zimmer et al., 1995; Worku, 1996). These results give age ranges of 694 ± 8 Ma for the youngest ANS ophiolite (Urd/Halaban; Stacey et al., 1984) to 870 ± 11 Ma for the oldest (Thurwah; Pallister et al., 1988). A mean of 781 Ma (1 standard deviation = 47 Ma) is obtained for 16 robust ophiolite ages (Fig. 4). Ophiolites formed during the first half of the time period encompassed by tectonic and magmatic activity of the Arabian-Nubian Shield. There is no obviously systematic geographic variation in the distribution of ANS

104

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

ophiolite ages; the oldest ophiolite is in the middle of the ANS and the youngest is near the eastern edge of ANS exposures.

5. OPHIOLITE COMPONENTS 5.1. Harzburgites Where the original protolith can be identified, the mantle peridotites associated with ANS ophiolites are predominantly tectonized harzburgites, lherzolite being rarely reported. Studies of ANS ophiolite fabrics are at a very early stage, but harzburgites and dunites of the Thurwah ophiolite have been tectonized at high temperatures (Nassief et al., 1984), and a similar high-temperature ductile fabric is reported from the Sol Hamed ophiolite (Fitches et al., 1983). The harzburgites represent residual mantle after extensive melting, whereas the dunites and wehrlites are cumulates or reflect melt-wallrock interactions. Harzburgites are mostly altered (Figs. 5A, B), but relict olivines and pyroxenes have been analyzed by electron microprobe for 5 ophiolites. Harzburgite associated with the Al Ays ophiolite contains olivine of Fo91–94, orthopyroxene of En89–91, and clinopyroxene of En48–53, Fs2.2–3.0, Wo44–49.2 (Chevremont and Johan, 1982a; Ledru and Auge, 1984). Harzburgite from the nearby Hwanet ophiolite has orthopyroxene of En88.9–91 and clinopyroxene of En48.9–50.3, Fs2.3–3.4, Wo46.3–48.9 (Chevremont and Johan, 1982a). Harzburgite associated with the Ess ophiolite contains olivine of Fo91–93, orthopyroxene of En88–92, and diopsidic clinopyroxene of En49–52, Fs2.4–3.1, Wo44.6–48.3 (Al-Shanti, 1982). Nassief et al. (1984) identified a mantle sequence that is up to 20 km thick for the Thurwah ophiolite, and harzburgite from this consists of 70–90% olivine (Fo89.5–93.4), 15–30% orthopyroxene (En90–92) and < 1% each of chromite and clinopyroxene. Mouhamed (1995) argued on the basis of CIPW normative compositions of Muqsim serpentinites along the AllaqiHeiani suture (Fig. 2) that these were originally harzburgites. Olivines from the Ingessana ophiolite are Fo91–97 (Price, 1984). These compositions are at the magnesium-rich end of peridotites, as shown on Fig. 6. Olivine compositions provide insights into the tectonic setting of ophiolites because magmagenesis in different tectonic settings reflects differing extents of melting. Because residual olivines become increasingly magnesian as melting progresses, residual mantle should have olivines that are more magnesian than the Fo88 of undepleted ‘pyrolitic’ upper mantle (Fig. 6). Bonatti and Michael (1989) suggest that mantle melting ranges from nearly zero for undepleted continental peridotites to about 10–15% melting for rifted margins to 10–25% melting associated with mid-ocean ridge (MOR) peridotites to 30% for peridotites recovered from forearcs, which generally form during the early stages in the evolution of the associated subduction zone (Bloomer et al., 1995). Mantle peridotites from back-arc basins were not available when this diagram was originally generated, but since that time mantle peridotites from the Mariana Trough active back-arc basin in the western Pacific have been studied (Ohara et al., 2002). These harzburgites have olivine compositions that are indistinguishable from MOR harzburgites and are distinctly less magnesian

5. Ophiolite Components

105

Fig. 5. Outcrop photographs of ANS ophiolites. (A) Barramiya, Egypt. Light areas are talc-carbonate alteration of harzburgitic ultramafics, darker areas are serpentinized ultramafics. White house for scale. (B) Thurwah, Saudi Arabia. Light areas are talc-carbonate alteration of harzburgitic ultramafics, darker areas are serpentinized ultramafics. Seated geologist (far left) for scale. (C) Carbonated ultramafics of the Bi’r Umq ophiolite (Saudi Arabia) thrust south over metasediments of the Mahd Group. Thrust contact is dashed. Three geologists climbing ridge for scale. (D) N-dipping layered gabbros and cumulate ultramafics on the south side of the Gerf ophiolite, SE Egypt/NE Sudan. Ridge is about 100 m tall. (E) Layered gabbros of the Onib ophiolite, NE Sudan. Light layers are anorthositic, darker layers are richer in mafic minerals. Card (∼ 15 cm long) for scale. (F) Pillowed basalts, Wadi Zeidun (near Fawkhir ophiolite), Egypt. Hammer for scale.

106

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Fig. 5. (Continued.)

than those of forearc peridotites. In this context, the Mg-rich nature of ANS ophiolite peridotite olivines is best interpreted to indicate that these are residual after extensive melting, similar to that observed for forearc peridotites. Inferences about the extent of melting based on olivine compositions are supported by abundant compositional data for spinels. Spinel resists alteration better than olivine, may be economically important, and thus are relatively well studied for ANS ophiolites (Fig. 7). Progressive melting of peridotites depletes Al relative to more refractory Cr in residual spinels, such that the Cr# of spinels (= 100Cr/Cr + Al) increases with melting

5. Ophiolite Components

107

Fig. 6. Composition of olivines in peridotites from various tectonic settings, modified after Bonatti and Michael (1989). The tectonic settings are arranged in an order in which melt depletion increases to the right. Note that the field for olivines from peridotites of 4 ANS ophiolites are Fo91 or are more magnesian. These compositions are most consistent with formation in a forearc setting.

(Dick and Bullen, 1984). As is the case for Fo content of harzburgitic olivines, the increase in Cr# of harzburgitic spinels reflects increasing degree of melting. Degree of melting is related in a general way to tectonic setting, at least for Cenozoic peridotites (Fig. 7A). Spinels from MOR peridotites generally have Cr# < 50 (although Barnes and Roeder, 2001 report that a subordinate proportion of MOR peridotites have Cr# up to 80). A limited dataset for spinels from back-arc basin peridotites indicates that these experienced extents of melting similar to MOR and thus have spinel Cr# that are similar to those of MOR peridotites (Ohara et al., 2002). Spinels in forearc harzburgites generally have higher Cr#

108

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Fig. 7. Composition of spinels in ANS ophiolitic peridotites compared with those in modern peridotites. Data are plotted on 100Cr/Cr + Al (Cr#) vs. 100Mg/Mg + Fe (Mg#) diagram, modified after Dick and Bullen (1984). (A) Spinels from ANS ophiolitic peridotites, location and size of rectangle determined by means and standard deviations listed in Table 1. (B) Fields defined by spinel compositions for likely ophiolitic analogues of Cenozoic age, modified after Bloomer et al. (1995). Note that spinels from peridotites from mid-ocean ridges and back-arc basin spreading axes characteristically have Cr# < 60, whereas spinels from forearc peridotites and boninites Cr# > 40. The vast majority of ANS ophiolitic spinels have high Cr#, suggesting these formed in a forearc setting.

5. Ophiolite Components

109

Table 2. Mean compositions of spinels in ANS harzburgites Locality Sudan Onib/Sol Hamed Oshib Rahib Ingessana

Mean Cr#

1 Std. Dev.

Mean Mg#

1 Std. Dev.

70.4 71.9 73.4 77.5

17.6 9.2 11.7 7.0

56.4 65.1 56.5 57.6

11.9 5.4 22.3 14.7

Egypt Various

52.3

5.8

56.6

4.9

Saudi Arabia Ess/Al ‘Ays Nabitah Al Amar Thurwah Bi’r Umq Tuluhah/Arjah Halaban

67.2 78.3 62.9 79.8 52.6 72.4 76

10.0 4.9 14.2 6.4 7.8 4.4 6.8

60.6 64.7 52.4 56.0 61.0 70.2 67.4

9.9 4.5 5.4 2.8 4.1 8.0 2.9

Ethiopia Adola Moyale

91.4 58.2

2.5 2.8

63.1 64.4

7.6 2.6

Kenya Sekerr Baragoi

88.1 82.6

1.3 1.1

68.7 56.7

7.9 5.1

Data sources: Onib-Sol Hamed, Sudan (N = 60: Abdel-Rahman, 1993; Hussein, 2000; Price, 1984); Oshib, Sudan (N = 15: Abdel-Rahman, 1993); Rahib, Sudan (N = 15: Abdel-Rahman, 1993); Ingessana-Kurmuk, Sudan (N = 97: Abdel-Rahman, 1993; Price, 1984); Egypt (various; 4 averages for spinels in harzburgite: Ahmed et al., 2001); Ess-Al ‘Ays, Saudi Arabia (N = 81: Al-Shanti and El-Mahdy, 1988; Al-Shanti, 1982; Chevremont and Johan, 1982a; Chevremont and Johan, 1982b; Ledru and Auge, 1984); Nabitah, Saudi Arabia (N = 9: Al-Shanti and El-Mahdy, 1988); Bi’r Umq, Saudi Arabia (N = 20: LeMetour et al., 1982); Thurwah, Saudi Arabia (N = 9: Al-Shanti and El-Mahdy, 1988); Tuluhah/Arjah, Saudi Arabia (N = 14: Al-Shanti and El-Mahdy, 1988); Halaban, Saudi Arabia (N = 8: Al-Shanti and El-Mahdy, 1988); Adola, Ethiopia (N = 10: Bonavia et al., 1993); Moyale, Ethiopia (N = 32: Berhe, 1988); Sekerr, Kenya (N = 123: Price, 1984); Baragoi, Kenya (N = 50: Berhe, 1988).

(up to 80) and spinels from boninites typically have Cr# of 70–90 (Fig. 7B), consistent inferences from olivine compositions that these are manifest residues and products of the highest degree of melting found for post-Archean igneous rocks. Table 2 summarizes spinel compositions from peridotites (mostly harzburgites) associated with 16 ANS ophiolites. These are mostly for spinels in harzburgites, but where this data was unavailable, compositions of podiform chromite were used. The data indicate that ANS peridotitic spinels have Cr# that are mostly > 60 (Egypt, Bi’r Umq, and Moyale being notable exceptions). Spinels from the Sekerr ophiolite (mean Cr# = 88) and the Adola ultramafic complex (mean Cr# = 91) are remarkably rich in chromium. Overall, the Cr#

110

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

of ANS ophiolitic peridotites appear most similar to those of modern forearc peridotites (Fig. 7B), although there is a hint of a regional gradient, from higher Cr# in the south to lower Cr# in the north. Podiform chromites from the Adola complex in southern Ethiopia are associated with harzburgites; Bonavia et al. (1993) did not explicitly identify this as part of an ophiolite, but they argued from the absence of phase layering and low Rh/Ir that the rocks did not form as part of a layered intrusion. These data supports an interpretation that the Adola mafic-ultramafic complex represents a highly metamorphosed, deformed ophiolite. This inference is supported by the interpretation of Yibas et al. (2003) that many mafic-ultramafic complexes in southern Ethiopia are ophiolites. 5.2. Transition Zone and Gabbros The transition zone lies between the petrologic Moho (defined at the base of the cumulate section and the top of the tectonized peridotites) and the seismic Moho (defined as the top of cumulate ultramafic section and the base of gabbros). Several ANS ophiolites have well-preserved transition zones characterized by interlayered pyroxenite, wehrlite, lherzolite, dunite, and/or chromite at the base (Fig. 5D) grading upwards into sections that are increasingly dominated by gabbro. Other ophiolites have thin to non-existent transition zones (e.g., Fawkhir; El-Sayed et al., 1999). The Onib ophiolite is characterized by an unusually thick (2–3 km) transition zone of interlayered cumulate ultramafics, podiform chromites, and layered gabbros (Fig. 3; Hussein et al., this volume; Kröner et al., 1987). The ultramafic rocks of the Sol Hamed ophiolite, NW of Onib, also seem to represent a similar crust-mantle transition zone, and are 80% dunite (with interbedded chromitites), with lesser proportions of wehrlite, harzburgite, and pyroxenite (Fitches et al., 1983; Price, 1984). Similarly, the Oshib ophiolite has a 2-km thick transition zone, which grades upwards from harzburgite tectonite through dunite and wehrlite followed by pyroxenite and cumulate gabbro (Abdel-Rahman, 1993). The transition zone of the Thurwah ophiolite contains about 1 km of cumulate dunite, lherzolite, and pyroxenite, in similar proportions (Nassief et al., 1984). Each rock layer is 1–10 m thick and is developed in the upward sequence dunite-lherzolite-pyroxenite. The Ess ophiolite contains a cumulate peridotite section that is about 400 m thick and consists of wehrlite and dunite (Shanti and Roobol, 1979). Significant bodies of dunite, often associated with podiform chromite, exist at the base of some gabbro sections (El-Bayoumi, 1983). Olivines in dunites from the Thurwah ophiolite are significantly more Fe-rich (Fo86.5–89.5; Nassief et al., 1984) than olivines of typical ANS harzburgites. The Al Ays ophiolite contains over 350 lenses of podiform chromitite within the dunite unit (Bakor et al., 1976). Chromites associated with dunites are often significantly more Cr-rich (Cr# = 65–85) than those associated with harzburgite (Cr# ∼ 50; Ahmed et al., 2001); the reason for this is not understood. Some chromites contain inclusions of olivine (Fo97–99) and diopside, along with hydrous phases, such as amphibole (edenite-tremolite) and phlogopite (Ahmed et al., 2001). The primary mineralogy of these inclusions is remarkably preserved, and studying inclusions in chromites and other resistant minerals is a promising avenue for understanding ANS ophiolites.

5. Ophiolite Components

111

Modal abundances of olivine decrease upsection as the abundance of especially clinopyroxene increases; correspondingly dunites are upwardly succeeded by wehrlites or lherzolites and then pyroxenites. Cumulate pyroxenites are dominated by diopside. Cumulate pyroxenites of the Onib transition zone are low in TiO2 (0.01–0.06%) and rich in Cr (1500– 3000 ppm; Kröner et al., 1987). Gabbro bodies up to 1 km thick are interleaved with the ultramafic cumulates at Igariri, part of the Oshib ophiolite (Abdel-Rahman, 1993). ANS ophiolites with well-developed transition zones may have formed by fast seafloor spreading, whereas those that do not have thick transition zones may represent seafloor produced by slow spreading (Dilek et al., 1998). Gabbros are ubiquitous and important components of ANS ophiolites. Original igneous textures are common but metamorphic recrystallization under greenschist to amphibolite facies conditions is ubiquitous. ANS ophiolitic gabbros are mostly pyroxene gabbro; olivine gabbro is much less common. Clinopyroxene generally dominates over orthopyroxene. Where ophiolitic gabbros are well preserved and studied, they are commonly layered, at least in part. Phase layering, evidenced by melanogabbro and anorthositic gabbro couplets are typically ∼ 5 cm thick and can be traced for meters. Layered metagabbros made up of alternating plagioclase (An63–83)- and amphibole-rich layers are part of the Ess ophiolite (Shanti, 1983). Plagioclase compositions in Sol Hamed gabbros change from An70–85 at the base to more sodic compositions upsection (Fitches et al., 1983). Amphibole may be uralitized pyroxene (El-Bayoumi, 1983). Similar uralitized clinopyroxene gabbros are reported for the Al Ays (al Wask) ophiolite. Upwards through the cumulate sequences, from the base of the transition zone into the layered gabbros, the sequence of rocks indicates a crystallization sequence of olivine ± chromite-clinopyroxene-plagioclase (Price, 1984), olivine ± chromite-clinopyroxene-orthopyroxene-plagioclase (Nassief et al., 1984), or, less commonly, olivine ± chromite-orthopyroxene-clinopyroxene-plagioclase (AbdelRahman, 1993). High level gabbros are more massive, lack layering and other evidence of crystal accumulation, and commonly include pegmatitic gabbro and isolated bodies of plagiogranite. ANS plagiogranites are high in SiO2 (70–77%) and low in K2 O (0.04–1.9%: Shanti, 1983; Abdel-Rahman, 1993). These mostly plot in the field of trondhjemite and tonalite on a normative Ab-An-Or diagram (Shanti, 1983). The high level gabbros are often intruded by mafic dikes, which represent the base of the sheeted dike complex. 5.3. Sheeted Dykes Sheeted diabase dikes are common components of ANS ophiolites. Where observed they typically transition downwards into the high-level gabbros and grade upwards into pillowed basalts. Sheeted dikes are reported from the following ophiolites: Rahib (Abdel-Rahman et al., 1990), Ess (Shanti and Roobol, 1979), Sol Hamed (Fitches et al., 1983), Ingessana (Price, 1984), Ghadir (El-Bayoumi, 1983), Gerf (Zimmer et al., 1995), and Thurwah (Nassief et al., 1984). Sheeted dykes for many other ANS ophiolites are not identified or are poorly developed (e.g., Al Ays and Fawkhir: Bakor et al., 1976; El-Sayed et al., 1999). It

112

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

is not yet clear whether this represents differential preservation of diagnostic dike-on-dike textures or represents original differences in crustal sections. 5.4. Pillowed Basalts and Other Volcanics Lavas with well-preserved pillow structures are diagnostic parts of ANS ophiolites (Fig. 5F). These are especially important because lava compositions provide valuable clues about tectonic setting and the nature of melt generation. This section thus concentrates on the chemical composition of ANS ophiolitic lavas. We recognize that the greenschist-facies metamorphism that these basalts have suffered probably disrupted their chemical composition, but it is likely that these effects may cancel each other out (e.g., Mg gain in one lava, Mg loss in another). For the purpose of understanding the compositional variability of ANS ophiolitic lavas, we compiled available chemical data for these. A total of 200 samples were used in the compilation, including 37 analyses from Egypt (El-Sayed et al., 1999; Stern, 1981; Zimmer et al., 1995), 52 from Sudan (Price, 1984; AbdelRahman, 1990, 1993; Harms et al., 1994; Hussein, 2000), 32 from Arabia (excluding Darb Zubaydah; Al-Shanti, 1982; Bakor et al., 1976; Kattan, 1983; Nassief et al., 1984; Shanti, 1983), and 29 from the Sekerr ophiolite, Kenya (Price, 1984). An additional 50 samples from Eritrea (Woldehaimanot, 2000) and Ethiopia (Wolde et al., 1996; Woldehaimanot and Behrmann, 1995) were used to compare the metavolcanic sequences of suspected ophiolites in these countries. These data are summarized in Table 3. Table 3. Mean composition of ANS ophiolitic pillowed basalts Egypt

Arabia

Sudan

Ethiopia & Eritrea N 37 32 52 50 50 ± 3 49 ± 3 49 ± 4 SiO2 (%) 51 ± 4 1.3 ± 0.4 1.0 ± 0.5 1.1 ± 0.6 0.7 ± 0.7 TiO2 9.8 ± 2.4 9.4 ± 2.2 10.0 ± 1.8 11.2 ± 3.0 FeO∗ MgO 6.2 ± 2.0 6.9 ± 1.7 7.5 ± 2.2 9.7 ± 5.9 0.18 ± 0.13 0.28 ± 0.31 0.15 ± 0.21 0.43 ± 0.8 K2 O Mg# 52 ± 8 56 ± 7 56 ± 9 58 ± 14 Sr (ppm) 147 ± 144 143 ± 47 191 ± 114 192 ± 205 Ba 84 ± 81 25 ± 21 70 ± 44 72 ± 102 Y 33 ± 13 21 ± 9 24 ± 7 13 ± 10 Zr 87 ± 36 65 ± 44 70 ± 37 42 ± 42 V 320 ± 90 230 ± 75 287 ± 75 84 ± 67 Cr 295 ± 150 326 ± 212 356 ± 238 467 ± 591 Ni 110 ± 56 99 ± 76 113 ± 84 155 ± 246

Sekerr DPL 16 46 ± 2 1.4 ± 0.3 9.9 ± 1.0 8.4 ± 3.7 0.14 ± 0.09 58 ± 8 257 ± 135 99 ± 86 25 ± 7 111 ± 20 222 ± 40 621 ± 535 304 ± 305

Sekerr UPL 13 47 ± 2 2.7 ± 0.7 11.2 ± 0.9 4.6 ± 1.2 0.48 ± 0.35 42 ± 7 466 ± 85 259 ± 241 33 ± 7 171 ± 15 227 ± 19 160 ± 154 69 ± 81

ANS mean 200 49 ± 4 1.2 ± 0.7 10.2 ± 2.3 7.6 ± 3.8 0.26 ± 0.46 55 ± 11 199 ± 161 85 ± 112 23 ± 12 76 ± 49 234 ± 108 380 ± 392 135 ± 174

Data sources: Egypt (El-Sayed et al., 1999; Stern, 1981; Zimmer et al., 1995), 52 from Sudan (Abdel-Rahman et al., 1990; Price, 1984; Abdel-Rahman, 1993; Harms et al., 1994; Hussein, 2000); Arabia (excluding Darb Zubaydah; Al-Shanti, 1982; Bakor et al., 1976; Kattan, 1983; Nassief et al., 1984; Shanti, 1983); Kenya (Price, 1984); Eritrea (Woldehaimanot, 2000) and Ethiopia (Wolde et al., 1996; Woldehaimanot and Behrmann, 1995).

5. Ophiolite Components

113

ANS ophiolitic lavas are dominantly basalts, with a mean of 49% SiO2 (1 standard deviation = 4% SiO2 ). There are some andesitic and even a few dacitic samples identified as part of the ophiolitic sequence (e.g., Fawkhir, Egypt). These lavas average 1.2% (±0.7%) TiO2 . ANS ophiolitic pillow basalts define a low-K suite, with mean K2 O = 0.26%. These major element characteristics are similar to a wide range of oceanic lavas, including MORB, some intra-oceanic arc lavas, and back-arc basin basalt (BABB). Ophiolitic ANS lavas are often fractionated, with Mg# (= 100Mg/Mg + Fe) = 55 ± 11. Mg# for basaltic melt in equilibrium with mantle peridotite is expected to be in the range 65–70. The mean of 135 ppm Ni (±174 ppm) and 380 ppm Cr (±392 ppm) is higher than would be expected for a mean Mg# = 55. Eritrean and Ethiopia suspected ophiolitic lavas typically contain less TiO2 , Y, and Zr and have higher K2 O and Mg#, Cr and Ni than most other ANS ophiolitic lavas. Upper pillow lavas from the Sekerr ophiolite are also distinct, with much higher TiO2 , Y, and Zr along with lower Mg#, Cr, and Ni than most other ANS ophiolitic sequences. The Sekerr UPL sequence may represent an alkalic succession, something that is rarely found for other ANS ophiolitics. ANS ophiolitic lavas are mostly tholeiitic but also include some calc-alkaline examples (Fig. 8). There is no obvious geographic variation. Samples from Kenya and Eritrea are dominated by tholeiites whereas those from Ethiopia are predominantly calc-alkaline. Most samples from Sudan, Egypt, and Arabia are tholeiitic, but with a significant proportion of calc-alkaline representatives as well. REE data for 67 samples from the ANS (including several from suspected ophiolites in Ethiopia and Eritrea) indicate that both LREE-enriched and LREE-depleted varieties exist. A simple way to see the extent to which samples are LREE-enriched or LREEdepleted is by use of the chondrite normalized Ce/Yb ratio, or (Ce/Yb)n . If (Ce/Yb)n > 1, the REE pattern is generally LREE-enriched (and the higher the ratio, the greater the LREE-enrichment). Similarly, if (Ce/Yb)n < 1, the sample is LREE-depleted. The mean (Ce/Yb)n for ANS ophiolitic lavas is 2.18, but with a very large standard deviation of 3.07. This largely results from one unusual sample or analysis (MV33 with (Ce/Yb)n = 21.9; El-Sayed et al., 1999), and omitting this sample lowers the mean (Ce/Yb)n to 1.89 ± 1.88. Another way to summarize the REE patterns is to look at Sm/Nd. The chondritic value, taken to approximate the bulk Earth, is ∼ 0.325, so values greater than this are LREEdepleted and values less than this are LREE-enriched. The mean for ANS ophiolitic lavas (0.30 ± 0.13) again indicates modest LREE-enrichment. Including data for Eritrea and Ethiopia lowers the mean Sm/Nd to 0.29 ± 0.10. A variety of discriminant diagrams can be applied to examine the tectonic affinities of ANS ophiolitic lavas. A plot of V vs. Ti (Fig. 9) shows that most samples have Ti/V between 20 and 50, and plot in fields defined by mid-ocean ridge and back-arc basin basalts. This includes most of the samples from Egypt, Sudan, and Sekerr, Kenya, downstream pillow basalts. A substantial portion of the ophiolitic basalts have Ti/V < 20 and so plot in the field of island arc tholeiites or calc-alkaline basalts. This includes a subordinate proportion of samples from Egypt and Sudan. No field for boninitic rocks is shown on these diagrams, but boninites have very low Ti and V and plot near the origin. Eritrea and Ethiopia samples

114

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Fig. 8. Magmatic affinities of ANS ophiolitic pillow lavas and suspected ophiolitic metavolcanics from Eritrea and Ethiopia. (A) TiO2 vs. FeO∗ /MgO. (B) SiO2 vs. FeO∗ /MgO. Field boundaries after Miyashiro (1975).

have low Ti/V and some have very low Ti contents, supporting suggestions that these are boninitic. Only a subset of the upstream pillow lavas from the Sekerr ophiolite plot in the field of within-plate basalts.

5. Ophiolite Components

115

Fig. 9. Ti-V discrimination diagram for basalts, after Shervais (1982). IAT = arc tholeiite, MORB & BABB = mid-ocean ridge basalt and back-arc basin basalt, WPB = within-plate basalt. Calc-alkaline basalts have low Ti concentrations and a wide range of Ti/V and plot in the grey field.

A plot of Cr vs. Y shows similar mixed tectonic affinities for ANS ophiolitic lavas (Fig. 10). Samples from Egypt, Sudan, and downstream pillow lavas from the Sekerr ophiolite mostly plot in the MORB field (note that there is no field for back-arc basin basalts on this diagram). Samples from Arabia plot in the MORB field and in the field for arc basalts or even on the low-Y side of the field for arc basalts. Samples from Eritrea and Ethiopia plot within the field for arc basalts and to the low-Y side of the arc field. Again, only a subset of the upstream pillow lavas from the Sekerr ophiolite plot in the field of within-plate basalts. Finally, Zr-based discriminant diagrams (Fig. 11) show the mixed affinities of ANS ophiolitic lavas. On a plot of Ti vs. Zr (Fig. 11A), the samples are remarkably coherent, clustering about a mean Ti/Zr = 97 ± 41 (excluding Ethiopia and Eritrea), decreasing slightly to Ti/Zr = 95 ± 41 when data from Ethiopia and Eritrea are included. In contrast to Fig. 8, a calc-alkaline trend is not seen on the Ti-Zr diagram. Fig. 11A suggests that the ophiolitic lavas are simply related by differing degrees of melting and fractionation, with Sekerr ophiolite samples representing low degrees of melting (and/or extensive fractionation), whereas samples plotting near the origin (most samples from Ethiopia-Eritrea and some samples from Arabia, Sudan, and Egypt) are relatively high degree melts. Most of the

116

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Fig. 10. Cr-Y discrimination diagram for basalts (Pearce, 1982). Note that there is no field for back-arc basin basalts.

Sekerr upstream pillow lavas and a sample from Al Ays (sample 5 of Bakor et al., 1976), a sample from the Uogame basalts of Eritrea (ER54 of Woldehaimanot, 2000), and OPL/2 from Onib (Abdel-Rahman, 1993) lie along the extension of the MORB field in a region where within-plate basalts plot. Zr/Y vs. Zr systematics reveal similar affinities and also shows that a large proportion of samples plot to the low-Zr side of fields defined by arc basalts, MORB, and within-plate basalts. This is the area where boninites plot, using the summary data of Crawford et al. (1989), and samples from Arabia, Sudan, and especially Ethiopia and Eritrea plot in this field. 5.5. Sediments Pelagic sediments that rest immediately on the uppermost part of ANS ophiolites include dolomite and ribbon chert (e.g., Al Ays: Bakor et al., 1976; Hussein, 2000). These limestones are inferred to be similar to modern pelagic carbonates, whereas the cherts are thought to be due to silica-rich emanations without the involvement of radiolaria. Metased-

5. Ophiolite Components

117

Fig. 11. Zr-based discriminant diagrams for ANS ophiolitic basalts. (A) Ti vs. Zr (Pearce and Cann, 1973). Field labels: I = arc tholeiites, II = MORB, III = calc-alkaline basalt. A fourth field, situated within I but obscured by data points is defined by all three fields. No field for within-plate lavas is shown, but these plot along the upper right extension of the MORB field. (B) Zr/Y vs. Zr (Pearce and Norry, 1979). Gray field shows composition of boninites, from Table 1-1 of Crawford et al. (1989).

iments overlying the Ess ophiolite include shale and minor conglomerate (Shanti and Roobol, 1979). Ophiolites in the Central Eastern Desert of Egypt are overlain by sediments that include diamictite, tuffaceous siltstones, and banded iron formation (Stern, 1981; Sims and James, 1982). Future ophiolite studies should take care to carefully characterize the nature of the sedimentary sequence immediately above the pillowed lavas.

118

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

All too little attention has been paid to the nature and composition of sediments that were deposited on ANS ophiolites, probably because investigations of these ophiolites have mostly been led by petrologists and igneous geochemists. Because the Neoproterozoic was an exceptional period in Earth history, characterized by extensive—perhaps global— glaciation, unparalleled excursions in seawater stable isotopic compositions, and explosive radiation of metazoa (Evans, 2000), sediments of this age are of increasing interest to the global geoscientific community. Efforts to understand the Neoproterozoic ‘Snowball Earth’ have concentrated to date on shallow water sequences, with very little of the record preserved in deep-sea sediments examined. Because sediments deposited on ANS ophiolites should preserve an excellent record of the composition of deep water of the Neoproterozoic ocean, these should become an increasing focus of research in the near future. 5.6. Alteration and Obduction-Related Metamorphism The ultramafic rocks associated with ANS ophiolites are generally highly altered, but it is often not known whether this alteration occurred before, during, or after emplacement. These rocks are largely converted to serpentinite or to mixtures of serpentine, talc, tremolite, magnesite, chlorite, magnetite, and carbonate. These rocks are variously called ‘talc-carbonate schists’, ‘Barramiya rocks’, or ‘listwaenite’. Strictly speaking, listwaenite should be reserved for rocks that are fuchsite-quartz-carbonate lithologies derived from ultramafic rocks by potassic and carbonate metasomatism (Halls and Zhao, 1995), but common usage in the ANS is more general. Silicified serpentinites are sometimes called ‘birbirites’ (Augustithus, 1965). Basta and Kader (1969) reported that lizardite is the main constituent of Egyptian serpentinites, whereas Akaad and Noweir (1972) identify antigorite as volumetrically dominant. The talc-carbonate rocks mainly consist of magnesite (± dolomite) and talc. The origin of the carbonate alteration fluids remains to be elucidated, but Stern and Gwinn (1990) argued on the basis of C and Sr isotopic studies that carbonate intrusions in the Eastern Desert of Egypt—which could be related to the carbonatizing fluids affecting ANS ultramafic rocks—are mixtures of mantlederived and remobilized sedimentary carbonate. Certainly the prevalence of carbonate alteration of ANS ophiolitic ultramafics suggests a tremendous flux of CO2 -rich fluids from the mantle during middle and late Neoproterozoic time (Newton and Stern, 1990; Stern and Gwinn, 1990). In contrast, Surour and Arafa (1997) argued that the ‘ophicarbonates’ of the Ghadir ophiolite are reworked oceanic calcites that formed after it was obducted. Regardless of how carbonatization of the ophiolitic ultramafics occurred, it has economic implications. A spatial and genetic relationship has been observed between carbonatized ultramafics, subsequent granite intrusions, and gold mineralization. Apparently the carbonatization preconcentrates gold up to 1,000 times that in the original ultramafic rocks, and interaction with hydrothermal systems associated with granite intrusions may further concentrate gold (Cox and Singer, 1986). Thrust contacts are documented at the base of some, but not all, ANS ophiolites (Fig. 5C). Metamorphic soles of ANS ophiolites have been studied along the west-

6. Isotopic Data

119

ern part of the Allaqi-Heiani ophiolite belt in SE Egypt, where El-Naby and Frisch (1999) inferred temperatures of up to 700 ◦ C and pressures up to 8 kbar. Other thrusts at the base of the ophiolites are associated with no obvious thermal effects. Chloritites are developed around the peripheries of some ultramafic masses (Takla, 1991; Takla et al., 1992). The metamorphic sole of the Halaban ophiolite consists of tightly folded amphibolites, hornblendites, gneisses, magmatirestites, rodingites and serpentinized ultramafics, all of which are cut by granitic dikes associated with partial melting (Al-Saleh et al., 1998). The metamorphic sole of the Halaban ophiolite in Arabia is divided into a western segment that is dominated by greenschist to amphibolite facies metasedimentary rocks, structurally overlain by higher grade amphibolite-facies meta-igneous rocks. Such inverted metamorphic zonation is common in sub-ophiolitic complexes (Al-Saleh et al., 1998). Most amphibolites and meta-sediments in the latter segment have experienced varying degrees of calcium metasomatism due to the release of excess calcium during serpentinization, and the original assemblages have often been rodingitized. The Halaban ophiolitic sole includes exotic blocks of serpentinite and metamorphosed ultramafic rock, interpreted to be derived from the basal peridotites of the Halaban Ophiolite that became fragmented as thrusting progressed and were later incorporated within the underlying amphibolites.

6. ISOTOPIC DATA There has been a modest amount of isotopic work conducted on ANS ophiolites, including two ophiolites in Arabia, a concentration of work around the Gerf ophiolite in SE Egypt, and data for the Adola mafic-ultramafic complex. The most reliable isotopic data for ANS ophiolites comes from Sm-Nd isotopic systematics. This is because alteration makes it difficult to rely on Sr and Pb isotopic compositions, whereas Sm and Nd are relatively immobile and corrections for in situ radiogenic growth is simple if the age is known. These results are summarized in Table 4, which shows that all ANS ophiolites studied to date have strongly positive εNd (+5.0 to +7.7). This indicates that these melts were generated from a long-depleted (high Sm/Nd) mantle source. The Nd isotopic data indicate an asthenospheric source and a juvenile, ensimatic setting. There is a hint that the mantle source Table 4. Neodymium isotopic composition of ANS ophiolites Ophiolite Ess (2) Al ‘Ays (Wask) Harga Zarqa (10) Heiani (7) Gerf (20) Adola

Age (Ma) 780 743 750 750 750 789

εNd(t) 6.9 7.6 7.6 ± 0.5 7.7 ± 0.8 6.8 ± 0.7 5.0

Numbers in parentheses refer to number of samples used to calculate mean value.

Reference Claesson et al. (1984) Claesson et al. (1984) Zimmer et al. (1995) Zimmer et al. (1995) Zimmer et al. (1995) Worku (1996)

120

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

for the northern ophiolites may have been more—or longer—depleted than those farther south, but the data at present are too sparse for this to be anything more than a suggestion. Os isotope data on chromites from the Al Ays ophiolite have recently been reported (Walker et al., 2002); these straddle the line defined for the evolution of the primitive mantle.

7. DISCUSSION Ophiolites of mid-Neoproterozoic age (690 to 890 Ma; mean = 781 ± 47 Ma) are abundant in NE Africa and Arabia. Ophiolites encompass an area of 3000 km N-S and > 1000 km E-W. These ophiolites are in various stages of dismemberment and alteration, but all of the diagnostic components can be found, including harzburgite, cumulate ultramafics, layered as well as higher level gabbro and plagiogranite, sheeted dikes, and pillowed basalt. Reconstructions of a few ophiolitic successions indicate a crustal thickness of 2.5 to 6 km. Many ANS ophiolites mark suture zones where smaller terranes coalesced; these suture zones indicate the location of fossil subduction zones. Some ANS ophiolites were emplaced while still hot enough to metamorphose underlying rocks, while others were emplaced cold. This gross tectonic setting is simplest to explain if ANS ophiolites generally were located on the hanging wall of a convergent plate margin and were emplaced when buoyant crust entered the subduction zone. Such an event could cause the subduction zone to fail, suturing the two terranes at the same time that the ophiolite was emplaced (Cloos, 1993). This describes a forearc setting for such ophiolites, an interpretation that finds increasing favor in the scientific community (Shervais, 2001). Mineral and lava compositions as well as limited isotopic data are consistent with the hypothesis that most ANS ophiolites formed in ‘suprasubduction zone’ (SSZ) settings. Most ANS ophiolites have the hallmarks of forearc ophiolites. Harzburgites are the most common type of mantle peridotite, and these contain magnesian olivines and spinels with compositions that indicate large extents of melting. Limited data for relict olivines in harzburgites show these to be significantly more Mg-rich than peridotites recovered from modern mid-ocean ridges and similar to olivines in harzburgites recovered from forearcs. Limited data for olivines from backarc basin peridotites are indistinguishable from MORB peridotites, arguing against a back-arc basin setting. This interpretation is consistent with spinel compositions. Cr# for spinels in ANS harzburgites are mostly > 60, again most like those recovered from modern forearcs and distinctly higher than those from mid-ocean ridges and the admittedly sparse database for backarc basin peridotites. ANS ophiolites are often associated with a thick (1–3 km) sequence of cumulate ultramafic rocks, which define a transition zone between the seismic and petrologic Mohos. These cumulates consist of a very high proportion of dunite but there are also a lot of pyroxene-rich lithologies. Chromites associated with dunites are often more Cr-rich than those in the underlying spinels. These cumulate ultramafics transition upwards into layered gabbro. Together these cumulate sequences indicate that the extensional magmatic systems represented by ANS ophiolites experienced significant fractionation. Several crystal-

7. Discussion

121

lization sequences are inferred from ANS transition zones and cumulate gabbro sections. Olivine and spinel are always first to crystallize: especially cpx-plag and cpx-opx-plag; opx-cpx-plag is rare. These are crystallization sequences (respectively) C, B, and A of Pearce et al. (1984), identified as hallmarks of SSZ ophiolites. This sequence contrasts with the crystallization sequence olivine-plagioclase-clinopyroxene which Pearce et al. (1984) interpreted to be diagnostic of ophiolites that formed at true mid-ocean ridges. The latter crystallization sequence has not been reported for ANS ophiolites. The overall sense of ANS ophiolites inferred from the transition zone and the gabbroic section is that of large, fractionating magma chambers that were repeatedly tapped and recharged but which generally evolved from a system dominated by primitive mafic magmas to ones dominated by highly evolved magmatic liquids. ANS ophiolitic lavas mostly define a subalkaline suite characterized by low K and moderate Ti contents that is moderately fractionated. They reveal both tholeiitic and calcalkaline affinities and include a significant if subordinate proportion of boninites. We find little in the structure or composition of ANS ophiolites to support the hypothesis that ANS crustal growth entailed widespread involvement of oceanic plateaus or large igneous provinces, although rare examples of within-plate lavas are identified. ANS ophiolitic lavas are quite fractionated (mean Mg# = 55) but have higher abundances of Cr (mean = 380 ppm) and Ni (mean = 135 ppm) than would be expected for this relatively low Mg#. ANS ophiolitic lavas include both LREE-depleted and LREE-enriched varieties, but as a group are slightly LREE-enriched: mean (Ce/Yb)n ∼ 2.2 and Sm/Nd ∼ 0.30. On a variety of discrimination diagrams, ANS ophiolitic lavas plot in fields for MORB, BABB, and arc tholeiites, along with a significant proportion of lavas with strong boninitic affinities. ANS lavas cluster reasonably tightly around Ti/Zr = 97, indicating that Ti-bearing phases did not precipitate early. These mixed subalkaline characteristics are characteristic of SSZ ophiolitic lavas, and the presence of boninitic lavas in particular supports a forearc origin. There is a tremendous amount of work that needs to be done on ANS ophiolites. Most of the information that we have on ANS ophiolites was collected in the 1980s and early 1990s and the field is ripe for new perspectives and for detailed studies using modern techniques. There have been few modern geochemical studies. Most of the trace element data summarized in this overview was generated with XRF techniques. We need to analyze ANS ophiolitic basalts using modern plasma analytical techniques. Modern ICP-MS techniques result in much better precision and accuracy for many more elements, including Nb and Ta, which are essential to understand tectonic setting of igneous rocks. Interestingly, one of the few ANS ophiolites that have been studied with modern geochemical techniques—the Gerf ophiolite—has strong trace element affinities to true MORB. The geochronologic and isotopic database is limited and additional data are needed if we are to understand when the crust and mantle lithosphere represented by these ophiolites were generated as well as when a given ophiolite was emplaced. Limited Sm-Nd isotopic data indicate derivation from depleted asthenospheric mantle. Isotopic data however do not resolve tectonic setting. There are no Hf isotopic data for these ophiolites, and such analyses should provide a valuable perspective on the evolution of the Lu-Hf isotopic sys-

122

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

tem in the Neoproterozoic mantle. We also need to better understand the composition of primary magmatic liquids; compositions of olivines and spinels in ANS harzburgites indicate unusually high degrees of melting so primary liquids should be very primitive but lava compositions are very fractionated. One way to resolve this apparent contradiction would be to analyze clinopyroxenes in ANS ophiolitic harzburgites using an ion microprobe to determine compositions of REE and other trace elements in equilibrium liquids. This would allow a more explicit linkage between the composition of ophiolitic lavas and the extent of melting in the associated mantle section to be examined. Another very useful strategy would be to identify and study inclusions in spinels, which have recently been discovered to preserve primary magmatic phases. One of the most interesting unresolved problems concerns the nature of the carbonate alteration of especially the ultramafics. Where did all this CO2 come from? Is there any relationship between this alteration and gold mineralization in the ANS? Several workers suggest a link between more depleted, boninitic ophiolites and gold mineralization; is there relationship between gold mineralization in the ANS and ophiolite type? Modern regions of seafloor spreading often are associated with hydrothermal vents and associated mineralization and biota, are such associations preserved in ANS ophiolites? Such studies may be especially important because ANS ophiolites may have formed at a time when a variety of independent geologic observations suggest that the surface of the earth was covered with ice (Evans, 2000), and life may have been restricted to such vents. Similarly, the association of banded iron formations overlying ANS ophiolites (Sims and James, 1982) may reflect interactions between deep water and hydrothermal activity during a snowball earth episode. Another aspect of ANS ophiolites worthy of study is a possible relationship between economic chromite deposits and lava compositions. In some ophiolites (e.g., Zambales, Philippines) the affinity of the magmatic section is a reliable guide to the grade of associated chromite deposits (Evans and Hawkins, 1989).

8. CONCLUSIONS The last three decades of research on ANS ophiolites allows us to sketch the broad outlines of how these complexes formed, but most of this work was completed a decade or more ago. New research initiatives that focus on ANS ophiolites are needed and these promise to be rewarding. A few examples of what needs to be done are presented below. ANS ophiolites change style on either side of the Bi’r Umq-Nakasib suture zone. Are these correlative? We are not certain that the allochthonous mafic-ultramafic complexes of Eritrea and Ethiopia should be considered as ophiolites, although we have argued for this. Compositions of spinels in harzburgite are a powerful tool for understanding the magmatic evolution and tectonic setting of ophiolites, but we have too few analyses for ophiolitic harzburgites from Egypt, Eritrea, and Ethiopia. Similarly, we need more mineral chemical data for primary ophiolite phases (olivine, clinopyroxene, plagioclase). We need new campaigns of petrologic and geochemical studies using the most modern analytical techniques

Acknowledgements

123

(ICP-MS, etc.) to determine the diagnostic trace element contents of ANS ophiolitic lavas. Some modern analytical techniques, such as using ion probe analyses of clinopyroxenes to independently infer trace element compositions of equilibrium liquids, have not yet been applied. More isotopic studies, especially Nd, Hf, and Os, are needed to understand the evolution of the mantle source region for these lavas. Similarly, we need more geochronological constraints on the age of ANS ophiolites. Ion probe ages of zircons would be especially useful because this best identifies inherited components but this technique has not been applied to ANS ophiolites. Precise determination of ophiolite age will also be critical for studies of deep water marine sequences immediately overlying the pillow lavas, a research direction that is needed to understand global change during Neoproterozoic time. We should identify the most complete ANS ophiolites and study them in detail. Interdisciplinary study of the best-preserved ophiolites by teams of structural geologists, petrologists, geochemists, and geochronologists focusing on the ophiolite itself in tandem with sedimentologists, geochemists, and paleontologists studying the overlying sedimentary succession should be encouraged. Efforts should be made to identify ancient hydrothermal vent deposits and investigate the associated fossilized biota. Economic considerations also favor renewed study of ANS ophiolites. The high grade of ANS chromites may be rich enough to mine. Given the observation that ANS gold mineralization often appears to be related to carbonate alteration of ANS ophiolitic peridotites, we can expect to better understand the former by focused studies of the latter. At present we have a very poor understanding of pervasive carbonate alteration of ANS ophiolitic peridotites. Research programs should be designed to better understand the age of this alteration and the origins of these fluids.

ACKNOWLEDGEMENTS Thanks to T. Tadesse (Ethiopian Geological Survey) and B. Woldhaimanot (U. Asmara) for their thoughts on the ‘ophiolites’ of Eritrea and Ethiopia. Thanks to J. Encarnacion and Y. Dilek for thoughtful reviews. This manuscript is dedicated to the memory of Ian Gass, who stimulated so many excellent studies of ANS ophiolites and whose enthusiasm for the ophiolites of the Arabian-Nubian Shield has yet to be matched.

REFERENCES Abdel-Rahman, E.M., et al., 1990. A new ophiolite occurrence in NW Sudan—constraints on late Proterozoic tectonism. Terra Nova 2, 363–376. Abdel-Rahman, M., 1993. Geochemical and geotectonic controls of the metallogenic evolution of selected ophiolite complexes from the Sudan. Berliner geowissenschaftliche Abhandlungen 145, 175. Abdelsalam, M.G., Stern, R.J., 1993. Tectonic evolution of the Nakasib suture, Red Sea Hills, Sudan: evidence for a late Precambrian Wilson cycle. Journal of the Geological Society of London 150, 393–404.

124

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Abdelsalam, M.G., Stern, R.J., 1996. Sutures and Shear Zones in the Arabian-Nubian Shield. Journal of African Earth Sciences 23, 289–310. Ahmed, A.H., Arai, S., Attia, A.K., 2001. Petrological characteristics of podiform chromitites and associated peridotites of the Pan African ophiolite complexes of Egypt. Mineralium Deposita 36, 72–84. Akaad, M.K., Noweir, A.M., 1972. Some Aspects of the Serpentinites and their Associated Derivatives along Qift-Quesir Road, Eastern Desert. Annals of the Geological Survey of Egypt 2, 251– 270. Al-Rehaili, M.H., Warden, A.J., 1980. Comparison of the Bi’r Umq and Hamdah ultrabasic complexes, Saudi Arabia. Institute of Applied Geology Bulletin 3 (4), 143–156. Al-Saleh, A.M., Boyle, A.P., Mussett, A.E., 1998. Metamorphism and 40 Ar/39 Ar dating of the Halaban ophiolite and associated units: evidence for two-stage orogenesis in the eastern Arabian shield. Journal of the Geological Society of London 155, 165–175. Al-Shanti, A.M., El-Mahdy, O.R., 1988. Geological Studies and Assessment of Chromite Occurrences in Saudi Arabia. KACST Project No. AT-6-094. Al-Shanti, M.M.S., 1982. Geology and mineralization of the Ash Shizm-Jabal Ess area. Ph.D. thesis. King Abdulaziz University, p. 291. Augustithus, S.S., 1965. Mineralogical and geochemical studies of the platiniferous dunite-birbiritepyroxenite complex of Yubdo/Birbir, W. Ethiopia. Chemie der Erde 24, 159–196. Bakor, A.R., Gass, I.G., Neary, C.R., 1976. Jabal al Wask, northwest Saudi Arabia: an Eocambrian back-arc ophiolite. Earth and Planetary Science Letters 30, 1–9. Barnes, S.J., Roeder, P.L., 2001. The Range of Spinel Compositions in Terrestrial Mafic and Ultramafic Rocks. Journal of Petrology 42, 2279–2302. Basta, E.Z., Kader, Z.A., 1969. The mineralogy of Egyptian serpentinites and talc-carbonates. Mineralogical Magazine 37, 394–408. Berhe, S.M., 1988. The Geologic and Tectonic Evolution of the Pan African/Mozambique Belt in East Africa. Ph.D. thesis. Open University, Milton Keynes, p. 265. 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. American Geophysical Union, Washington, DC, pp. 1–30. Bonatti, E., Michael, P.J., 1989. Mantle peridotites from continental rifts to ocean basins to subduction zones. Earth and Planetary Science Letters 91, 297–311. Bonavia, F.F., Diella, V., Ferrario, A., 1993. Precambrian Podiform Chromitites from Kenticha Hill, southern Ethiopia. Economic Geology 88, 198–202. Chevremont, P., Johan, Z., 1982a. The Al Ays ophiolite complex. BRGM-OF-02-5. Saudi Arabia Deputy Ministry for Mineral Resources. Chevremont, P., Johan, Z., 1982b. Wadi al Hwanet-Jabal Iss ophiolite complex. BRGM-OF-02-14. Saudi Arabia Deputy Ministry for Mineral Resources. Church, W.R., 1988. Ophiolites, structures, and micro-plates of the Arabian-Nubian shield: a critical comment. In: Greiling, S.E.-G.a.R.O. (Ed.), The Pan-African Belt of Northeast Africa and Adjacent Areas. Veiweg, Braunschweig/Wiesbaden, pp. 289–316. Claesson, S., Pallister, J.S., Tatsumoto, M., 1984. Samarium-neodymium data on two late Proterozoic ophiolites of Saudi Arabia and implications for crustal and mantle evolution. Contributions to Mineralogy and Petrology 85, 244–252.

References

125

Cloos, M., 1993. Lithospheric buoyancy and collisional orogenesis: Subduction of oceanic plateaus, continental margins, island arcs, spreading ridges, and seamounts. Geological Society of America Bulletin 105, 715–737. Cox, D.P., Singer, D.A., 1986. Grade and tonnage model of porphyry Cu-Au. In: Cox, D.P., Singer, D.A. (Eds.), Mineral Deposit Models. U.S. Geological Survey Bulletin, 110–114. Crawford, A.J., Falloon, T.J., Green, D.H., 1989. Classification, petrogenesis and tectonic setting of boninites. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Cambridge Univ. Press, Cambridge, pp. 1–49. Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. Geological Society of America Bulletin 109, 16–42. 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. 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, P.T.D., Karson, J.A., Lagabrielle, Y. (Eds.), Faulting and Magmatism at Mid-Ocean Ridges. AGU, Washington, DC, pp. 219–265. Dixon, T.H., 1981. Gebel dahanib, a late Precambrian layered sill of komatiitic composition. Contributions to Mineralogy and Petrology 76, 42–52. Drury, S.A., Filho, C.R.d.S., 1998. Neoproterozoic terrane assemblages in Eritrea: review and prospects. Journal of African Earth Sciences 27, 331–348. El-Bayoumi, R.M., 1983. Ophiolites and mélange complex of Wadi Ghadir Area, Eastern Desert, Egypt. Bulletin of the Faculty of Earth Sciences, King Abdulaziz University, Jiddah 6, 329–342. El-Naby, H.A., Frisch, W., 1999. Metamorphic sole of Wadi Haimur-Abu Swayel ophiolite: Implications on late Proterozoic accretion. In: Wall, H.D., Greiling, R.O. (Eds.), Aspects of Pan-African Tectonics. Heidelberg, Germany, pp. 9–14. El-Sayed, M.M., Furnes, H., Mohamed, F.H., 1999. Geochemical constraints on the tectonomagmatic evolution of the late Precambrian Fawakhir ophiolite, Central eastern Desert, Egypt. Journal of African Earth Sciences 29, 515–533. Evans, C., Hawkins, J.W., 1989. Compositional heterogeneities in upper mantle peridotites from the Zambales Range Ophiolite, Luzon, Philippines. Tectonophysics 168, 23–41. Evans, D., 2000. Stratigraphic, geochronological, and paleomagnetic constraints upon the Neoproterozoic climatic paradox. American Journal of Science 300, 347–433. Fitches, W.R., et al., 1983. The Late Proterozoic ophiolite of Sol Hamed, NE Sudan. Precambrian Research 19, 515–533. Halls, C., Zhao, R., 1995. Listwaenite and related rocks: perspectives on terminology and mineralogy with reference to an occurrence at Cregganbaun, Co. Mayo, Republic of Ireland. Mineralium Deposita 30, 303–313. Harms, U., Darbyshire, D.P.F., Denkler, T., Hengst, M., Schandelmeier, H., 1994. Evolution of the Neoproterozoic Delgo suture zone and crustal growth in Northern Sudan: Geochemical and radiogenic isotope constraints. Geologische Rundschau 83, 591–603. Henjes-Kunst, F., Altherr, R., Baumann, A., 1994. Evolution and composition of the lithospheric mantle underneath the western Arabian Peninsula; constraints from Sr-Nd isotope systematics of mantle xenoliths. Contributions to Mineralogy and Petrology 105, 460–472. Hussein, I.M., 2000. Geodynamic evolution of the Pan-African crystalline basement in the northern Red Sea Hills, Sudan, with special emphasis on the Wadi Onib ophiolite and geology to the west of Port Sudan. Joahnnes Gutenberg Universität-Mainz, Mainz.

126

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Hussein, I.M., Kröner, A., Durr, S., 1983. Wadi Onib-A Dismembered Pan-African Ophiolite in the Red Sea Hills of Sudan. Bulletin of the Faculty of Earth Sciences, King Abdulaziz University, Jiddah 6, 320–327. Hussein, I.M., Kröner, A., Reischmann, T., this volume. The Wadi Onib mafic-ultramafic complex: a Neoproterozoic supra-subduction zone ophiolite in the northern Red Sea Hills of the Sudan. Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004. Neoproterozoic ophiolites in the Arabian Shield: Field relations and structure. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162. Kattan, F.H., 1983. Petrology and Geochemistry of the Tuluhah Belt, North East Arabian Shield. M.S. thesis. King Abdulaziz University, Jiddah. Knoll, A.H., 2000. Learning to tell Neoproterozoic Time. Precambrian Research 100, 3–20. Kröner, A., et al., 1987. Pan-African Crustal Evolution in the Nubian segment of Northeast Africa. In: Kröner, A. (Ed.), Proterozoic Lithosphere Evolution. American Geophysical Union, Washington, DC, pp. 235–257. Kröner, A., Todt, W., Hussein, I.M., Mansour, M., Rashwan, A.A., 1992. Dating of late Proterozoic ophiolites in Egypt and Sudan using the single grain zircon evaporation technique. Precambrian Research 59, 15–32. Kusky, T.M., Abdelsalam, M., Tucker, R., Stern, R., 2003. Evolution of The East African and Related Orogens, and the Assembly of Gondwana. Special Issue of Precambrian Research 123, 81–344. Ledru, P., Auge, T., The Al Ays ophiolitic complex; petrology and structural evolution. BRGM-OF04-15. Saudi Arabia Deputy Ministry for Mineral Resources. LeMetour, J., Johan, V., Tegyey, M., Relationships between ultramafic-mafic complexes and volcanosedimentary rocks in the Precambrian Arabian Shield. BRGM-OF-12-15. Saudi Arabia Deputy Ministry for Mineral Resources. McGuire, A.V., Stern, R.J., 1993. Granulite xenoliths from western Saudi Arabia: The lower crust of the Late Precambrian Arabian-Nubian Shield. Contributions to Mineralogy and Petrology 114, 395–408. Meert, J.G., 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics 362, 1–40. Miyashiro, A., 1975. Classification, Characteristics, and Origin of Ophiolites. Journal of Geology 83, 249–281. Mouhamed, M.F.S., 1995. Geology, Geochemistry and Structure of Gabal Muqsim Area and Environs, South Eastern Desert, Egypt. Forschungszentrum Jülich, Jülich. Nassief, M.O., Macdonald, R., Gass, I.G., 1984. The Jebel Thurwah upper Proterozoic ophiolite complex, western Saudi Arabia. Journal of the Geological Society of London 141, 537–546. Newton, R.C., Stern, R.J., 1990. A Late Precambrian CO2 “Event”. Geological Society of America, Dallas, p. 190. Ohara, Y., Stern, R.J., Ishii, T., Yurimoto, H., Yamazaki, T., 2002. Peridotites from the Mariana Trough: First look at the mantle beneath an active backarc basin. Contributions to Mineralogy and Petrology 143, 1–18. Pallister, J.S., Stacey, J.S., Fischer, L.B., Premo, W.R., 1988. Precambrian ophiolites of Arabia: Geologic settings, U-Pb geochronology, Pb-isotope characteristics, and implications for continental accretion. Precambrian Research 38, 1–54. Patchett, P.J., Chase, C.G., 2002. Role of transform continental margins in major crustal growth episodes. Geology 30, 39–42. Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plate boundaries. In: Thorpe, R.S. (Ed.), Andesites. Wiley, Chichester, pp. 525–548.

References

127

Pearce, J.A., Cann, J.R., 1973. Ophiolite origin investigated by discriminant analysis. Earth and Planetary Science Letters 12, 339–349. Pearce, J.A., Lippard, S.J., Roberts, S., 1984. Characteristics and tectonic significance of suprasubduction zone ophiolites. In: Kokelaar, B.P., Howell, M.F. (Eds.), Marginal Basin Geology. Geological Society of London, pp. 77–94. Pearce, J.A., Norry, M.J., 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33–47. Price, R.C., 1984. Late Precambrian Mafic-Ultramafic Complexes in Northeast Africa. Ph.D. thesis. Open University, Milton Keynes, p. 325. Quick, J.E., 1990. Geology and origin of the Late Proterozoic Darb Zubaydah ophiolite, Kingdom of Saudi Arabia. Geological Society of America Bulletin 102, 1007–1020. Quick, J.E., 1991. Late Proterozoic transpression of the Nabitah fault system—implications for the assembly of the Arabian shield. Precambrian Research 53, 119–147. Quick, J.E., Bosch, P.S., 1989. Tectonic history of the northern Nabitah fault zone, Arabian Shield, Kingdom of Saudi Arabia. USGS-TR-08-2. Saudi Arabia Directorate General of Mineral Resources. Rittmann, A., 1958. Geosynclinal volcanism, Ophiolites and Barramiya rocks. The Egyptian Journal of Geology 2, 61–66. Schandelmeier, H., et al., 1994. Atmur-Delgo suture: a Neoproterozoic oceanic basin extending into the interior of northeast Africa. Geology 22, 563–566. Shanti, M., 1983. The Jabal Ess ophiolite complex. Bulletin of the Faculty of Earth Sciences, King Abdulaziz University, Jiddah 6, 289–317. Shanti, M., Roobol, M.J., 1979. A late Proterozoic ophiolite complex at Jabal Ess in northern Saudi Arabia. Nature 279, 488–491. Shervais, J.W., 1982. Ti-V plot and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59, 101–118. Shervais, J.W., 2001. Birth, death, and resurrection; the life cycle of suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems 2, 2000GC000080. Sims, P.K., James, H.L., 1982. Banded iron-formations of late Proterozoic age in the central Eastern Desert, Egypt; geology and tectonic setting. Economic Geology 79, 1777–1784. Stacey, J.S., Stoeser, D.B., Greenwood, W.R., Fischer, L.B., 1984. U-Pb zircon geochronology and geological evolution of the Halaban-Al Amar region of the Eastern Arabian Shield, Kingdom of Saudi Arabia. Journal of the Geological Society of London 141, 1043–1055. Stein, M., Goldstein, S.L., 1996. From plume head to continental lithosphere in the Arabian-Nubian Shield. Nature 382, 773–778. Stern, R.J., 1981. Petrogenesis and tectonic setting of late Precambrian enzymatic volcanic rocks, Central Eastern Desert of Egypt. Precambrian Research 16, 195–230. Stern, R.J., 1994. Arc Assembly and continental collision in the Neoproterozoic East African Orogen: Implications for the assembly of Gondwanaland. Annual Reviews of Earth and Planetary Sciences 22, 319–351. Stern, R.J., 2002. Crustal evolution in the East African Orogen: a neodymium isotopic perspective. Journal of African Earth Sciences 34, 109–117. Stern, R.J., Gwinn, C.J., 1990. Origin of Late Precambrian Intrusive Carbonates, Eastern Desert of Egypt and Sudan: C, O, and Sr Isotopic Evidence. Precambrian Research 46, 259–272. Stern, R.J., et al., 1990. Orientation of late Precambrian sutures in the Arabian-Nubian shield. Geology 18, 1103–1106.

128

Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

Sultan, M., Arvidson, R.E., Duncan, I.J., Stern, R.J., Kaliouby, B.E., 1988. Extension of the Najd shear system from Saudi Arabia to the central Eastern Desert of Egypt based on integrated field and Landsat observations. Tectonics 7, 1291–1306. Sultan, M., Arvidson, R.E., Sturchio, N.C., 1986. Mapping of serpentinites in the Eastern Desert of Egypt by using Landsat thematic mapper data. Geology 14, 995–999. Surour, A.A., Arafa, E.H., 1997. Ophicarbonates: calcified serpentinites from Gebel Moghara, Wadi Ghadir area, Eastern Desert, Egypt. Journal of African Earth Sciences 24, 315–324. Takla, M.A., 1991. Chloritites at the contacts of some ophiolitic ultramafics, Eastern Desert, Egypt. Egyptian Mineralogist 3, 151–165. Takla, M.A., Basta, F.F., Surour, A.A., 1992. Petrology and Mineral Chemistry of Rodingites associating the Pan-African Ultramafics of Sikait-Abu Rusheid area, South Eastern Desert, Egypt. Geology of the Arab World, 491–507. 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. Wolde, B., Asres, Z., Desta, Z., Gonzalez, J.J., 1996. Neoproterozoic zirconium-depleted boninite and tholeiitic series rocks from Adola, southern Ethiopia. Precambrian Research 80, 261–279. Woldehaimanot, B., 2000. Tectonic setting and geochemical characterization of Neoproterozoic volcanics and granitoids from the Adobha Belt, northern Eritrea. Journal of African Earth Sciences 30, 817–831. Woldehaimanot, B., Behrmann, J.H., 1995. A study of metabasite and metagranite chemistry in the Adola region (south Ethiopia): Implications for the evolution of the East African orogen. Journal of African Earth Sciences 21, 459–476. Worku, H., 1996. Geodynamic Development of the Adola Belt (Southern Ethiopia) in the Neoproterozoic and Its Control on Gold Mineralization. Verlag Dr. Köster, Berlin, p. 156. Yibas, B., Reimold, W.U., Anhaeusser, C.R., Koeberl, C., 2003. Geochemistry of the mafic rocks of the ophiolitic fold and thrust belts of southern Ethiopia: constraints on the tectonic regime during the Neoproterozoic (900–700 Ma). Precambrian Research 121, 157–183. Yihunie, T., 2002. Pan-African deformations in the basement of the Negele area, southern Ethiopia. International Journal of Earth Science 91, 922–933. Zimmer, M., Kröner, A., Jochum, K.P., Reischmann, T., Todt, W., 1995. The Gabal Gerf complex: A Precambrian N-MORB ophiolite in the Nubian Shield, NE Africa. Chemical Geology 123, 29–51.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

129

Chapter 4

NEOPROTEROZOIC OPHIOLITES IN THE ARABIAN SHIELD: FIELD RELATIONS AND STRUCTURE PETER R. JOHNSONa , FAYEK H. KATTANa AND AHMED M. AL-SALEHb a Saudi Geological Survey,

PO Box 54141, Jiddah 21514, Saudi Arabia Department, King Saud University, PO Box 2009, Riyadh, Saudi Arabia

b Geology

Ophiolites make up a small but tectonically important part of the Arabian shield. Where most complete, they consist of serpentinized peridotite, gabbro, dike complex, basalt, and pelagic rocks. However, because of folding and shearing, the majority of the ophiolites lack one or more of these diagnostic lithologies. Nonetheless, the incomplete assemblages are identified as ophiolites because they minimally include peridotite and gabbro, in many cases are associated with basalt, and in all cases show evidence of emplacement by thrusting and shearing rather than intrusion. The ophiolites range in age from ∼ 870 Ma to ∼ 695 Ma, documenting a 200-million year period of oceanic magmatism in the Arabian shield, and are caught up in ∼ 780 Ma to ∼ 680 Ma suture zones that reflect a 100-million year period of terrane convergence. All the ophiolites are strongly deformed, metamorphosed, and altered by silicification and carbonatization. Low-grade greenschist facies metamorphism predominates, but in places the rocks reach amphibolite grade. Alteration resulted in the development of listwaenite, particularly in shear zones, and locally the only evidence that mafic-ultramafic rocks underlie a given area is the presence of upstanding ridges of listwaenite that are resistant to erosion. S/C fabrics are widespread and indicate that the ophiolites were affected by both strike-slip and vertical displacements. Variations in senses of shear observed along and across the strike evidence considerable strain partitioning during deformation. However, prevailing senses of shear can be discerned for several of the ophiolites that, in conjunction with other structural observations, indicate the main shear trajectories of the shear zones containing the ophiolites. Jabal Ess, Jabal Tharwah, and Bi’r Umq ophiolites were emplaced during periods of dextral transpression on the Yanbu and Bi’r Umq sutures, respectively. The Bi’r Tuluhah ophiolite was emplaced during sinistral transpression of the Hulayfah-Ad Dafinah-Ruwah suture, and the Halaban ophiolite was emplaced during west-directed convergence on the Halaban suture. DOI: 10.1016/S0166-2635(04)13004-1

130

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

1. INTRODUCTION Neoproterozoic mafic-ultramafic complexes make up less than 1% of the surface area of the Arabian shield but, starting with the pioneering work of Al-Shanti and Mitchell (1976) and Bakor et al. (1976), they figure prominently in discussions of the origins of the shield because of their possible tectonic significance as remnants of oceanic crust and indicators of arc-arc suturing (Pallister et al., 1987; Stoeser and Camp, 1985). Certainly not all mafic-ultramafic complexes in the region are ophiolites—some are nondiagnostic lenses of sheared serpentinite, some are intrusions in the base of volcanic arcs, and some are layered intrusions—and care must be taken to avoid misidentification, misinterpretation, and spurious correlations (Church, 1988, 1991). Nevertheless, a significant number of complexes have the hallmarks of ophiolites. They are widespread in the shield along shear zones (Fig. 1) and, together with stratigraphic, geochronologic, and structural data, provide evidence of active ocean-floor magmatism in association with development of the tectonostratigraphic terranes in the shield and the process of suturing during terrane amalgamation (Pallister et al., 1988; Johnson and Woldehaimanot, 2003; Genna et al., 2002). The mineralogy, chemistry, and tectonic settings of Arabian and Nubian shield ophiolites are reviewed by R.J. Stern and colleagues in a companion chapter in this volume. The purpose of this report is to describe the lithology, structure, and field relations of selected Arabian shield ophiolites, thereby providing examples of Neoproterozoic ophiolites and illustrating the outcrop characteristics and degrees of dismemberment and structural complexity that may be expected of Neoproterozoic ophiolites elsewhere. Of the ophiolites selected, Jabal Ess lies on the Yanbu suture at the join between the Midyan and Hijaz terranes in northwestern Saudi Arabia (Fig. 1). The Jabal Tharwah and Bi’r Umq ophiolites lie on the Bi’r Umq suture between the Hijaz and Jiddah terranes. The Bi’r Tuluhah ophiolite is at the northern end of the Hulayfah-Ad Dafinah-Ruwah suture joining the Hijaz-JiddahAsir terranes and the Afif terrane. The Halaban and Jabal al Uwayjah ophiolites are parts of the Halaban suture between the Afif and Ad Dawadimi terranes, and the Jabal Tays ophiolite is within the Ad Dawadimi terrane east of the Halaban suture. The Jabal Ess, Jabal Tharwah, Bi’r Umq, Bi’r Tuluhah, Halaban, and Jabal al Uwayjah ophiolites are believed to be rooted in the shear zones with which they are associated and, as such, mark the sites of consumption of oceanic crust. The Jabal Tays ophiolite, in contrast, appears to be part of a structurally detached ophiolite allochthon far traveled from its root zone.

2. JABAL ESS OPHIOLITE The Jabal Ess ophiolite (Figs. 2, 3) comprises mantle peridotite, isotropic and layered gabbro, a dike complex, pillow basalt, and pelagic sediments metamorphosed in the greenschist facies and is the most complete ophiolite in the Arabian shield (Al-Shanti, 1982). It covers an area of approximately 30 km east-west, and 5 km north-south, and has a minimum estimated thickness of about 3 km. The ophiolite crops out in hills that rise 300 m

2. Jabal Ess Ophiolite

131

Fig. 1. Mafic-ultramafic complexes and major faults and shear zones in the Arabian shield, Saudi Arabia with an inset showing terranes and interterrane ophiolite-decorated sutures and shear zones. Complexes that satisfy criteria as ophiolites, are named; other mafic-ultramafic complexes (mostly serpentinite lenses along fault zones) are shown by initials. Complexes: A = Al Amar; D = Ad Dafinah; H = Hamdah; HA = Hakran; I = Ibran; M = Muklar; N = Nabitah; R = Rahah; RD = Ar Ridaniyah; TA = Tabalah-Tarj; TW = Tawilah; UF = Umm Fawrah. Terranes: Af = Afif; As = Asir; D = Ad Dawadimi; H = Hijaz; Hi = Hail; J = Jiddah; M = Midyan; R = Ar Rayn. Boxes outline areas of Figs. 2, 9, and 11.

above the surrounding valley bottoms and is well dissected by east- and northeast-flowing drainages, which provide good exposures of the underlying geology. To the east, the northern and southern boundary faults of the ophiolite converge, and the ophiolite tapers and ceases to be recognizable. To the west, the ophiolite is cut by the northwest-trending sinistral fault system of the Da’bah and Durr shear zones. Mafic-ultramafic rocks continue to the south as the Sahluj mélange (named here after Jabal Sahluj) and the Jabal Wask ophiolite (Fig. 2). Together with the Jabal Ess ophiolite, these mafic-ultramafic rocks are

132

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 2. Simplified geologic map and geochronologic data for the Jabal Ess-Jabal Wask ophiolite zone, which marks the Yanbu suture in the northwestern Arabian shield. Mapping after Kemp (1981) and Hadley (1987). Geochronologic data after Kemp et al. (1980), Ledru and Augé (1984), Claesson et al. (1984), and Pallister et al. (1988). Box shows area of Fig. 3.

parts of the zone of deformed rocks that constitute the Yanbu suture in Saudi Arabia, and its extension in Northeast Africa, the Allaqi-Sol Hamid suture (Kröner et al., 1987; Abdelsalam and Stern, 1995; Johnson and Woldehaimanot, 2003). (For details on the Jabal

2. Jabal Ess Ophiolite

133

Fig. 3. Simplified map and cross section of the Jabal Ess ophiolite (after Al-Shanti, 1982; Chevrèmont and Johan, 1982b; and this report). Geochronologic data from Claesson et al. (1984); Pallister et al. (1988).

Wask ophiolite, see Bakor et al., 1976; Chevrèmont and Johan, 1982a; Ledru and Augé, 1984.) Peridotite is mainly exposed on the southern slope of Jabal Ess in the central part of the ophiolite shown in Fig. 3. It is strongly altered and is chiefly black, massive serpentinite in which original textures are rarely preserved although sufficient primary features remain to locally indicate the presence of harzburgite, subordinate tec-

134

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

tonized dunite, and cumulate wehrlite, orthopyroxenite, and serpentinite (Al-Shanti, 1982; Chevrèmont and Johan, 1982b). The harzburgite and dunite are concentrated in a zone of mantle peridotite as much as 500 m wide in outcrop. The harzburgite contains 5–20% bastite pseudomorphs after euhedral orthopyroxene in a serpentinized olivine ground mass (Shanti and Roobol, 1979). Dunite contains bastatized pyroxene ghosts and local disseminated chromite and podiform chromite lenses 20 cm across. Enstatite banding and trains of ovoid, stretched chromian spinel define a metamorphic foliation and lineation, which are suggestive of high-temperature subsolidus deformation possibly as a result of plastic mantle flow beneath a spreading ridge (Pallister et al., 1988). Olivine in the peridotite is magnesian rich with fosterite in the range Fo91 –Fo92.7, orthopyroxene is close to enstatite, and clinopyroxene is mainly diopside (Chevrèmont and Johan, 1982b; Shanti, 1983). Chromiferous spinel is similar to that from modern forearc peridotites (Stern et al., 2004) and is present as anhedral grains enclosing olivine and less commonly as euhedral grains interstitial to the cumulate olivine in the dunite or enveloped in orthopyroxene in harzburgite (Chevrèmont and Johan, 1982b). The cumulate rocks, inferred by Al-Shanti (1982) to be a unit about 400 m wide overlying the harzburgite, consists of serpentinized wehrlite and orthopyroxenite in layers a meter or so thick intercalated with serpentinite several meters thick. Gabbro is predominantly a dark, relatively featureless, massive rock, but locally has well-developed igneous lamination and rhythmic alternations of melanocratic gabbro and leucocratic anorthosite in layers as much as 20 cm thick (Fig. 4C). The gabbro is metamorphosed and, where strongly deformed, is mylonitized and brecciated, particularly in a narrow zone southeast of Jabal Ess where gabbro is exposed between peridotite and the dike complex and is tectonically intercalated with peridotite (Shanti and Roobol, 1979). The sheeted dike complex is a unit as much as 600 m wide composed of metadolerite dikes 30 cm to 2 m wide. Its contacts with gabbro, below, and pillow basalt, above, are transitional. Outcrop features of the complex are commonly obscured by extensive desert varnish but, where exposure is favorable, the complex is seen to consist entirely of dikes that have fine-grained chilled margins and fine- to medium-grained cores (Shanti and Roobol, 1979). The basalt unit includes pillow basalt (Fig. 4D), subordinate massive basalt flows as much as 10 m thick, and very sparse basalt breccia. It is estimated to be up to 300 m thick but because of folding and fault repetition is exposed over a width of nearly 2.5 km (AlShanti, 1982). Thin-skinned pillow lava characterized by amygdaloidal cores and rims of chloritized and (or) spherulitic basalt predominates. Khaki, locally siliceous shale and laminated chert crop out as interbeds 50 m thick in the pillow basalt and as isolated, strongly

Fig. 4. Features of the Jabal Ess ophiolite. (A) View of Jabal Ess from the south showing north-dipping shear surfaces and (along ridge line) a north-dipping sheet of carbonated and silicified peridotite. Relief about 150 m. (B) South-dipping thrusts in serpentinite mélange at the southern margin of the ophiolite. (C) Rhythmic layering in metagabbro showing anorthosite intercalated with melanocratic gabbro. (D) Pillow basalt. (Photos C and D after Al-Shanti, 1982.)

2. Jabal Ess Ophiolite

135

136

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

sheared exposures at the northern boundary fault of the ophiolite (Shanti and Roobol, 1979), and are interpreted as pelagic sediments at the top of the ophiolite succession. The complex is steeply dipping, and the exposures are effectively a cross-section through the ophiolite. The gross distribution of rock types suggests an ophiolite succession younging from south to north but the succession is disrupted by deformation. Igneous layering in the peridotite dips 40◦ –90◦ , mostly to the south; the basalt is folded into a series of anticlines and synclines; and the gabbro is repeated by folding and/or thrusting north of the basalt. Shear zones, characterized by serpentinite schist, secondary listwaenite, magnesite, and mélange, are abundant and dip between 30◦ and 90◦ south and north. Mélange, composed of angular to subrounded blocks of massive serpentinite, gabbro, dolerite, and basalt up to 50 m in diameter in a yellow to black serpentinite schist matrix, is particularly conspicuous as a subvertical shear zone as much as 500 m wide in the southern part of the ophiolite. A north-dipping unit of chert and listwaenite (Fig. 4A) marks the shear zone at the northern boundary of the ophiolite on Jabal Ess. South-dipping shear zones are common in the northern unit of gabbro and in the southern part of the ophiolite (Fig. 4B), and a subvertical shear zone forms the southernmost boundary of the ophiolite. S/C shear fabrics are widespread. A dextral sense of horizontal shear predominates, but variation in the sense of shear along and across the ophiolite indicates that there was a large degree of strain partitioning in the ophiolite during deformation. The northern boundary fault, minor shear zones in the northern gabbro, and the shear zone at the southern margin of the ophiolite are dextral (Fig. 5A). A shear zone about 100–200 m south of the northern boundary fault about 1 km east of Jabal Ess summit is sinistral (Fig. 5B). The south dipping shear zones in mélange in the southern part of the ophiolite are both dextral and sinistral (Figs. 5C, D). Indicators of the sense of vertical movement on the shear zones have not been observed, but it is conceivable that south-dipping shears throughout the ophiolite are north-vergent thrusts. Worldwide, suture zones commonly display combinations of horizontal shearing and thrusting that reflect deformation during transpression. The structures at Jabal Ess, suggestive of north-vergent thrusting and regional dextral horizontal shear, are consistent with development under conditions of dextral transpression. The Jabal Ess ophiolite is directly dated by means of a 780 ± 11 Ma U-Pb zircon age obtained from gabbro in the eastern part of the ophiolite (Pallister et al., 1988) and a 782 ± 36 Ma Sm-Nd mineral model age obtained from the same gabbro sample (Claesson et al., 1984). A younger U-Pb zircon age of 706 ± 11 Ma obtained from trondhjemite that intrudes already serpentinized and sheared gabbro provides a minimum age for ophiolite formation (Pallister et al., 1988). Both U-Pb ages are model ages, obtained by forcing the lower intercept through a fixed point of 15 ± 15 Ma, a procedure commonly applied to U-Pb geochronologic data in the Arabian shield (Cooper et al., 1979).

3. THARWAH OPHIOLITE COMPLEX The Tharwah ophiolite complex (Nassief, 1981; Nassief et al., 1984; Pallister et al., 1988) consists of mafic-ultramafic rocks preserved as a stack of steeply dipping, northwest-

3. Tharwah Ophiolite Complex

137

Fig. 5. Typical shear fabrics in the Jabal Ess ophiolite. (A) Dextral shear fabric in shear zone at the northern margin of the ophiolite. Knife for scale, about 10 cm long. (B) Sinistral shear fabric in serpentinite mélange south of the northern margin of the ophiolite. Hammer for scale. (C and D) Dextral and sinistral shear fabrics, respectively, about 5 m apart along strike in serpentinite mélange close to the southern margin of the ophiolite. Pen for scale in (C), hammer for scale in (D).

and southeast-vergent thrust sheets exposed over an area 13 km east-west and 6 km northsouth (Fig. 6). Together with adjacent pelagic rocks, the ophiolite is part of the Labunah thrust zone (Ramsay, 1986) and lies in the zone of deformed rocks that constitutes the southwestern part of the Bi’r Umq suture (Pallister et al., 1988; Johnson et al., 2002). The ophiolite is exposed in hills rising 150–200 m above the adjacent Red Sea Coastal Plain. Weathering is locally intense and most rock surfaces are coated in desert varnish that, compounded by pervasive metamorphism and shearing, makes rock identification difficult. The succession is disrupted and locally inverted. Serpentinized depleted-mantle harzburgite and subordinate dunite together with minor lenses and dike-like bodies of lherzolite and gabbro make up the central part of the complex. Harzburgite contains relict olivine (Fo89.5–93.4) (70–90 mode%), bastite pseudomorphs of orthopyroxene (En90–92) (15–30%), chromite (< 1%), and clinopyroxenes (< 1%). Dunite is largely serpentinized olivine. Chemically, the chromites resemble chromitiferous spinels in modern-day forearcs (Stern et al., 2004). The rocks are tectonized and have a strong, high-temperature foliation

138

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 6. Simplified map and cross section of the Jabal Tharwah ophiolite (after Nassief, 1981; Nassief et al., 1984; Ramsay, 1986; Pallister et al., 1988; Johnson, 1998). Geochronologic data after Pallister et al. (1988).

3. Tharwah Ophiolite Complex

139

Fig. 7. Features of the Jabal Tharwah ophiolite. (A) Cumulate peridotite showing interlayering of dunite and pyroxenite (circle encloses bent-over person for scale). (B) Rhythmic layering in gabbro (pen for scale 12 cm long). (C) Sinistral shear fabric in the interior of the ophiolite (hammer for scale).

composed of orthopyroxene and chromite grains (Nassief et al., 1984). Cumulate ultramafic rocks crop out in the northern part of the complex as a unit of dunite, lherzolite, and pyroxenite intercalated in layers 1–10 m thick (Fig. 7A). The cumulate rocks are as much as 2.9 km thick, but are probably thickened by deformation from an original thickness of about 1 km (Nassief et al., 1984). Oriented pyroxene produces a weak igneous lamination, but a high-temperature deformational foliation of the type displayed by the peridotite is absent. Olivine is less magnesium rich than in the mantle peridotite (Fo86.5–89.5) and clinopyroxene is chiefly diopside. Orthopyroxene, mostly present as exsolution laminae in clinopyroxene, is largely replaced by bastite but is preserved locally as grains of En82–89 (Nassief et al., 1984). Layered to locally massive gabbro is present in the north and south of the complex. It is metamorphosed in the greenschist facies, but has well-developed igneous lamination, cm- to m-scale plagioclase- and pyroxene-rich phase laying (Fig. 7B), and, to a lesser de-

140

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

gree, grain-size layering (Nassief et al., 1984). Sheeted dikes crop out in fault-bounded units 10 km long and over 300 m thick at the southern margin of the ophiolite. The dikes are not as well exposed as the dike complex in the Jabal Ess ophiolite and, because of shearing and alteration, their protoliths are not always evident. However, Nassief et al. (1984) report that, in places, dikes comprise 50–90% of the outcrop and are observed to be 1– 1.5 m wide, separated by screens of altered gabbro and basalt. Pillow basalt occurs in part of the dike complex. The rocks are strongly sheared and altered but bulbous pillow forms are still discernible (Nassief et al., 1984). Fine-grained argillaceous and cherty sedimentary rocks, carbonates, and basalt interpreted to be pelagic, ocean-floor deposits at the top of the ophiolite are faulted against the main mass of the Tharwah ophiolite south of Jabal Tharwah along the Thamrih fault and appear to be in depositional contact with the ophiolite northeast of Wadi Qirba’. Small lenses and veins of gabbroic pegmatite and leucodiorite or trondhjemite, probably representing late-fractionated derivatives of the gabbroic magma, intrude massive gabbro at the extreme northern edge of the complex along the Qirba’ fault. Layering in the cumulate unit is locally moderately inclined (Fig. 7A), but most structures in the Tahrwah ophiolite are steep. The Qirba’ fault dips 50◦ –80◦ to the southeast; the Thamrih fault and shear zones within the complex are subvertical. The pelagic rocks south of the complex make up the southwest plunging (40◦ –50◦) Farasan synform (Fig. 6). This has a steeply dipping northwest limb, a more gently inclined southeast limb, and is truncated by a southeast-vergent thrust along the southern flank of Jabal Farasan. The Ukaz fault at the southern boundary of the Labunah thrust zone is an oblique dextral, hangingwall-up-to-the south steep reverse fault that juxtaposes the pelagic rocks with the Samran group (Johnson, 1998). An exceptional gently inclined fault exposed at a location about 6 km northeast of Jabal Farasan may be a remnant of an original thrust dipping 35◦ –43◦ to the northwest. The structure of the Tharwah ophiolite and the Labunah thrust zone is believed to reflect two phases of progressive deformation (Johnson, 1998). The early phase caused northeast- and southwest-trending tight to isoclinal folding, the development of beddingparallel shear surfaces, and thrusting. Folding during the second phase created the Farasan synform, folded and steepened early thrusts and shear surfaces, and refolded early isoclinal folds and lineations. The rocks were pervasively affected by non-coaxial strain during both phases of deformation and S/C fabrics, asymmetrical extensional-shear bands, winged porphyroclasts, and quartz-mosaic ribbons are widespread (Johnson, 1998). The Qirba’ fault shows evidence of both sinistral and dextral horizontal as well as top-to-the-northwest reverse-slip movements; the Ukaz fault shows top-to-the-southeast and dextral horizontal movements; whereas shears interior to the ophiolite are commonly sinistral (Fig. 7C). Johnson (1998) proposes that the Tharwah ophiolite is a flower structure that developed in a zone of dextral transpression (see the cross section in Fig. 6) and, as in the case of the Jabal Ess ophiolite, the variations in sense of shear indicate considerable strain partitioning during its formation. Zircon grains from gabbro in the northern and southern parts of the ophiolite yield a near-concordant U-Pb age of 870 ± 11 Ma (Pallister et al., 1988). Other gabbro zircon

4. Bi’r Umq Ophiolite Complex

141

fractions are highly discordant, yielding 207 Pb/206Pb model ages of about 1250 Ma. The 870 Ma result is not robust, but if provisionally accepted as a crystallization age, suggests ocean-floor magmatism 100 million years earlier than the Jabal Ess magmatism. The 1250 Ma age is likely to be an artifact caused by assimilation of xenocrystic zircons, similar to the explanation of anomalously old ages for some Bi’r Umq ophiolite zircon samples (Calvez et al., 1985).

4. BI’R UMQ OPHIOLITE COMPLEX The Bi’r Umq ophiolite complex consists of serpentinized and carbonate-altered peridotite, gabbro, and mélange (undivided in Fig. 8), and a 1500-m thick succession of spilitic metabasalt, chert, and metatuff assigned to the Sumayir formation (Al-Rehaili, 1980; Al-Rehaili and Warden, 1980; Le Metour et al., 1982; Kemp et al., 1982; Pallister et al., 1988). The complex crops out in an area of about 60 km by 20 km at the northeastern end of the Bi’r Umq suture (Johnson et al., 2002). The ultramafic rocks and gabbro are concentrated in the south close to the Bi’r Umq and Wobbe faults and are a disproportionately small component of the ophiolite in comparison with other ophiolites described here. The ultramafic rocks and gabbro form a chain of discontinuous hills that have moderate relief of 50–75 m and are partly held up by more resistant listwaenite and chert. The Sumayir formation crops out in low-lying exposures north of the Bi’r Umq fault and is extensively covered by colluvium and alluvium. Mélange is mostly in discontinuous exposures along the Bi’r Umq fault. The ophiolite is truncated by the Arj fault on the west (west of the area shown in Fig. 8), a sinistral strike-slip structure belonging to the Najd fault system. The Raku-Mandisa faults truncate the ophiolite on the east. The Raku fault is a dextral shear of uncertain origin; because of poor exposure little is known about the Mandisa fault other than its trace, identified by a narrow, linear zone of listwaenite. Peridotite at Bi’r Umq is interpreted by Le Metour et al. (1982) to be an ultramafic cumulate consisting of dunite and subordinate, locally cumulus harzburgite. The rocks were pervasively sheared during ophiolite emplacement (Le Metour et al., 1982) and are extensively serpentinized, carbonated, and silicified, which results in the common development of Cu- and Ni-rich listwaenite along shear zones. Olivine in the dunite is thoroughly replaced by serpentine and is only recognized as ghost pseudomorphs. Harzburgite contains cumulus serpentinized olivine and intercumulus bastite-altered orthopyroxene (Le Metour et al., 1982). Small intrusions of hypabyssal trondhjemite, plagiogranite (termed keratophyre by Pallister et al., 1988), diorite, hornblende gabbro, metadiabase, and basalt occur at, or in a separate thrust slice south of, the Bi’r Umq fault (Le Metour et al., 1982; Pallister et al., 1988). The Sumayir formation is predominantly a homogeneous, monotonous unit of fine-grained greenstone derived from basalt flows and tuffs and subordinate pillow basalt and basaltic breccia (Al-Rehaili and Warden, 1980). Diagnostic of their metamorphic grade, the rocks contain sodic plagioclase (An5–20), secondary green tremolite, chlorite, epidote, iron oxide, and carbonate, with local relict clinopyroxenes. Minor metasedimentary units in the greenstone consist of thin-bedded felsic tuff, limestone,

142

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 8. Simplified map and cross section of the Bi’r Umq ophiolite and adjacent areas. Mapping after Al-Rehaili (1980); Kemp et al. (1982); Le Metour et al. (1982); Pallister et al. (1988); Johnson et al. (2002). Geochronology after Dunlop et al. (1986); Pallister et al. (1988).

chert, and siltstone, locally altered to mafic and felsic schist and amphibolite. Mélange consists of blocks of serpentinite, spilitic basalt, dolerite, and gabbro a few to several hundred meters across in a sheared serpentinite matrix.

5. Bi’r Tuluhah Ophiolite

143

Regionally, the Bi’r Umq ophiolite trends east-west, but is broadly folded about a northsouth axis and in detail trends west-southwest in the west and east-southeast in the east (Fig. 8). The Bi’r Umq fault at its southern margin is a steeply (50◦ –70◦ ) north-dipping reverse fault, and the Wobbe fault is a moderately (40◦ –50◦ ) southwest-dipping reverse fault. The northern Shuwaykah fault is believed to be a high-angle southeast-dipping reverse fault (Johnson et al., 2002). The dip across the Bi’r Umq complex changes from steeply northwest in the south to steeply southeast in the north and the complex is interpreted by Al-Rehaili (1980) to be a large asymmetric synform, although a flower structure of the type shown in the cross section in Fig. 8 is our preferred interpretation. Pillow structures indicate that basalt units are right way up (Al-Rehaili and Warden, 1980). Shear fabrics and the down-dip plunge of stretching lineations indicate an early phase of top-to the south reverse dip-slip movement on a south-vergent thrust along the Bi’r Umq fault; later movement included dextral and sinistral horizontal shear (B. Blasband, written communication, 2001). The kinematics of the Wobbe fault are unknown. The ophiolite is directly dated by a three-point U-Pb zircon model age of 838 ± 10 Ma obtained from diorite in the ophiolite close to the Bi’r Umq fault (Pallister et al., 1988). Trondhjemite and a pyroxene separate obtained from nearby gabbro yield a composite Sm-Nd isochron age of 828 ± 47 Ma (Dunlop et al., 1986), and Sumayir-formation basalt yields a three-point Rb-Sr whole-rock isochron of 831 ± 47 Ma (Dunlop et al., 1986). Single-point zircon model ages of 764 ± 3 Ma and 782 ± 5 Ma obtained from plagiogranite (or keratophyre) that cuts already serpentinized and carbonated peridotite and is interpreted to be a post-serpentinization and post-obduction intrusion constrain the minimum age of ophiolite emplacement (Pallister et al., 1988).

5. BI’R TULUHAH OPHIOLITE The Bi’r Tuluhah ophiolite crops out in the northern (Hulayfah) part of the HulayfahAd Dafinah fault zone in the north-central part of the Arabian shield (Fig. 9). The rocks are strongly folded and sheared and together with rocks of the Nuqrah formation constitute a subvertical brittle-ductile shear zone that resulted from sinistral transpression during suturing between the Afif and Hijaz terranes (Quick and Bosch, 1989; Johnson and Kattan, 2001). The suture continues as an ophiolite-decorated shear zone over 500 km to the south and southeast, and is one of the longer sutures recognized in the Arabian shield. Lithologic contacts in the fault zone are mostly faults so that original stratigraphic and structural relations are obscure, but an ophiolite is identified at Bi’r Tuluhah on the basis of the presence of amphibolite, serpentinized peridotite, layered gabbro, and noncumulus gabbro (Delfour, 1977). The ophiolite, the central part of which is shown in Fig. 10, is about 30 km long in a north-south direction and 6 km wide, and crops out in low-relief hills rising 10–40 m above the wadi plain (Kattan, 1983). The rocks are strongly weathered and intense coatings of desert varnish commonly cover outcrop surfaces. Volcanic and volcaniclastic rocks of the Hulayfah formation flank the fault zone on the west and epiclastic rocks and bimodal basalt and rhyolite of the Shammar group and Shammar

144

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 9. Simplified geologic map showing the location of the Bi’r Tuluhah, Darb Zubaydah, and Wadi Khadra ophiolitic complexes along and east of the Hulayfah-Ad Dafinah fault zone (HDFZ), part of the Hulayfah-Ad Dafinah-Ruwah suture. The fault zone is overlain and intruded by post-amalgamation basins and granites and displaced by Neoproterozoic III northwest-trending Najd faults. Map after Johnson and Kattan (2001). Box outlines area of Fig. 10.

5. Bi’r Tuluhah Ophiolite

145

Fig. 10. Simplified map of the Bi’r Tuluhah ophiolite and adjacent areas. Map after Kattan (1983); Le Metour et al. (1983); Quick and Bosch (1989); and Johnson et al. (1989). Geochronologic data after Calvez et al. (1984); Stuckless et al. (1984); Calvez and Kemp (1987); Pallister et al. (1988). HFZ = Hulayfah fault zone.

146

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

granites overlie and intrude the fault zone on the east. Post-Shammar northwest-trending sinistral Najd faults dislocate the ophiolite and fault zone on the north and south. The most extensive unit in the ophiolite is peridotite, which crops out in a zone 2.5 km wide north and south of Bi’r Tuluhah. The rocks are strongly serpenitinized but harzburgite and dunite protoliths are recognized. The rocks are mylonitized and have a strong cataclastic texture although ghost equant grains and a relict granoblastic texture are recognized in thin sections (Kattan, 1983). Serpentinized harzburgite comprises about 15% orthopyroxene, 80% serpentinized olivine presudomorphs, and minor chromite and magnetite. In hand specimen, serpentinized dunite is fine grained and dark gray to green, and in thin section is an aggregate of serpentine minerals that locally have a well-developed boxwork texture derived from the original olivine (Kattan, 1983). It contains anhedral grains of chromite and magnetite, and small lenses of massive chromite. As shown by Stern et al. (2004), the chromites plot on the low-Mg side of the field of present-day forearc chromian spinels, suggesting a suprasubduction origin for the ophiolite. Layered gabbro and dunite with minor wehrlite, lherzolite, websterite, and olivine clinopyroxenite layered on scales of centimeters to meters occur in a narrow band west of the serpentinized peridotite (Le Metour et al., 1983). Serpentinization obscures primary textures, and it is not clear whether the layered rocks are mantle tectonites or ultramafic cumulates (Quick and Bosch, 1989). A narrow zone of mafic plutonic rocks farther west consists of gabbro and diorite in the north, in the northern half of Fig. 10, and fault slivers of layered gabbro in the south. The northern gabbro and diorite either intrude or are faulted against the ophiolitic rocks, and may postdate ophiolite magmatism (Le Metour et al., 1983). The southern layered gabbro has a cumulus texture in pyroxene- and hornblenderich phases and may be a cumulate part of the ophiolite succession. Amphibolite, treated by Delfour (1977) as part of the ophiolite, crops out east of the peridotite. Fine-grained amphibolite is strongly schistose and lacks clear textural or relict mineralogic indications of its protoliths. Coarse-grained amphibolite appears to be the result of epidote-amphibolite facies metamorphism of gabbro and diabase (Quick and Bosch, 1989). Massive, locally pillowed metabasalt and chert together with fine-grained sandstone, keratophyre, and interbedded felsic tuff and minor basalt make up the volcanicvolcaniclastic rocks of the Nuqrah formation located on either side of the mafic-ultramafic units along the fault zone (Le Metour et al., 1983; Quick and Bosch, 1989). Considered in isolation, it is conceivable that the basalt and chert represent pelagic rocks at the top of the ophiolite succession but their association with felsic tuffs and sandstone suggest that they are part of a suprasubduction volcanic arc. The metabasalt is a fine-grained, light gray to gray-green rock, the original structure and texture of which are virtually obliterated by metamorphism. The rock is identified as basalt in the field by its mafic composition, generally massive appearance, and the local presence of pillow structure. In thin section, the basalt has a strongly developed, fine-grained metamorphic foliation composed of saussuritized plagioclase, epidote, carbonate, chlorite, clinozoisite, and iron oxides (Kattan, 1983). Dikes of diabase, gabbro, plagiogranite, and diorite cut all the serpentinized ultramafic rocks, and plutons of diorite and quartz diorite intrude the southern part of the ophiolite.

6. Halaban Ophiolite

147

Structurally, the Bi’r Tuluhah ophiolite is a set of fault-bounded lenses. Together with the flanking Nuqrah formation, the rocks are pervasively cleaved and sheared, and deformation is spread across the entire width of the Hulayfah fault zone, although narrow zones of ultramylonite, schist, and carbonate-altered serpentinite identify discrete shears within the fault zone. All shear surfaces are subvertical, and any low-angle thrusts that may have been originally present have been obliterated or steepened by subsequent deformation. Model U-Pb zircon ages of 847 ± 14 Ma and 823 ± 11 Ma obtained from plagiogranite (trondhjemite) dikes that intrude serpentinized harzburgite in the center of the ophiolite (Pallister et al., 1988) provide a minimum age for the ophiolite. The dikes have rodingite margins indicating intrusion prior to complete serpentinization and they are interpreted by Pallister and colleagues as forming late during ophiolite magmatism. Together with a UPb zircon age of 839 ± 23 Ma obtained from Nuqrah formation rhyolite 40 km east of the Bi’r Tuluhah ophiolite (Calvez et al., 1984), the Bi’r Tuluhah model ages suggest that oceanic crust and volcanic arc rocks were actively forming in the region between 840 and 820 Ma. The age of the Hulayfah formation is weakly constrained by a U-Pb zircon age of 720 ± 10 Ma obtained from tonalite that intrudes the formation west of the Hulayfah fault zone (Calvez et al., 1984). An approximate U-Pb zircon age of 710 Ma obtained from a quartz diorite pluton intruded into the southern part of the ophiolite and fault zone (Fig. 10) constrains the minimum age for ophiolite deformation and alteration (J.S. Stacey, personal communication, cited by Quick, 1991) and approximate U-Pb and Rb-Sr ages between 630 Ma and 615 Ma obtained from the post-amalgamation Shammar group and Shammar “stitching” granites (Stuckless et al., 1984; Calvez and Kemp, 1987) give a minimum age for completion of suturing along the fault zone.

6. HALABAN OPHIOLITE The Halaban ophiolite is a zone of mafic-ultramafic rocks exposed north and south of Halaban in the eastern part of the Arabian shield (Figs. 11, 12) and located along the Halaban suture at the join between the Afif and Ad Dawadimi terranes. The Ad Dawadimi terrane is strongly deformed and treated by some authors as, itself, part of a larger suture zone referred to as the Al Amar suture (Stoeser and Camp, 1985). The ophiolite consists of metagabbro and subordinate serpentinite. It is bounded on the west by the HalabanZarghat fault zone, a complex structure including a west-vergent thrust in the south and a down-to-the-west normal fault in the north, and on the east by the Eastern shear zone (Fig. 12). The rocks west of the ophiolite include mafic plutons, orthogneiss, and amphibolite referred to as the Suwaj domain and late Neoproterozoic sedimentary rocks of the Jibalah group deposited in the Antaq basin. Rocks to the east are low-grade metasedimentary units of the Abt formation and post-amalgamation granitoids. The ophiolitic rocks crop out in hills with relief of about 50 m and are generally well exposed. Unfortunately, the margins of the ophiolite tend to coincide with valleys so that structural details at the boundaries of the ophiolite are largely obscured. From the point of view of the Penrose definition, the rocks in the vicinity of Halaban do not include mantle peridotite, a dike

148

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 11. Simplified geologic map and geochronologic data for the Halaban, Jabal Tays, and Jabal al Uwayjah ophiolite complexes in the eastern Arabian shield. Map after Delfour (1979); Manivit et al. (1985); and this report. Geochronologic data after Calvez et al. (1984); Stacey et al. (1984). Abbreviations: AFZ = Al Amar fault zone; EF = East fault (magnetically inferred); HYFZ = Hufayrah fault zone; HZF = Halaban-Zarghat fault zone; ARFZ = Ar Rika fault zone; WF = West fault (magnetically inferred).

complex, or pillow basalt and, at best, are an incomplete ophiolite. However, north of the area described here, the on-strike continuation of the Halaban rocks includes peridotite, gabbro, serpentinite, listwaenite, and basalt (Al-Shanti and El-Mahdy, 1988). It is possi-

6. Halaban Ophiolite

149

Fig. 12. Simplified map and cross section of the Halaban ophiolite complex and adjacent units, showing the locations and results of geochronologic dating. Map after Delfour (1979), Al-Saleh (1993), and this report; geochronologic data after Stacey et al. (1984), Al-Saleh (1993), and Al-Saleh et al. (1998).

150

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

ble that the rocks in the vicinity of Halaban village are the mafic plutonic section of an ophiolite, whereas the mantle part is preserved in the north. The Halaban rocks are predominantly pale green, well-foliated metagabbro (Al-Saleh et al., 1998). Al-Shanti and El-Mahdy (1988) interpret the foliation as igneous layering and describe microscale igneous lamination and large-scale rhythmic layering caused by differences in mineralogy, grain size, and texture. Some of the foliation, however, is clearly secondary in origin, with oriented metamorphic minerals and quartz ribbons, ductile folding, as well as S/C shear fabrics and Al-Saleh et al. (1998) interpret much of the foliation to be the result of sea-floor metamorphism under greenschist- to amphibolite-facies conditions. The widespread development of actinolite, chlorite, clinozoisite, and albite is inferred to reflect ubiquitous, low-grade, low-temperature, off-axis metamorphism, whereas local amphibolitization of gabbro is inferred to reflect metamorphism close to the spreading axis in conjunction with shearing. Primary feldspar in the gabbro is commonly saussuritized and clinopyroxene tends to be replaced by chlorite and quartz. Lenses of massive black serpentinite and serpentinized lherzolite and olivine websterite occur sporadically along the western margin and axis of the gabbro (Fig. 12). The serpentinite lenses have sharp contacts with surrounding gabbro and are interpreted as ultramafic diapirs emplaced in the gabbro from an originally lower stratigraphic position in the ophiolite (Al-Saleh et al., 1998). Chromite from a small pod south of Halaban village plots close to the fields of chromian spinels from boninite and forearc ophiolites (Stern et al., 2004). South of Halaban village the gabbro outcrops taper and are structurally underlain by metamorphosed mafic and ultramafic rocks belonging to a sub-ophiolitic metamorphic complex (Al-Saleh et al., 1998). The eastern part of the metamorphic complex contains abundant blocks of serpentinite, 1–20 m across, sheathed by soapstone and small lenses of chromite in a matrix of orthoamphibolite and rodingite. It forms an inhomogeneous unit that may represent an obduction-related mélange. To the west, the metamorphic complex becomes more felsic. It contains no ophiolitic material and was probably largely derived from diorite and tonalite belonging to the Suwaj magmatic arc. The metamorphosed rocks were affected, particularly at their contact with the Halaban gabbro, by partial melting, which resulted in the development of migmatitic gneiss composed of coarse-grained hornblendite, amphibolite and gneissic gabbro and diorite intruded by numerous veins and irregular lenses of trondhjemite. Petrologic studies indicate that the mafic paleosome of the gneiss was partially melted under hydrous conditions; the neosome segregations are chiefly quartz and andesine plagioclase (Al-Saleh et al., 1998). The Eastern shear zone consists of strongly deformed gabbro, talc schist mélange, and pelitic schist exposed in a zone as much as 1 km wide. The rocks are tectonically intercalated with each other or are present as a mélange comprising irregular, scattered blocks of the various rock types in an anthophyllite-talc schist matrix (Fig. 13A). Gabbro in the Eastern shear zone commonly has a mylonitic texture and is cut by shear zones several centimeters thick that contain S/C fabrics. Altered basalt consists of pumpellyite pseudomorphs of original plagioclase phenocrysts, chlorite, quartz, and hematite, and the pelitic schists are rich in Ca and Mg silicates and believed to be derived from deep-marine argillaceous carbonates (Al-Saleh, 1993).

6. Halaban Ophiolite

151

Fig. 13. Features of the Halaban ophiolite. (A) Mélange from the Eastern shear zone. (B) Brittleductile shear in the partial melt zone showing top-up-to-the west displacement.

The steep dips of schistosity and shear surfaces indicate that the Eastern shear zone is subvertical. The western boundary of the ophiolite west of Halaban, in contrast, is inferred to be east dipping in conformity with east-dipping foliations and shear surfaces in the gabbro in proximity to the boundary. Sense-of-shear indicators in the partial melt zone beneath the gabbro (Fig. 13B) suggest that this western boundary was affected by west-

152

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

directed shearing and the boundary is interpreted by us as a west-vergent thrust that placed the Halaban ophiolite above the Suwaj domain. Radiometric age determinations are reported from many units in the area. A U-Pb model zircon age of 694 ± 8 Ma obtained from a hypersthene gabbro in the southern part of the ophiolite about 10 km NNW of Halaban (Stacey et al., 1984) constrains the magmatic age of the ophiolite. Several 40 Ar/39Ar ages of about 680 Ma obtained from amphibolites of the sub-ophiolitic metamorphic complex and from metamorphic hornblendes in the ophiolite gabbro are interpreted to reflect rapid cooling and obduction of the ophiolite (Al-Saleh et al., 1998). U-Pb and 40 Ar/39Ar dates obtained from Suwaj diorite suggest that the Suwaj domain developed between 681 Ma and 675 Ma (Stacey et al., 1984; Al-Saleh et al., 1998).

7. JABAL TAYS OPHIOLITE The Jabal Tays ophiolite crops out in the central part of the Ad Dawadimi terrane, 75 km east-southeast of the Halaban ophiolite. The exposures form a group of prominent hills that have a local relief of 220 m rising to a summit of 1057 m above sea level at Jabal Tays, and are surrounded by low-relief exposures of low-grade sandstone, siltstone, conglomerate, and limestone of the Abt formation (Fig. 14). Isolated bodies of gabbro and mafic dikes exposed south of the area shown in Fig. 14 may be detached parts of the ophiolite (AlShanti and Gass, 1983), but their exact relation to the Jabal Tays exposures are not clear at this stage because of surficial cover. Mafic-ultramafic rocks at Jabal Tays include a large amount of undifferentiated serpentinite, subordinate amounts of gabbro intruded by mafic dikes, mélange, serpentinite schist, and listwaenite. Gabbro is variably serpentinized but is fresh enough that igneous lamination and cyclic layering of melanocratic, olivine- and pyroxene-rich gabbro and anorthosite are recognized (Al-Shanti and Gass, 1983). Plagioclase and clinopyroxene in the gabbro have a cumulate texture; orthopyroxene is mostly replaced by chlorite. The serpentinite, which makes up the bulk of Jabal Tays, is variably sheared and typically consists of relatively massive serpentinite cut by shear zones marked by serpentinite schist. The serpentinite protoliths have not been identified but are presumably varieties of mafic and ultramafic rocks. Mélange is uniformly developed at the outer margins of the ophiolite as a zone up to 500 m wide. It comprises irregular blocks of gabbro and massive to schistose serpentinite from a few centimeters to tens of meters across in a serpentinite and talc-schist matrix. Along the western side of Jabal Tays, the mélange creates a distinctive rugged terrain in which the mélange clasts weather out as protuberances (Fig. 15A). Carbonate alteration is widespread in the area, but is particularly prominent in west-dipping shear zones on the southern flank of Jabal Tays, on which basis the mountain is interpreted to be a stack of west-dipping thrusts. The external contacts of the ophiolite are poorly and discontinuously exposed, but the manner in which the outer contact and mélange zone wraps around Jabal Tays suggests that the ophiolite is a synform (Fig. 14). Where exposed, the outer, structurally lower contact is a shear zone 1–5 m thick discordant with respect to the underlying Abt formation.

7. Jabal Tays Ophiolite

153

Fig. 14. Geologic map and cross section of the Jabal Tays ophiolite complex. Map after Al-Shanti and Gass (1983) and this report. Geochronologic data after Al-Shanti et al. (1984).

154

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 15. Features of the Jabal Tays and Jabal Uwayjah ophiolites. (A) View to the southwest of the mélange zone on the western side of the Jabal Tays ophiolite. (B) Dextral shear in the footwall of the Jabal Tays basal thrust. (C) Strongly developed deformational (?) foliation in Jabal Uwayjah gabbro. (D) West-dipping shear surfaces in Jabal Uwayjah serpentinized peridotite. (E) Shear fabric in serpentinized peridotite from locality (D) looking south, showing top-to-left, that is hanging-wall-up-to-east, displacement.

8. Jabal al Uwayjah Ophiolite

155

Mesoscale folding in the basal shear zone is indicated by changes in dip of shear surfaces from flat lying (25◦ –35◦ ) to subvertical over distances of a few meters. S/C shear fabrics indicate dextral-horizontal (Fig. 15B) and hanging-wall up-to-the-east vertical movements along the eastern and southeastern parts of the shear zone, suggestive of the ophiolite being part of an easterly vergent nappe. This inferred transport direction is compatible with an E-W elongated chromite lineation described from deformed gabbro in the central part of the ophiolite (Al-Shanti and Gass, 1983) and with the southwesterly plunge of stretched pebbles observed by us in Abt formation conglomerate in the footwall of the ophiolite on the southeastern flank of Jabal Tays. Whether the mélange along the western margin of the ophiolite is part of the basal thrust upturned by synclinal folding or a secondary mélange created along the north-trending steep fault that appears to truncate the ophiolite on the west is not yet established. The mafic-ultramafic rocks of the Jabal Tays ophiolite are not directly dated. Their minimum age is weakly constrained by an Rb-Sr whole-rock isochron of 620 ± 40 Ma obtained from trondhjemite that intrudes the mélange zone (Al-Shanti et al., 1984). However, granitoids elsewhere in the Ad Dawadimi terrane are dated 670–640 Ma (Stacey et al., 1984), and the Rb-Sr age is too young to be a meaningful constraint on the ophiolite. By comparison with the Halaban ophiolite, the Jabal Tays body is probably more likely to be about 680 Ma.

8. JABAL AL UWAYJAH OPHIOLITE The Jabal al Uwayjah ophiolite is exposed at the eastern edge of the Arabian shield in a group of isolated hills of low relief (< 80 m). Because of extensive cover by Quaternary eolian sand, pediment gravel, and wadi alluvium (Fig. 16), exposure is poor, and this characteristic in combination with little recent mapping means that the ophiolite is the least well known in the shield. Overall, it appears to occupy an area of about 45 km N-S and 12 km E-W on the shield but, judging by its aeromagnetic signature, continues 40 km to the southeast beneath the Permian, making it one of the larger ophiolites in the region. The contacts of the ophiolite are concealed and its structure is obscure, but prominent magnetic lineaments suggest major faults occur along the axis and western margin of the ophiolite. Permian sandstone and limestone are unconformable on the ophiolite on the east and amphibolite-grade metadiorite, metagabbro, and amphibolite, and garnet-amphibole gneiss of the Ghadaniyah complex (701 ± 5 Ma; Agar et al., 1992) flank the ophiolite on the west. The Ghadaniyah complex is lithologically and geochronologically similar to the Suwaj domain rocks in the Halaban area, and the two are correlated by Johnson (1996), as is implied by use of the same graphic symbol for the two rock units in Fig. 11. The most extensive exposures of the ophiolite are in low hills north and south of Jabal al Uwayjah and include serpentinized peridotite, pyroxenite, metagabbro, undifferentiated serpentinite, and minor metabasalt and metaandesite (Manivit et al., 1985). Pyroxenite has relict orthopyroxene (enstatite) and clinopyroxene (augite and diallage), and metagabbro, which is fine grained, strongly foliated (Fig. 15C), and closely resembles the Halaban

156

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Fig. 16. Map of the Jabal al Uwayjah ophiolite and adjacent units. Abbreviations: EF = Eastern fault; HYFZ = Hufayrah fault zone; WF = Western fault. Map after Bois and Shanti (1970), Brosset (1974), Manivit et al. (1985), and this report.

gabbro, contains relict olivine and pyroxenes pervasively altered to epidote and serpentine. Serpentinite is a black to green rock composed of antigorite, talc, and relict pyroxene. A fine-grained brick-red colloidal and ferruginous siliceous unit located at the contact be-

9. Summary and Discussion

157

tween the ophiolite and Permian rocks probably represents silicification of serpentinite as a result of weathering. A low-lying area west of the Jabal al Uwayjah hills is virtually devoid of exposures other than discontinuous north-trending ridges of carbonate. Some of the ridges are fine-grained gray or variegated white and gray, massive to thinly layered marble that resembles sedimentary marbles in other parts of the shield. Others, however, are listwaenite, which suggests that, despite the lack of outcrop, bedrock includes a significant amount of ultramafic rock. Structures in the Jabal al Uwayjah ophiolite are predominantly north trending. They include cleavage in serpentinite and gabbro, foliation in metagabbro, linear outcrops of listwaenite and carbonate, and magnetically inferred faults, two of which dominate the region. The western fault separates the ophiolite from the Ghadaniyah complex. The eastern fault, which locally coincides with ridges of listwaenite, separates the low hills north and south of Jabal al Uwayjah from the area of poor exposure to the west. West-dipping shear zones in serpentinite at the western edge of these hills contain S/C fabrics indicating top-to-the-east reverse slip (Figs. 15D, E), which suggests that the eastern fault may be an east-vergent thrust. The Jabal al Uwayjah ophiolite is not directly dated but on the basis of lithology is correlated by Brosset (1974) with the Halaban ophiolite. We concur with this correlation, which is consistent with the correlation mentioned above between the Ghadaniyah complex and Suwaj domain on the western flanks of the Jabal al Uwayjah and Halaban ophiolites, respectively, and provisionally infer that the Jabal al Uwayjah ophiolite is about 680 Ma.

9. SUMMARY AND DISCUSSION Ophiolites are widespread in the Arabian shield. However, as is evident from this review, they are ubiquitously deformed, with the consequence that typical ophiolite successions are not preserved at every occurrence. Nevertheless, sufficient diagnostic lithologic criteria are present to confidently conclude that numbers of the mafic-ultramafic complexes of the shield are indeed ophiolites. Of these, Jabal Ess is one of the most complete (Table 1); Jabal Tays the least complete. Jabal al Uwayjah is the least well known. Available geochronologic data indicate that the ophiolites developed over a 200-million year period delimited by Jabal Tharwah (∼ 870 Ma) and Halaban (∼ 695 Ma). Jabal Tharwah and Halaban are, in fact, the oldest and youngest ophiolites known in the entire Arabian-Nubian shield (Stern et al., 2004). The western and eastern geographic locations of the Tharwah and Halaban ophiolites conceivably suggest an eastward migration of oceanic floor magmatism in this period in the Arabian shield. However, in the larger setting of the entire region of juvenile Neoproterozoic rocks represented by the Arabian and Nubian shields, there is no unidirectional time-space distribution. Jabal Tharwah is in the middle of the combined Arabian-Nubian shield and younger ophiolites occur to the east, west, north, and south. Although not a specific topic of this review, it is also evident, in conjunction with a range of structural and stratigraphic information about the adjacent rocks, that ophiolites

158

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Table 1. Summary table showing lithologic components and estimated magmatic ages of selected Arabian shield ophiolites Ophiolite Jabal Ess Jabal Tharwah Bi’r Umq Bi’r Tuluhah Halaban Jabal Tays Jabal al Uwayjah √

Age (to nearest 5 Ma) 780 870 840–830 845–825 695

Peridotite

Gabbro

Dikes

Basalt

√ √ √ √ √ minor √ serpentinite √

√ √

√ √

? √ √ √ √

? ? ? ? ?

√ √ √ √ √ √

in the north

Pelagic sediments √ √ √ ? ? ? ?

= lithology observed; ? = lithology not observed to date.

in the Arabian shield occur along suture zones. The structure and geochronology of the ophiolites are therefore important constraints on the history of suturing. Dating of the Jabal Tharwah-Bi’r Umq ophiolites and associated intrusions imply ocean-floor magmatism in the northwestern part of the shield between ∼ 870–830 Ma and is consistent with development of the Bi’r Umq suture at about 780–760 Ma (Johnson et al., 2002). The 706 Ma age of post-ophiolite trondhjemite at Jabal Ess is consistent with convergence along the Yanbu suture and its Northeast African extension between ∼ 700 Ma and ∼ 600 Ma. The Bi’r Tuluhah ophiolite is the oldest known example of oceanic-floor magmatism along the Hulayfah-Ad Dafinah-Ruwah suture (Johnson and Kattan, 2001), and the Halaban ophiolite constrains oceanic magmatism and suturing in the eastern shield at ∼ 695 Ma and ∼ 680 Ma, respectively. Structurally, all the ophiolites are complex, and exhibit multiple phases of folding and shearing. Most structures are steep, and unlike some of the Neoproterozoic ophiolites in the Nubian shield (Abdelsalam and Stern, 1993; Schandelmeier et al., 1994), the Arabian examples have few preserved low-angle thrusts. The only candidates for original thrusts identified to date are low- to moderately inclined shears in the southern part of Jabal Ess, in eastern Jabal Tharwah, at the western contact of Halaban, along parts of the Jabal Tays basalt contact, and at the Eastern fault at Jabal al Uwayjah. Other shear zones are subvertical, either because they are folded thrusts, similar to the folding evident in the basal thrust at Jabal Tays, or are steep shear zones that developed during other phases of deformation. Overall, the available structural evidence is permissive of modeling the Jabal Ess ophiolite as a stack of steepened north-vergent thrusts and horizontal shear zones that resulted from a period of dextral transpression. The Jabal Tharwah and Bi’r Umq ophiolites are possibly both flower structures related to southeast- and northwest-vergent thrusting along the Bi’r Umq-Nakasib suture (Johnson et al., 2002). The Bi’r Tuluhah ophiolite is preserved in a subvertical shear zone that forms the northern part of the Hulayfah-Ad Dafinah-Ruwah shear zone created during sinistral transpression. The Halaban ophiolite is part of a westvergent allochthon thrust over the eastern margin of the Afif terrane at the Halaban suture. The Jabal al Uwayjah ophiolite is an extension of the Halaban ophiolite, detached from the Halaban rocks by the sinistral and top-to-the-north Hufayrah fault zone (Fig. 11), and its

References

159

location is evidence that the Halaban suture continues to the edge of the shield and beyond, beneath Permian rocks that flank the shield. Available information about the Arabian shield ophiolites varies in quality and quantity. Unfortunately, none are known in sufficient detail to fully determine their tectonic setting. Additional, and in some cases, original, petrologic, geochemical, geochronologic, and structural research are required. The ophiolites have a small surface area, but are critical for our understanding of the tectonic history of the shield, and warrant ongoing study and exploration. They testify to the juvenile tectonic environment of the Arabian-Nubian shield; they document the creation of oceanic floor following the breakup of Rodinia; and in their deformed and metamorphosed state they record stages in the subduction and closure of the Mozambique Ocean concurrent with the amalgamation and suturing of the tectonostratigraphic terranes that make up the shield.

REFERENCES Abdelsalam, M.G., Stern, R.J., 1993. Structure of the late Proterozoic Nakasib suture, Sudan. Journal of the Geological Society of London 150, 1065–1074. Abdelsalam, M.G., Stern, R.J., 1995. Sutures and shear zones in the Arabian-Nubian Shield. Journal of African Earth Sciences 23, 289–310. Agar, R.A., Stacey, J.S., Whitehouse, M.J., 1992. Evolution of the southern Afif terrane—a geochronologic study. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report DGMR-OF-10-15, p. 41. Al-Rehaili, M.H., 1980. Geology of the mafic-ultramafic complex of Bi’r Umq area. M.Sc. thesis. King Abdulaziz University, Jiddah, p. 160. Al-Rehaili, M.H., Warden, A.J., 1980. Comparison of the Bi’r Umq and Hamdah ultrabasic complexes, Saudi Arabia. Institute of Applied Geology Bulletin 3 (4), 143–156. Al-Saleh, A.M., 1993. Origin, age and metamorphism of the Halaban ophiolite and associated units: implications for the tectonic evolution of the eastern Arabian Shield. Ph.D. thesis. University of Liverpool, p. 274. Al-Saleh, A.M., Boyle, A.P., Mussett, A.E., 1998. Metamorphism and 40 Ar/39 Ar dating of the Halaban ophiolite and associated units: evidence for two-stage orogenesis in the eastern Arabian shield. Journal of the Geological Society of London 155, 165–175. Al-Shanti, A.M., El-Mahdy, O.R., 1988. Geological studies and assessment of chromite occurrences in Saudi Arabia. King Abdulaziz City for Science and Technology Project No. AT-6-094 Final Report, p. 165. Al-Shanti, A.M., Gass, I.G., 1983. The Upper Proterozoic ophiolite mélange zones on the easternmost Arabian shield. Journal of the Geological Society of London 140, 867–876. Al-Shanti, A.M.S., Abdel-Monem, A.A., Marzouki, F.H., 1984. Geochemistry, petrology and Rb-Sr dating of trondhjemite and granophyre associated with Jabal Tays ophiolite, Idsas area, Saudi Arabia. Precambrian Research 24, 321–334. Al-Shanti, A.M.S., Mitchell, A.H.G., 1976. Late Precambrian subduction and collision in the Al Amar-Idsas region, Arabian Shield, Kingdom of Saudi Arabia. Tectonophysics 30, T41–T47. Al-Shanti, M.M.S., 1982. Geology and mineralization of the Ash Shizm-Jabal Ess area. Ph.D. thesis. King Abdulaziz University, Jiddah, p. 291.

160

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Bakor, A.R., Gass, I.G., Neary, C.R., 1976. Jabal al Wask, northwest Saudi Arabia: an Eocambrian back-arc ophiolite. Earth and Planetary Earth Sciences Letter 30, 1–9. Bois, J., Shanti, M., 1970. Mineral resources and geology of the As Sakhin quadrangle, photomosaic sheet 130. Bureau de Recherches et Géologiques et Minières Technical Record 70-JED-6, scale 1:100,000. Brosset, R., 1974. Geology and mineral exploration of the Umm Sulaym quadrangle, 22/45C. Bureau de Recherches Géologiques et Minières Technical Record 74-JED-9, scale 1:100,000. Calvez, J.-Y., Kemp, J., 1987. Rb-Sr geochronology of the Shammar group in the Hulayfah area, northern Arabian Shield. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-07-11, p. 22. Calvez, J.-Y., Alsac, C., Delfour, J., Kemp, J., Pellaton, C., 1984. Geochronological evolution of western, central, and eastern parts of the northern Precambrian shield, Kingdom of Saudi Arabia. Faculty of Earth Sciences, King Abdulaziz University, Jiddah, Bulletin 6, 24–48. Calvez, J.-Y., Delfour, J., Kemp, J., Elsass, P., 1985. Pre Pan-African inherited zircons from the northern Arabian shield. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-05-13, p. 22. Chevrèmont, P., Johan, Z., 1982a. The Al Ays ophiolite complex. Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-02-5, p. 65. Chevrèmont, P., Johan, Z., 1982b. Wadi al Hwanet-Jabal Iss ophiolite complex. Deputy Ministry for Mineral Resources Open-file Report BRGM-OF-02-14, p. 30. Church, W.R., 1988. Ophiolites, structures, and micro-plates of the Arabian-Nubian shield: a critical comment. In: El-Gaby, S., Greiling, R.O. (Eds.), The Pan-African Belt of Northeast Africa and Adjacent Areas. Veiweg, Braunschweig/Wiesbaden, pp. 289–316. Church, W.R., 1991. Discussion of ophiolites in northeast and east Africa: implications for Proterozoic crustal growth. Journal of the Geological Society of London 148, 600–601. Claesson, S., Pallister, J.S., Tatsumoto, M., 1984. Samarium-neodymium data on two late Proterozoic ophiolites of Saudi Arabia and implications for crustal and mantle evolution. Contribution to Mineralogy and Petrology 85, 244–252. Cooper, J.A., Stacey, J.S., Stoeser, D.B., Fleck, R.J., 1979. An evaluation of the zircon method of isotopic dating in the southern Arabian craton. Contributions to Mineralogy and Petrology 68, 429–439. Delfour, J., 1977. Geology of the Nuqrah quadrangle, 25E, Kingdom of Saudi Arabia. Saudi Arabian Directorate General of Mineral Resources Geologic Map GM 28, 1:250,000 scale. Delfour, J., 1979. Geologic map of the Halaban quadrangle, sheet 23G, Kingdom of Saudi Arabia. Saudi Arabian Directorate General of mineral Resources Geologic Map GM-46, scale 1:250,000. Dunlop, H.M., Kemp, P., Calvez, J.-Y., 1986. Geochronology and isotope geochemistry of the Bi’r Umq mafic-ultramafic complex and Arj group volcanic rocks, Mahd adh Dhahab quadrangle, central Arabian Shield. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-07-7, p. 38. Genna, A., Nehilg, P., Le Goff, E., Guerrot, C., Shanti, M., 2002. Proterozoic tectonism of the Arabian Shield. Precambrian Research 117, 21–40. Hadley, D.G., 1987. Geologic map of the Sahl Al Matran quadrangle, sheet 26C, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Geoscience Map GM-86, scale 1:250,000. Johnson, P.R., 1996. Geochronologic and isotopic data for rocks in the east-central part of the Arabian shield: stratigraphic and tectonic implications. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report USGS-OF-96-3, p. 47.

References

161

Johnson, P.R., 1998. The structural geology of the Samran-Shayban area, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Technical Report USGS-TR-98-2, p. 45. Johnson, P.R., Kattan, F., 2001. Oblique sinistral transpression in the Arabian shield: the timing and kinematics of a Neoproterozoic suture zone. Precambrian Research 107, 117–138. Johnson, P.R., Woldehaimanot, B., 2003. Development of the Arabian-Nubian Shield: perspectives on accretion and deformation in the northern East African Orogen and the assembly of Gondwana. Geological Society of London Special Publication 206, 289–325. Johnson, P.R., Abdelsalam, M., Stern, R.J., 2002. The Bi’r Umq-Nakasib shear zone: Geology and structure of a Neoproterozoic suture in the northeastern East African Orogen, Saudi Arabia and Sudan. Saudi Geological Survey Technical Report SGS-TR-2002-1, p. 33. Johnson, P.R., Quick, J.E., Kamilli, R.J., 1989. Geology and mineral resources of the Bi’r Tuluhah quadrangle, Kingdom of Saudi Arabia. Saudi Arabian Directorate General of Mineral Resources Technical Record USGS-TR-09-1, p. 42. Kattan, F.H., 1983. Petrology and geochemistry of the Tuluhah belt, northeast Arabian shield. M.S. thesis. King Abdulaziz University, Jiddah, p. 111. Kemp, J., 1981. Geologic map of the Wadi al Ays quadrangle, sheet 25C, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Geologic Map GM 53, scale 1:250,000. Kemp, J., Gros, Y., Prian, J.-P., 1982. Geologic map of the Mahd adh Dhahab quadrangle, sheet 23E, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Geologic Map GM 64, scale 1:250,000. Kemp, J., Pellaton, C., Calvez, J.-Y., 1989. Geochronological investigations and geologic history of the Precambrian of northwestern Saudi Arabia. Saudi Arabian Directorate General of Mineral Resources Open-File Report BRGM-OF-01-1, p. 120. Kröner, A., Greiling, R., Resichmann, T., Hussein, I.M., Stern, R.J., Dürr, S., Krüger, J., Zimmer, M., 1987. Pan-African crustal evolution in the Nubian segment of Northeast Africa. In: Kröner, A. (Ed.), Proterozoic Lithospheric Evolution. In: Geodynamic Series, vol. 17. American Geophysical Union, pp. 235–257. Ledru, P., Augé, T., 1984. The Al Ays ophiolitic complex; petrology and structural evolution. Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-04-15, p. 57. Le Metour, J., Johan, V., Tegyey, M., 1982. Relationships between ultramafic-mafic complexes and volcanosedimentary rocks in the Precambrian Arabian Shield. Deputy Ministry for Mineral Resources Open-File Report BRGM-OF-12-15, p. 90. Le Metour, J., Johan, V., Tegyey, M., 1983. Geology of the ultramafic-mafic complexes in the Bi’r Tuluhah and Jabal Malhijah areas. Deputy Ministry for Mineral Resources Open-Field Report BRGM-OF-03-40, p. 47. Manivit, J., Pellaton, C., Vaslet, D., Le Nindre, Y.-M., Brosse, J.-M., Fourniguet, J., 1985. Geologic map of the Wadi al Mulayh quadrangle, sheet 22H, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Geoscience Map GM-92, scale 1:250,000. Nassief, M.O., 1981. Geology and petrology of the Jabal Thurwah area, Western Province, Saudi Arabia. Ph.D. thesis. University of Lancaster, p. 180. Nassief, M.O., Macdonald, R., Gass, I.G., 1984. The Jabal Thurwah upper Proterozoic ophiolite complex, western Saudi Arabia. Journal of the Geological Society of London 141, 537–546. Pallister, J.S., Stacey, J.S., Fischer, L.B., Premo, W.R., 1987. Arabian Shield ophiolites and late Proterozoic microplate accretion. Geology 15, 320–323. Pallister, J.S., Stacey, J.S., Fischer, L.B., Premo, W.R., 1988. Precambrian ophiolites of Arabia: Geologic settings, U-Pb geochronology, Pb-isotope characteristics, and implications for continental accretion. Precambrian Research 38, 1–54.

162

Chapter 4: Neoproterozoic Ophiolites in the Arabian Shield: Field Relations and Structure

Quick, J.E., 1991. Late Proterozoic transpression on the Nabitah fault system–implications for the assembly of the Arabian Shield. Precambrian Research 53, 119–147. Quick, J.E., Bosch, P.S., 1989. Tectonic history of the northern Nabitah fault zone, Arabian Shield, Kingdom of Saudi Arabia. Directorate General of Mineral Resources Technical Record USGSTR-08-2, p. 87. Ramsay, C.R., 1986. Geologic map of the Rabigh quadrangle, sheet 22D, Kingdom of Saudi Arabia. Saudi Arabian Deputy Ministry for Mineral Resources Geoscience Map GM-84, scale 1:250,000. Schandelmeier, H., Wipfler, E., Küster, D., Sultan, M., Becker, R., Stern, R.J., Abdelsalam, M.G., 1994. Atmur-Delgo suture: a Neoproterozoic oceanic basin extending into the interior of northeast Africa. Geology 22, 563–566. Shanti, M., 1983. The Jabal Ess ophiolite complex. Faculty of Earth Sciences, King Abdulaziz University, Jiddah, Bulletin 6, 289–317. Shanti, M., Roobol, M.J., 1979. A late Proterozoic ophiolite complex at Jabal Ess in northern Saudi Arabia. Nature 279, 488–491. Stacey, J.S., Stoeser, D.B., Greenwood, W.R., Fischer, L.B., 1984. U-Pb zircon geochronology and geologic evolution of the Halaban-Al Amar region of the eastern Arabian shield, Kingdom of Saudi Arabia. Journal of the Geological Society of London 141, 1043–1055. Stern, R.J., Johnson, P.R., Kröner, A., Yibas, B., 2004. Neoproterozoic ophiolites of the ArabianNubian Shield. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 95–128. Stoeser, D.B., Camp, V.E., 1985. Pan-African microplate accretion in the Arabian shield. Geological Society of America Bulletin 96, 817–826. Stuckless, J.S., Hedge, C.E., Wenner, D.B., Nkomo, I.T., 1984. Isotopic studies of postorogenic granites from the northeastern Arabian Shield. Saudi Arabian Deputy Ministry for Mineral Resources Open-File Report USGS-OF-04-42, p. 40.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

163

Chapter 5

THE WADI ONIB MAFIC-ULTRAMAFIC COMPLEX: A NEOPROTEROZOIC SUPRA-SUBDUCTION ZONE OPHIOLITE IN THE NORTHERN RED SEA HILLS OF THE SUDAN I.M. HUSSEIN, A. KRÖNER AND T. REISCHMANN Institut für Geowissenschaften, Universität Mainz, 55099 Mainz, Germany

The Wadi Onib mafic-ultramafic complex represents the best preserved, though tectonically dismembered, Neoproterozoic (Pan-African) ophiolite in the northern Red Sea Hills of the Sudan. Forming part of a regionally distinct, southwest to northeast trending ophiolite-decorated shear belt (Onib-Sol Hamed suture) it consists, from bottom to top, of a basal peridotite unit, an exceptionally thick (2–3 km) transitional zone (TZ) of interlayered cumulates, an isotropic gabbroic mass with plagiogranite bodies, and a sheeted basic dyke complex. The highest stratigraphic section of the ophiolitic sequence is represented by pillowed basaltic lavas (containing fragmentary lenses of ribbon chert and/or graphitic to shaly carbonates) which are tectonically juxtaposed against the basal peridotite. Whereas the basal unit is strongly serpentinized and/or carbonatized, the transitional zone comprises abundant and well preserved pyroxenites, some of which are enriched in Cr relative to TiO2 . The TZ also shows a polycyclic cumulate arrangement that possibly originated from multiple magma pulses rather than from tectonic interslicing. Moreover, mineral grading, gravity stratification and a spectrum of folds with varying geometrical dispositions and amplitudes within discrete layers as well as a vertical metamorphic zonation (suggesting seafloor hydrothermal processes) are evident within the Onib ophiolitic sequence. In particular, the volcanic component is Ti-rich, has a transitional IAT/MORB character and is indistinguishable from anomalous MORB and/or marginal basin basalts. Thus, the Onib is envisaged to be of arc/back-arc (marginal) basin affiliation, and it is classified as a supra-subduction zone (SSZ) rather than normal MORB-type ophiolite. It was generated at 808 ± 14 Ma as documented by a plagiogranite single zircon Pb-Pb age. The ophiolitic sequence probably resulted from parental magma(s) generated through multistage partial fusion of mantle peridotite. 1. INTRODUCTION The Northern Red Sea Hills (NRSH) of the NE Sudan (Fig. 1) cover some 100,000 km2 between latitudes 19◦ N and 23◦ N and longitude 34◦ 30 E and the Red Sea. They form part DOI: 10.1016/S0166-2635(04)13005-3

164

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Fig. 1. Simplified geological map of part of the Nubian segment of the Arabian-Nubian shield (after Kröner et al., 1987) showing location of the Wadi Onib ophiolite (arrow), northern Red Sea Hills, Sudan.

of the Neoproterozoic (Pan-African) Nubian segment of the Arabian-Nubian shield (ANS) that straddles the Red Sea perimeter in NE Africa and Arabia. The ANS is the northern continuation of the Mozambique belt and, together, they have been referred to as the East African Orogen (EAO) (Stern, 1994). The ANS represents an excellent example of the Pan-

2. Previous Work in the Onib Region

165

African orogenic cycle that has long been recognized as a period of major crustal accretion (Kröner, 1984), where continental, island-arc, and oceanic terranes were brought together (Gass, 1981; Kröner et al., 1991; Reischmann and Kröner, 1994; Kusky et al., 2003) to form the crystalline basement of the African continent as part of the late Neoproterozoic supercontinent Gondwana. The Nubian segment of the ANS contains numerous ophiolite occurrences (Fitches et al., 1983; Hussein et al., 1984; Kröner et al., 1987; Abdel-Rahman, 1993; Zimmer et al., 1995; Reischmann, 2000; Hussein, 2000; Stern et al., 2004; Johnson et al., 2004). These classify as Tethyan-type (found in shallow structural positions with a preponderance towards the margin of the older continent—the Nile craton to the west) or as Cordilleran-type (interspersed within steep suture zones separating the arc terranes) (Reischmann, 2000). Modern-style plate tectonic processes are perceived to have operated within the NRSH, and this is because the bulk of the region is made up of voluminous arc-related volcanoplutonic terranes that are separated by a number of distinct high-strain ophiolite-decorated sutures (Kröner et al., 1987; Hussein, 2000; Kusky and Ramadan, 2002). Lateral accretion and suturing thus contributed significantly to the evolution of the NRSH. The Wadi Onib mafic-ultramafic complex constitutes one of the best preserved, though tectonically fragmented, ophiolitic sequence within the ANS and makes up the major part of the prominent, southwest to northeast oriented Onib-Sol Hamed suture. An improved understanding of the composition and history of this complex has important regional tectonic significance and helps to understand global crustal evolution in the Neoproterozoic.

2. PREVIOUS WORK IN THE ONIB REGION Early geological data on the Wadi Onib and adjacent Deraheib regions were provided by Bagnall (1955) and Gabert et al. (1960) who thought that gabbroic and serpentinitic rocks of the NRSH represent intrusive bodies and/or dykes. The latter authors, in particular, have shown most of the Onib area to be underlain by greenstones which they incorporated into the Nafirdeib Series, a lithostratigraphic term (Ruxton, 1956) embracing widespread volcano-sedimentary sequences, the type area of which is Wadi Nafirdeib in the northern Red Sea Hills. In the late 1970’s a joint French/Sudanese mineral exploration project carried out the first systematic regional geological mapping and mineral prospecting of the Deraheib region. Within the scope of this project Stolojan et al. (1978) and Hussein et al. (1978) delineated the Onib mafic-ultramafic complex and, in the western part of the area, they described a large number of serpentinized and/or heavily carbonatized ultramafic bodies that are interspersed within a prominent shear belt named by Kröner et al. (1987) as the Hamisana Shear Zone (HSZ). Stolojan et al. (1978) and Hussein et al. (1978) reported on the mineral potential of the Onib/Hamisana region and also proposed that the Onib complex is comparable to the Jebel Sol Hamed ophiolite described by Hussein (1977) and Fitches et al. (1983). In particular, in a summary of mineral assessment activities of the French/Sudanese Project, de Bretizel (1980) noted that the Onib complex is a probable ophiolitic remnant.

166

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Hussein et al. (1984), Kröner (1985), and Kröner et al. (1987) reappraised the Onib and other regions in the Red Sea Hills and recognized pillow basalts and sheeted basic dykes within the Onib mafic-ultramafic suite. These authors advocated that the Onib complex represents a well preserved, though tectonically dismembered, Pan-African ophiolite.

3. ANALYTICAL PROCEDURES Based on detailed petrography, representative suites of ultramafic rocks, gabbros, basic dykes and basaltic pillow lavas as well as chromitiferous ores were selected for major and trace element analysis. Each rock sample was pulverized (< 0.06 mm) using an agate vessel fitted into a Siebtechnik mill. The powders were then decomposed into stock solutions which were analysed for major elements using directly coupled (DC) plasma emission spectrometry. The rock powders were also compressed into duplicate pellets and were analysed for trace and light rare earth elements (LREE) using a Siemens SRS 200 X-ray fluorescence spectrometer. Analytical details are given in Laskowski and Kröner (1984). Polished and thin sections were prepared from selected mafic and ultramafic rocks as well as chromite ores. The sections were used for major element analysis of primary and secondary silicate mineral phases and chromite ores using a Camebax Microbeam electron microprobe fitted with four wavelength dispersive spectrometers, one energy dispersive spectrometer and the capability of electron microscope scanning. The accelerating voltage was 15 kV, and the beam current was 10 nA. All measurements were corrected by the ZAF procedure (Atomic Number, Absorption Fluorescence correction), and oxygen contents for the oxides were calculated by stoichiometry. The standard deviation for all components measured is about 1%. For whole-rock major element analyses the DC plasma emission spectrometer was calibrated with artificial standards and checked against international rock standards (NIM-G (granite), NIM-S (syenite), NIM-N (norite), NIM-D (diorite), NIM-P (pyroxenite), BHVO1 (basalt), DTS-1 (dunite), PCC-1 (proxenite), RGM-1 (rhyolite), GSP-1 (granodiorite), SDC-1 (mica schist), STM-1 (syenite), JB-1 (basalt), BR (basalt) and GA (granite)) (see also Krüger, 1982). The accuracy of the method was about 2%, whereas the precision was about 1%. For major elements determined on the electron microprobe, international rock standards (BCR-1 and G-2) were measured as unknowns. The accuracy of the measurements was mostly better than 1%. The trace element analyses by X-ray fluorescence spectrometry were calibrated against international standards. For details see Tables 1–6 in Hussein (2000).

4. OUTLINE OF THE GEOLOGY OF THE ONIB OPHIOLITE The Wadi Onib ophiolitic complex (Figs. 1–5) forms a large (ca. 80 km × 10 km), smoothly curved exposure and can be reached from Port Sudan via a number of desert

4. Outline of the Geology of the Onib Ophiolite

167

Fig. 2. Geological map of the Wadi Onib ophiolite (after Hussein, 2000).

tracks that run westwards from an all-season coastal road along the Red Sea. Satellite imagery as well as geological mapping (Hussein et al., 1984; Kröner et al., 1987; Stolojan et al., 1978; Hussein et al., 1978) reveal that the Onib complex is enclosed within multiply deformed, weakly metamorphosed arc-related volcano-sedimentary sequences and that the contact between the two suites is a thrust fault everywhere. Though the basalt-andesite arc assemblages locally show rapid facies changes along a predominantly southwest to northeast bedding/foliation strike, they generally display comparable lithotypes. To the east of the Onib complex, there are widespread basaltic andesite/andesite lava flows in addition to restricted dacitic and/or quartz-crystal tuffs. The lavas are accompanied by abundant andesite-derived clastic sediments, tuffaceous greywackes, shales,

168

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Fig. 3. Distribution of ophiolitic lithologies along Wadi Onib (A) and Wadi Sudi (B). Locations of both Wadis are indicated in Fig. 2.

4. Outline of the Geology of the Onib Ophiolite

169

Fig. 3. (Continued.)

polymictic intra-arc conglomerates (some of these contain rare ophiolitic clasts) as well as extensive limestone interbands that locally show stromatolitic structures. The conglomeratic and limestone beds in particular may be traced along strike for tens of kilometres (e.g., between wadis Onib and Sudi). To the southwest of Onib the arc assemblage includes sheared lavas, greywackes and siliceous sediments. Within the Hamisana shear zone to the west the arc sequences are mylonitized to the extent that, at places, it becomes difficult to identify protoliths in the field. The Onib ophiolite and arc-related volcano-sedimentary assemblages are intruded by widespread syntectonic granitoids of batholithic proportions (composite dioritic/granodioritic/granitic plutons) and/or by younger (post-tectonic) gabbroic/granitoid stocks.

170

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Fig. 4. Composite stratigraphic column for the Wadi Onib ophiolite (compiled from Kröner et al., 1987 and Hussein, 2000).

In places, the younger magmatic rocks form distinct ring structures (for details see Hussein, 2000).

5. OPHIOLITE LITHOLOGIES AND LITHOSTRATIGRAPHY The Onib complex displays an orderly succession that defines a Penrose-style (Anonymous, 1972) ophiolitic suite (Fig. 4). Though swelling, pinching out and shearing, thrusting and interslicing of magmatic layers prevail and an entirely intact ophiolitic sequence is not preserved, five ophiolitic lithostratigraphic units have been reconstructed through detailed

5. Ophiolite Lithologies and Lithostratigraphy

171

Fig. 5. Schematic cross-sections showing disposition of lithologic components of the Onib ophiolite between Wadi Ader Aweb (1) via Wadi Sudi (2) and Onib (3) and as far as Wadi Hofra (4), northern Red Sea Hills, Sudan. A section across the Jebel Sol Hamed ophiolitic complex (5) farther northeast (see sketch map, top left) is shown for comparison (modified after Hussein, 2000).

field work carried out along major dry river beds (viz. Sudi and Onib) and/or side channels. The sequential ophiolitic stratigraphy begins, at the bottom, with a discontinuous, ca. 20 km × 2 km, extensively serpentinized and/or variably carbonatized basal peridotite. This unit is succeeded by an exceptionally thick (ca. 2–3 km) transitional zone (TZ) of interlayered cumulates including layered melano-to-mesogabbros, serpentinized dunitic and pyroxenitic peridotites in addition to well preserved olivine- and/or chromian spinel-bearing pyroxenites and wehrlitic transitions. At upper levels the TZ displays a gradual diminution of both mafic gabbroic and ultramafic rocks. Thereupon, a large (ca. 80 km × 2–6 km) gabbroic mass, dominated by crudely to well layered leucogabbros, appears in the magmatic sequence. At upper sections the layered gabbros intermingle with high-level isotropic gab-

172

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

bro, hornblende gabbro and quartz-diorite (plagiogranite). These rocks were frequently injected by single, multiple and/or criss-cross microgabbroic, microdioritic and/or mafic to intermediate dykes. In particular, a sheeted mafic dyke complex that displays asymmetrical chilling of dyke margins occurs in lower Wadi Sudi. A block of mafic dyke-intruded basaltic pillow lava containing rare fragments of pelagic sediments (ribbon chert and/or graphitic to shaly carbonates) represents the highest stratigraphic section of the ophiolite sequence. The latter unit is tectonically juxtaposed against the basal peridotite of uppermost Wadi Onib and tributaries. 5.1. The Basal Ultramafic Unit The Onib basal unit is dominated by an extensively serpentinized peridotite that hardly displays any of the diagnostic features of metamorphic tectonites (e.g., gneissic foliation and linear fabrics due to subparallel orientation of stretched orthopyroxene porphyroblasts). It is best exposed around uppermost Wadi Sudi and environs where the unit and adjacent strongly sheared and/or mylonitized lavas, greywackes and siliceous sediments of the arc assemblage are in tectonic contact along a vertically to steeply eastward-dipping thrust fault. The serpentinized peridotites are generally fine-grained, smooth, massive, distinctly obdurate, dark greyish-black and/or greenish in colour. Whitish magnesite-vein infillings, reddish nets of iron stains in addition to medium brown colouration are common across these ultramafic rocks. Some 1.5 km downstream from the tectonic contact zone a mélange-like magmatic breccia and carbonatized serpentinites form part of the basal peridotitic suite. In particular, the breccia contains numerous disoriented blocks of up to several metres in diameter. Occurring within a weathered, fragile, yellowish-brown, mediumgrained pyroxenitic matrix, the blocks include greyish-brown, fine-grained serpentinized dunite and an exceptionally fresh and remarkably coarse-grained protogranular pyroxenite which shows pegmatoid textures at places. Hand specimens from the pyroxenitic blocks are steel-black to bluish-black on fresh surface and brown to yellowish on weathered surface. Longer dimensions of pyroxenes in these rocks can reach 3–5 cm in length and some 0.5 to 1.5 cm in width. We interpret this rock as a localized primary magmatic breccia. Farther north (upper Wadi Onib and adjacent areas) the basal unit is faulted against the pillow lava block. The contact zone is characterized by strong shearing and/or transformation of some rocks into greasy, brownish-white, talcose material. It is crossed by a few mafic dykes which seem to post-date the deformation of the pillow lavas. The Onib basal ultramafic mass locally shows clear magmatic contacts between fresh to mildly serpentinized interlayers. Magmatic interfingering is evident where dunitic schlieren penetrate pyroxenitic host rocks. 5.2. The Transition Zone (TZ) Transition zones in layered mafic-ultramafic complexes (Wilson, 1959) usually represent marker horizons for a gradual change from cumulate peridotites to cumulate gabbros. Various estimates show that they rarely exceed 750 m in thickness (Coleman, 1977). The

5. Ophiolite Lithologies and Lithostratigraphy

173

Onib TZ, however, may thicken up to 2–3 km at places. In the main part it consists of a well preserved, lensoid pyroxenitic suite that grades into a crudely layered to a massive melano- to mesogabbroic assemblage containing frequent serpentinite layers and some restricted anorthositic lenses. The TZ ultramafic bodies vary in thickness and lateral extent from metres to kilometres in scale but average hundreds of metres in their long dimensions. Swelling out and/or sudden disappearance of the cumulate layers are common. In places, sharp magmatic contacts between ultramafic lenses and gabbroic host rocks occur on a metric scale as well. The pyroxenitic suite of the TZ is dominated by greyish-green, medium to coarse-grained olivine-rich pyroxenites and wehrlitic transitions, which form rugged hillocks and merge into the basal unit dominated by the serpentinized peridotite. In the same area the peridotitic rocks contain sporadic podiform (high-Cr) and disseminated (high-Al) chromite lenses and selvages that locally show size-grading and/or sharp magmatic layering (Fig. 6C); here the mineral stratification attests to and confirms the cumulative nature of the hosting pyroxenites and of the TZ. A pyroxenitic rock from the area is almost an ore-bearing host and is found to contain almost 40% Al2 O3 when chemically analysed. The widely distributed serpentinite bodies within the melano- to mesogabbros are finegrained, massive and range in colour from dark-brown to greyish-green. Their magmatic contacts with the gabbros are generally sharp though tectonic slices of serpentinite, detached from the basal peridotite unit, may belong to the TZ. Furthermore, it is not uncommon to observe gabbroic and serpentinitic rocks tightly folded together, a criterion that equally supports the primary nature of the TZ cumulates. Shear zones lacing deeper-level serpentinitic bodies of the TZ contain conspicuous, excessively carbonatized and/or silicified, yellowish-brown, massive, amorphous and inhomogeneous dyke-like bodies that range from centimetres to hundreds of metres in extent. Such silica-carbonate rocks are also known from the Jebel Sol Hamed ophiolitic complex near the Red Sea coast to the northeast (Hussein, 1977). Carbonatization (listwanitization) and/or silicification of the ultramafic rocks of the TZ and elsewhere are believed to be of metasomatic origin. Breakdown of serpentinite into carbonatized rock may have taken place through complicated processes involving release of carbon dioxide and its combination with oxygen and magnesium to form MgCO3 out of an essentially magnesium-aluminum-silicate serpentinitic structure. Alumina and silica will be removed as silicates and free quartz which is sometimes found as tiny aggregates or veinlets. Some of the iron, which originally existed in the olivine and/or pyroxene crystals, may be oxidized to form a pervasive reddish-brown colouration of these rocks. According to Buisson and Leblanc (1986) hydrothermal alteration at moderate temperatures (150–300 ◦ C) by Na- and Cl-brines derived from mantle material and interaction with seawater is perceived to be responsible for the development of carbonatized bodies from their ultramafic protoliths. The latter authors highlight that the listwanitization processes may be responsible for some gold mineralization (in quartz gangue), sulphides and arsenides known from the Bou Azzer complex, a Pan-African ophiolite in southern Morocco (Bodinier et al., 1984).

174

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

5.3. The Gabbroic Unit Upwards in the magmatic sequence the TZ grades, through a gradual diminution of ultramafic layers, into a large (ca. 80 km × 6 km) gabbroic mass that forms the bulk of the Onib ophiolite complex. At deeper levels of the gabbro suite some dark, medium to coarsegrained, crudely layered, partly massive pyroxene-gabbros predominate that may thicken up to 1.5 km and may contain localized melagabbroic/anorthositic layers. The pyroxene gabbro was subjected to metamorphic recrystallization involving sea-floor (hydrothermal) metamorphism. Strong shearing has disrupted the continuity of individual gabbroic layers which cannot be followed for more than a few hundreds of metres along the roughly southwest to northeast strike of magmatic layering. In turn, the pyroxene-gabbros grade upwards into well layered, medium to coarsegrained, light grey to greenish-grey, plagioclase-rich leucogabbros that dominate the magmatic sequence farther east. In places, the leucogabbros contain rare serpentinitic lenses (tectonic slices?) and/or deformed pegmatoid wedges. They frequently display magmatic layering (Fig. 6E), gradual vertical change of cumulus mineral phases in overturned mineral graded stratification as well as combinations of continuous and sharp mineral grading with indications of cross stratification. The gross igneous stratification allows recognition of way-up criteria. This suggests that although there are northwest and southeast-dipping igneous laminations (variable orientations are due to large open folds deforming older, isoclinal folds and due to abundant low angle thrusts), the overall stratigraphic vergence is towards the east (i.e., the overall younging of the gabbroic layer is in that direction; see also Hussein et al., 1984). The layered gabbroic suite merges gradually upwards into a non-cumulate, isotropic gabbro which forms low hillocks to the SW and W of the Sudi well (Fig. 3B). Isotropic gabbro, however, is missing from the Wadi Onib area to the north, most probably due to the inconsistency of stratigraphic horizons across the entire ophiolitic sequence. The contact zone between the layered gabbroic pile and the isotropic gabbro is difficult to ascertain. High-level hornblende gabbro and quartz-diorite are associated with the isotropic

Fig. 6. (A) Photomicrograph (crossed nicols) of an Onib olivine-bearing pyroxenite (transitional zone of middle-upper Wadi Sudi) with abundant high-Al chromian spinel (dark subhedral crystals at centre). Length of photograph is approximately 2.5–3 mm. (B) Photomicrograph (crossed nicols, ca. 3.5 mm in length) of fresh Onib pyroxenite (uppermost Wadi Sudi) depicting twinning as well as deformed exsolution lamelle in clinopyroxene. (C) Mineral size-grading in chromite from Onib ophiolite (upper-middle Wadi Sudi). (D) Photomicrograph (crossed nicols) showing chromite in serpentinized peridotite (upper-middle Wadi Sudi). Note crude layering and/or pull-apart segmentation displayed by some of the chromite granules. Length of photo is about 2.5 mm. (E) Mineral grading in typically layered Onib ophiolitic gabbro (lower Onib gorge). (F) Tight isoclinal folding indicating high temperature ductile deformation in layered Wadi Onib ophiolitic gabbro (upper-middle Wadi Sudi). (G) Disposition of feeder (sheeted) dykes and lavas of Onib ophiolite (area of uppermost Wadi Onib). (H) Well-preserved Onib pillow lavas (uppermost Wadi Onib). Note tightly packed pillow lobes surrounded by chilled (fine-grained) interpillow matrix.

5. Ophiolite Lithologies and Lithostratigraphy

175

176

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

gabbro. Restricted leucocratic, dyke-like bodies and thin plagioclase-rich veins also occur in places across this gabbroic lithotype. The leucocratic rocks contain minor plagiogranite bodies. The Onib ophiolitic gabbroic mass, in general, contains numerous mafic dykes, the frequency of which decreases westwards, a feature which implies a genetic link with the sheeted dyke complex farther to the east. 5.4. Sheeted Dyke Complex Sheeted doleritic dyke-in-dyke units are common within ophiolite complexes. They have been almost universally considered as convincing indicators of seafloor spreading at zones of plate accretion (Moores, 1982) and, consequently, provide evidence that ophiolitic sections represent ancient oceanic lithosphere. The sheeted dykes are generally characterized by asymmetric chilling of their margins. This is conceived to be the result of injections along a single fracture where each successive dyke is emplaced within the middle of the previously solidified dyke (Moores and Vine, 1971). A wide range of single, multiple, locally criss-crossing or swarmy doleritic, microgabbroic, microdioritic and porphyritic plagioclase andesite dykes and sills occur within the Wadi Onib complex. Emplaced at varying trends, they range in width from several cm to metres and extend in lengths over tens of metres. In general terms, all dykes experienced varying degrees of deformation (e.g., contortion and shearing) as well as low grade metamorphic alteration. Of special interest are dykes that form a sheeted block about 1.5 km west of the Sudi well (Figs. 2 and 3B). These are mafic dykes that invade each other and/or isotropic, microgabbroic to dioritic assemblages. The dykes have intrusive contacts, frequently carry screens of microgabbro, show doleritic textures and at places have chilled margins. The chilled margins do not exceed 10 cm in width, and asymmetric chilling is common. The sheeted dyke block is over one kilometre in length and widens up to 300 m. It merges into a transition from microgabbro to diorite which, in turn, grades into isotropic gabbro that passes westwards into the plagioclase-rich, well layered leucogabbroic mass outlined earlier. At the eastern side, the sheeted dyke complex is bounded by deformed arc-related lavas and volcaniclastic sediments through a sheared contact mostly obliterated by rock debris and superficial deposits. Strong shearing along this contact zone has transformed all rocks into schists. The sheeted dykes of Wadi Sudi are generally oriented north-northwest to south-southeast, approximately perpendicular to the principally southwest to northeast magmatic layering of the igneous cumulate rocks. However, due to post-emplacement deformation, this is by no means the only trend for the dykes which are frequently irregular in pattern. Along the western side of the ophiolite, specifically near the western tectonic contact between the basal ultramafic layer and the pillow lavas, there are also closely spaced basic dykes possibly of the sheeted dyke complex (Fig. 6G). The pillow lavas (upper Wadi Onib) were injected by such dykes which are subvertical. Over 13 east to west dykes can be seen within several metres across the stratigraphically upper level but structurally lowermost extrusive section. These magmatic injections range from 50 to 150 cm in width and mea-

6. Internal Structure and Deformation History of the Onib Ophiolite

177

sure up to several or tens of metres in length. Their contacts with the pillow lavas are sharp and intrusive, but with negligible chilling along their margins. They are interpreted here as feeders for the lavas since both assemblages are similarly deformed. 5.5. Pillow Lavas and Sedimentary Cap Direct evidence for the existence of pillow lavas on top of the sheeted dyke complex is lacking. It may be that the pillow lavas of the ophiolite have been tectonically imbricated with the arc lavas and, therefore, the units have become indistinguishable. However, to the west there is a discontinuous, ca. 300–800 m wide volcanic strip that stretches SWNE parallel to the overall lithological layering of the ophiolite (Fig. 2). It is tectonically juxtaposed against the basal, serpentinized peridotite and/or cumulate gabbroic layers and is sharply intruded by batholithic granitoid masses farther west. The volcanic exposure comprises massive and pillowed basalt, breccias, sills, and many dykes, some of which are feeders as outlined earlier. These rocks are dark to greyishgreen, extensively sheared and amphibolitized locally where intruded by the batholithic granitoids. Near the contact zone, distinct, tightly interlocked pillow structures occur in several basaltic lava outcrops. Subcircular to lenticular pillows (Fig. 6H) dominate over other forms which take the shape of circular, sickle-shaped, roughly triangular and other structures. Pillow margins, or inter-pillow sheaths, are thin (3 mm to 1 cm in width) and are essentially formed of bluish-grey or greenish chloritic aggregates and/or quartz material. Individual pillows have voids and vugs, some of which are filled with siliceous and carbonate material. The latter material, filling the vesicles and hollow cores, may be used as an indirect criterion to estimate the depth of lava extrusion. Moore (1979) advocates that the primary control of vesicularity of subaqueously erupted basalt is the depth (and hydrostatic pressure) of eruption. Shallow eruption produces more (and larger) vesicles not only because exsolved volatiles expand with lower pressure, but also because exsolution of volatiles dissolved in the melt is more complete. In the Sol Hamed ophiolite, the analogue and probable extension of the Wadi Onib ophiolite, pillow lavas were estimated to have been extruded at a water depth of about 1–2 km (Price, 1984). In addition, the upper Onib sheared pillow lavas also contain restricted, fragmentary bluish to brown ribbon chert and black, graphitic, shaly carbonate lenses. The pelagic sediments are several centimetres to a metre long and up to 5 cm thick. They generally occur as thin infillings and have experienced intense shearing together with their host pillow lavas. It may also be that the sediments were ripped off from a much thicker cap that once covered the pillow lavas.

6. INTERNAL STRUCTURE AND DEFORMATION HISTORY OF THE ONIB OPHIOLITE Structures developed during ductile flow and late stage congealing of magma before and during final crystallization (Davis, 1984) are reasonably well preserved within the

178

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Onib deeper sections (basal unit, TZ and layered gabbros). The principal primary structures within the lower ophiolitic sections are the cumulate fabric and lithological layering. In the TZ pyroxenites, the mineral-size grading of chromite (Figs. 6C and D) is evidence for cumulate formation within a magma chamber. Irregular, anastomosing, magmatic interlayering of peridotite with pyroxenite/dunite schlieren occurs locally in upper Wadi Onib. Sinuous contacts within the schlieren constitute flow lines that facilitated the development of a high-temperature ductile fabric. Fitches et al. (1983) also recorded a penetrative flattening fabric within ultramafic and layered gabbroic rocks of the Jebel Sol Hamed ophiolite complex. They interpreted the structure as a high-temperature ductile fabric, assumed to be related to horizontal ductile shearing in the uppermost mantle. Moreover, the cyclic nature of the cumulate rocks of the TZ and contiguous layered gabbros can be conceptualized in the framework of magmatic differentiation. Pearce et al. (1984a) described such an order of lithologies in ophiolites which they interpreted as having formed in a supra-subduction zone (SSZ) setting. However, Shervais (2001) advocated that SSZ ophiolites display a consistent sequence of events during their formation and evolution; entailing birth, youth, maturity, death and, finally, resurrection—a progression implying that ophiolite formation is not a stochastic event but is a natural consequence of the SSZ tectonic setting. Evidence of secondary structures in the Onib complex is provided by abundant folds ranging in size from microscopic crenulations (lineations) through meso- to macroscopic structures. From field observations it has been possible to separate at least three phases of deformation. This is indicated by folds with varying orientations and amplitudes within discrete layers. High-temperature isoclinal folding is recorded within interlayered peridotites and gabbros (Fig. 6F). This is probably one of the earliest structures and may have formed whilst the rocks were still hot and partially consolidated. At places mm-thick bands of metamorphic mineral growths (foliation) can be seen crossing the isoclinally folded magmatic layering. The younger bands of metamorphic minerals (e.g., amphiboles) may have developed during greenschist- to lower amphibolite-facies metamorphism. The syn-igneous (?) folds are best preserved in the lower levels of the TZ (upper-middle Wadi Sudi). Their dimensions vary from a few to tens of centimetres. The folds are generally asymmetrical to parallel, tight, and are commonly accompanied by high-temperature ductile shearing. At places there is clear transposition of folded igneous layers due to strong stretching, and the folds are intrafolial. They normally disappear along strike of the igneous layers. Tight to open, asymmetric folds, plunging at 30◦ to 60◦ northeast, are well developed in the Onib TZ and overlying layered gabbro. These folds are cut by penetrative schistosites, some of which are oriented parallel to the generally southwest to northeast trending fold axes. In the middle of Wadi Onib, layering of the gabbro is bent into a clearly asymmetric fold with limbs separated by about 90◦ . Farther west (upper middle, left side of Wadi Onib—Fig. 3A) highly contorted serpentinitic peridotite interlayered with gabbro can be seen within a section of 300–400 m. The folds in this area are tight, a few to several metres in amplitude, and make up a repetitive synform/antiform structure. They are locally disrupted by steep shear zones along which dislocations of layers can be seen. The folds also affected some of the mafic dykes crossing the plutonic ophiolite lithologies. The asymmet-

7. Petrography of the Onib Ophiolite Sequence

179

ric, tight-to-open folds that deform the igneous layering of the cumulates are believed to have developed during early phases of ophiolite obduction. At a few places early (?) shear zones cross the limbs of such folds, and the shear zones are locally annealed by mm- to cm-wide fillings of oxidized pyrite crystals (e.g., contorted gabbro some 5 km NW of Sudi well—Fig. 3B). The shear zones themselves are cut by later thrusts which are mainly shallow to subhorizontal and are apparently responsible for the interstacking and thickening of the ophiolite complex. Some late, gentle open folds are tens of metres to km in size and post-date all the structures outlined above. They occur in both the ophiolitic sequence and the adjacent VAA strata and may belong to syn-to-post-ophiolite emplacement deformation events.

7. PETROGRAPHY OF THE ONIB OPHIOLITE SEQUENCE Although the Onib complex was affected by serpentinization, metasomatic and ocean floor hydrothermal alteration in addition to regional metamorphic re-equilibration, detailed petrographic and geochemical studies made it possible to characterize several rock types that form the major lithological units outlined earlier. Details pertaining to the lithotypes of the plutonic and volcanic components are summarized below. 7.1. Dunite and Harzburgite Dunitic exposures within the basal ultramafic unit are inferred, indirectly, from the degree of carbonatization and/or the development of magnesite (vein-infillings and/or encrustations) in addition to a yellow-brown colouration on weathered surfaces. Harzburgitic rocks were not mapped as separate bodies in the Onib ophiolite. However, it is not uncommon to encounter partially serpentinized harzburgitic kernels that “swim” within strongly serpentinized assemblages. Such rocks are medium to coarse-grained in texture and range in colour from steel grey to black, dark brown, and pale green at places. Harzburgitic rocks in ophiolites are generally interpreted as representing refractory residues of a depleted mantle (Gass, 1980). They may also be essential components of a polycyclic igneous assemblage that shows distinct cumulate textures (e.g., size grading and parallelism of mineral phases such as chromite), kelyphytic growth of adcumulates as well as elongate, euhedral to subhedral and irregular, sharp grain-to-grain boundaries of some of the mineral components such as olivine and/or orthopyroxene. Serpentine derived from dunitic rocks has low relief, fine-grained texture and shows nets of mesh- and/or bladed as well as hour-glass structures (lizardite? and/or antigorite?). It constitutes some 95 to 99% of the rock, the rest of which includes scanty relicts of olivine and/or rare, pseudomorphosed orthopyroxene in addition to minor amounts of magnetite and/or chrome spinel. Some of the serpentine phases are crossed by silky serpentine and/or chloritic veinlets that probably developed during post-obduction regional metamorphism. Magnetite, in particular, occurs as a secondary product of serpentinization (black streaks, feathery granular aggregates and shapeless to irregular spots) or as rare, subhedral,

180

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

smoothly outlined accessories that could be of primary origin. Its disposition helps to indirectly recognize outlines of precursor olivine and/or orthopyroxene crystals. Subhedral and/or polygonal grain-to-grain boundaries of olivine are frequently preserved by magnetite trails and trains; this suggests a cumulate origin for the Onib serpentinized dunites. The magnetite grains also form linear tracts of exsolution lamellae of pseudomorphosed orthopyroxene, a phenomenon suggesting high temperatures and slow cooling. Furthermore, the accessory chromite grains range in colour from black to various tints of coffee-brown to reddish, signifying alteration into ferrichrome. Pull-apart structures and/or criss-crossed cracks have formed within the spinel granules some of which contain inclusions of flaky plates of serpentine, a phenomenon indicating that olivine was probably occluded as inclusions. Elongation and tadpole structures shown by some of the chrome accessories, in particular, are due to strong ductile deformation. These are high-temperature fabrics and were also reported from ophiolite complexes elsewhere (e.g., see Burgath and Weiser, 1980; Nicolas and Al Azri, 1991; Li et al., 2002; Huang et al., 2004). The harzburgites, on the other hand, consist of strongly to completely serpentinized olivine (up to 85%) and orthopyroxene (10–15%) in addition to substantial percentages of magnetite and minor, but ubiquitous, accessory chromite granules. Though the chromite grains are predominantly equant, euhedral and/or subhedral, they display elongation as well as pull-apart (?) segmentation and development of tapering ends in places. Relict olivine and/or orthopyroxene crystals showing poikilitic and/or kelyphitic interrelations are evident. Olivine engulfed by orthopyroxene (or bastitic pseudomorphs) and, in places, contrasts between large and small serpentine flakes are interpreted as indicators of original textures. Rare clinopyroxene crystals were locally encountered among the dominating mineral phases. In general terms, the serpentinized harzburgitic rocks consist essentially of meshtextured lizardite?, bladed antigorite? and tabular, or fibro-lamellar, bastitic? aggregates. Where bladed antigorite? phases become dominant, this feature may be correlated with advanced stages of tectonic mobilization which normally leads to recrystallization of the early serpentine phases as proposed by Prichard (1979). Serpentinization of orthopyroxene usually takes place through consumption from the outer margins inwards until the crystals are pseudomorphosed or left as tiny patches (oikocrysts?) surrounded by serpentine flakes. In addition, micro-structures related to high-temperature ductile deformation are common (e.g., as segmentation and development of tapering ends of elongate chromite grains as well as smooth, S-shaped bending and/or curvature of orthopyroxene crystals). Furthermore, in one harzburgitic section a triple-phase structural relation is indicated whereby anomalously yellow and/or greenish serpentine fibrils (probably chrysotile? grown along slip movement directions) cross serpentine blades which themselves penetrate into orthopyroxene crystals. In the oceanic realm the orthopyroxenes were probably subjected to further alteration through hydrothermal solutions at elevated temperatures to produce serpentine phases (e.g., bastite and/or modifications?). During ophiolite obduction, however, tiny blades of serpentine recrystallized, and these were then crossed by late-stage chrysotile? serpentine that shows the least indications of deformation. The latter serpentine phases, in particular, may well be related to post-obduction regional metamor-

7. Petrography of the Onib Ophiolite Sequence

181

phism which resulted in greenschist-facies conditions imprinted on the ophiolitic sequence and enclosing arc-related strata. 7.2. Pyroxenite and Transitions The Onib pyroxenitic assemblage includes exceptionally coarse-grained clinopyroxenite of the basal unit in addition to olivine-rich, spinel-bearing pyroxenite (Fig. 6A) and wehrlitic/websteritic pyroxenite and/or other transitions within the overlying cumulate TZ. The Sudi clinopyroxenite displays very coarse-grained, holocrystalline (hypodiomorphic granular) textures and consists predominantly of augitic clinopyroxene showing two-directional exsolution lamellae which are deformed (Fig. 6B) into zig-zag patterns. Clinopyroxene includes blebs of secondary amphibole. Bending and twisting of the minerals and their lamellae strongly suggest that polyphase deformation has affected the Sudi clinopyroxenite; at least one phase is thought to be related to ductile deformation, indicated by tight isoclinal folding of magmatic layering. Microprobe analyses of the exsolution lamellae within clinopyroxene indicate compositions close to chromian augite and/or ferroaugite (Deer et al., 1966). Only localized blebs within the augitic crystals show relatively high amounts of sodium, indicating local inversion of pyroxene into amphibole. Opaques within clinopyroxene are magnetite and, less commonly, chromian spinel. Within the lower sections of the TZ the most abundant pyroxenitic lithofacies is olivinerich spinel-bearing pyroxenite. This rock, which is inferred to be a cumulate, is medium to coarse-grained equigranular and consists of clinopyroxene in the main part (∼ 60–75%) as well as varying proportions of orthopyroxene and/or olivine (which is variably serpentinized) and accessory amounts of chromite and magnetite. Clinopyroxene and orthopyroxene reveal a common mosaic fabric. The pyroxene crystals are characterized by schiller structure and mutual intergrowths at places. Deformational effect are shown by granulation, cracking, twisting and/or minor dislocations of exsolution and twin lamellae, bending of entire crystals as well as wavy, oscillatory and fan extinction. Other textural features are relict embayments, occluded relations (where recrystallized pyroxene grains surround serpentine mantles that include clinopyroxene remnants) and deformed lamellae that are not associated with kelyphitic recrystallization. Optical and microprobe data from olivine-rich, spinel-bearing pyroxenites indicate a predominance of augite, diopside-augite, endiopside and, possibly, pigeonite and subcalcic augite at places (clinopyroxene) as well as enstatite bronzite and/or hypersthene. Olivine is either recrystallized into smaller grains and/or partially serpentinized. Wehrlitic transitions from olivine-rich spinel-bearing pyroxenite are not uncommon. Microscopic investigations of wehrlite indicate that it consist essentially of olivine (∼ 65%), a substantial amount of clinopyroxene (∼ 30%), with or without trace amounts of orthopyroxene. These mineral phases are accompanied by accessory amounts of spinel that may constitute up to 2% of the mode at places. Olivine is variably serpentinized. Whereas some olivine grains were replaced by serpentine along cleavage fractures and borders, others were totally consumed. In places it is possible to recognize olivine pseudomorphs which are enclosed in fresh, though strained, clinopyroxene within wehrlitic sec-

182

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

tions. Spinel is varying in abundance and type from one wehrlite to another. Dark opaque subhedral (local euhedral) chromite grains and clots are common. Another accessory phase is characteristically greenish and translucent under plane-polarized light and isotropic under cross-polarized light. This is an aluminous spinel which is abundant within olivinepyroxenite in the SW section of the TZ. In addition to spinel, black magnetite dust and granules are commonly seen within fractures of serpentinized olivine or within partings of orthopyroxene. Other pyroxenite assemblages within the TZ include websterite and websteritic-olivineclinopyroxenite. From a modal point of view websterite consists of comparable proportions of ortho- and clinopyroxene. The websteritic olivine-clinopyroxenite contains orthopyroxene, clinopyroxene and olivine in decreasing abundance. 7.3. Podiform Chromite The deeper levels of the Onib TZ comprise sporadically distributed, disseminated or speckled (high-Al) and restricted massive (high-Cr) chromian spinel occurrences. Petrographic studies of the disseminated chromite under transmitted and reflected light reveal that the ore occurs in interlocked to disparate anhedral to subhedral, brownish to greenish grains with silicate phases filling the interspaces. The ore shows relict cumulate textures (including poikilitic relations, i.e., chromian spinel enclosed by pyroxene crystals) though it is unquestionably deformed (cracked and microbrecciated grains are evident at places). It varies from a low, accessory level up to 30% or more of the olivine-pyroxenitic host rock (see Fig. 6A). Martitization (haematitization of magnetite veinlets associated with greenish and/or brownish spinel—the colours are internal reflections of the latter mineral) is common. This is possibly due to the deformational effect and/or metamorphic transformation that has affected the primary mineralogy and fabric of the ophiolite. On the other hand, the high-Cr chromian spinel is far richer (forming up to 80% of the host serpentinite) than the disseminated (high-Al) variety and, under the ore microscope, clearly forms a chromitite. This is generally massive and is made up of crudely layered, compactly interlocked and/or interspaced subhedral to anhedral, dark-brown to amber coloured (plain polarized light) grains and aggregates. The chromite is intricately intergrown with serpentine flakes which contain tiny magnetite dust in places. Ferrichrome rims and/or martitized magnetite alteration zones within the chromite grains are common. A rather rare mineral associated with the chromite is some metallic-yellow, anisotropic suspected sulphide (hazelwoodite nickel and/or chalcopyrite?) which is noticeable across some polished chromitite sections. Platinoid granules (?) are also not ruled out from the suspect sulphides. The subparallel disposition (alternation of ore and silicate mineral sheaths), the existence of some euhedral grain forms and occluded textural relations with the serpentine phases principally indicate a cumulate origin for the massive chromite as also advocated for the disseminated variety. The Onib chromitite, however, is believed to have been subjected to a complex, postaccumulation deformational history. Occasional pull-apart segmentation structures (e.g., Fig. 6D), brecciation and invasion of some chromite grains by strings of serpentine and/or

7. Petrography of the Onib Ophiolite Sequence

183

carbonate infillings as well as the presence of smoothly-bounded, curvilinear to irregular clots and one-side-tapering (development of tadpoles?) chromite grains of 1 to 2 mm in length, are common. The tadpoling of the chromite granules is interpreted as an indication of high-temperature ductile deformation. It is broadly comparable to flattening and elongation of chromite parallel to foliation and lineation reported by Li et al. (2002) from an Archaean ophiolitic peridotite in northern China. These authors attribute the phenomena to intensive high-temperature shear strain and that the textures probably record the plastic flow of the upper mantle, now mainly preserved in the core of tectonic blocks. Tadpole grains (i.e., boudinaging) of chromite can be brought about by solid state flow as documented and interpreted by Burgath and Weiser (1980) who studied primary features and genesis of podiform chromites from Greek ophiolites. These authors considered chromite stretching as the first step of boudinage which, on a large scale, can explain the formation of the dissected, lensoid podiform chromite deposits in the ophiolites of Greece. Holtzman (2000) investigated mantle chromitite pods in the Oman ophiolite and explained that at some time after chromite has crystallized, it begins to deform by fracture and possibly creep mechanisms in a matrix deforming mainly by dislocation creep. He summarized that among the only strain markers in peridotite, the chromitite pods record the kinematics of corner flow and rheological transition from asthenosphere to lithosphere. Thus, as chemical heterogeneities, they provide opportunities for insight into mechanical and petrologic processes of melt migration in the upper mantle. 7.4. Melano- and Mesocratic Gabbro These rocks are essentially pyroxene-gabbros with inferred gabbronoritic gradations. They are dark in colour and medium to coarse-grained in texture. Mineralogically they are simple, due to a notable absence of olivine as well as considerable recrystallization of pyroxene into uralitic amphibole and plagioclase into saussuritic material. For example, melanocratic gabbro from the middle to uppermost Wadi Onib consists of some 70% uralite, ∼ 25% saussurite and minor amounts of zoisite and chlorite as well as traces of magnetite and sphene. Mesocratic gabbro is much more abundant than the mela facies and contains up to 45% saussuritic aggregates, the main components of which are sericite, albite, carbonates and/or laths of ink-blue epidote (zoisite) that has straight extinction and residual twin lamellae. Whereas uralite (tremolite/actinolite?) occurs as fibrous blades which are non-uniformly bleached and show interpenetration, the saussurite is cloudy to turbid and rarely contains original plagioclase. From this it is clear that although the Onib gabbros are altered, they preserved their primary textures; this indicates greenschist to lower amphibolite-facies metamorphism. Such mineralogical and textural criteria are indicators of in-situ metamorphism (i.e., hydrothermal transformation of minerals taking place within the oceanic crust). This was also reported from neighbouring ophiolites, e.g., Wadi Ghadir, SE Desert of Egypt (El Bayoumi, 1980) and Gerf some 80 km to the north of Onib (Zimmer, 1989).

184

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

7.5. Leucogabbro The bulk of the Onib gabbroic mass is made up of well layered, predominantly leucogabbroic varieties. Microscopic investigations revealed a typical cumulate nature. The rock shows considerable variations in texture and/or mineral assemblage, though the simple paragenesis saussurite/uralite is retained almost everywhere. However, the average rock from the layered plagioclase-rich leucogabbro consists of more than 60% saussuritic material, about 30% uralitic amphibole, and the rest of the mode comprises the usual ink-blue zoisite, sericite, albite patches, clear specks of quartz (?), carbonate aggregates, iron oxide rods and blebs as well as epidote clusters and apatite needles in decreasing abundance. Plagioclase is strongly altered into ink-blue zoisite and retains some twinning. Its An-content is about 60%. 7.6. Isotropic Gabbro Coalescing occurrences of isotropic gabbro, hornblende-gabbro and/or gabbro-diorite occur mainly to the west and southwest of the Sudi well (Fig. 3B). The rocks are dark to greenish-grey, massive, fine- to medium grained, essentially holocrystalline (hypodiomorphic to allotriomorphic granular) and show less deformation than the layered gabbroic cumulates. The average isotropic gabbro consists essentially of somewhat equal proportions (∼ 40%) of uralitic amphibole and saussurite with subordinate amounts of remnant clinopyroxene. Chlorite, zoisite, quartz, apatite and iron granules constitute accessory phases. The appearance of quartz and apatite may be due to metamorphism of plagioclase or may be explained as reflecting crystallization (as interstitial material) during late stages of magmatic activity when temperatures were significantly lower than those maintained during formation of the cumulates. The isotropic gabbro is the only gabbroic assemblage in the Onib area that contains up to 7% Qtz and between 3 and 26% Ab in its CIPW normative composition. Its An is conversely the lowest (∼ 37%) and Hy is ∼ 15%, similar to the other gabbroic facies analyzed (Hussein, 2000). The normative features are not far from the modal mineral characteristics, though alteration is ubiquitous. In general terms, the isotropic gabbro shows a shift towards a sodic plagioclase-bearing gabbroic/dioritic association that is frequently encountered along the marginal zones of the uppermost gabbroic sections of many ophiolitic complexes. 7.7. Plagiogranite Plagiogranite or trondhjemite (low colour index quartz-plagioclase rocks) is frequently found as irregular bodies within, or marginal to, the gabbroic suites of ophiolitic complexes. These rocks may occur as small dykes and veinlets amongst high-level gabbrodioritic complexes or as wedges within sheeted dykes. Plagiogranite is a late-stage differentiate of qtz-normative basaltic magma (Coleman and Peterman, 1975). Chemically, an ideal plagiogranite is impoverished in K2 O (< 1%) and Rb (< 5 ppm) but high in Sr when

7. Petrography of the Onib Ophiolite Sequence

185

compared to continental granophyres. Because plagiogranites are end members of the differentiation products of the ophiolitic suite they are enriched in SiO2 , Na2 O, FeO, and TiO2 . Pearce et al. (1984a) asserted that plagiogranites of all tectonic subgroups (normal ocean ridges, anomalous ocean ridges, back-arc basin ridges and supra-subduction zone ridges) have hornblende as the dominant ferromagnesian mineral, plot as quartz-diorite or tonalite in the Streckeisen (1976) diagram and may be metaluminous or peraluminous in character. In the Wadi Onib ophiolite, plagiogranites can easily be confused in the field with postophiolite arc-related batholithic granitoid occurrences. However, the batholithic granitoids are biotite and/or hornblende-bearing; they have larger dimensions and are, more or less, mafic dyke-intruded. They also have sharp contacts with the country rocks. The plagiogranite, on the other hand, contains little or no biotite and occurs as dyke-like and/or irregular bodies within the gabbroic domain. In addition, staining and petrographic studies of a number of thin sections from suspected plagiogranites have revealed no or only traces of alkali feldspar. Under the microscope the Onib plagiogranite is holocrystalline, allotriomorphic granular and consists essentially of plagioclase, quartz, hornblende and uralite. In addition, there is chlorite, epidote, sericite, zoisite, apatite and iron ore granules that occur in trace abundance. Plagioclase appears to be albite to oligoclase in composition. Modal albitic plagioclase is clearly reflected by some 40% of normative Albite (Hussein, 2000). Single zircons from a sample of plagiogranite collected ca. 5 km west-northwest of the Sudi well yielded a magmatic crystallisation age of 808 ± 14 Ma (Kröner et al., 1992). 7.8. Sheeted Dykes Petrographic studies and interpretations of mineral parageneses and fabrics of the Onib inter-ophiolite dykes indicate that they are extensively deformed and recrystallized through hydrothermal ocean-floor alteration as well as through regional metamorphism that accompanied ophiolite obduction and orogeny (Hussein, 2000). The ophiolite-related dykes are tholeiitic (normative hypersthene-bearing) and texturally vary between basalt, dolerite and microgabbro. Sheeted dykes are found west of Sudi well and within the pillow lavas which are tectonically juxtaposed against the basal ultramafic unit in uppermost Wadi Onib. Both occurrences have comparable mineral parageneses and textural relations. The Sudi dykes and Onib counterparts are greyish-green in colour, medium to fine-grained in texture and predominantly consist of saussuritized, partly zoned tabular to fragmental plagioclase crystals with subordinate actinolite/tremolite. These are associated with trace amounts of remnant pyroxene and ubiquitous alteration products, mainly ink-blue zoisite that accompanies cloudy, cryptocrystalline saussuritic material, chlorite, epidote, carbonate and albite. The sheeted dykes, though altered and deformed, still preserve subophitic textures (with plagioclase crystals providing the framework and pyroxene crystals and derivatives building the interstitial pore-filling assemblages). However, the dykes of Wadi Sudi have less zoisite than those from uppermost Wadi Onib. Whereas the former have about 19% normative hypersthene the latter, in general, contain less normative hypersthene. Additionally, the Sudi sheeted dykes are more deformed than those in Wadi Onib and show tendencies of upper

186

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

greenschist/lower amphibolite-facies grade of metamorphism. This is thought to be an indication of the different levels of injection of the dykes; the Sudi dykes are believed to be stratigraphically lower than those of the Onib area. This inference is also corroborated by field criteria (e.g., in uppermost Wadi Onib the dykes inject pillow lavas whereas at Sudi they mainly inject microgabbros). 7.9. Pillow Lavas The Wadi Onib lavas have been extensively deformed and metamorphosed. Though the mineral assemblages are broadly similar, there are variable textures. Thin sections from margins of individual pillows, for example, show cryptocrystalline greenish chlorite which has probably replaced a hyaloclastic matrix. On the other hand, thin sections from the inner portions of the usually subrounded to elongated pillows still display some aphyric to microporphyritic textures, in places forming subophitic intergrowths between plagioclase laths and/or phenocrysts and partially to completely uralitized pyroxene flakes and blades. In one section, abundant uralitic amphibole alternates with fragmental, cloudy plagioclase crystals and all form sub-parallel structures (a sort of metamorphic foliation). Another rock shows more plagioclase than amphibole. However, the plagioclase is highly charged with zoisite and, to a lesser extent, with sericite. These secondary minerals form roughly subparallel bands. The latter may be indications of residual flow structures rather than a metamorphic foliation. The predominance of plagioclase and alteration products in some of the Onib lavas may indicate a fractionation phenomenon (fractionation leads to differentiated melts). Nonetheless, the majority of the lavas show pillow structures, textures and mineralogies of submarine tholeiites. 7.10. Sequence of Crystallization and Metamorphism Field and microscopic observations as well as structural interpretations enable tentative inferences pertaining to the possible course of mineral crystallization within the magma chamber(s) generating the Onib ophiolite. Granularity, crystallinity, grain shape and size as well as fabric relations of mineral phases that constitute the plutonic components reveal a broad zonation whereby the ophiolite became more fractionated in the west and east respectively. Olivine, chrome spinel, pyroxene and calcic plagioclase are the most abundant phases of the deeper sections of the plutonic cumulate mass, mainly in the basal peridotite and the adjacent TZ. Higher up in the magmatic sequence, the layered gabbroic rocks become enriched in plagioclase. Finally, at the highest levels, some isotropic gabbros bear sodic plagioclase, primary amphibole and minor amounts of quartz. The latter becomes an essential component of the differentiation end product, the plagiogranite. Preserved textures in the peridotite/gabbro component point to cumulate processes. Oikocrysts of olivine and orthopyroxene swimming within serpentine meshes and/or blades, euhedral to subhedral chrome spinel grains engulfed by or engulfing pyroxene crystals at places (poikilitic relations), the presence of two generations of pyroxene and olivine (large and small), the binding of plagioclase tabular crystals by intercumulus pyroxenes

8. Geochemistry of the Wadi Onib Ophiolite

187

(subophitic to ophitic textural fabric) and many other microstructural criteria suggest that a complex evolution took place before the cumulate pile was built up. The exceptionally thick TZ contains abundant clinopyroxene. Both olivine and orthopyroxene are thought to have been trapped early in the basal peridotite. This inference is corroborated by the fact that olivine is not distinguished as a component of the gabbroic rocks which are found to be essentially pyroxene and/or plagioclase-rich with noritic and anorthositic affinities at places, the latter being adcumulate. However, due to the advanced stage of alteration in the cumulate component (serpentinization in the peridotite and saussuritization and/or uralitization of the gabbros) it was difficult to establish a single order of crystallization, though combinations of olivine-clinopyroxene-orthopyroxene-plagioclase, olivine-clinopyroxeneplagioclase-orthopyroxene and/or olivine-orthopyroxene-clinopyroxene-plagioclase sequences are thought to have been maintained. It is probable that a parental magma was generated through partial melting of mantle peridotite beneath a spreading ridge and evolved towards a basaltic and, in places, even andesitic composition through fractional crystallization. It has produced large quantities of olivine, spinel, clinopyroxene and calcic plagioclase within the TZ. Further fractionation has given rise to high-level gabbroic/dioritic emplacements and, finally, to the uppermost intrusive/extrusive carapace of dykes and pillow lavas. Laurent et al. (1980) asserted that the composition of a parental magma and its changes during differentiation are the main factors controlling the composition of the cumulus minerals and the order of crystallization. Furthermore, detailed petrographic studies indicate that the Onib ophiolite was subjected to varying degrees of metamorphic overprinting, partly due to hydrothermal circulation. Criteria that enable us to make such an inference are: (1) vertical zonation of metamorphic grades and termination of metamorphic effects with depth (namely the pyroxenites and/or wehrlites of the TZ are the best preserved amongst the cumulate rocks), (2) albitization and saussuritization of plagioclase, (3) uralitization of pyroxene, (4) development of lizarditic serpentine (possibly early phases), (5) alteration of ilmenitic accessories into sphene and magnetite into haematite, (6) preservation of igneous textures although the primary minerals are more or less replaced (e.g., zoisite is well developed and retains twin lamellae of precursor plagioclase phases), and (7) rarity of schistosity and linear fabric within the gabbros, though saussuritized and/or uralitized.

8. GEOCHEMISTRY OF THE WADI ONIB OPHIOLITE Whole-rock and mineral data for the Onib ophiolitic sequence are presented in Tables 1–3 and Figs. 7–11. Due to the cumulate nature of the Onib plutonic sequence, emphasis is placed on the geochemistry of the volcanic suite. Although the latter was subjected to sea-floor hydrothermal alteration and/or subsequent regional metamorphism most major and trace elements still reflect the composition of the Onib parental magmas (Hussein, 2000). Major element abundances, variations and ratios are used to establish broad-scale chemical groupings, whereas trace elements are used to highlight igneous processes and pertinent palaeotectonic environments.

ONH18 Gabbro 48.21 0.18 24.57 1.08 3.4 0.06 5.7 14.86 2.42 < 0.01 0.05 0.65 0.49 0.09 0.16

ONH41 Gabbro 45.65 0.08 20.71 1.48 3.88 0.07 10.66 15.59 0.66 < 0.01 0.01 1.89 1.88 0.11 0.01

Total

100.28

101.27

100.79

Rb Sr Nb Zr Y V Co Cr Ni Cu Zn

<2 167 <3 5 5 126 43 273 78 12 32

<2 32 <3 8 9 134 25 161 74 53 29

<2 25 <3 4 3 77 47 350 145 15 25

ONH54 Gabbro 47.72 0.17 19.14 1.36 3.49 0.06 9.48 15.64 0.5 0.02 0.1 1.89 1.68 0.01 0.21 99.58 <2 217 <3 4 6 130 33 1362 110 20 18

OND14/1A Gabbro 45.94 0.11 18.97 2.07 4.79 0.10 10.42 15.06 0.54 < 0.01 0.04 1.67 1.00 0.12 0.67 99.83 <2 149 <3 4 5 91 50 503 127 <3 34

OND14/1B Gabbro 46.58 0.15 18.24 1.82 4.74 0.09 10.17 15.35 0.54 < 0.01 0.01 1.26 1.18 0.16 0.08 99.11 <2 157 <3 5 6 123 44 655 110 21 43

OND14/3 Gabbro 47.87 0.19 16.82 2.05 3.8 0.08 11.67 16.66 0.40 0.01 0.02 0.26 0.17 0.01 0.17 99.84 <2 159 <3 6 6 133 44 295 108 108 27

OND44/1A Gabbro 47.92 0.19 17.63 2.16 2.25 0.06 10.65 16.97 0.49 0.01 0.01 0.83 0.71 0.07 0.12 99.24 <2 160 <3 6 6 143 35 628 111 52 18

OND57/1 Gabbro 44.95 0.21 16.63 3.88 5.59 0.13 12.8 14.66 0.58 0.1 0.01 1.11 0.72 0.11 0.39 100.76 <2 84 <3 5 7 188 57 199 9 108 27

OND57/1B Gabbro 45.55 0.06 27.92 0.82 2.40 0.04 4.62 15.87 1.35 < 0.01 0.02 1.28 1.21 0.06 0.07 99.99

OND57/5 Gabbro 47.64 0.28 13.81 4.00 4.70 0.14 12.09 17.74 0.35 0.04 < 0.01 0.68 0.68 0.09 n.d. 101.56

<2 <2 204 87 <3 <3 3 6 2 8 54 253 19 48 57 390 26 73 14 5 43 39 (continued on next page)

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

SiO2 TiO2 Al2 O3 FeO Fe2 O3 MnO MgO CaO Na2 O K2 O P 2 O5 L.O.I. H2 O+ H2 O − CO2

ONH17 Gabbro 46.91 0.13 19.88 1.86 4.75 0.08 9.28 16.01 0.61 0.03 0.04 0.58 0.23 0.12 0.35

188

Table 1. Major and trace element analyses of representative rocks from the Wadi Onib ophiolite, northern Red Sea Hills, Sudan (after Hussein, 2000)

SiO2 TiO2 Al2 O3 FeO Fe2 O3 MnO MgO CaO Na2 O K2 O P 2 O5 L.O.I. H2 O+ H2 O − CO2

OND22/2 Dyke 54.10 1.85 15.41 4.23 5.03 0.09 5.25 7.82 2.87 1.64 0.67 1.07 0.54 0.07 0.53

Total

100.1

Rb Sr Nb Zr Y V Co Cr Ni Cu Zn

31 995 10 162 19 188 33 137 50 21 120

ONH31 Dyke 54.23 1.89 15.74 4.64 3.62 0.09 4.65 7.62 2.93 1.46 0.71 2.02 1.83 0.09 0.19

ONH33 Dyke 56.40 0.74 15.2 4.50 1.52 0.07 6.81 6.31 4.42 0.75 0.19 2.13 2.04 0.07 0.09

99.69

99.11

22 913 12 166 21 227 34 143 50 21 107

9 438 5 76 12 147 31 541 184 29 80

ONH53 Dyke 53.75 1.83 14.99 4.47 4.37 0.09 5.55 7.67 2.68 1.69 0.62 1.83 1.72 0.08 0.11 99.62 33 891 13 170 21 202 34 196 49 42 77

SDD53/4 Dyke 52.47 1.78 16.00 5.33 3.16 0.09 4.83 4.57 4.28 1.00 0.63 5.09 4.91 0.11 0.18 99.34 12 618 10 223 30 168 37 38 68 23 93

SDD111 Dyke 47.99 0.52 19.81 4.10 4.11 0.12 6.83 12.64 0.65 < 0.01 0.11 3.06 2.88 0.11 0.18 100.05 <2 402 <3 2 6 250 37 143 41 141 70

SDD112 Dyke 58.06 0.55 16.72 2.57 3.75 0.09 3.43 7.79 2.48 0.61 0.12 2.73 2.73 0.04 n.d.

SD188 Dyke 53.71 0.47 15.22 4.17 4.78 0.13 7.56 6.67 3.82 0.20 0.11 2.51 2.45 0.05 0.06

SD189 Dyke 55.54 0.62 14.04 4.13 3.10 0.10 9.17 7.12 2.86 0.38 0.12 1.45 1.38 0.09 0.07

98.94

99.4

98.72

10 274 3 86 18 184 21 184 16 3 61

3 106 <3 27 14 185 41 371 73 56 64

ONH42/3 Dyke 54.15 1.84 15.5 6.32 2.68 0.09 4.93 8.61 3.32 1.11 0.67 1.28 1.2 0.17 0.08

8. Geochemistry of the Wadi Onib Ophiolite

Table 1. (Continued)

100.67

189

4 14 372 974 3 11 83 162 19 20 175 217 34 33 633 147 30 51 <3 22 67 115 (continued on next page)

190

Table 1. (Continued)

Total Rb Sr Nb Zr Y V Co Cr Ni Cu Zn

97.12 3 179 17 107 29 264 51 548 247 29 83

101.17 15 406 23 175 32 290 56 328 160 85 125

98.87 3 319 4 82 23 225 38 175 68 108 60

98.42 2 98 <3 22 17 217 61 108 396 58 56

99.55 <2 106 15 100 23 251 50 524 248 61 80

99.55 4 186 17 173 31 219 57 265 145 48 175

100.49 < 1.0 163 <3 38 2 < 11.0 < 4.0 < 3.0 < 2.0 < 2.0 < 2.0

ONH42 OND58/2 ONH42/4 SD121 Dunite Dunite Pyroxenite Pyroxenite 40.43 42.55 50.46 51.64 0.01 0.02 0.02 0.03 0.12 0.40 1.26 3.79 3.19 1.39 1.12 2.74 7.85 6.15 2.29 2.63 0.11 0.07 0.07 0.09 34.88 37.64 23.01 21.05 0.26 0.02 17.62 17.05 < 0.01 < 0.01 0.85 < 0.01 < 0.01 0.01 < 0.01 < 0.01 0.01 0.01 0.02 < 0.01 12.18 11.18 3.27 1.06 12.01 11.12 3.11 1.06 0.04 0.01 0.70 0.10 0.17 0.06 0.16 n.d. 99.08 <2 <2 <3 4 <2 22 110 3829 1255 30 48

99.45 <2 <2 3 3 <2 30 97 2287 2049 25 30

100.69 <2 16 3 3 3 113 59 5238 713 53 17

100.18 <2 6 <3 4 3 157 56 5121 446 52 34

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

SiO2 TiO2 Al2 O3 FeO Fe2 O3 MnO MgO CaO Na2 O K2 O P 2 O5 L.O.I. H2 O+ H2 O − CO2

ONH43/3 ONH47 ONH69/2 OND87/1 OND62/12 SD208 SD241 Pillow lava Pillow lava Pillow lava Pillow lava Pillow lava Pillow lava Plagiogranite 46.99 48.85 48.4 49.41 51.86 46.54 77.41 1.58 2.52 1.26 0.44 1.52 2.74 0.03 15.43 14.02 19.19 13.33 13.89 13.95 12.55 6.39 10.02 5.85 3.34 4.63 8.66 0.08 4.34 3.13 2.97 5.89 5.51 6.11 0.23 0.14 0.27 0.12 0.13 0.12 0.15 0.01 9.56 7.57 6.88 10.10 8.63 4.48 1.16 10.00 9.83 9.26 13.26 8.21 5.84 2.74 0.71 1.30 1.69 0.55 2.16 3.02 4.73 0.17 0.88 0.58 0.08 0.1 0.2 0.05 0.17 0.28 0.15 0.02 0.17 0.29 0.02 1.61 2.48 2.47 1.80 2.74 7.38 1.46 1.48 2.07 2.35 1.60 2.56 3.88 0.91 0.03 0.02 0.05 0.07 0.01 0.19 0.02 0.13 0.41 0.12 0.20 0.18 3.5 0.55

n SiO2 TiO2 Al2 O3 FeO Fe2 O3 MnO MgO CaO Na2 O K2 O Cr2 O3 NiO

SD148 ACR 5 0.15 0.37 15.59 22.8 7.48 0.5 7.29 0.03 0.04 0.01 45.06 0.12

SD180A MCR 8 0.02 0.44 17.17 18.63 10.39 0.39 10.29 0.01

Total

99.44

99.59

Si Ti Al Fe2+ Fe3+ Mn Mg Ca Na K Cr Ni

0.04 0.07 4.87 5.06 1.49 0.11 2.88 0.01 0.02 0.02 9.44 0.03

Cations Fm

0.04 42.15 0.06

ONH65A MCR 3 0.02 0.26 8.49 12.27 4.13 0.48 13.97

62.08 0.1 101.8

ONH65B MCR 9 0.01 0.17 14.81 11.65 2.53 0.45 14.98

SD180 SCR 6

SD60/1 SCR 5

64.25 14.1 3.05 0.01 18.27

63.32 13.4 3.14 0.11 18.48

56.97 0.12

0.23 0.06

0.76 0.07

101.69

99.97

6A MCR 9 0.1 0.06 10.01 18.6 2.43 0.6 9.37 0.01

SD148 Cpx 16 55.24 0.08 1.02 2.05

ONH42/4 Cpx 8 48.23 0.03 0.63 11.92

SD163 Cpx 5 56.19 0.02 0.48 1.67

SD133 Cpx 11 52.79 0.12 5.30 3.99

SD148 Opx 3 53.98 0.01 0.89 7.12

SD133 Opx 8 55.33 0.04 5.01 9.29

SD148 Ol 10 41.59 0.02 0.02 11.24

SD163 Ol 5 41.87 0.01

0.06 16.33 20.98 0.04 n.d. 0.15 0.04

0.11 17.77 24.59 0.07 0.07 0.25 0.04

0.12 16.80 21.66 0.01 0.01 0.50 0.02

0.18 36.05 0.43 0.01 0.02 0.25 0.03

0.19 30.40 0.64 0.02 0.01 0.41 0.03

0.18 46.78 0.02

0.15 47.75 0.01

0.01 58.29 0.04

0.12 17.30 24.00 0.20 0.01 0.56 0.03

0.01 0.02

0.03 0.16

99.28

99.52

100.61

98.41

101.26

101.32

98.97

101.37

99.88

1.99 0.01 0.04 0.06

1.75

2.00

1.89

1.92

1.91

0.03 0.75

0.02 0.05

0.22 0.12

0.04 0.18

0.20 0.26

0.89 0.81

0.94 0.94

0.89 0.83 0.01

0.01 1.88 0.01

0.01 1.56 0.02

0.01

0.01

0.01

0.01

0.05 2.56 2.62 0.8 0.1 5.32

0.03 4.32 2.41 0.47 0.09 5.52

15.48 2.41 0.47 0.02 5.6

15.35 2.31 0.51 0.02 5.67

0.03 0.01 3.16 4.15 0.49 0.12 3.75

8.61 0.02

12.54 0.02

11.14 0.03

0.04 0.01

0.12 0.02

12.29 0.01

0.01

24.04

24.10

24.01

24.01

24.03

24.00

24.01

3.97

4.23

3.96

3.97

4.05

3.97

0.64

0.58

0.34

0.31

0.4

0.29

0.54

0.06 0.48 0.48 0.03

0.15 0.41 0.44 0.14

0.05 0.48 0.49 0.03

0.12 0.45 0.48 0.06

0.09

0.15 0.01 0.84 0.15

0.1 5.2 4.03 2.01 0.13 3.98

11.12

0.93 0.92 0.01

0.02

8. Geochemistry of the Wadi Onib Ophiolite

Table 2. Electron microprobe analyses of selected minerals from the Onib ophiolite and adjacent ophiolitic complexes (after Hussein, 2000)

101.1

1.02

1.02

0.23

0.23

1.72

1.76

0.01

Wo En Fs

0.90 0.09

Fo Fa

2.98

3.01

0.12 0.87 0.12

0.12 0.88 0.12

191

ONH 65 A, B Hamisana area W of Onib, 6A Wadi Amur area, SD Sudi, ACR accessory chromite, MCR massive chromite, SCR speckled chromite.

192

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Table 3. Trace element analyses of selected rocks from the Wadi Onib ophiolite by spark source mass spectrometry (from Zimmer, 1989)

K Rb Sr Y Zr Nb Sn Sb Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Yb Lu Hf Pb Bi Th U

ONH 69/2 Pillow lava 4600 5.8 330 26 76 1.9 1.2 0.74 0.11 94 4.1 11 1.7 9.4 2.4 0.95 3.2 0.58 2.8 0.71 2.3 1.9 0.35 1.9 0.87 − 0.15 0.055

SD 189 Sheeted dyke 3651 3.8 369 16 85 1.1 0.96 0.13 0.95 161 4.9 14 1.67 7.6 1.8 0.57 2.7 0.48 2.4 0.54 1.7 1.1 0.15 2 1.64 0.023 0.77 0.43

OND 14/3 Gabbro 310 0.43 180 7.4 2.7 1.1 0.59 0.67 − 9.9 0.43 0.63 0.12 1.2 0.66 0.32 0.61 0.13 1.3 0.16 0.46 0.53 0.059 0.22 0.28 − 1 −

SD 121 Ultramafic 34 0.098 7.1 1.6 − − − − 0.047 1.6 0.052 0.18 0.022 0.17 0.087 0.042 0.16 0.04 0.31 0.078 0.23 0.19 0.034 − 0.07 − − −

8.1. Synopsis of the Ultramafic and Gabbroic Rocks Major element abundances of selected, serpentinized dunites and pyroxenites from the Onib ophiolite are presented in Table 1. The dunites are moderate in silica (40–42%), high in MgO (34–37%), very low in TiO2 (0.01–0.02%), lacking Na2 O and K2 O, and are poor in P2 O5 . The rocks are also rather high in Cr (2287–3829 ppm) and Ni (1255–2049 ppm). The pyroxenites, on the other hand, are moderate in silica (50–51%), very low in TiO2 (0.02–0.03%), high in CaO (17%), almost depleted in both alkalies and P2 O5 and are high in Cr (5121–5238 ppm). Furthermore, Hussein (2000) reported that subtypes of TZ pyroxenites are high in Al2 O3 (8–24%). They are also very low in TiO2 , alkalies and P2 O5 relative to their Cr-contents which may reach values as high as 6392 ppm in places. These

8. Geochemistry of the Wadi Onib Ophiolite

193

Fig. 7. Jensen triangular diagram (Jensen, 1976) of Al-Mg-(Fe + Ti) showing compositional range of analysed plutonic and extrusive rocks from the Onib ophiolite. Plot includes post-ophiolite granitoids for comparison (modified after Hussein, 2000).

compositional features strongly suggest that the above rocks are derived from a mantle source, and that they are predominantly cumulates (e.g., Fig. 9A). The gabbroic suite shows a wide range of compositional variation (see Table 1). Its pronounced heterogeneity inhibits significant utilization of its major and trace element chemistry for petrogenetic considerations. From the overall major element abundances (Hussein, 2000) the Onib gabbros (cumulates in the main part) appear to have gained Na and lost Ca during ocean-floor hydrothermal alteration. They also show a narrow range (0.3 to 0.4) for the mafic index (FeO∗ /FeO∗ + MgO) and rather low TiO2 contents. A plagiogranite (SD 241) associate of the gabbros is notably high in SiO2 (77%), moderate in Al2 O3 (12.55%), extremely low in K2 O (0.05%), TiO2 (0.03%) and Sr (163 ppm). Generally, however, the Onib gabbroic rocks are comparable to gabbroic suites from classic ophiolitic complexes as described by Coleman (1977) and Laurent et al. (1980). 8.2. Lavas and Dykes Basaltic pillow lavas of uppermost Onib gorge, the sheeted dyke complex of lower Wadi Sudi and numerous basic dykes form the uppermost unit of the ophiolitic sequence. Detailed petrographic and geochemical studies (Hussein, 2000) reveal that the lavas and dykes

194

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Fig. 8. (A) Trace element abundance patterns for selected Wadi Onib dykes and pillow lavas. Data normalized to N-type MORB (Pearce, 1980). Normalizing factors (ppm or percentages where indicated): Sr = 120; K2 O = 0.15%; Rb = 2.0; Ba = 20; Th = 2; Ta = 0.18; Nb = 3.5; Ce = 10; P2 O5 = 0.12%; Zr = 90; Hf 2.4; Sm 3.3; TiO2 = 1.5%; Y = 30; Sc = 40; and Cr = 250 (modified after Hussein, 2000). (B) Chondrite-normalized REE patterns of a suite of rocks from the Wadi Onib ophiolite (modified after Zimmer, 1989).

display distinct mineralogical, textural and chemical variations as well as variable sea-floor alteration and subsequent low-grade regional metamorphic overprinting. Whole-rock major and trace element abundances (Tables 1 and 3) and pertinent correlations with emphasis on immobile trace (e.g., Zr, Y, Nb), and major (e.g., TiO2 , P2 O5 ) elements make it possi-

8. Geochemistry of the Wadi Onib Ophiolite

195

ble to draw inferences on classification, primary igneous processes (parental magma and differentiation trends) as well as possible eruptive environment(s) and tectonic setting. The geochemical criteria of the extrusive component are summarized below (see also Figs. 7, 8, 10 and 11). The Onib volcanic rocks show considerable variations with respect to SiO2 -contents (46 to 54 wt%) but are in good agreement with average ophiolitic lavas with reference to Al2 O3 (14–16%), FeO (total iron as FeO∗ is 6–8%), MnO (ca. 0.9%), MgO (ca. 6%) and CaO (ca. 9%) (see classification in Barker, 1983). In contrast, Na2 O- and K2 O-values vary notably, a feature that is probably related to post-igneous metasomatism. The Onib lavas and dykes are generally rich in TiO2 (> 1%), although a number of dykes have TiO2 -contents of < 1%. The high-Ti extrusive subgroup is also elevated in P2 O5 (ca. 0.7%) in contrast to the Ti-poor varieties whose P2 O5 values are in the range of 0.1 to 0.2%. These compositional features are equally mimicked by the incompatible trace element Zr, known to be resistant to alteration (Alabaster, 1982, and references cited therein). It varies systematically and significantly with igneous processes such as partial melting and fractionation. An increase in TiO2 and Zr within the Onib volcanic suite suggests that some fractionation has taken place, most probably within an open system magma chamber. The Onib classifies as a high-Ti-ophiolite and compares well with other complexes (e.g., Troodos, Cyprus; Semail, Oman; Jabal Al Wask, Saudi Arabia; Wadi Ghadir, Egypt; Bou Azzer, Morocco (see also Zimmer et al., 1995, and references cited therein as well as review in Stern et al., 2004). The Onib lavas and dykes are generally hypersthene- and quartz-normative and tholeiitic, although some of the high-Ti and most of the low-Ti dykes show a tendency towards the calc-alkaline series (Fig. 7). Trace element geochemistry also suggests that the Onib extrusive component has a transitional IAT/MORB character (Fig. 11) with a tendency towards within-plate volcanism. This phenomenon is most probably due to enrichment of some assemblages in TiO2 relative to increasing Zr contents. In essence, a positive interelement correlation displayed by the Onib lavas and dykes in some covariation diagrams (Hussein, 2000) may indicate their cogenetic character. The transitional island arc/MORB character of the Onib lavas and dykes has implications for the possible eruptive environment of the ophiolitic sequence. Although it is well-known that ophiolites are generated in different oceanic environments such as midoceanic ridges, small (back-arc or fore-arc) ocean basins, leaky transform faults and, perhaps, immature island arcs (Shervais, 2001; Serri, 1981), difficulties exist in distinguishing geochemical signatures of these settings. This is because the evolution of such ophiolite complexes involves considerable geochemical and petrological complexities such as mantle heterogeneities, magma mixing, contamination and fractionation processes. In order to identify the possible magma type from which the Wadi Onib ophiolite was generated, a number of covariation plots (Hussein, 2000) employing trace and/or REE concentrations were considered (e.g., see Tables 1 and 3 as well as Figs. 8A and B). In the latter plot it is evident that the Onib lavas and dykes are enriched in elements of low ionization potential relative to MORB. Such a feature is thought to be caused by selective enrichment of the magma sources due to aqueous fluids released from a downgoing slab

196

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

during subduction (Pearce et al., 1981). The enrichment of the Onib extrusive suite in, for example, Nb, Ce, Pa and Zr makes it comparable to counterparts in marginal basins (e.g., Scotia Sea) and, more specifically, to the “Geotimes” lavas of the Semail ophiolite, Oman, described by Alabaster (1982) as well as to dykes and lavas from the Wadi Ghadir ophiolite, SE Egypt (Kröner et al., 1987). Using major element data Kröner (1985) suggested that the Ghadir and Onib resemble rocks from back-arc basins. In broad terms, magmas from which the Onib evolved are comparable to those giving rise to marginal (back-arc) basalts and/or to anomalous MORB. 8.3. Mineral Chemistry Data on mineral chemistry enhance our understanding of the chemical composition and structure of melt(s) partitioning conditions in mineral lattices and cooling histories of magmas (Nisbet and Pearce, 1977). Bearing this in mind and in order to corroborate and/or constrain the major and trace element data, microprobe work was carried out on primary and secondary mineral phases of the Onib ophiolite complex (Hussein, 2000). Here, we summarize these data by presenting and discussing a selected batch of analyzed pyroxene, olivine and chromian spinel (Table 2). The Onib clinopyroxenes (mainly from wehrlitic olivine-bearing pyroxenites of the exceptionally thick transitional zone of interlayered cumulates) are characterized by a high CaO-content (ca. 24%) and SiO2 ranging from 48 to 54%. They are very low in alkalies (K2 O = 0.00–0.02%, Na2 O = 0.01–0.9%) and have average NiO- and Cr2 O3 -contents of around 0.05%. TiO2 is about 0.1%, and Al2 O3 varies from 1 to 5%. The Mg/(Mg + Fe) ratio is about 0.85, and their composition (En45–49 Wo41–50 Fs6–14 ) is predominantly diopsidic augite. The Onib orthopyroxenes co-existing with clinopyroxene have MgO-values in the range of 30–36%, ∼ 54% SiO2 , 0.02% TiO2 and 1–5% Al2 O3 . Their alkalies are similarly low as in the clinopyroxenes, and their NiO and Cr2 O3 are moderate (0.03% and 0.3–0.4%,

Fig. 9. (A) Cr2 O3 versus NiO diagram for Wadi Onib ophiolitic ultramafic rocks in comparison to counterparts from Jebel Sol Hamed (NE) and Ingessana Hills (far in SE Sudan). Open circles represent Onib ultramafic rocks, filled circles stand for Sol Hamed serpentinites. Square, stars and triangles are for Ingessana harzburgites, cumulate dunites and mantle dunites respectively. Sol Hamed and Ingessana data are from Price (1984). Broken line separating fields of mantle (A) and cumulate (B) rocks from the Bay of Islands ophiolite is after (Malpas, 1978). (B) Cr × 100/(Cr + Al) versus Mg × 100/(Mg + Fe2 ) diagram showing position of chromites from the Onib ophiolite and related occurrences with respect to compositional fields for chromian spinels in alpine-type peridotites, abyssal peridotites, layered intrusions and S.E. Alaskan intrusives (fields from Dick and Bullen, 1984). Other chromite data from various ophiolites are shown for comparison. A, B and C correspond to Sol Hamed and Ingessana (Sudan) and Semail (Oman) ophiolites respectively (after Price, 1984). D (field enclosed by broken line) is for some 1000 chromitiferous ore samples from Sakhkot-Qila ophiolite, NW Pakistan (after Ahmed, 1984).

8. Geochemistry of the Wadi Onib Ophiolite

197

198

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Fig. 10. TiO2 versus mafic index diagram for classifying Wadi Onib rocks. Mafic index is F/(F + M) which is FeO/(FeO + MgO) where FeO is total iron as FeO. Dotted line separates high- and low-Ti ophiolites (after Serri, 1981). Inset shows fields of selected ophiolitic complexes and is indicated for comparison (after Zimmer et al., 1995).

respectively). They classify as bronzite and/or clinoenstatite (En84–90 Wo1 Fs9–15 ). Their Mg/(Mg + Fe) ratio is about 0.85, similar to that of the clinopyroxenes. Broadly speaking, the Onib clinopyroxenes are fresh and characteristically calcic. They belong to a tholeiitic (subalkaline) magma series, as also indicated by their whole-rock chemistry. Moreover, variations in Al2 O3 -contents may be related to compositional differences in the original melt(s) from which these minerals crystallized. The high-Al2O3 clinopyroxenes, in particular, may be depth-related since they formed at high pressure conditions. By and large, the Onib pyroxenes compare well with those from typical ophiolitic cumulates as described by Coleman (1977) as well as with pyroxenes from adjacent ophiolitic complexes such as Ingessana, NE Sudan (Price, 1984), Thurwah, Saudi Arabia (Nassef, 1982, cited in Price, 1984) and Semail, Oman (Alabaster, 1982). Olivine is more sensitive to alteration (serpentinization) than is clinopyroxene. Therefore, recrystallized olivine as well as suspected adcumulus and/or clearly serpentinized phases have not been considered in the average compositions presented in Table 2. The

8. Geochemistry of the Wadi Onib Ophiolite

199

Fig. 11. Zr versus TiO2 covariation diagram showing tectonic setting of Wadi Onib ophiolite. Other ophiolites such as Gerf to the north of Onib (Zimmer, 1989) as well as Semail, Oman, and Troodos, Cyprus (Pearce, 1980) are indicated for comparison. Fields of various eruptive environments (arc lavas, MORB and within-plate lavas) are from Pearce (1980). Lower Oman (Geotimes unit) and lower Troodos are ophiolitic sections used in delineating the respective fields in the diagram (modified after Hussein, 2000).

Onib olivines are forsteritic in composition (Fo88 ), have SiO2 of about 41%, MgO of 47%, contain negligible amounts of Al2 O3 , TiO2 , CaO and Cr2 O3 , and contain practically no alkalies. NiO-contents vary from 0.02 to 0.2%, a relatively high value that reflects the magnesian nature of olivine. Mg/(Mg + Fe), MgO/FeO and FeO + MgO values of olivine are in the range of 0.81, 4 and 58, respectively. These are compatible with those from ultramafic rocks of cumulate ophiolite sequences in general (Coleman, 1977). The remarkable homogeneity of olivine suggests that relatively stable conditions within the host magma chamber must have prevailed for at least enough time for olivine to crystallize from a fairly magnesian-rich liquid, before some replenishments of the magma chamber occurred. Chromian spinel is either high in Cr or high in Al (i.e., the Onib chromite is bimodal, see Table 2 and Fig. 9B). The high-Cr spinel is in the range of 42–45% Cr2 O3 , 18–22% FeO, 15–17% Al2 O3 , 7–10% MgO, negligible SiO2 , CaO, Na2 O, K2 O, TiO2 , and 0.02–0.2% NiO. The Cr# (Cr/(Cr + Al)) varies between 62 and 65, and the Mg# (Mg/(Mg + Fe)) is between 36 and 50. The relatively high FeO-content (∼ 22%) is explained in terms

200

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

of ferritchromite and/or martitization effects. Ferritchromite development is a common phenomenon within chromites of ophiolites. Ahmed (1984) suggested from the SakhakotGila ophiolitic complex of NW Pakistan that ferritchromite development begins in the spreading centre environment and continues through serpentinization of the chromite host rocks. Conversely, high-Al chromian spinel shows a remarkable depletion in Cr2 O3 (ca. 0.5%) and a pronounced enrichment in Al2 O3 (ca. 64%). It contains no SiO2 , TiO2 , CaO, Na2 O and K2 O but relatively high FeO (ca. 14%) and MgO (ca. 18%). Its NiO-content is indistinguishable from that of the Cr-rich counterpart. The high-Al spinel shows a Mg# of approximately 70 and has a Cr# of 30–80. Chromian spinels represent sensitive petrogenetic indicators (Dick and Bullen, 1984). Nicolas and Al Azri (1991) suggested that the source of chromite during melting is diopside and that within ophiolitic transition zones, below the layered crustal gabbros, a temperature drop and oxygen fugacity increase (oxygen fugacity could possibly be related to a minor contamination by seawater) would allow for appearance of chromite on the basaltic liquidus. Edwards et al. (2000) advocated that most major podiform chromitite bodies are Cr-rich and occur in the harzburgitic mantle section and the mantle-crust transition zone of ophiolites formed in supra-subduction zone environments. The latter authors also conclude that the formation of most, if not all, melts capable of forming podiform chromitites requires the involvement of water in the zone of melt generation. Water not only promotes partial melting of refractory peridotites, it also dissolves in the resulting melt, and this modification, in turn, promotes greater solubility of Cr. The presence of water will greatly increase diffusion rates, enhancing the dissolution of unstable elements and growth of stable ones. Consequently, a migrating melt will become progressively more polymerized as it moves to lower pressure, cools, and Si enters the melt because of the conversion of wall-rock pyroxene to olivine. As a result, Cr will become insoluble, and nucleation of chromian spinel will take place. It has also been assumed that high alumina chromites are more likely to develop under higher pressures than high chromium chromites (e.g., Irvine, 1967, cited on p. 133 of Coleman, 1977; Dickey, 1975). Similar conditions may have also been maintained during magmatic segregation of the Onib bimodal chromitites both of which indicate a cumulate rather than stratiform affiliation. According to Dick and Bullen (1984) who classified Alpine-type peridotites and associated volcanics into three empirical subdivisions (1–3), the Onib podiform chromites (Fig. 9B) are comparable to type 2 with a Cr# of < 60 and > 60. These authors advocate that type 2 Alpine-type peridotites and associated volcanics formed in mid-ocean ridges, inter-arc rifts as well as back-arc settings.

9. DISCUSSION AND GEODYNAMIC CONSIDERATIONS Following the preceding discussion and available data it is clear that the Onib is a full-scale ophiolite (Hussein et al., 1984; Kröner et al., 1987; Abdel-Rahman, 1993; Reischmann, 2000; Kusky and Ramadan, 2002; Kröner et al., 1992). Furthermore, it occurs within and forms the major part of the Onib-Sol Hamed ophiolite-decorated suture that

10. Summary and Conclusions

201

delineates a major structural contact between voluminous, arc-related volcano-plutonic terranes of Gebeit (E) and Gabgaba (W) (Kröner et al., 1987). In an attempt to classify the ophiolitic complexes of the Nubian shield Reischmann (2000) advocated the Onib to be of Cordilleran (steeply dipping) rather than of low-angle-thrust type, an assertion corroborated by structural criteria. Single zircon Pb-Pb dating on plagiogranite (Kröner et al., 1992) indicated that the Onib ophiolite was generated about 808 Ma ago. The ophiolite retains a wide range of stratigraphic, structural, compositional and depositional features that attest to its evolution within an oceanic environment, particularly within a marginal (back-arc) setting. The ophiolite complex is surrounded by arc-derived lavas and sediments characterized by an abundance of andesitic components. Locally, the arc-related volcanogenic sediments contain intraformational conglomerates with ophiolitic clasts. These depositional features suggest that the ophiolite was probably folded, exposed, became a source for conglomeratic clasts and submerged again. Thus, it was involved in a complex evolution of intra-oceanic island arc systems. Deeper levels of the exceptionally thick (ca. 2.5 km) transitional zone (TZ) of interlayered cumulates contain podiform, bimodal chrome spinel (massive high-Cr and disseminated high-Al varieties). There is also an abundance of olivine-bearing pyroxenites relative to olivine-bearing gabbros (troctolites) within the TZ cumulates. Olivine could have been extracted earlier in excess and, consequently, beheaded pyroxene and plagioclase which predominate upwards in the magmatic sequence. These features signify a complex mineral segregation history that took place within cooling magma reservoir(s) from which the ophiolitic rocks were generated. In addition, the Onib lavas and dykes are predominantly Ti-rich, have a transitional IAT-MORB character and are indistinguishable from anomalous MORB and/or marginal basin basalts (see also Figs. 7–11). The overall field, stratigraphic, structural and compositional criteria suggest that the ophiolitic complex is comparable to supra-subduction zone (SSZ) ophiolites described by Pearce et al. (1984b) and, more recently, by Shervais (2001). The evolution of the Onib ophiolite was closely related to an intra-oceanic island arc system that developed above a westerly-dipping subduction zone. In essence, Shackleton (1994) concluded that ophiolite obduction in the Nubian shield at large was generally northwestwards and that subduction is inferred to have been in the same direction (see, also Hussein, 2000 and references cited therein). Thus, our geodynamic model entails island arc formation, growth and intra-arc rifting, leading to inception and widening of an intra-arc marginal basin which eventually closed. Part of its oceanic crust was obducted onto the surrounding remnant and/or evolving island arc systems between 808 Ma and 750–720 Ma ago.

10. SUMMARY AND CONCLUSIONS The Wadi Onib mafic-ultramafic complex is an exceptionally well-preserved though tectonically fragmented ophiolitic sequence in the northern Red Sea Hills (NRSH) of the NE Sudan. It is characterized by a number of internal stratigraphic, structural and compositional features some of which are: (A) an exceptionally thick (ca. 2.5 km) transitional zone

202

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

of interlayered and laminated cumulates showing gravity stratification; (B) an abundance of olivine-bearing pyroxenites and wherlitic (?) transitions relative to troctolites; (C) podiform bimodal chromite (high Cr/high-Al); (D) isoclinal folding of magmatic layering (intrafolial folding) suggesting high-temperature ductile deformation; (E) sea water/rock interaction (hydrothermal) alteration. The Onib lavas and dykes are predominantly Ti-rich, have a transitional IAT-MORB character and are indistinguishable from anomalous MORB and/or marginal basin basalts. The Onib ophiolite was subjected to successive metamorphic episodes which are difficult to distinguish. Vertical zonation of metamorphic grades, albitization and/or saussuritization of plagioclase, uralitization of pyroxene as well as preservation of igneous textures in spite of replacement of primary minerals indicate that sea-floor hydrothermal alteration must have taken place prior to basin closure and ophiolite obduction. Exceptionally thick transition zones have also been reported from other ophiolitic sequences (e.g., see Kröner et al., 1987; Hussein, 2000), and such transitional zones of interlayered cumulates and associated podiform chromite seem to be characteristic of marginal basin ophiolites (Hawkin and Evans, 1983). All these data imply a back-arc (marginal) basin setting in a supra-subduction zone (SSZ) environment. The ophiolite sequence was generated some 808 Ma ago. Partial melting conditions were most probably of multi-stage, and the magma chamber was of an open system type. Thus, the evolution of the ophiolite sequence attests to the fact that modern-type plate tectonic regimes can be extended back into the early Neoproterozoic. The ophiolite forms the major part of the Onib-Sol Hamed ophiolite-decorated suture the northeastern extension of which concurs with the Jabal Al Wask ophiolite complex in Saudi Arabia.

ACKNOWLEDGEMENTS We gratefully acknowledge funding provided by the Volkswagen-Stiftung and the German Science Foundation (DFG). In addition, we thank the Geological Research Authority of the Sudan (GRAS) for logistic support during field work in the Red Sea Hills. Stefan Dürr and Ralph Koennecke contributed to the success of the field work during initial stages of the project. Martina Zimmer kindly made available some unpublished data. Gisa Prescott is thanked for typing part of the text and Petra Koppenhöfer for drafting the figures. Furthermore, our cordial thanks are extended to T.M. Kusky and R.J. Stern for reviewing the manuscript and making constructive suggestions.

REFERENCES Abdel-Rahman, E.M., 1993. Geochemical and geotectonic controls of the metallogenic evolution of selected ophiolite complexes from the Sudan. Berliner Geowissenschaftliche Abhandlungen, Reihe A (145), Fachbereich Geowissenschaften, Freie Universität Berlin, p. 175. Ahmed, Z., 1984. Stratigraphic and textural variations in the chromite composition of the ophiolitic Sakhakot-Qila Complex, Pakistan. Economic Geology 79, 1334–1359.

References

203

Alabaster, T., 1982. The interrelationship between volcanic and hydrothermal processes in the Oman ophiolite. Unpublished Ph.D. thesis. Open University, Milton Keynes, UK, p. 408. Anonymous, 1972. Penrose Field Conference. Geotimes 17, 24–25. Bagnall, P.S., The geology of the Deraheib area (sheet 35 L). Unpublished Report. Geological Survey Department, Khartoum, Sudan. Barker, D.S., 1983. Igneous Rocks. Prentice–Hall, Englewood Cliffs, NJ. Bodinier, J.L., Dupuy, C., Dostal, J., 1984. Geochemistry of Precambrian ophiolites from Bou Azzer, Morocco. Contributions to Mineralogy and Petrology 87, 43–50. Buisson, G., Leblanc, M., 1986. Gold-bearing listwaenites (carbonatized ultramafic rocks) from ophiolite complexes. In: Gallagher, M.J., Ixer, R.A., Neary, C.R., Prichard, H.M. (Eds.), Metallogeny of Basic and Ultrabasic Rocks. Institute for Minerals and Metallurgy, London, pp. 121–131. Burgath, K., Weiser, T., 1980. Primary features and genesis of Greek podiform chromite deposits. In: Panyiotou, A. (Ed.), Ophiolites (Proceedings of the International Ophiolite Symposium, Cyprus, 1979). Cyprus Geological Survey Department, Nicosia, pp. 675–690. Coleman, R.G., 1977. Ophiolites: Ancient Oceanic Lithosphere. Springer-Verlag, New York, Berlin, Heidelberg, p. 229. Coleman, R.G., Peterman, Z.E., 1975. Oceanic Plagiogranite. Journal of Geophysical Research 80, 1099–1108. Davis, G.H., 1984. Structural Geology of Rocks and Regions. John Wiley & Sons, New York, p. 492. de Bretizel, P., 1980. Results of three-year exploration programme (1977–1980) (vol. A), Cooperation Programme of Mineral Research in the Sudan, Red Sea Hills. Unpublished Report 80 SUD 001. Bureau de Recherches Géologiques et Minière (Orléans) and Geological and Mineral Resources Department (Khartoum). Deer, W.A., Howie, R.A., Zussman, J., 1966. An Introduction to the Rock-Forming Minerals, 11th ed. Longman, London, p. 528. 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. Dickey, J.S. Jr., 1975. A hypothesis of origin for podiform chromite deposits. Geochimica et Cosmochimica Acta 39, 1061–1074. Edwards, S.J., Pearce, J.P., Freeman, J., 2000. New insights concerning the influence of water during the formation of podiform chromitite. In: Yildirim, D., Moores, E., Elthon, D. (Eds.), Ophiolites and Oceanic Crust: New Insights from Field Studies and the Oceanic Drilling Program. Geological Society of America Special Paper 349, 139–147. El Bayoumi, R.M.A., 1980. Ophiolites and associated rocks of Wadi Ghadir, east of Gebel Zabara, Eastern Desert, Egypt. Unpublished Ph.D. thesis. University of Cairo, Egypt, p. 227. Fitches, W.R., Graham, R.H., Hussein, I.M., Ries, A.C., Shackleton, R.M., Price, R.C., 1983. The late Proterozoic ophiolite of Sol Hamed, NE Sudan. Precambrian Research 19, 385–411. Gabert, G., Ruxton, B.P., Venzlaff, H., 1960. Über untersuchungen im Kristallin der nordlichen Red Sea Hills im Sudan. Geologisches Jahrbuch 77, 241–270. Gass, I.G., 1981. Pan-African (upper Proterozoic) plate tectonics of the Arabian-Nubian Shield. In: Kröner, A. (Ed.), Precambrian Plate Tectonics. Elsevier, Amsterdam, pp. 387–405. Gass, I.G., 1980. The Troodos Massif; its role in the unravelling of the ophiolite problem and its significance in the understanding of constructive plate margin processes. In: Panayiotou, A. (Ed.), Ophiolites (Proceedings of the International Ophiolite Symposium, Cyprus, 1979). Geological Survey of Cyprus, Nicosia, pp. 23–35. Hawkins, J.W., Evans, C.A., 1983. Geology of the Zambales Range, Luzon, Philippine Islands; ophiolite derived from an island arc-back arc basin pair. In: Hayes, D.E. (Ed.), The Tectonic Evolution

204

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

of the Southeast Asian Seas and Islands, Part 2. In: Geophysical Monographs, vol. 27. American Geophysical Union, pp. 95–123. Holtzman, B., 2000. Gauging stress from mantle chromitite pods in the Oman Ophiolite. In: Yildirim, D., Moores, E., Elthon, D., Nicolas, A. (Eds.), Ophiolites and Oceanic Crust: New Insight from Field Studies and the Oceanic Drilling Program. Geological Society of America Special Paper 349, 149–159. Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithospheric mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337. Hussein, I.M., 2000. Geodynamic evolution of the Pan-African crystalline basement in the northern Red Sea Hills, Sudan, with special emphasis on the Wadi Onib ophiolite and the geology to the west of Port Sudan. Unpublished Ph.D. thesis. University of Mainz, Germany, p. 325. Hussein, I.M., Khalifa, M.I., Lataillade, J.C., Stolojan, N., Ayme, Y., Bellivier, F., 1978. Geology and study of chromite occurrences of the sheet Deraheib between 21◦ –22◦ N and 34◦ 30 –36◦ E. Unpublished Report 78 SUD 002. Cooperation Programme of Mineral Research in the Sudan, Red Sea Hills, Bureau de Recherches Geologiques et Miniere (Orléans) and Geological and Mineral Resources Department (Khartoum). Hussein, I.M., 1977. Geology of the Halaib area of the northern Red Sea Hills, Sudan, with special reference to the Sol Hamed basic complex. Unpublished M. thesis. Portsmouth Polytechnic, UK, p. 175. Hussein, I.M., Kröner, A., Dürr, S., 1984. Wadi Onib; a dismembered Pan-African ophiolite in the Red Sea Hills of Sudan. Bulletin of Faculty of Earth Sciences, King Abdul Aziz University, Jiddah 6, 319–327. Jensen, L.S., 1976. Ontario Division of Minerals Miscellaneous Paper 66, p. 22. Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004. Neoproterozoic ophiolites in the Arabian shield: field relations and structure. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162. Kröner, A., 1985. Ophiolites and the evolution of tectonic boundaries in the late Proterozoic ArabianNubian Shield of Northeast Africa and Arabia. Precambrian Research 27, 277–300. Kröner, A., 1984. In: Klerkx, J., Michot, J. (Eds.), Géologie Africaine—African Geology. Musée Royale de l’Afrique Centrale, Tervuren, Belgium, pp. 23–28. Kröner, A., Greiling, R., Reischmann, T., Hussein, I.M., Stern, R.J., Krüger, J., Dürr, S., Zimmer, M., 1987. Pan-African Crustal Evolution in the Nubian Segment of Northeast Africa. In: Geodynamics Series, vol. 17. American Geophysical Union, pp. 235–257. Kröner, A., Linnebacher, P., Stern, R.J., Reischmann, T., Manton, W., Hussein, I.M., 1991. Evolution of Pan-African island arc assemblages in the southern Red Sea Hills, Sudan, and in southwestern Arabia as exemplified by geochemistry and geochronology. Precambrian Research 53, 99–118. Kröner, A., Todt, W., Hussein, I.M., Mansour, M., Rashwan, A.A., 1992. Dating of late Proterozoic ophiolites in Egypt and the Sudan using the single grain zircon evaporation technique. Precambrian Research 59, 15–32. Kusky, T.M., Abdelsalam, M., Stern, R.J., Tucker, R.D., 2003. Evolution of the East African and related orogens, and the assembly of Gondwana. Precambrian Research 123, 81–85. Kusky, T.M., Ramadan, T.M., 2002. Structural controls on Neoproterozoic mineralization in the South Eastern Desert, Egypt; an integrated field, Landsat TM, and SIR-C/X SAR approach. Journal of African Earth Sciences 35, 107–121.

References

205

Krüger, J., 1982. Ein Beitrag zur Geologie, Tektonik und Geochemie der Svecofenniden westlich von Falun, Zentral Schweden, sowie der Eignungstest eines DC-Plasmen Spektrometers zur Gesteinsvollanalyse. Unpublished Diploma thesis. University of Mainz, Germany, p. 125. Laskowski, N., Kröner, A., 1984. Geochemical characteristics of Archaean and late Proterozoic to Palaeozoic fine-grained sediments from Southern Africa and significance for the evolution of the continental crust. Geologische Rundschau 74, 1–9. Laurent, R., Delaloye, M., Wagner, J.J., Vuagnat, M., 1980. Composition of parental basaltic magma in ophiolites. In: Panayiotou, A. (Ed.), Ophiolites (Proceedings of the International Ophiolite Symposium, Cyprus, 1979). Cyprus Geological Survey Department, Nicosia, pp. 172–180. Li, J., Kusky, T.M., Huang, X., 2002. Archean podiform chromitites and mantle tectonites in ophiolitic melange, North China Craton; a record of early oceanic mantle processes. Geological Society of America Today 12 (7), 4–11. Malpas, J., 1978. Magma generation in the upper mantle, field evidence from ophiolite suites, and application to the generation of oceanic lithosphere. Philosophical Transactions of the Royal Society of London A 288, 527–546. Moore, J.G., 1979. Vesicularity and CO2 in mid-ocean ridge basalt. Nature 282, 250–253. Moores, E.M., 1982. Origin and emplacement of ophiolites. Review in Geophysics and Space Physics 20, 735–760. Moores, E.M., Vine, F.J., 1971. The Troodos massif, Cyprus and other ophiolites as oceanic crust; evaluation and implications. Transactions of the Royal Society of London A 268, 443–466. Nicolas, A., Al Azri, H., 1991. Chromite-rich and chromite-poor ophiolites; the Oman case. In: Peters, T.J., Nicolas, A., Coleman, R.G. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere (Proceedings of the Ophiolite Conference, Ministry of Petroleum and Minerals, Muscat, Oman, 7–18 January 1990). Kluwer Academic, Dordrecht, pp. 261–274. Nisbet, E.G., Pearce, J.A., 1977. Clinopyroxene composition in mafic lavas from different tectonic settings. Contributions to Mineralogy and Petrology 63, 149–160. Pearce, J.A., 1980. Geochemical evidence for the genesis and eruptive setting of lavas from Tethyan ophiolites. In: Panayiotou, A. (Ed.), Ophiolites (Proceedings of the International Ophiolite Symposium, Cyprus, 1979). Cyprus Geological Survey Department, Nicosia, pp. 261–272. 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 Transaction of the Royal Society of London A 300, 299–317. Pearce, J.A., Harris, N.B.W., Tindle, A., 1984a. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology 25, 956–983. Pearce, J.A., Lippard, S.J., Roberts, S., 1984b. Characteristics and tectonic significance of suprasubduction zone ophiolites. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basins Geology. Geological Society of London Special Publication 16, 77–94. Price, R.C., 1984. Late Precambrian mafic-ultramafic complexes in northeast Africa. Unpublished Ph.D. thesis. Open University, Milton Keynes, UK, p. 325. Prichard, H.M., 1979. A petrographic study of the process of serpentinisation in ophiolites and the ocean crust. Contributions to Mineralogy and Petrology 68, 231–241. Reischmann, T., 2000. Ophiolites and island arcs in the Late Proterozoic—Nubian Shield. Ofioliti 25, 1–13. Reischmann, T., Kröner, A., 1994. Late Proterozoic island arc volcanics from Gebeit, Red Sea Hills, north-east Sudan. Geologische Rundschau 83, 547–563. Ruxton, B.P., 1956. The major rock groups of the northern Red Sea hills, Sudan. Geological Magazine 93, 314–330.

206

Chapter 5: The Wadi Onib Mafic-Ultramafic Complex

Serri, G., 1981. The petrochemistry of ophiolite gabbroic complexes; a key for the classification of ophiolites into low-Ti and high-Ti types. Earth and Planetary Science Letters 52, 203–212. Shackleton, R.M., 1994. Review of late Proterozoic sutures, ophiolitic melanges and tectonics of eastern Egypt and North-east Sudan. Geologische Rundschau 83, 537–546. Shervais, J.W., 2001. Birth, death, and resurrection; the life cycle of suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems 2. 2000GC000080. Stern, R.J., 1994. Arc-assembly and continental collision in the Neoproterozoic East African Orogen; implications for the consolidation of Gondwanaland. Annual Review of Earth and Planetary Science 22, 319–351. Stern, R.J., Johnson, P.R., Kröner, A., Yibas, B., 2004. Neoproterozoic ophiolites of the ArabianNubian Shield. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 95–128. Stolojan, N., Khalifa, M.I., Hussein, I.M., Bellivier, F., Ayme, Y., Chez, Y., Gabralla, A.F., El Sammani, Y., Husson, Y., 1978. Geology and mineral prospection of sheets Deraheib between 21◦ – 22◦ N and 36◦ 30 –36◦ E and Derudeib between 17◦ 30 –18◦ N and 36◦ –36◦ 45 E as well as photogeological interpretation of the Mersa Shaab between 22◦ –22◦ 30 N and 35◦ –35◦ 45 E. Unpublished Report 78 SUD 001. Cooperation Programme of Mineral Research in Sudan, Bureau de Recherches Géologiques et Minière (Orléans) and Geological and Mineral Resources Department (Khartoum). Streckeisen, A., 1976. To each plutonic rock its proper name. Earth Science Reviews 12, 1–33. Wilson, R.A.M., 1959. The geology of the Xeros-Troodos area. Cyprus Geological Survey Department Memoir 1, 135. Zimmer, M., 1989. Der Gebel Gerf-Komplex (Arabisch-Nubischer Schild): Petrographische und geochemische Untersuchungen eines spätproterozoischen Ophiolites. Unpublished Ph.D. thesis. University of Mainz, Germany, p. 192. Zimmer, M., Kröner, A., Jochum, K.P., Reischmann, T., Todt, W., 1995. The Gabal Gerf Complex; a Precambrian N-MORB ophiolite in the Nubian Shield, NE Africa. Chemical Geology 123, 29– 51.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

207

Chapter 6

TECTONO-MAGMATIC EVOLUTION, AGE AND EMPLACEMENT OF THE AGARDAGH TES-CHEM OPHIOLITE IN TUVA, CENTRAL ASIA: CRUSTAL GROWTH BY ISLAND ARC ACCRETION J.A. PFÄNDERa,b,1 AND A. KRÖNERa a Institut

für Geowissenschaften, Universität Mainz, D-55099 Mainz, Germany Chemie, Postfach 3060, D-55020 Mainz, Germany

b Max-Planck-Institut für

The Agardagh Tes-Chem ophiolite in Tuva, Central Asia, is part of the Central Asian Mobile Belt which formed during subduction-accretion processes lasting from the early Neoproterozoic to the late Palaeozoic. The ophiolite was obducted onto the TuvaMongolian Massif (microcontinent?) in the early Palaeozoic towards the SE along N- and NW-dipping faults and is embedded within a tectonic mélange and thus is part of an accretionary wedge. Dating of three small zircon fractions from a plagiogranite by the evaporation technique yielded a 207 Pb/206Pb age of 569.6 ± 1.7 Ma, which reflects the crystallization age of the plutonic section of the ophiolite. Geochemical data reveal an island arcrelated origin for the ophiolite, where typical island arc volcanic rocks predominate over MORB-like pillow lavas. In contrast to the highly incompatible element-enriched volcanic rocks, all plutonic rocks of the ophiolite are depleted, and mineral compositions of ultramafic cumulates indicate the presence of boninitic parental melts. The ophiolite therefore consists of an association of island arc and back-arc related sequences that have been amalgamated during subduction-accretion and collisional obduction. Isotopic and trace element data reveal the existence of a depleted and refractory mantle source beneath Central Asia, from which the volcanic and plutonic rocks of the ophiolite were formed. However, source contamination took place by sediment subduction before the parental melts of the island arc volcanic rocks were formed.

1. INTRODUCTION A large number of late Mesoproterozoic to late Palaeozoic ophiolites are known in Central Asia and are thought to have formed in what is generally known as the Palaeo-Asian Ocean (Khain et al., 2003), but only a few have been investigated structurally and geochemically. The geochronological data available for the ophiolites of western Mongolia 1 Present address: Institut für Mineralogie, Universität Münster, Corrensstr. 24, D-48149 Münster, Germany.

DOI: 10.1016/S0166-2635(04)13006-5

208

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

Table 1. Ophiolites in Central Asia, ages and initial ε Nd values Name Agardagh Tes-Chem ophiolite Ozernaya ophiolite zone Dariv ophiolite Khantaishir ophiolite Bayankhongor ophiolite

Location Tuva

Age 569.6 ± 1.7

Western Mongolia Western Mongolia Western Mongolia Central Mongolia

Upper εNd (i) +8.5

Reference This study

+8.0

Kovalenko et al. (1996)

–

Khain et al. (2003)

U-Pb zircon

–

Khain et al. (2003)

Sm-Nd internal isochron

+11.9

Kepezhinskas et al. (1991)

522 ± 13 to 527 ± 43 571 ± 4

Type Single zircon 207 Pb/206 Pb age Sm-Nd internal isochron U-Pb zircon

568 ± 4 569 ± 21

display surprisingly similar ages (Table 1) (Khain et al., 2003; Buchan et al., 2002), although they are distributed over an area of more than ∼ 500,000 km2 (Khain et al., 2003; Kepezhinskas, 1986; Buchan et al., 2001). One of the northernmost ophiolites of known age in Central Asia is the Agardagh Tes-Chem ophiolite in the Tuva Autonomous Republic of Russia (Fig. 1). This ophiolite and its surrounding regions are part of the Neoproterozoic to Palaeozoic Central Asian Mobile Belt (CAMB), also known as Altaids in the Western literature (Fig. 1 inset) which comprises an area of more than ∼ 5.3 million km2 (Sengör et al., 1993), corresponding to about 12% of the present day‘s surface of Asia, or ∼ 3.5% of the present day’s solid surface of the Earth. According to various subduction-accretion models (Sengör et al., 1993; Mossakovskii et al., 1993; Sengör and Natalin, 1998), roughly half of this amount of newly formed Neoproterozoic to Palaeozoic crust is juvenile and was directly extracted from the mantle by supra-subduction zone melting and subsequent differentiation. It would be expected that such large volumes of melt extraction should leave a distinct thermal and chemical signature on the mantle. Therefore, the ophiolites accreted to the CAMB during crustal evolution play a key role in understanding the tectono-magmatic processes, which were responsible for crust-mantle differentiation during Neoproterozoic and Palaeozoic times. This chapter outlines the structure and composition of the Agardagh Tes-Chem ophiolite and places constraints on its age and tectono-magmatic evolution. A model is presented which explains the highly variable geochemical characteristics of different mafic rock types.

2. GEOLOGY The Agardagh Tes-Chem ophiolite marks the northwestern border of the TuvaMongolian Massif (TMM), which comprises several intrusive and metamorphic complexes. These complexes consist of metatonalites, gneisses, migmatites, amphibolites, marbles, quartzites, low-grade metasedimentary rocks and minor ultramafic lenses, and radiometric ages range between 536 and 464 Ma (Salnikova et al., 2001). Kyanitebearing gneisses in the Moren Complex of the TMM indicate high-pressure metamor-

2. Geology

209

Fig. 1. Geological map showing main lithotectonic units and structural interpretations. (a) Tes-Chem (NE of Kyzyl road) and Karachat (SW of Kyzyl road) units, (b) Agardagh unit (map kindly provided by C. Oidup, Tuvinian Institute for the Exploration of Natural Resources, Kyzyl, Tuva). Upper inset: location of working area in Asia (CAMB = Central Asian Mobile Belt). Lower inset: enlargement of the working area showing the position of the lithotectonic units relative to each other.

210

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

Fig. 2. Pillow lavas in the Tes-Chem unit along the road to Kyzyl.

phic conditions at around 536 Ma (Salnikova et al., 2001), whereas the amalgamation of different lithotectonic units occurred somewhat later at around 497 Ma. To the north of the TMM lies the E-W striking Tannuola ridge, which represents a middle to late Ordovician island-arc assemblage (Zonenshain et al., 1990; Fedorovskii et al., 1995; Kuzmichev et al., 2001), that evolved shortly after formation and obduction of the ophiolite. The ophiolite can be subdivided into three major units which are separated by thrusts and strike slip faults (Fig. 1): The Agardagh zone in the southwest is dominated by variably serpentinized ultramafic rocks (dunites, harzburgites, wehrlites and pyroxenites), with minor occurrences of gabbroic differentiates. A small plagiogranite body from the Agardagh zone was sampled for zircon chronology. The Karachat zone in the central part consists of a single intrusive body comprising hornblende-bearing gabbros, olivine-gabbros, gabbronorites, gabbrodiorites and some wehrlites in the basal section. The Tes-Chem zone builds the northeastern part of the ophiolite belt and comprises a variety of fine-grained gabbroic and doleritic rocks (microgabbros) associated with massive basalts and some pillow lavas (Fig. 2). The Tes-Chem section is overlain by conglomerates, sandstones and shallow-marine limestones, indicating a relatively shallow marine environment (Fig. 3). These carbonate-clastic sequences are regarded as late Neoproterozoic to early Cambrian cover rocks (Zonenshain et al., 1990; Fedorovskii et al., 1995). The entire ophiolite is embedded within a tectonic mélange, consisting predominantly of metasedimentary rocks, limestone slices and fragments of ultramafic rocks, sheared pillow lavas and massive basaltic andesites. Cherts are common in the Agardagh mélange (Kuskunuk formation), whereas they are scarce in the mélange associated with the Karachat- and Tes-Chem units (Seligskaja formation). Biostratigraphically the mélanges are dated to be early Cambrian

3. Field Relationships

211

Fig. 3. Massive, shallow marine limestones, sandstones with intercalated basaltic flows and conglomerates overlying the sheeted dyke complex in the Tes-Chem unit. View is to the SW, field of view at the horizon is about 1200 m.

in age, whereas the Kuskunuk formation is slightly older than the Seligskaja formation (C. Oidup, personal communication, 1999).

3. FIELD RELATIONSHIPS Field observations suggest that the southwestern part of the Tes-Chem unit represents the lower to intermediate part of a sheeted dyke complex. This is evident from the widespread occurrence of microgabbroic and doleritic dykes and the overall absence of chilled margins, typically found only in higher levels of a sheeted dyke complex (Nicolas and Boudier, 1991). Further evidence comes from frequent late-magmatic hornblende-rich gabbroic and gabbro-pegmatitic rocks and felsic differentiates forming small, isolated patches within the microgabbros and dolerites. Farther to the northeast, basaltic rocks typically forming massive flows become more frequent, whereas well preserved dykes are absent. Pillow lavas occur both embedded within the massive basalts and within the mélange. The plutonic section in the central part of the ophiolite (Karachat unit in Fig. 1a) covers an area of about 5 × 2 km2 and consists of layered and isotropic gabbroic rocks ranging from primitive olivine-gabbros and gabbronorites to hornblende-bearing gabbros, gabbrodiorites and minor diorites. Wehrlite layers up to several metres thick occur predominantly in the southwestern part of the Karachat complex, but are also observed throughout the entire intrusive body as lenses varying in thickness between several centimetres to several decimetres. Where present, layering in the gabbroic rocks is striking about N-S and dipping to the E by about 60◦ –65◦ . The occurrence of more evolved, quartz-bearing gabbros

212

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

Fig. 4. Foliations (upper plots) and lineations (lower plots) in metasedimentary rocks of the mélange associated with the Agardagh Tes-Chem ophiolite (lower hemisphere, equal area projection).

and gabbrodiorites in the northwestern part of the Karachat section indicates a top-bottom relationship with the upper, isotropic part of the inferred magma chamber located in the northwestern part, whereas more primitive, clinopyroxene-rich cumulates occur predominantly in the southeastern parts of the Karachat zone (Fig. 1a). To the northeast, the gabbros continue to the Tes-Chem unit (Fig. 1a), where the internal stratigraphy is repeated. The top-bottom relationships and repeated mafic to felsic successions require both the Karachat section as well as the entire ophiolite zone to be internally dismembered and tectonically thickened during obduction. Foliations and lineations measured in the metasedimentary rocks of the mélange (Fig. 4) indicate that thrusting occurred predominantly to the southeast along north- and northwest dipping faults (Figs. 4 and 5). Subsequent normal faulting roughly corresponding to E-W extension is mainly observed in Tes-Chem zone metasedimentary rocks (Fig. 4) and possibly separated the Karachat and Tes-Chem units from the Agardagh zone. 4. ZIRCON GEOCHRONOLOGY AND ND ISOTOPIC SYSTEMATICS The age of the ophiolite complex is constrained by the dating of three small zircon grain fractions from a plagiogranite body. Zircon analyses were carried out by the single zircon

4. Zircon Geochronology and Nd Isotopic Systematics

213

Fig. 5. Schematic profile through the Tes-Chem unit (along the road to Erzin, see Fig. 1a; not to scale). The occurrence of conglomerates and sandstones (see Fig. 3) associated with the sheeted dike complex suggests a shallow marine basin. Table 2. Isotopic data from evaporation of small zircon fractions from Plagiogranite sample Tu 5621 Zircon morphology and colour

Grain No.

Mass scans∗

Euhedral, clear, transparent, stubby

1 2 3

67 118 78

Evaporation temperature (◦ C) 1596 1599 1597

Mean (207 Pb/206 Pb) ratio† and error 0.059062 ± 61 0.059044 ± 30 0.059099 ± 27

age and error 569.4 ± 2.3 568.7 ± 1.1 570.8 ± 1.0 569.6 ± 1.7‡

Mean of grains 1–3 Clear, transp., stubby

207 Pb/206 Pb

4

56

1598

0.099935 ± 78

1623 ± 2

∗ Number of 207 Pb/206 Pb ratios evaluated for age assessment. † Observed mean ratio corrected for non-radiogenic Pb using the model of Stacey and Kramers (1975). Error is

2σ m based on counting statistics, given are the last two digits. ‡ Mean of three ages, error is 2σ . m

evaporation technique (Kober, 1987) with slight modifications (Kröner et al., 1991; Kröner and Hegner, 1998). Measurement of the Curtin University SHRIMP II zircon standard CZ3 resulted in a 207 Pb/206 Pb age of 564.8 ± 1.4, consistent with the adopted age of 564 Ma for this standard (Pidgeon et al., 1994). External reproducibility of the evaporation technique has been evaluated by repeated measurements of zircons from the Phalaborwa Complex (South Africa) over 12 months, resulting in an age of 2051.8 ± 0.4 Ma (2σm ) and consistent with a conventional U-Pb age of 2052.2 ± 0.8 Ma (2σ ; W. Todt, personal communication, 1999). Zircon morphology, measured isotopic ratios and calculated ages are given in Table 2. Fig. 6 shows a histogram of all measured 207Pb/206 Pb ratios from three grain fractions, resulting in a mean age of 569.6 ± 1.7 Ma. Cathodoluminescence images of two representative zircons from the plagiogranite sample are shown in Fig. 7. Both zircons are euhedral with distinct magmatic zoning. Cores are absent, suggesting a primary magmatic origin

214

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

Fig. 6. Histogram showing measured 207 Pb/206 Pb ratios and corresponding mean age derived from evaporation of three small zircon fractions of plagiogranite sample Tu 5621. Error is 2σ -mean (see Table 1).

Fig. 7. Cathodoluminescence images showing two euhedral and well zoned, typical zircons from the plagiogranite sample Tu 5621.

4. Zircon Geochronology and Nd Isotopic Systematics

215

Fig. 8. Sm-Nd isotopic composition of gabbroic rocks from the Karachat unit (grey points are discarded from regression due to wall rock assimilation). From Pfänder (unpublished data).

without inherited components. This is consistent with field relationships, where the plagiogranite is associated with isotropic gabbros, and is therefore regarded as a late stage fractionation product of a common parental magma. The NdDM model age of the plagiogranite sample is 880 Ma, assuming a linear mantle evolution (Goldstein et al., 1984). This model age is significantly older than the crystallization age and is best explained by a source contamination model (Pfänder et al., 2002), in which older crustal material was entrained into the mantle source of the ophiolite by sediment subduction. The entrainment of old crustal material into the mantle source is underlined by a significantly older zircon grain with an age of 1623 ±2 Ma (Table 2), which was isolated from the same plagiogranite sample. A slightly younger age than the zircon evaporation age for the ophiolite results from whole-rock Sm-Nd isotopic data of gabbroic rocks including the plagiogranite sample (Fig. 8). Regression of virtually cogenetic samples selected by field relationships and traceelement compositions yielded an age of 546 ± 18 Ma and an initial ε Nd value of 6.2 ± 0.6 (Fig. 8). However, due to likely assimilation of pre-existing arc crust during the final crystallization of the gabbroic rocks (Pfänder, unpublished data), the zircon age is a more reliable estimate for the age of the ophiolite. The Sm-Nd model age of granitic conglomerate pebbles from the Tes-Chem unit is ∼ 995 Ma (147Sm/144 Nd = 0.1978; 143 Nd/144 Nd(today) = 0.513045). Possibly, these granite pebbles are derived from the metasedimentary and intrusive rocks of the nearby Moren and Erzin complexes, which contain detrital zircons with U-Pb ages between 690 and 900 Ma (Salnikova et al., 2001). The presence of the conglomerates in tectonic contact with the basalt and sheeted dyke units of the ophiolite further supports the existence of a narrow ocean basin that received detritus from adjacent continental terrains.

216

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

5. PETROGRAPHY The majority of volcanic rocks of the ophiolite are Ol- and Qz-normative high-alumina basalts and basaltic andesites, with minor andesites and dacites, all characterised by a calcalkaline affinity. These rocks underwent low-grade metamorphism and are more or less altered, thus only relict minerals are present in a few cases, indicating that most of the samples have been augite- and plagioclase-phyric with abundant magnetite. Mg-numbers of relict augites are between 0.5 and 0.7, whereas relict plagioclase ranges from albite to labradorite (An65). Augite to hornblende transitions indicate greenschist-metamorphic conditions. Whole-rock Mg-numbers of the volcanic rocks vary between 0.48 and 0.73, whereas low Cr and Ni concentrations indicate substantial olivine-spinel and clinopyroxene fractionation (Pfänder et al., 2002). Gabbros are commonly hornblende-bearing, with abundant clinopyroxene now forming hornblende. Green, Ca-rich magnesio-hornblende occurs as interstitial or poikilitic phases enclosing clinopyroxene. Magmatic twins in clinopyroxene are scarce, whereas abundant orthopyroxene exsolutions indicate subsolidus re-equilibration. Magnetite and ilmenite are common phases in nearly all gabbroic rocks, indicating a high oxygen fugacity. Due to slow cooling, zonation in plagioclases is absent. Anorthite-contents are generally high and range from An67 to An92. Ultramafic rocks comprise residual mantle rocks (harzburgites and dunites) and ultramafic cumulates (pyroxenites and wehrlites), where harzburgites and dunites are strongly serpentinized. Serpentinite from harzburgite is characterised by retrograde transitions of orthopyroxene to bastite and by asbestos veins. Nephrite is common and indicates epizonal metamorphic conditions. Listwaenite occurs in the contact zone between serpentinized dunite and younger granite intrusions in the southwestern part of the Agardagh zone (Fig. 1b). The residual mantle rocks (harzburgites and dunites) are characterised by Cr-rich spinel compositions (Cr# up to 0.82) and high degrees of melt extraction up to ∼ 25%. The pyroxenites and wehrlites crystallised from a refractory, boninitic primary melt as indicated by extremely low trace element concentrations in clinopyroxene (about 0.1-times chondritic for light-rare earth elements; see (Pfänder, unpublished data).

6. GEOCHEMISTRY Trace element compositions of volcanic and gabbroic rocks are shown in Fig. 9a. Among the volcanic rocks two groups are distinguished, most likely indicating two different geodynamic settings. The majority of rocks are highly enriched in incompatible trace elements, whereas moderately incompatible elements are depleted relative to N-MORB. This clearly indicates an island arc origin for this group, which is confirmed by pronounced negative Nb and positive Pb and Sr anomalies. This further indicates the presence of fluids and subducted sediments from the downgoing slab within these melts. Pfänder et al. (2002) suggested that the island arc volcanic rocks formed during an early stage of subduction by low-degree melting from a depleted mantle source, containing subducted sediments. The

6. Geochemistry

217

Fig. 9. Primitive mantle-normalised trace element composition of volcanic (a) and gabbroic rocks (b) from the Agardagh Tes-Chem ophiolite. Note different scales and slightly different order of elements (from Pfänder et al., 2002; Pfänder, unpublished data; N-MORB data from Hofmann, 1988).

second group comprises N-MORB like massive basalts and low-grade pillow lavas. These rocks are attributed to the back-arc region of the ophiolite, where they formed without an appreciable influence of the downgoing slab from a similar depleted mantle source by

218

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

higher degrees of melting. Initial ε Nd values are lower in the island arc rocks (around +5.5) than in the back-arc basin basalts (+7.8 to +8.5; Pfänder et al., 2002). This clearly indicates the existence of a highly depleted mantle source beneath the back-arc region of the Agardagh Tes-Chem island arc system. In contrast to the volcanic rocks, the gabbros are strongly depleted in incompatible trace elements (Fig. 9b) but also have negative Nb- and positive Ba- and Sr-anomalies, consistent with an island arc related origin. All gabbros except the two most “enriched” samples are cumulates that crystallised from a parental melt which was similar to N-MORB with respect to the moderate incompatible trace elements (Sm-Yb), but which was enriched in highly incompatible and large-ion lithophile elements. In terms of their initial ε Nd values the gabbroic rocks are relatively uniform with values between +4.8 and +7.1.

7. TECTONO-MAGMATIC EVOLUTION Field relationships and geochemical data strongly suggest the evolution of the Agardagh Tes-Chem ophiolite to have been linked to a magmatic arc system, active around 570 Ma ago. Fig. 10 shows a speculative geodynamic model suitable to explain both field observations as well as geochemical data from volcanic and plutonic rocks of the ophiolite. According to this model, subduction was initiated within the Palaeo-Asian ocean during late Neoproterozoic times before 570 Ma (Fig. 10a). Triggered by fluids released from the downgoing slab, initial melting of an already depleted, refractory mantle source led to the formation of primary melts (Fig. 10b). Due to the presence of subducted, continentderived sediments in the mantle source, and because of low melting degrees and postmelting fractional crystallization, these melts became enriched in large-ion lithophile and incompatible trace elements (Pfänder et al., 2002), now forming the volcanic rocks of the ophiolite. Although speculative, the subducted continental sediments may have been derived from the nearby Tuva-Mongolian Massif (TMM in Fig. 10) SE of the active arc system. Simultaneously, or even after initial arc magmatism, the ultramafic cumulates associated with the Agardagh zone and the gabbroic rocks of the Karachat zone were formed “behind” the arc system, where the influence of slab-derived components was lower. Due to the boninitic nature of the parental melts of the ultramafic cumulates and gabbros (Pfänder, unpublished data), their formation requires an already depleted, sub-arc mantle wedge as well as sufficient heat to permit remelting of refractory harzburgite. This heat was most likely provided by an active back-arc spreading centre propagating towards the arc system. The existence of this back-arc spreading centre is confirmed by the occurrence of MORB-like pillow lavas having high ε Nd values up to +8.5. Ongoing subduction then finally led to the collision of the island arc system with the passive continental margin of the Tuva-Mongolian Massif in the SE (Figs. 10c, d). As subduction was roughly to the NW, obduction and thrusting of the ophiolite was directed to the SE. The accretion process may have occurred at around 536 Ma, which is the age of the earliest metamorphic event in the Tuva-Mongolian Massif (Salnikova et al., 2001), and also coincides with the age range

7. Tectono-Magmatic Evolution

219

Fig. 10. Geodynamic model showing evolution of the Agardagh Tes-Chem ophiolite. (a) Initial stage of subduction and initial magmatism (enriched volcanic rocks); (b) mature stage of subduction with formation of more depleted melts forming ultramafic cumulates and gabbroic intrusions; (c) collision and obduction of ophiolite onto Tuva-Mongolian massif (TMM); and (d) isostatic uplift and possibly extension of thickened crust (ADZ = Agardagh zone, TCZ = Tes-Chem zone; for details see text).

estimated for obduction of the Bayankhongor ophiolite in central Mongolia (540–450 Ma, Buchan et al., 2002). After collision, isostatic uplift of the thickened crust, possibly triggered by the breakoff of the subducted slab (Fig. 10d), led to E-W extension and normal faulting as preserved in the metasedimentary rocks of the mélange.

220

Chapter 6: Tectono-Magmatic Evolution, Age and Emplacement of the Agardagh Tes-Chem Ophiolite

ACKNOWLEDGEMENTS We greatly acknowledge the support of our Russian colleagues during the field season in Tuva, namely I. Kozakov, A. Kotov, E. Salnikova and V. Kovach from St. Petersburg and V. Lebedev, C. Oidup and A. Songorakova from Kyzyl (Tuva). C. Oidup and the Tuvinian Institute for the Exploration of Natural Resources are particularly thanked for providing us with their geological map of the Agardagh unit. REFERENCES Buchan, C., Cunningham, D., Windley, B.F., Tomurhuu, D., 2001. Structural and lithological characteristics of the Bayankhongor ophiolite zone, Central Mongolia. Geological Society of London 158, 445–460. Buchan, C., Pfänder, J.A., Kröner, A., Brewer, T., Cunningham, D., Windley, B.F., Tomurhuu, D., Tomurtogoo, O., 2002. Timing of accretion and collisional deformation in the Central Asian orogenic belt; implications of granite geochronology in the Bayankhongor ophiolite zone. Chemical Geology 192, 23–45. Fedorovskii, V.S., Khain, E.V., Vladimirov, A.G., Kargopolov, S.A., Gibsher, A.S., Izokh, A.E., 1995. Tectonics, metamorphism, and magnetism of collisional zones of the Central Asian caledonides. Geotectonics 29, 193–212. Goldstein, S., O’Nionsm, R.K., Hamilton, P.J., 1984. Age and evolution of the continental crust. Earth and Planetary Science Letters 70, 221–236. Hofmann, A.W., 1988. Chemical differentiation of the Earth; the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297–314. Kepezhinskas, K.B., 1986. Structural-metamorphic evolution of late Proterozoic ophiolites and Precambrian basement in the Central Asian foldbelt of Mongolia. Precambrian Research 33, 209– 223. Kepezhinskas, P.K., Kepezhinskas, K.B., Puchtel, I.S., 1991. Lower Paleozoic oceanic crust in Mongolian Caledonides: Sm-Nd isotope and trace element data. Geophysical Research Letters 18, 1301–1304. Khain, E.V., Bibikova, E.V., Salnikova, E.B., Kröner, A., Gibsher, A.S., Didenko, A.N., Degtyarev, K.E., Fedotova, A.A., 2003. The Palaeo-Asian ocean in the Neoproterozoic and early Proterozoic: new geochronologic data and palaeotectonic reconstructions. Precambrian Research 122, 329– 358. Kober, B., 1987. Single-zircon evaporation combined with Pb+ emitter bedding for 207 Pb/206 Pb-age investigations using thermal ion mass spectrometry, and implications to zirconology. Contributions to Mineralogy and Petrology 96, 63–71. Kovalenko, V.I., Puchtel, I.S., Yarmolyuk, V.V., Zhuravlev, D.Z., Stosch, H., Jagoutz, E., 1996. The Sm-Nd isotope system of ophiolites in the Ozernaya Zone, Mongolia. Stratigrafiya, Geologicheskaya Korrelyatsiya 4, 107–113. Kröner, A., Byerly, G.R., Lowe, D.R., 1991. Chronology of early Archean granite-greenstone evolution in the Barberton Mountain Land, South Africa, based on precise dating by single zircon evaporation. Earth and Planetary Science Letters 103, 41–54. Kröner, A., Hegner, E., 1998. Geochemistry, single zircon ages and Sm-Nd systematics of granitoid rocks from the Gory Sowie (Owl Mts), Polish West Sudetes; evidence for early Palaeozoic arcrelated plutonism. Journal of the Geological Society of London 155, 711–724.

References

221

Kuzmichev, A.B., Bibikova, E.V., Zhuralev, D.Z., 2001. Neoproterozoic (approximately 800 Ma) orogeny in the Tuva-Mongolia massif (Siberia); island arc-continent collision at the northeast Rodinia margin. Precambrian Research 110, 109–126. Mossakovskii, A.A., Ruzhentsev, S.V., Samygin, S.G., Kheraskova, T.N., 1993. Geodynamics of the Central-Asian Paleozoic oceans. Geotektonika 6, 3–33. Nicolas, A., Boudier, R., 1991. In: Peters, T., Nicolas, A., Coleman, R.G. (Eds.), Textures and Fabrics in the Mantle-Crust Transition Zone of Ophiolites; Implications on Physical State of Magma Chambers. Ministry of Petroleum and Minerals, Muscat, Sultanate of Oman. Pfänder, J.A., Jochum, K.P., Kozakov, I., Kröner, A., Todt, W., 2002. Coupled evolution of back-arc and island arc-like mafic crust in the late-Neoproterozoic Agardagh Tes-Chem ophiolite, Central Asia; evidence from trace element and Sr-Nd-Pb isotope data. Contributions to Mineralogy and Petrology 143, 154–174. Pidgeon, R.T., Furfaro, D., Kennedy, A., Nemchin, A.A., Van Bronswjk, W., Todt, W., 1994. Calibration of the CZ3 zircon standard for the Curtin SHRIMP II. United States Geological Survey Circular 1107, 251. Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters 26, 207–221. Salnikova, E.B., Kozakov, I.K., Kotov, A.B., Kröner, A., Todt, W., Nutman, A., Yakovleva, S.K., Kovach, V.P., 2001. Age of Paleozoic granites and metamorphism in the Tuvino-Mongolian massif of the Central Asian mobile belt; loss of a Precambrian microcontinent. Precambrian Research 110, 143–164. Sengör, A.M.C., Natalin, B.A., 1998. In: Yin, A., Harrison, T.M. (Eds.), Magmatic Fronts and StrikeSlip Faulting in Orogens Dominated by Accretionary Prism Material (Examples from the Altaids). Cambridge Univ. Press, Cambridge, UK. Sengör, A.M.C., Natalin, B.A., Burtman, V.A., 1993. Tectonic Evolution of Altaids. Nature 364, 299–307. Zonenshain, L.P., Kuzmin, M.I., Natapov, L.M., 1990. Geological History of the USSR; Approach with Different Positions, Reply to Critique. In: Geodynamics Series, vol. 21. American Geophysics Union.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

223

Chapter 7

ORIGIN AND EMPLACEMENT OF ARCHEAN OPHIOLITES OF THE CENTRAL OROGENIC BELT, NORTH CHINA CRATON TIMOTHY M. KUSKYa , JAINGHAI LIb , ADAM GLASSc AND X.N. HUANGb a Department

of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63103, USA b Department of Geological Sciences, Peking University, Beijing, China c Department of Geology, Cardiff University, Scotland, UK

Understanding Archean crustal and mantle evolution hinges upon identification and characterization of oceanic lithosphere. We have discovered and reported a complete, albeit dismembered and metamorphosed Archean ophiolite sequence in the North China Craton. Banded iron formation structurally overlies several tens of meters of variably deformed pillow lavas and mafic flows. These are in structural contact with a 2 km thick mixed gabbro and sheeted dike complex with gabbro screens, exposed discontinuously along strike for more than 20 km. The dikes consist of metamorphosed diabase, basalt, hb-cpx-gabbro, and pyroxenite. Many have chilled margins developed on their NE sides, indicating one-way chilling. The dike/gabbro complex is underlain by several kilometers of mixed isotropic and foliated gabbro, which develop compositional layering approximately 2 km below the sheeted dikes, and then over several hundred meters merge into strongly compositionally layered gabbro and olivine-gabbro. The layered gabbro becomes mixed with layered pyroxenite/gabbro marking a transition zone into cumulate ultramafic rocks including serpentinized dunite, pyroxenite and wehrlite, and finally into strongly deformed and serpentinized olivine and orthopyroxene-bearing ultramafic rocks interpreted as depleted mantle harzburgite tectonites. A U/Pb zircon age of 2.505 Ga from gabbro of the Dongwanzi ophiolite makes it the world’s oldest recognized, laterally-extensive complete ophiolite sequence. Characteristics of this remarkable ophiolite may provide the best constraints yet on the nature of the Archean oceanic crust and mantle, and offer insights to the style of Archean plate tectonics and global heat loss mechanisms. The Dongwanzi ophiolite is one of the largest well-preserved greenstone belts in the Central Orogenic belt that divides the North China craton into eastern and western blocks. More than 1000 other fragments of gabbro, pillow lava, sheeted dikes, harzburgite, and podiform-chromite bearing dunite occur as tectonic blocks (tens to hundreds of meters long) in a biotite-gneiss and BIF matrix, intruded by tonalite and granodiorite, in the Zunhua structural belt. Blocks in this metamorphosed Archean ophiolitic mélange preserve deeper levels of oceanic mantle than the Dongwanzi ophiolite. The ophiolite-related DOI: 10.1016/S0166-2635(04)13007-7

224

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

mélange marks a suture zone across the North China Craton, traced for more than 1600 km along the Central orogenic belt. Many of the chromitite bodies are localized in dunite envelopes within harzburgite tectonite, and have characteristic nodular and orbicular chromite textures, known elsewhere only from ophiolites. The chromites have variable but high chrome numbers (Cr/Cr + Al = 0.74–0.93) and elevated P, also characteristic of suprasubduction zone ophiolites. The high chrome numbers, coupled with TiO2 wt% < 0.2 and V2 O5 wt% < 0.1 indicate high degrees of partial melting from a very depleted mantle source and primitive melt for the chromite. A Re-Os model age from the chromites indicates an age of 2547 ± 10 Ma (Kusky et al., 2004), showing that they are the same age as the Dongwanzi ophiolite. The ultramafic and ophiolitic blocks in the Zunhua mélange are therefore interpreted as dismembered and strongly deformed parts of the Dongwanzi ophiolite. We suggest the name “Dongwanzi ophiolite belt” for these rocks. Neoarchean (2.50 Ga) high-pressure granulites form a belt more than 700 km long along the western side of the Central Orogenic Belt. Several Neoarchean sedimentary basins consisting of conglomerate, graywacke and shale are located along the eastern side of the Central Orogenic Belt, and are interpreted as remnants of a foreland basin. The three belts record the Neoarchean subduction and collision between the western and eastern blocks of the North China Craton.

1. INTRODUCTION Understanding early Earth evolution and obtaining reliable geologic markers of temporal changes in specific tectonic environments has proven elusive. Remarkably few constraints exist on the nature of Archean oceanic crust and mantle, and even whether or not Earth lost heat in the Archean by plate tectonic mechanisms such as creating and cooling oceanic lithosphere, as it does today. Although many arguments have been made to support the operation of plate tectonics in the Archean, particularly in the Slave, Superior, Yilgarn, Kaapvaal, and Zimbabwe cratons, no well-documented and complete ophiolites were known from the geological record until recently. We reported a complete, though dismembered and metamorphosed Archean ophiolite sequence in a Neoarchean orogenic belt (2.60–2.50 Ga) of the North China Craton (Kusky et al., 2001). This remarkable ophiolite may offer the best constraints yet on the nature of the late Archean oceanic crust and mantle. The ophiolite is associated with an intensely sheared linear belt of ophiolitic mélange in which blocks of pillow lava, diabase dikes, gabbro, dunite, harzburgite, and podiform chromitite occur in a biotite-gneiss and banded iron formation (BIF) matrix (Li et al., 2002). Many parts of the mélange belt were intruded by tonalitic and granodioritic magmas, then deformed and metamorphosed again. In this contribution we document the field, structural, geochronological, and mineralogical characteristics of this ophiolite and related mélange. This contributes to a better understanding of Archean crustal and mantle evolution, which is helpful to document the very way in which the Earth lost heat during early times of high heat production.

1. Introduction

225

Ophiolites are a distinctive association of allochthonous rocks interpreted to form in a variety of plate tectonic settings, including oceanic spreading centers, back-arc basins, forearcs, and arcs (Anonymous, 1972; Moores, 1982, 2002; Nicolas, 1989; Parson et al., 1992; Sylvester et al., 1997; Dilek et al., 2000; Karson, 2001). A typical, complete ophiolite, grades downward from pelagic sediments into a mafic volcanic complex generally made of mostly pillow basalts, underlain by a sheeted dike complex. These are underlain by isotropic and layered gabbro exhibiting cumulate textures, then tectonized peridotite, resting above a thrust fault marking the contact with underlying rock sequences. As described below, the 2505 Ma Dongwanzi ophiolite contains all of the rock associations of a typical ophiolite, and is thus a complete ophiolite sensu stricto. There are many variations in the rock sequence preserved in ophiolites, some related to initial variations between the crustal sequences formed in different tectonic settings (e.g., Moores, 2002), and other variations related to emplacement-related deformation (e.g., Kusky and Vearncombe, 1997). In fact, most ophiolites are not typical (Karson, 2001; Moores, 2002), and lack one or more of the above units, or have additional unusual rock units such as silicic intrusives, hornblende-gabbros, or high-magnesium lavas. Ophiolites are one of the hallmarks of collisional mountain belts interpreted to mark the sites along which oceanic basins have closed, and therefore to demonstrate lateral motion between plates. The next oldest nearly-complete ophiolites recognized in the geological record are the 1960 Ma Jourma Complex, Finland (Kontinen, 1987; Peltonen and Kontinen, 2004), and the 1998 Ma Purtuniq ophiolite, Cape Smith Belt (Scott et al., 1991, 1992), and the 1730 Ma Payson ophiolite, Arizona (Dann, 1991, 1997a, 1997b, 2004). The apparent lack of complete or nearly complete ophiolites in the older geological record has prompted many theories that plate tectonics may have operated in fundamentally different ways in Earth’s early evolution (Karson, 2001; Sleep and Windley, 1982; Bickle et al., 1994; Abbott, 1996; Moores, 2002). However, numerous Archean greenstone belts contain two or more parts of the full ophiolite sequence (Harper, 1985; de Wit et al., 1987, 1992; Kusky, 1987, 1990, 1991; Brandl and de Wit, 1997). This led others to theorize that parts of many greenstone belts may be dismembered ophiolites formed in a manner analogous to younger dismembered ophiolites (Kusky and Polat, 1999). Similarly, most Proterozoic and Phanerozoic ophiolites including the Jourma, Payson, and Purtuniq complexes are dismembered, or partial sequences (Kontinen, 1987; Scott et al., 1991; Kröner, 1985; Berhe, 1990; Johnson et al., 2004; Stern et al., 2004). The documentation of a complete and laterallyextensive Archean ophiolite sequence 500 Ma older than the best Proterozoic examples is important to our understanding of Earth evolution and how the planet may have lost heat during early times of high heat production. This discovery also has implications for the development of Earth’s early biosphere, since some of the chert and BIF probably formed near sea-floor hydrothermal vents, and may host early life forms (e.g., Rasmussen, 2000; Reed, 2002). The main obstacle to recognizing Archean ophiolitic rocks is polyphase deformation and metamorphism, which may rework them into belts of mafic schist or gneiss. It is important to delineate sites of preservation of ophiolites during regional tectonic analy-

226

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

sis, and to identify ancient ophiolites by their characteristic rock assemblages, including pillow lavas, sheeted dikes, layered gabbro, and podiform chromitite. Detailed structural mapping is important for documenting relationships between ancient ophiolites and country rocks. Geochemistry, petrology, and magmatic stratigraphy can further resolve the early tectonic setting of the ophiolite, whether arc, suprasubduction zone (SSZ), or mid-ocean ridge (MOR). In this paper, we review our current state of knowledge of the field, structural, geochronological, mineralogical, and chromite chemistry of this ophiolite and related mélange. We assess what these data mean in terms of Archean crustal and mantle evolution, thermal state of the early Earth, and ideas about Precambrian plate tectonics. We discuss the implications these observations have for our understanding of fundamental properties of Archean crust and mantle, and planetary evolution. Understanding Archean oceanic processes in the North China craton provides a valuable contrast with similar processes recorded in younger ophiolites, indicating how mid-ocean ridge processes may have evolved from a period of high heat production to one of lower heat production. The current review presented in this contribution represents the results of the first-phase reconnaissance and 1:100,000 scale mapping, and is presented as our team embarks on a new effort to complete more detailed mapping over the course of the next several years. Therefore, some of the details and outstanding questions will hopefully be answered by future work, and some interpretations are likely to change as more data becomes available.

2. REGIONAL GEOLOGY AND TECTONIC DIVISIONS OF THE NORTH CHINA CRATON The North China Craton (Fig. 1) occupies about 1.7 million square kilometers in northeastern China, Inner Mongolia, the Yellow Sea, and North Korea. It is bounded by the Qinling-Dabie Shan orogen to the south, the Yinshan-Yanshan orogen to the north, the Longshoushan belt to the west and the Qinglong-Luznxian and Jiao-Liao belts to the east (Bai and Dai, 1996, 1998). The North China Craton (NCC) includes a large area of intermittently-exposed Archean crust (Fig. 1), including ca. 3.8–2.5 Ga gneiss, TTG, granite, migmatite, amphibolite, ultramafite, mica schist and dolomitic marble, graphitic and sillimanititc gneiss (khondalites), banded iron formation (BIF), and metaarkose (Jahn and Zhang, 1984a, 1984b; Bai et al., 1992; Wu et al., 1998; Zhao, 1993; Jahn et al., 1987; Bai, 1996; He et al., 1991, 1992; Shen et al., 1992; Wang et al., 1997). The Archean rocks are overlain by the 1.85–1.40 Ga Mesoproterozoic Changcheng (Great Wall) system (Li et al., 2000a, 2000b). In some areas in the central part of the North China Craton, 2.40– 1.90 Ga Paleoproterozoic sequences deposited in cratonic graben are preserved (Kusky and Li, 2003). We divide the North China craton into two major blocks (Fig. 2) separated by the Neoarchean Central Orogenic Belt in which virtually all U-Pb zircon ages (upper intercepts) fall between 2.55 and 2.50 Ga (Kröner et al., 1998, 2002; Li et al., 2000b; Wilde et al., 1998; Zhao, 2001; Kusky et al., 2001). The Western Block, also known as the

2. Regional Geology and Tectonic Divisions of the North China Craton

227

Fig. 1. Map of the North China craton, showing the distribution of Archean, Proterozoic, and younger rocks. Compiled from numerous sources.

Ordos Block (Bai and Dai, 1998; Li et al., 1998), is a stable craton with a thick mantle root, no earthquakes, low heat flow, and a lack of internal deformation since the Precambrian. In contrast, the Eastern Block is atypical for a craton in that it has numerous earthquakes, high heat flow, and a thin lithosphere reflecting the lack of a thick mantle root. The North China Craton is one of the world’s most unusual cratons in that it had a thick tectosphere (subcontinental lithospheric mantle) developed in the Archean, which was present through the Ordovician as shown by deep xenoliths preserved in Ordovician kimberlites (Gao et al., 2002). However, the eastern half of the root appears to have delaminated or otherwise disappeared during Paleozoic, Mesozoic, or Cenozoic tectonism. This is demonstrated by Tertiary basalts that bring up mantle xenoliths of normal “Tertiary mantle” with no evidence of a thick root (e.g., Menzies et al., 1993; Griffin et al., 1998; Zheng et al., 1998; Gao et al., 2002). The processes responsible for the loss of this root are enigmatic but are probably related to the present day high-heat flow, Phanerozoic basin dynamics and orogenic evolution.

228

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 2. Major tectonic divisions of the North China craton, into the Western and Eastern blocks, Central orogenic belt, and Paleoproterozoic orogenic belt (after Li et al., 2000a).

The Central Orogenic belt (COB) includes belts of TTG, granite, and supracrustal sequences metamorphosed from granulite to greenschist-facies. It can be traced for about 1,600 km from west Liaoning to west Henan (Fig. 2). Widespread high-grade regional metamorphism including migmatization occurred throughout the Central Orogenic belt between 2.60 and 2.50 Ga, with final uplift of the metamorphic terrain at ca. 1.9–1.80 Ga associated with extensional tectonism (Li et al., 2000a) or a collision on the northern margin of the NCC (Kusky and Li, 2003). Amphibolite to greenschist grade metamorphism predominates in the southeastern part of the COB (Fig. 2), but the northwestern part of the COB is dominated by granulite-facies to amphibolite facies rocks, including some high-pressure assemblages (10–13 kbars at 850 ± 50 ◦ C; Li et al., 2000b; Zhao et al., 2001; see additional references in Kröner et al., 2002). The high-pressure assemblages can be traced for more than 700 km along a linear belt trending ENE. Internal (western) parts of the orogen are characterized by thrust-related horizontal foliations, flat-dipping shear zones, recumbent folds, and tectonically interleaved high-pressure granulite migmatite and metasediments. It is widely overlain by sediments deposited in graben and continental shelf environments, and intruded by several dike swarms (2.40–2.50 Ga, 1.80–1.90 Ga). Several large anorogenic granites with ages of 2.20–2.00 Ga are identified within the belt. Recently, two linear units have been documented within the belt, including a high-pressure granulite belt in the west (Li et al., 2000a) and a foreland-thrust fold belt in the east (Li et al., 2002). The high-pressure granulite belt is separated by normal-sense shear zones from the western block, which is overlain by thick metasedimentary sequences

3. Ophiolites of the Central Orogenic Belt, North China Craton

229

(khondalite, younger than 2.40 Ga, and metamorphosed at 1,862.7 ± 0.4; A. Kröner, personal communication, 2003). The Hengshan high-pressure granulite belt is about 700 km long, consisting of several metamorphic terrains, including the Hengshan, Huaian, Chengde, and west Liaoning complexes (Fig. 2). The HPG commonly occurs as inclusions within intensely sheared TTG (2.60–2.50 Ga) and granitic gneiss (2.50 Ga), and are widely intruded by Kgranite (2.20–1.90 Ga) and mafic dike swarms (2.40–2.45 Ga, 1.77 Ga) (Li et al., 2000b; Kröner et al., 2002). Locally, khondalite and turbiditic slices are interleaved with the highpressure granulite rocks, suggesting thrusting. The main rock type is garnet-bearing mafic granulite with characteristic Pl-OPX corona around the garnet (Figs. 3a–c), which show rapid exhumation-related decompression. The isothermal decompressive P-T-t path can be documented within the rocks, the peak PT is in the range of 1.2–1.0 GPa, at 700–800 ◦ C. At least three types of REE patterns are shown by mafic rocks of the high-pressure granulites, from flat to LREE-moderately enriched, indicating a tectonic setting of active continental margin or island arc (Li et al., 2002). The high-pressure granulites were formed through subduction-collision, followed by rapid rebound-extension, recorded by mafic dike swarms of 2.50–2.40 Ga and graben-related sedimentary sequences in the Wutai Mountain-Taihang Mountain areas (Kusky and Li, 2003). The Qinglong foreland basin and fold-thrust belt (Fig. 2) is north to northeast-trending, and is now preserved as several relict folded sequences (Qinglong, Fuping, Hutuo, and Dengfeng). Its general sequence from bottom to top can be further divided into three subgroups of quartzite-mudstone-marble, turbidite, and molasse, respectively (Fig. 4). The lower subgroup of quartzite-mudstone-marble is well preserved in central sections of the Qinglong foreland basin (Taihang Mountain), with flat-dipping structures, interpreted as a passive margin developed prior to 2.5 Ga on the eastern block. It is overlain by lower-grade turbidite and molasse type sediments (Figs. 3d–h). The western margin of Qinglong foreland basin is intensely reworked by thrusting and folding, and is overthrust by the overlying orogenic complex (TTG gneiss, ophiolitic, accretionary sediment). To the east its deformation becomes weaker in intensity (Fig. 5). The Qinglong foreland basin is intruded by a gabbroic dike complex consisting of 2.40 Ga diorite, and overlain by graben-related sediments and flood basalts. In the Wutai and North Taihang basins, many ophiolitic blocks are recognized along the western margin of the foreland thrust-fold belt. These consists of pillow lava, gabbroic cumulates, and harzburgite. The largest ophiolitic thrust complex imbricated with foreland basin sedimentary rocks is up to ten kilometers long, preserved in the Wutai-Taihang Mountains (Wan et al., 1998).

3. OPHIOLITES OF THE CENTRAL OROGENIC BELT, NORTH CHINA CRATON We have identified several dismembered Archean ophiolites in the Central orogenic belt, including some in Lioning Province, at Dongwanzi, north of Zunhua, and at Wutai Mountain (Fig. 2). The best studied of these are the Dongwanzi and Zunhua ophiolitic terranes, which are the main focus of this review. The Zunhua Structural belt (ZSB) of

230

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 3. Photographs of rocks in the Central Orogenic Belt. (a) High-grade gneiss with large garnet porphyroblasts exhibiting retrograde high-P granulite textures, Hengshan Complex; (b) mafic boudin in granulite gneiss, Hengshan Complex; (c) garnet-rich granulite gneiss, Hengshan Complex; (d) foreland basin flysch from the Qinglong basin; (e) conglomerate of the Hutuo Group, Wutai Mountains; (f) graywacke/shale beds in Hutuo Group, interpreted as flysch; (g) west-dipping flysch of Hutuo Group, Wutai Mountains (small temple for scale is 1.5 m tall); (h) thrust slice of arkose (Lower Wutai Group), interpreted as stable continental margin imbricated with foreland basin flysch, Wutai Mountains.

3. Ophiolites of the Central Orogenic Belt, North China Craton

231

Fig. 4. Stratigraphic column showing Group and Formation names for different rock assemblages in the Wutai Mountain area, along with their tectonic interpretation. Modified after Tian (1991). Abbreviations for formation names: BYK, Banyukou Formation; JGK, Jingangku Formation; ZW, Zhuangwang Formation; WX, Wenxi Formation; BZY, Baizhiyan Formation; HMY, Hongmenyan Formation.

the Eastern Hebei Province (Fig. 6) preserves a cross section through most of the northeastern part of the Central Orogenic belt. This belt is characterized by highly strained gneiss, banded iron formation (BIF), 2.60–2.50 Ga greenstone belts and mafic to ultramafic complexes in what Li et al. (2002) interpret as a high-grade ophiolitic mélange. The belt is intruded by widespread 2.60–2.50 Ga tonalite-trondhjemite gneiss (TTG), granites (2.50 Ga) and is cut by ductile shear zones (Li et al., 2000a; Wu et al., 1998; Kusky et al., 2001). The Neoarchean high-pressure granulite belt (Chengde-Hengshan HPG) strikes through the northwest part of the belt. The Zunhua structural belt is thrust over the Neoarchean Qianxi-Taipingzhai granulite-facies terrain (2.50 Ga), consisting of enderbitic to charnockitic gneiss forming several small dome-like structures southeast of the Zunhua belt. The Zunhua structural belt clearly cuts across the dome-like Qian’an-

232

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 5. Schematic structural cross section through the Hengshan Complex, Wutai Complex, and Fuping Complex, showing Hengshan thrust over Wutai, and Wutai thrust over the Fuping Complex. Modified after Tian (1991).

Qianxi structural patterns to the east. The Qian’an granulite-gneiss dome (3.80–2.50 Ga) forms a large circular dome in the southern part of the area (Fig. 6), and is composed of tonalitic-trondhjemitic gneiss, and biotite-granite. Mesoarchean (2.80–3.00 Ga) and Paleoarchean (3.50–3.85 Ga) supracrustal sequences (Shen et al., 1992) outcrop in the eastern part of the region (Fig. 6). The Qinglong Neoarchean (2.70–2.50) amphibolite to greenschist facies supracrustal sequence strikes through the center of the area, and is interpreted here to be a foreland fold-thrust belt, intruded by large volumes of 2.40 Ga diorite in the east. The entire North China craton is widely cut by at least two Paleoproterozoic mafic dike swarms (2.50–2.40 Ga, 1.80–1.70 Ga), associated with regional extension (Li et al., 2000a). Mesozoic-Cenozoic granite, diorite, gabbro and ultramafic plugs occur throughout the NCC, and form small intrusions in some of the belts. The largest well-preserved sections of the Dongwanzi ophiolite are located approximately 200 km NE of Beijing in the northeastern part of the Zunhua structural belt, near the villages of Shangyin and Dongwanzi (Figs. 6 and 7). It consists of prominent amphibolitefacies mafic-ultramafic complexes (Fig. 8) in the northeast sector of Zunhua structural belt, previously mapped as belts of amphibolite-facies layered intrusion and associated rocks (Shen et al., 1992; Zhang et al., 1986, 1991). The southern end of the Dongwanzi ophiolite belt near Shangyin is complexly faulted against granulite-facies gneiss, with both thrust faults and younger normal faults present. The main section of the ophiolite dips steeply

3. Ophiolites of the Central Orogenic Belt, North China Craton

233

Fig. 6. Map of part of the Eastern Hebei Province, North China Craton, showing the location and tectonic setting of the Dongwanzi ophiolite. Note abundant similar belts of amphibolite-granulite grade metabasites in the Zunhua structural belt. The Dongwanzi ophiolite is intruded by the (deformed) ca. 2.4 Ga Cuizhangzi gneiss (Wang and Zhang, 1995) and has a Sm-Nd whole rock isochron model age of 2756 ± 177 Ma (Wu and Geng, 1991). The ophiolite is approximately the same age as the 2.7–2.5 Ga Qinglong supracrustal sequence, consisting of interbedded metagraywackes and shales.

NW, is approximately 50 km long, and is 5–10 km wide (Figs. 6 and 7). We have obtained a U/Pb-zircon age of 2505 ± 2 Ma for two gabbro samples from the Dongwanzi ophiolite (Kusky et al., 2001). Additional preliminary U/Pb zircon geochronology from dikes cutting the Central Belt of the Dongwanzi ophiolite has revealed the presence of a fraction of clear euhedral zircons with ages falling between 300 and 200 Ma (R. Tucker, personal communication, 2002). Zhai et al. (personal communication, 2002) report an 40 Ar/39 Ar plateau (metamorphic) age of 1.8 Ga from amphiboles from the same outcrop that the 300 Ma U/Pb ages come from, which led us to re-examine the central belt of the Dongwanzi ophiolite. We have determined that parts of the central belt are intruded by a mafic/ultramafic Mesozoic pluton with related dikes (Fig. 9), and that this pluton is slightly larger than we

234

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 7. Modified reconnaissance map of the Dongwanzi ophiolite, showing belts of pillow lavas, sheeted dikes, gabbro complex, cumulate ultramafics, and harzburgite tectonite. Modified after Kusky et al. (2001).

originally suggested (Kusky et al., 2001; Huson et al., 2004). More work is needed to determine if the central belt is intruded by even more younger rocks related to the Dushan composite plutonic complex. The base of the ophiolite is strongly deformed, and intruded by the 2391 ± 50 Ma Cuizhangzi diorite-tonalite complex (Zhang et al., 1986). The Dongwanzi ophiolite is associated with a number of other amphibolite-facies belts of mafic plutonic and extrusive igneous rocks (Fig. 6) in the Zunhua structural belt. These mafic to ultramafic slices and blocks can be traced regionally over a large area from Zunhua to West Liaoning (about 200 km). Much of the ZSB is interpreted as a high-grade ophiolitic mélange (Li et al.,

3. Ophiolites of the Central Orogenic Belt, North China Craton

235

Fig. 8. Photographs of rocks of the Dongwanzi ophiolite. (a) Pillow lavas, (b) interpillow chert and hyaloclastite, (c) pillow lava, (d) high-level sheeted dikes, (e) layered gabbro, (f) thinly layered gabbro.

2002), with numerous tectonic blocks of pillow lava, BIF, dike complex, gabbro, dunite, serpentinized harzburgite, and podiform chromitite in a biotite-gneiss matrix (Fig. 6), intruded extensively by tonalite and granodiorite. Cross-cutting granite has yielded an age of 2400 ± 15 Ma (Wu et al., 1998). We suggest that the blocks in the mélange correlate with the Dongwanzi and other ophiolitic fragments in the ZSB. This correlation is supported by recent Re-Os age determinations on several of these blocks, revealing that they are 2.54 Ga old. Existing age constraints need to be supplemented by new U/Pb ages on

236

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

other units of the ophiolite, cross-cutting granites, and other mafic belts in mélange that may be correlative with the Dongwanzi ophiolite.

4. FIELD DESCRIPTION OF OPHIOLITIC UNITS The upper part of the Dongwanzi ophiolite consists of pillow lavas, pillow breccias, and interflow sedimentary rocks including rare chert, banded iron formation (with local sulfides), and metapelites (Figs. 7 and 8). There are other extensive thick fault-bounded units of banded iron formation located near the top of the ophiolite. The pillow lavas, pillow breccias, and interpillow sediments are rarely well-preserved, extensively altered to chlorite and hornblende schists, and most are significantly deformed. Pillows are in many cases difficult to recognize (Kusky and Li, 2002), and are also interbedded with more massive flows, and cut by sills (Fig. 8). Where well preserved, the pillows show typical cuspate lower boundaries and lobate upper contacts, defining stratigraphic younging. In some places pillows are delineated by 2–3 cm thick epidote-rich selvages surrounding fine-grained, 0.5–1.0 m wide pillow cores. In some locations metamorphism has caused the pillow cores to become coarsely-crystalline with large hornblende crystals overgrowing earlier magmatic textures. Elongate pillow lava tubes are preserved in a few places, such as 1 km west of the village of Shang Yin (Fig. 7). There, the pillow lavas are faulted against granulite facies tonalitic gneiss (ca. 2.50 Ga), and locally occur as inclusions in a younger tonalite (Fig. 8). Interpillow chert is also present but rare at Shang Yin. The pillow lavas have flat MORB-like to suprasubduction zone types of REE signatures (Kusky and Li, 2002; Kusky et al., 2004). The dike complex of the Dongwanzi ophiolite is at least 2 km thick and extends discontinuously along strike for more than 20 km (Figs. 8 and 9). Several, hundred meter long outcrops exhibit 100% dikes, but in most places the dikes intrude gabbro, and exhibit mutually cross-cutting relationships with the gabbro. In contrast, the largest-known sheeted dike complex in a Proterozoic ophiolite is the 700 m long by 150 m wide dike complex in the ca. 1998 Ma Purtuniq ophiolite (Scott et al., 1991). The Dongwanzi dike complex exhibits predominant one-way chilling of diabase dikes that are chilled on their NE sides, but are intruded by other dikes along their SW margins. More than 70% of the dikes in the sheeted complex exhibit one-way chilling, with hundreds of dikes measured. Gabbro screens are common in the dike complex, with diabase dikes exhibiting double and single-chill margins chilled against the gabbro. The number and thickness of gabbro screens generally increases downward in the ophiolite, marking a gradual transition from the dike complex

Fig. 9. Geological map of the lower part of the transition zone from layered gabbro to cumulate ultramafic rocks in the Dongwanzi ophiolite. Note also the lenses of harzburgite tectonite, gabbro, and the intrusion of two sets of mafic dikes that appear related to the sheeted dike complex that is stratigraphically above this unit. Large diorite intrusion on SW side of map area also intrudes the Changcheng system (1.85–1.40 Ga). Modified after Kusky et al. (2001).

4. Field Description of Ophiolitic Units

237

238

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 10. Photographs of mutually cross cutting gabbro and diabase dikes from the central belt.

into the gabbroic fossil magma chamber. In many locations the gabbro also exhibits internal chill margins, with coarse-grained gabbro becoming finer-grained toward chill margins that also show a preferential one-way chilling along their NE margins, and intrusion by other gabbroic dikes along their SW margins (Fig. 8). This transition from dikes to gabbro is reminiscent of parts of the Semail ophiolite, where dikes grade into and were probably fed from magmatically layered gabbro (Nicolas and Boudier, 2000). The gabbro is variable in mineralogy and texture, with some phases being very feldspar rich, while others show abundant, several cm long crystals of clinopyroxene largely altered to hornblende. The mineralogical composition of the gabbroic rocks ranges from diorite through true gabbro, hornblende gabbro, hornblende-pyroxene gabbro, to pyroxenite. Olivine is extremely rare. Basaltic-diabasic dikes commonly cut the gabbro, but in places gabbro also cuts and includes xenoliths of diabasic dikes suggesting that the gabbro and diabase are synchronous or comagmatic phases, and both are related to magmatic extension. In places, diabase dikes can be shown to root in the gabbro complex, as they form from the coalescence of many small melt veinlets that are disseminated through the gabbro (Fig. 10). The gabbro and diabase dike complex is cut by abundant epidote-clinozoisite veins, and massive sulfides (Cu, Fe) similar to sea-floor metamorphism veins in other ophiolites (Sylvester et al., 1997; Harper, 1999). The presence of these veins in the gabbro gives us an idea of the depth to which the sea floor hydrothermal system extended in this ophiolite. Other pyroxenite

4. Field Description of Ophiolitic Units

239

Fig. 11. Photograph of cross-cutting pyroxene-apatite dike that has yielded a preliminary date of ca. 300 Ma (R. Tucker, personal communication, 2002), from the central belt.

through granodiorite veins and dikes also intrude the ophiolite and dike complex (Fig. 11), but these have yielded a preliminary U/Pb age of 300 Ma (R. Tucker, personal communication) and are therefore probably related to the nearby Paleozoic-Dushan plutonic complex, and not the ophiolite as originally proposed (Kusky et al., 2001). Accordingly, we have revised our map of the central (Fig. 9) belt to show as accurately as known the distribution of younger intrusive rocks. Some normal faults have developed along the sheeted dike margins, which tilted the gabbroic screens between dikes. To the east, metagabbro and pyroxenite locally occurs as inclusions within Neoarchean gneiss. Sheeted dikes and gabbroic rocks are locally interleaved with the Mesoproterozoic Chang Cheng Series quartzite (1.85–1.40) and repeated by Mesozoic thrusts. The angular discordance between two sets of gabbro dikes observed in many outcrops is similar to dike relationships observed in many younger ophiolites. The two sets of dikes could be formed before and after rotation of normal fault blocks during sea-floor extensional tectonics. Alternatively, the second set of dikes might be related to a younger episode of arc or other type of volcanism. A third possibility is that some of the dikes are Mesozoic, as confirmed by the 300 Ma preliminary age obtained from a dike at Dongwanzi village. We recognize so far that the Dongwanzi ophiolite is cut by at least four deformed dike swarms (three mafic dike swarms, including pyroxenite dikes, the two sets of sheared gabbroic dikes, and diabase dikes), followed by one granodiorite-pyroxenite suite (dated at ∼ 300 Ma by R. Tucker).

240

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Since we now recognize that parts of the ophiolite are cut by Mesozoic granodiorite, gabbro, and pyroxenite, and one undeformed mafic to ultramafic dike swarm also intrudes the Mesoproterozoic sequence, additional field efforts need to be aimed at differentiating Mesozoic from older Archean units through detailed mapping and geochronology. When additional work aimed at more-clearly identifying the ages and characteristics of specific dike swarms is complete, some fundamental questions about the Archean oceanic thermal gradient may be answered. For instance, how do the thickness and textures of the Archean dikes vary with depth, and are these changes related to the thermal profile in the Archean oceanic environment? This is especially critical because temperature gradients change dramatically from the base of the dike complex to the gabbro unit in ophiolites (Nicolas, 1989). The gabbro complex of the Dongwanzi ophiolite is up to 5 km thick and grades downward from isotropic gabbro, into faintly layered gabbro, into strongly layered gabbro, then into a mafic/ultramafic transition zone of mixed layered gabbro and cumulate ultramafic rocks (Figs. 7–9). The gabbro complex is metamorphosed to amphibolite facies. Layering in the gabbro ranges from several cm-thick discontinuous layers, through decameter thick layers, to faint layers several meters thick. These include cm to meter thick alternations between clinopyroxene and plagioclase rich layers, with individual layers varying in composition between anorthosite and clinopyroxenite (Fig. 9). In some locations in the thinly layered gabbro the gabbro exhibits folds that could be slumps resulting from layers that accumulated along walls of the magma chamber sliding down to the chamber floor. Thin, relatively homogeneous gabbro dikes a few cm to decimeters thick intrude the layered gabbro at small angles to the compositional layering, and pods of gabbro-pegmatite occur locally. The gabbro and mixed cumulate gabbro and ultramafic parts of the ophiolite grade gradually downward into a unit of about 50% coarse-grained gabbro and 50% pyroxenite (Fig. 9). This unit represents the “transition zone” between mafic rocks above, and ultramafic cumulates and depleted mantle below. These layers are locally cut by metabasaltic dikes with unknown ages, but are mineralogically and texturally similar to mafic dikes at higher levels of the ophiolite. The leucogabbro is locally strongly foliated, and the gabbro at this level contains many pods of ultramafic rocks. The lower part of the cumulate mafic/ultramafic complex consists almost entirely of ultramafic cumulates (Fig. 9). Orthocumulate pyroxenite, rare dunite, harzburgite (Fig. 12), and other peridotites are interlayered with olivine-pyroxene gabbro, and olivine gabbro layered cumulates. In a few localities layers grade from a dunite base up into wehrlite and clinopyroxenite tops. Thin layers and disseminated grains of chromite are present but rare (Fig. 9). However, we have documented numerous podiform chromitites in 200–400 m large tectonic blocks in mélange underlying the ophiolite, traced to the southwest 60 km along strike (see Huang et al., 2004). Small amounts of strongly tectonized and serpentinized ultramafic rocks have been found along the exposed base of the Dongwanzi ophiolite (Fig. 9). These rocks, showing field evidence for high-temperature deformation, include strongly foliated and lineated dunite and layered olivine-orthopyroxene harzburgite tectonite (Fig. 12), some containing

4. Field Description of Ophiolitic Units

241

Fig. 12. Photographs of ultramafic tectonite rocks from the southern part of the central belt.

individual septa of olivine and orthopyroxene crystals, and ultramafic mylonites. The pyroxene crystals in the peridotites are commonly aligned, and bands of chromite in dunite probably formed during early, high-temperature deformation. The ultramafic tectonites are extensively serpentinized making identification of primary mineralogy difficult. Despite the pervasive serpentinization, this unit is clearly more deformed than overlying units, it contains depleted harzburgite (with orthopyroxene in contrast to clinopyroxene dominated units in the crustal section), and preserves evidence for early, high-temperature deformation. We provisionally interpret this as the lower, residual or depleted mantle part of the ophiolite, from which the overlying magmatic rocks were extracted.

242

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

5. CONTACT RELATIONSHIPS The base of the Dongwanzi ophiolite is an early shear zone, locally in contact with biotite-gneiss of the Zunhua structural belt. It has been extensively intruded by the 2391 ± 50 Ma Cuizhanghizi diorite-tonalite complex (Zhang et al., 1991) and other tonalitic gneiss to the east, then deformed again after intrusion of the diorite complex. Diorite forms dikes, layer-parallel sills, and is strongly deformed. The base of gabbro/layered ultramafic cumulates and the ultramafic tectonites has progressively more diorite intruding it downward until the Dongwanzi ophiolite is only represented as ultramafic and garnetamphibolite inclusions of tens of cm to several meters in scale in the dioritic to tonalitic gneiss to the east (Kusky et al., 2001). The lower contact relationships require additional detailed mapping, especially because the area of the Zunhua structural belt southwest of the Dushan granite contains a large number of structurally-bounded lozenges of gabbro and ultramafic rock that may be dismembered parts of the base of the ophiolite. Rocks of the Zunhua Structural belt therefore preserve a much more extensive record of mantle processes than originally estimated (Li et al., 2002). The upper contact of the main thrust sheet containing the Dongwanzi ophiolite is also a fault, but it is at least in part Proterozoic or Mesozoic, as the ophiolite is imbricated with Mesoproterozoic quartzites. Regional relationships show that this is a Mesozoic thrust belt that imbricates the Dongwanzi ophiolite with Mesoproterozoic sediments (Xu, 1990; Li et al., 2000a; Davis et al., 1996; Ziegler et al., 1996). Additional work is needed to separate Mesozoic and Proterozoic structural elements from the older Archean record.

6. HIGH-GRADE OPHIOLITIC MÉLANGE IN THE ZUNHUA STRUCTURAL BELT The Zunhua structural belt (ZSB) is mainly an amphibolite-facies terrain separated from an Archean granulite-gneiss dome (3.85–2.50 Ga) of the Eastern Block by a major thrustsense shear zone (Figs. 6 and 13). It contains NE-striking, intensely strained gneiss and amphibolites exhibiting tight composite folds and ductile shear zones. Various thrust slices, such as TTG gneiss, mafic plutonic rocks, supracrustal sequences, mafic volcanics, BIF, garnet-bearing gneiss, and granites are tectonically intercalated with each other (Kusky et al., 2001; Wu et al., 1998; Wu and Zhong, 1998; Shen et al., 1992; Zhang et al., 1991; Fang et al., 1998). More than 1000 mafic to ultramafic boudins, ranging from several meters to several kilometers in length are recognized in the ZSB. Locally, layered gabbro and cumulate rocks form belts up to several kilometers long (Shen et al., 1992; Zhang et al., 1991; Zhang et al., 1986; Fang et al., 1998), that are intruded by ca. 2.56–2.50 Ga tonalitic gneiss (Wu and Geng, 1991; Wu et al., 1998). Mafic volcanics in the ZSB commonly have an oceanic affinity as shown by their flat REE patterns to LREE-depleted patterns, similar to basalts from suprasubduction and mid-oceanic ridge settings (Zhao, 1993; Wu et al., 1998; Kusky and Li, 2002). These mafic and ultramafic boudins occur in a fine-grained biotitegneiss matrix, and are interpreted as tectonic blocks in ophiolitic mélange (Li et al., 2002). Many parts of the mélange were intruded by tonalite and granodiorite, and deformed again.

6. High-Grade Ophiolitic Mélange in the Zunhua Structural Belt

243

Fig. 13. Detailed map of the western part of the Zunhua structural belt, showing the location of podiform chromite and dismembered ophiolitic ultramafic pods.

In the Paleoproterozoic, the ZSB was intruded by mafic dike swarms that are now metamorphosed, and ca. 2400 ± 15 Ma granite (Wu and Geng, 1991). Finally, it was overlain unconformably by unmetamorphosed Mesoproterozoic sequences (younger than 1.85 Ga), demonstrating that major tectonothermal episodes associated with the Zunhua structural belt occurred before 1.85 Ga. Mapping of ultramafic to mafic blocks within the Zunhua structural belt has resulted in the recognition of various rock types typical of oceanic crust and mantle. We interpret some large boudins as strongly dismembered fragments of an originally more continuous sequence, tectonically transposed with Neoarchean biotite-gneiss, BIF, and tonalite. The boudins are preserved as blocks in a high-grade metasedimentary and tonalitic gneiss (Li et al., 2002). The Dongwanzi ophiolite may represent one such unusually large and complete block, particularly the main part that contains sheeted dikes, gabbro, and cumulate ultramafic rocks. The smaller tectonic blocks and pods display typical ophiolitic rock types, including partly serpentinized harzburgite, peridotite tectonite, dunite, serpentinite, podiform chromitite, hornblendite, wehrlite, pyroxenite, metagabbro, cumulates and pillow lavas, massive metabasalt, and greenschist. Locally, well-preserved sheeted dikes of several meters width are recognized. Garnet-amphibolite blocks are also quite common. Granite intruding the mafic to ultramafic rocks have yielded ages of 2400 ± 15 Ma age (Wu and Geng, 1991). The mafic to ultramafic blocks preserve delicate textures and structures that are characteristic of ophiolites (Nicolas, 1989). In particular, the podiform chromitites provide very important information on the nature of the Archean oceanic mantle. Commonly,

244

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

it is generated at high-temperatures (1200–1300 ◦ C), and can escape late metamorphism and intense deformation. Ultramafic to mafic pods and tectonic blocks are stretched and occur within a strongly deformed matrix consisting of foliated and sheared, fine-grained biotite-gneiss and hornblende-gneiss with some layers of amphibolite and BIF (Figs. 13 and 14). These blocks are intensely sheared and tectonically transposed along their margins. In contrast, internal structures of the blocks commonly show distinct foliation, layering, and fold patterns, discordant to the external foliations outside the blocks in the surrounding shear zones. Gabbro and pyroxenite boudins exhibit well-preserved relict cumulate textures (Fig. 15). Metamorphic tectonite fabrics are well-preserved within the cores of blocks and pods, which are defined by oriented orthopyroxene porphyroclasts, strings of chromite, and elongated ribbons of olivine. Polyphase and isolated foliation patterns are recognized within larger peridotite blocks. The early tectonic fabrics defined by compositional layering include chromite seams, disseminated chromite, oriented nodular and orbicular chromite, and flattened antinodular chromite. In younger ophiolite complexes, these textures are commonly interpreted to form during high-temperature plastic deformation in the upper mantle associated with oceanic spreading (Nicolas and Azri, 1991; Zhao, 1993; LeBlanc and Nicholas, 1992; Leblanc, 1997; Matsumoto and Arai, 1999). The early high-temperature tectonite fabrics in the peridotite are cut by late steeplydipping shear zones, which are commonly parallel to tectonic contacts with country rocks. Serpentinization is concentrated along late shear zones and fractures cutting across the earlier foliation. Within these shear zones, ultramafic protoliths are separated into numerous small-scale pods and lenses of tens of centimeters to several meters scale, which are flattened and stretched. Many preserve asymmetric shapes. The presence of more than two sets of tectonic fabrics suggests that the harzburgitic mantle tectonites have been overprinted by late obduction-related deformation.

7. DESCRIPTION OF ROCK TYPES IN THE BLOCKS OF MÉLANGE Peridotites are mainly composed of serpentinized olivine, relict orthopyroxene, and minor chromite and magnetite (Fig. 16). Stretched orthopyroxene grains form augen up to 2–3 mm in diameter enclosed in a serpentinite matrix, and display ribbon-shaped tails. Some orthopyroxene porphyroclasts preserve embayed outlines associated with corrosion by melt. Relict olivine aggregates show elongated geometry (Fig. 16). Minor subhedral to euhedral chromites are present. Some peridotite tectonites show penetrative foliations and stretching lineations. High-temperature metamorphic tectonite fabrics defined by aligned and stretched olivine are well-preserved in cores of blocks (Fig. 16). Serpentinized dunite in the Zunhua structural belt typically exhibits well-defined layering defined by needle-like chromite grains. Relict olivine forms extended ribbons with asymmetrical geometry. Augen of olivine exhibit deformation kink bands (Fig. 16). Small amounts of chromite (< 5%) are euhedral to subhedral. These fabrics are attributed to man-

7. Description of Rock Types in the Blocks of Mélange 245

Fig. 14. Outcrop map of Zunhua structural belt podiform chromite area, showing locations of rock types in metasedimentary and metatonalitic gneiss. Modified from Li et al. (2002).

246

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 15. Podiform chromite photos. (a) Flattened lens of serpentinized harzburgite; (b) Serpentinized harzburgite block in mélange, surrounded by weathered biotite gneiss; (c) Gabbroic lens (coarse-grained, bottom of photo) within sheared biotite gneiss; (d) Ophiolitic mafic to ultramafic boudins as inclusion with granitic gneiss; (e) Pillow lava.

tle flow at high temperatures (Li et al., 2002). Olivine crystals are commonly serpentinized, with magnetite distributed along the foliation planes. Boudins and tectonic blocks of various types of gabbro range in size from 20 × 15 cm to 10 × 100 m. They occur within intensely sheared garnet-bearing gneiss (Fig. 15). A 400 × 60 cm block of cumulate gabbro and layered gabbro is identified within augen gneiss. Rarely, pods of dunite are recognized within olivine gabbro. These rocks are gen-

7. Description of Rock Types in the Blocks of Mélange

247

Fig. 16. Details of peridotite mineralogy and textures. (a, b) Polished surfaces of hand specimens illustrating principal microstructures and textures of flattened harzburgite core with rings of serpentinite. Early high-temperature foliations are preserved in cores of pods; (c) Harzburgite showing Opx, chromite, and olivine crystals; (d) Asymmetrical olivine porphyroclast with recrystallized tail within harzburgite tectonite; (e, f) Kink bands within relict olivine in serpentinized harzburgite.

erally less-deformed than the ultramafic complex. Cumulate textures including alternating pyroxene-rich and plagioclase-rich banding are locally preserved. The layers are generally about 2 to 5 cm thick, characterized by alternating internal texture and mineral composition, representing primary magmatic layering. They clearly underwent amphibolite-facies metamorphism as pyroxenite layers are commonly transformed into foliated hornblendite along their margins.

248

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Sheeted and multiple dike complexes are recognized from several places in the Zunhua structural belt. The width of individual half dikes are generally tens of centimeters. Many dikes have only one chilled margin, which is a consequence of repeated intrusions in the center of a single opening fissure. The chilled margins are recognized as 2–3 cm thick boundaries defined by strong alignment of fine-grained hornblende (Fig. 14). One dike complex can be traced for more than 100 m along strike. The mineral assemblage is made up of plag + cpx + hb, characteristic of amphibolite facies conditions. In one outcrop, more than ten dikes have been recognized over a distance of 4 m, with individual dikes being about 5–30 cm wide. They preserve differences in grain size, and several have slightly different mineralogy than adjacent dikes, suggesting a complex or polyphase history. Spectacular pillow lava structures are preserved locally in weakly deformed mafic volcanic domains (Figs. 14 and 15). Pillows vary in size from tens of centimeters to one meter, and are interbedded with amygdular massive basalt. The pillow lavas are pervasively altered to albite and chlorite assemblages, associated with hydrothermal alteration. Rarely, the pillows preserve dark cryptocrystalline margins, representing original glassy selvages. Younging toward the northwest is indicated by the shapes of pillows. Layers of pillow breccia and volcanoclastic sediment are intercalated with the pillow basalt, and these units are metamorphosed into plagioclase-biotite schist and biotite schist. Some ultramafic lenses are intercalated with pillow lavas, suggestive of large amounts of shearing. At least six large and numerous smaller chromite-rich peridotite massifs are recognized within the eastern Zunhua Structural Belt (Figs. 13 and 14). The chromitites are concordant with the foliation of the enclosing harzburgites, are commonly lens-shaped, and hosted in dunite envelopes within intensely serpentinized harzburgite. Serpentinized pods and lenses show concentric rings with systematic variations in mineral composition and textures. Outer rings are commonly 2–10 cm thick and composed of serpentine, whereas inner cores preserve dunite or massive harzburgite with clear tectonic foliations. Narrow deformed pyroxenite and gabbroic dikes (1–10 cm wide) with branches are recognized within serpentinized harzburgites, which are interpreted as melt channels or trapped parental melts to oceanic basalt (Fig. 17). Tectonic fabrics defined by chromite are well-preserved in the cores of the serpentinized pods. Open folds, tight folds and rootless folds are identified with the chromitite bands (Fig. 17). Dunite envelopes are commonly one to three centimeters wide, and separate chromite from harzburgite. They are a common feature of podiform chromites, and they are known to form almost exclusively in the mantle or crustmantle transition zone of suprasubduction zone (harzburgite type) ophiolites of different ages (Nicolas and Azri, 1991; Zhou et al., 1996; Bai et al., 1992; Edwards et al., 2000; Li et al., 2002). Most of the chromitites are strongly deformed by plastic flow, although nodular chromitites are locally preserved, especially in discordant pods. Nodular, orbicular, banded, massive, antinodular, and disseminated chromitite textures are all present (Fig. 17), and in many places grade into each other. Discordant structures are preserved within weaklystrained domains. Nodular textures consist of small balls of chromite in a dunite matrix, whereas orbicular chromitites consist of thin-rings of chromite surrounding cores of dunite. Nodular and orbicular chromite textures are only known to form in ophiolitic settings.

7. Description of Rock Types in the Blocks of Mélange

249

Fig. 17. Polished surfaces of handspecimens illustrating principal microstructures and textures of chromitites ores. Scale bar is 1 cm in all photos. (a) Banded chromitites with dunite envelop; (b) Rootless fold showing thickening of chromitite band; (c) Chromite vein within serpentinized dunite; (d) Nodular chromites in weakly-deformed domain; (e) Nodular and orbicular chromitites; (f, g) Antinodular chromite; (h) Layered antinodular chromites.

250

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Nodular and orbicular chromites, with diameters of 2–10 mm and occasionally larger than 10 mm, are generally flattened into elliptical shapes, and some nodules form flattened rings. They occur together or separately. Some nodular and orbicular chromites grade into massive chromites, veins, or disseminated ores. Nodular and orbicular textures are the most typical magmatic structures of ophiolitic chromitites (Nicolas, 1989; Nicolas and Azri, 1991). Pull-apart textures are common in the massive, banded, and antinodular chromitite deposits, and similar textures have been attributed to high-temperature (> 1000 ◦ C) mantle deformation in other ophiolites (e.g., Holtzman, 2000; Nicolas, 1989). Abundant olivine occurs as rounded inclusions in the chromitite, although they are widely altered into serpentinites. Orthopyroxene porphyroclasts show asymmetrical recrystallized tails indicating high-temperature shearing. Rounded to flattened chromite balls are characteristic of nodular chromitite in the Zunhua structural belt, with the diameter of the chrome nodules ranging from 0.2 to 1 cm. Commonly, the nodules impinge into orbicules, deforming the initially spherical rings. The nodules and orbicules show patterns of flattening and mutual impression along their contacts with each other (Figs. 17 and 18), suggesting that they settled while they were still soft. Most nodules with flattened geometry are oriented parallel to the foliation. Locally, nodules are sorted into layers by their sizes. These features are interpreted to be a result of rapid deposition of chromite nodules while they were still plastic (Lago et al., 1982). The nodules and orbicules commonly exhibit stretching fabrics interpreted to have formed when they were still in liquid form (Li et al., 2002). They are elongated by plastic strain and show a preferred orientation, forming easily recognized lineations in many samples. The outer boundary of single nodules are typically smooth and rounded (Figs. 17 and 18). In contrast, their inner parts display individual chromite grains that grew inward. The development of nodular chromites record dynamic magmatic flow or partial melting conditions, needed to keep chromite suspended and growing concentrically into the magma (Edwards et al., 2000). The delicate magmatic structures preserved show that they have not been significantly deformed after their formation, which is attributed to their rigid nature. In some cases, nodules grade into antinodules in the same hand-specimen. They record magmatic growth and settling in the upper mantle (e.g., Edwards et al., 2000). These original magmatic structures are commonly destroyed or become incomplete as the shear strain increases. For example, nodular and orbicular textures are strongly stretched and transposed into layering or antinodular chains. Compared with nodules, orbicules are more strongly stretched, their ratio of X/Z are up to 5:1. Minor orthopyroxene occurs as porphyroclasts. Rounded inclusions of olivine are recognized within chromite grains, and some inclusions of olivine show kink bands (Fig. 16), recording plastic deformation before or during growth of the chromite. We attribute this deformation to upper mantle flow in the oceanic mantle. Antinodular chromitites contain 30–50% chromite. They consist of rounded or flattened dunite aggregates surrounded by chromite chains or networks (Fig. 17). Olivine composes the cores surrounded by chromite net-like bands, suggestive of strong flattening or extension. Flattened antinodular texture is typical of plastic deformation in oceanic mantle, which is a result of straining of weaker olivine inclusions in a chromite-rich matrix

7. Description of Rock Types in the Blocks of Mélange

251

Fig. 18. Microscopic textures of chromite ores. (a) Flattened orbicule of chromite (reflected light); (b) Flattened nodules of chromite and igneous contact between two nodules (reflected light); (c) Flattening of contact between orbicule and nodule of chromite; (d) Flattening of chrome nodule with asymmetrical tail; (e) Inclusion of orthopyroxene (OPX) and olivene (OL) showing kink bands within chromites; (f) Deformation textures of chromitite ore showing stretched olivine crystals in dunite core (strain ellipse outlined in black) with antinodular chromite; (g) Pulled apart chromites; (h) Cumulate layer of chromite nodules.

252

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

(Nicolas, 1989). Alignment of needle-like chromite also indicates strong shearing. Disseminated chromitites contain 10–30% chromite, with grain sizes of chromite being larger than 0.1 mm. The disseminated chromites locally grade into nodular chromite, and also locally have nodules with diameters of 2–5 mm scattered throughout the disseminated ores. In disseminated chromitites, chromite grains are uniformly scattered in an olivine matrix. Layered and banded textures consist of anastomosing chains of chromite surrounding ovoids of olivine, which were generated by shearing, rather than crystallization and accumulation. They are characterized by intense shearing. Locally, they are cross-cut by late, less than one cm wide veins of chromite, suggesting that at least two phases of chromites were originally present. Banded and massive chromitites show characteristic banding or layering, with 2–4 cm thick bands of chromite (Fig. 17). In a few places they are crosscut by pyroxenite and dunite dikes. Some chromite layers occur as rootless to tight folds or asymmetric lensoidal boudins, and other layers and lenses consist of nodules. Minor grains of orthopyroxene are preserved. Olivine is commonly serpentinized. Massive ores contain at least 70% fine-grained chromite distributed homogeneously in a serpentinite matrix. Massive chromitites may grade into disseminated chromites, whereas other layers display podiform geometry with indicators of pre-full crystallization shearing. Pull-apart textures are filled by olivine (Li et al., 2002), forming indicators of magmatic shearing with the massive chromites.

8. CHEMISTRY OF CHROMITITE DEPOSITS Chromite from two samples of chromitite in the Zunhua ophiolitic mélange (dn, dn2), and one from dunite with disseminated chromite from the Dongwanzi ophiolite (dd) were analyzed by WDS on a Cameca S × 100 at the Manchester Electron Microprobe Facility (MEMF), UK. The operating conditions for WDS analysis were an acceleration voltage of 20 kV, a counting time of 20 s, and a 20 nA beam for Cr, Al, Mg and Fe with a counting time of 80 s with a 100 nA beam for Zn, Co, Ti, V, Mn, Ni and Si. Natural oxide and synthetic standards were used, with the manufacturers ZAF correction program. Iron was determined as FeOt and FeO and Fe2 O3 were calculated by stoichiometry using the method of Droop (1987). Detection limits are between 10 and 20 ppm. The data is shown in Table 1. Twenty three additional samples from the Zunhua chromitites were analyzed using the WDS techniques on a Cameca-50 at the Geological Institute Academia Sinica in Beijing. These data are shown in Table 2. The chromite from the ultramafic fragments in the mélange beneath Dongwanzi ophiolite has very refractory composition, with Cr#s Cr/(Cr + Al) of 0.74 to 0.93, TiO2 at less than 0.3 wt%, and V2 O5 wt% at less than 0.1 for the samples analyzed; with the exception of sample 1525 which has a Cr# of 0.42. Excluding sample 1525 there are two distinct groups; massive chromitite and nodular chromite in dunite (samples dn and dn2) and disseminated chromite in dunite (dd). These groups have almost identical TiO2 wt% but the disseminated chromite has a lower Cr# and higher V2 O5 content. The Fe2 O3 wt% in the samples analyzed is variable, with a range from 3.2 to 37.2, though with the exception of

8. Chemistry of Chromitite Deposits

253

Table 1a. Major element concentrations of chromite measured on the Manchester electron microprobe facility Sample Oxide wt% Al2 O3 Cr2 O3 dn2 5.05 59.67 dn2 5.53 60.52 dn2 5.18 59.15 dn2 5.19 59.00 dn2 5.32 59.65 dn2 4.91 58.67 dn2 5.24 58.41 dn2 5.19 58.18 dn2 5.29 56.92 dn2 4.50 58.14 dn2 5.36 59.35 dn2 5.15 60.59 dn2 5.11 58.97 dn2 4.84 59.61 dn dn dn dn dn dn dn dn dn dn dn dn dn dn dn

4.88 4.68 5.06 4.98 4.94 4.97 4.81 4.92 4.93 4.91 4.95 4.92 4.80 5.01 4.91

dd dd dd dd dd dd dd dd dd dd dd dd dd dd

10.75 10.72 10.83 10.06 10.33 10.77 9.88 10.35 10.41 10.63 10.20 10.91 10.78 10.19

Total MgO 6.25 7.73 4.93 4.79 5.95 4.28 3.54 4.06 3.36 4.51 5.41 6.29 4.41 4.73

MnO 0.49 0.45 0.48 0.49 0.49 0.51 0.51 0.51 0.53 0.51 0.50 0.49 0.52 0.51

TiO2 0.14 0.14 0.15 0.12 0.18 0.13 0.13 0.14 0.10 0.10 0.15 0.10 0.15 0.09

ZnO 0.05 0.05 0.06 0.05 0.04 0.05 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.06

SiO2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.15 0.00 0.00 0.00

V2 O5 0.03 0.04 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03

NiO Fe2 O3 0.07 5.89 0.07 5.18 0.06 5.36 0.06 5.31 0.07 5.41 0.07 5.91 0.06 5.25 0.06 5.87 0.05 7.02 0.07 7.24 0.07 4.91 0.07 5.01 0.07 5.49 0.05 5.24

FeO 22.99 20.83 25.01 25.14 23.57 25.91 27.04 26.28 27.43 25.50 24.50 22.98 25.81 25.15

100.63 100.54 100.41 100.17 100.73 100.47 100.27 100.37 100.80 100.64 100.48 100.77 100.62 100.32

61.32 8.48 62.05 8.39 60.94 8.49 60.78 8.38 60.94 8.51 61.39 8.64 62.09 8.72 61.76 8.98 61.94 8.71 61.45 8.72 62.32 9.02 61.86 8.83 62.72 10.17 62.30 9.06 61.87 8.82

0.48 0.48 0.48 0.48 0.48 0.48 0.47 0.47 0.47 0.47 0.47 0.48 0.42 0.46 0.46

0.16 0.15 0.17 0.16 0.17 0.17 0.16 0.16 0.17 0.16 0.16 0.14 0.16 0.17 0.16

0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.05 0.06

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.15 0.00 0.00 0.00 0.00 0.11 0.08 0.00

0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03

0.13 0.11 0.13 0.13 0.12 0.11 0.13 0.12 0.10 0.12 0.11 0.11 0.12 0.12 0.12

5.52 4.65 5.80 5.49 5.59 4.68 4.49 4.27 4.32 5.10 3.94 4.93 4.13 3.98 4.31

19.52 19.48 19.59 19.49 19.38 18.97 18.96 18.68 18.95 19.00 18.40 18.89 16.81 18.56 18.66

100.55 100.06 100.74 99.98 100.20 99.49 99.88 99.60 99.67 100.02 99.46 100.23 99.50 99.80 99.39

46.20 46.29 46.17 47.23 46.87 44.99 46.16 45.63 46.40 44.16 44.88 46.45 47.01 47.00

0.56 0.54 0.53 0.57 0.58 0.58 0.58 0.58 0.55 0.59 0.60 0.55 0.53 0.58

0.19 0.20 0.20 0.17 0.18 0.19 0.11 0.20 0.18 0.15 0.11 0.20 0.19 0.17

0.22 0.20 0.20 0.22 0.23 0.23 0.21 0.22 0.19 0.23 0.22 0.20 0.18 0.22

0.03 0.01 0.00 0.00 0.00 0.02 0.03 0.02 0.02 0.01 0.00 0.01 0.01 0.02

0.08 0.08 0.09 0.08 0.08 0.08 0.08 0.09 0.08 0.08 0.08 0.09 0.08 0.08

0.13 0.13 0.13 0.11 0.11 0.13 0.12 0.13 0.13 0.11 0.07 0.13 0.12 0.11

11.45 11.81 11.83 11.61 12.07 12.43 12.74 12.38 11.99 13.33 12.72 11.64 11.54 11.65

25.45 25.23 24.95 25.69 26.39 26.33 26.00 26.40 25.05 27.29 27.42 25.19 24.50 26.07

100.03 100.40 100.30 100.55 101.40 100.15 100.45 100.34 100.25 100.32 99.80 100.66 100.71 100.72

4.97 5.20 5.38 4.83 4.56 4.40 4.56 4.35 5.25 3.75 3.51 5.31 5.77 4.63

254

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Table 1b. Trace element concentrations of chromite measured on the Manchester electron microprobe facility Sample Molar concentration in ions Al Cr Mg Mn Ti Zn dn2 0.21 1.66 0.33 0.01 0.00 0.00 dn2 0.23 1.66 0.40 0.01 0.00 0.00 dn2 0.22 1.66 0.26 0.01 0.00 0.00 dn2 0.22 1.67 0.25 0.01 0.00 0.00 dn2 0.22 1.66 0.31 0.01 0.00 0.00 dn2 0.21 1.66 0.23 0.02 0.00 0.00 dn2 0.22 1.66 0.19 0.02 0.00 0.00 dn2 0.22 1.65 0.22 0.02 0.00 0.00 dn2 0.23 1.62 0.18 0.02 0.00 0.00 dn2 0.19 1.65 0.24 0.02 0.00 0.00 dn2 0.22 1.66 0.28 0.01 0.00 0.00 dn2 0.21 1.68 0.33 0.01 0.00 0.00 dn2 0.22 1.66 0.23 0.02 0.00 0.00 dn2 0.20 1.68 0.25 0.02 0.00 0.00

Si 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00

V 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Ni 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Fe3+ 0.15 0.13 0.14 0.14 0.14 0.16 0.14 0.16 0.19 0.19 0.13 0.13 0.14 0.14

Fe2+ 0.66 0.60 0.73 0.74 0.68 0.76 0.80 0.77 0.81 0.75 0.71 0.66 0.76 0.74

O 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00

Total Ratios Cr# Mg# 3.04 0.89 0.33 3.04 0.88 0.40 3.04 0.88 0.26 3.04 0.88 0.26 3.04 0.88 0.31 3.04 0.89 0.23 3.04 0.88 0.19 3.04 0.88 0.22 3.05 0.88 0.18 3.05 0.90 0.24 3.04 0.88 0.29 3.04 0.89 0.33 3.04 0.89 0.24 3.04 0.89 0.25

Fe3+ # 0.076 0.066 0.070 0.069 0.070 0.077 0.069 0.077 0.092 0.094 0.064 0.064 0.071 0.068

dn dn dn dn dn dn dn dn dn dn dn dn dn dn dn

0.20 0.19 0.21 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.19 0.20 0.20

1.68 1.71 1.67 1.68 1.68 1.69 1.70 1.69 1.70 1.69 1.71 1.69 1.71 1.70 1.70

0.44 0.44 0.44 0.44 0.44 0.45 0.45 0.46 0.45 0.45 0.47 0.46 0.52 0.47 0.46

0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.14 0.12 0.15 0.14 0.14 0.12 0.12 0.11 0.11 0.13 0.10 0.13 0.11 0.10 0.11

0.56 0.56 0.56 0.56 0.55 0.55 0.54 0.53 0.54 0.54 0.53 0.54 0.48 0.53 0.54

4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00

3.04 3.04 3.04 3.04 3.04 3.04 3.03 3.03 3.03 3.04 3.03 3.04 3.03 3.03 3.03

0.89 0.90 0.89 0.89 0.89 0.89 0.90 0.89 0.89 0.89 0.89 0.89 0.90 0.89 0.89

0.44 0.44 0.44 0.44 0.44 0.45 0.45 0.46 0.45 0.45 0.47 0.46 0.52 0.47 0.46

0.070 0.059 0.073 0.070 0.071 0.060 0.057 0.055 0.055 0.065 0.050 0.063 0.053 0.051 0.055

dd dd dd dd dd dd dd dd dd dd dd dd dd dd

0.45 0.45 0.45 0.42 0.43 0.46 0.42 0.44 0.44 0.45 0.44 0.45 0.45 0.43

1.30 1.30 1.29 1.33 1.31 1.28 1.31 1.29 1.31 1.26 1.29 1.30 1.31 1.32

0.26 0.28 0.28 0.26 0.24 0.24 0.24 0.23 0.28 0.20 0.19 0.28 0.30 0.25

0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02

0.01 0.01 0.01 0.00 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.01 0.00 0.00

0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.01

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.30 0.30 0.30 0.30 0.31 0.32 0.33 0.32 0.31 0.35 0.33 0.30 0.29 0.30

0.73 0.72 0.71 0.74 0.75 0.76 0.75 0.76 0.72 0.79 0.80 0.72 0.69 0.75

4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00

3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08

0.74 0.74 0.74 0.76 0.75 0.74 0.76 0.75 0.75 0.74 0.75 0.74 0.75 0.76

0.27 0.28 0.29 0.26 0.24 0.24 0.25 0.23 0.28 0.20 0.19 0.28 0.30 0.25

0.144 0.148 0.148 0.146 0.151 0.157 0.160 0.156 0.151 0.168 0.162 0.145 0.144 0.146

15z12

z2

15z15

1535

1552

1553

1555

1566

1572

1577

1590

1515

1589

1578

1579

1584

SiO2 (%) TiO2 Al2 O3 Fe2 O3 (T) MnO MgO CaO Na2 O K2 O P2 O5 Cr2 O3 NiO (ppm) FeO (%) L.O.I (%) Ba (ppm) Co Cu Ga V Zn

39.4989 0.002 0.0733 6.9663 0.0522 37.8122 0.0854 0.0182 0.0071 0.0121 0.7135 5798 1.6 13.61 0 107.101 0.45 1.0319 18.0361 74.0762

39.2334 0.0049 0.0766 5.2075 0.0611 42.2416 0.4468 0.0201 0.0084 0.0103 0.7032 3917 1.48 11.98 0 95.8663 0.3309 5.0339 14.0914 29

39.8596 0.0048 0.1023 6.3428 0.0552 37.8945 0.0583 0.0671 0.0092 0.0123 0.7862 5430 0.52 13.76 4.5636 128.9148 2.1103 4.2656 14.2615 49.5923

40.3738 0.0278 0.8517 10.8189 0.1035 35.314 0.0664 0.015 0.0148 0.0159 1.1852 2785 4.3 10.66 6.3761 78.8935 4.1154 4.2826 29.5437 93.9131

36.1557 0.0111 0.5285 4.9546 0.0544 37.6019 0.7603 0.0354 0.008 0.0107 5.9187 3842 0.72 12.87 13.1976 96.266 2.0759 3.0222 24.9442 30.822

38.56 0.0063 0.0732 5.6639 0.0616 38.5175 0.3381 0.0031 0.007 0.013 0.7744 4116 0.55 15.09 4.7537 101.535 1.9214 4.7669 12.8521 40.7071

33.1993 0.012 0.5382 4.6792 0.0758 37.6144 0.1986 0.0147 0.0062 0.0107 7.4619 4117 0.65 16.16 3.2319 96.5728 0.6746 6.1056 25.1216 36.6401

42.0188 0.0236 0.5628 13.0095 0.1903 33.5305 0.1212 0.0673 0.0091 0.0203 0.529 3622 6.15 9.69 385.001 91.9813 13.9635 5.5092 25.4958 164.5144

39.418 0.0083 0.2647 5.7205 0.0302 38.7221 0.0472 0.0122 0.0093 0.0092 1.1029 3787 1 13.59 5.8972 73.7451 1.419 0.9256 21.9646 48.7478

34.6402 0.0221 0.7825 5.3911 0.0482 35.5799 0.0991 0.0472 0.0077 0.0093 10.0148 3693 0.9 12.33 8.61 95.8065 0 4.7213 26.1656 44.3615

36.1302 0.0167 0.509 8.295 0.043 35.6311 0.0734 0.0503 0.0098 0.0078 5.8001 3795 1.6 12.36 8.5551 88.6797 0 2.5772 22.6376 59.5131

32.0265 0.05 1.3361 8.959 0.1505 31.7702 1.3642 0.0399 0.0077 0.0071 13.4612 3076 2.89 10.22 38.8209 108.7291 0 6.4718 49.0554 112.784

35.718 0.006 0.0932 10.1639 0.0656 35.251 1.7351 0.0516 0.0062 0.0087 0.8231 4546 2.3 14.55 11.3105 80.3125 0 4.0258 14.6973 30.7199

38.439 0.0084 0.2498 5.3177 0.1038 37.7137 0.9868 0.0054 0.0074 0.0089 0.8352 3623 1.33 15.32 7.3739 83.9053 1.9759 2.625 12.5445 43.555

33.6712 0.0259 0.8663 6.1065 0.0664 34.6841 0.2764 0.0306 0.0057 0.007 11.2609 3189 0.95 12.01 35.7902 93.1489 2.2691 5.9127 32.4333 59.6983

39.9353 0.0077 0.1652 6.3332 0.0346 37.5135 0.076 0.0071 0.008 0.0063 1.3413 3589 1.35 13.5 8.7355 75.9036 2.3433 5.2211 16.2245 60.8765

8. Chemistry of Chromitite Deposits

Table 2a. Major element concentrations of chromite measured on the Academia Sinica electron microprobe facility Sample No.

255

256

Table 2b. Trace element concentrations of chromite measured on the Academia Sinica electron microprobe facility z2 1.936 3.285 91.961 0.596 23.979 0.263 0.273 0.079 0.014 0.48 0.84 0.088 0.273 0.04 0.012 0.048 0.004 0.028 0.006 0.019 0.004 0.026 0.008 0.007 0.058 2.212 0.148 0.203

15z12 1.404 7.423 112.673 0.6 4.08 0.84 0.312 0.029 0.12 0.57 0.829 0.138 0.557 0.105 0.031 0.118 0.014 0.085 0.02 0.073 0.012 0.115 0.023 0.001 0.019 3.691 0.079 0.935

15z15 1.228 11.114 126.154 1.266 2.799 0.945 0.322 0.126 0.216 0.773 0.983 0.218 0.903 0.168 0.038 0.172 0.018 0.106 0.025 0.077 0.011 0.092 0.016 0.003 0.018 1.123 0.09 1.481

1515 2.344 54.139 125.001 0.484 43.455 0.952 1.689 0.271 0.015 0.688 1.622 0.217 0.931 0.214 0.237 0.21 0.025 0.145 0.025 0.082 0.009 0.065 0.008 0.033 0.038 48.536 0.092 0.165

1535 3.943 31.714 77.773 2.651 3.74 4.567 1.992 2.152 0.042 1.954 4.464 0.574 2.468 0.689 0.184 0.82 0.129 0.769 0.155 0.482 0.074 0.522 0.079 0.112 0.862 16.165 0.733 0.177

1544 1.494 140.925 227.269 5.599 1.194 1.036 11.454 1.004 0.134 0.849 1.32 0.189 0.727 0.138 0.03 0.144 0.023 0.166 0.033 0.114 0.015 0.11 0.015 0.481 0.371 6.651 0.426 0.291

1552 1.804 21.58 101.468 2.321 20.631 0.263 0.597 0.209 0.137 0.348 0.751 0.065 0.224 0.034 0.025 0.044 0.006 0.03 0.006 0.017 0.003 0.021 0.003 0.049 0.096 1.253 0.16 0.145

1553 1.964 4.742 94.487 0.56 7.319 0.471 0.234 0.061 0.052 0.27 0.459 0.061 0.204 0.04 0.013 0.044 0.007 0.043 0.009 0.033 0.007 0.053 0.012 0.005 0.035 0.98 0.146 0.172

1555 1.679 14.263 100.47 3.443 23.001 2.693 0.315 0.077 0.064 4.918 10.846 0.718 2.38 0.458 0.064 0.462 0.062 0.35 0.07 0.218 0.036 0.246 0.043 0.007 0.05 6.68 3.088 0.221

1566 5.012 23.874 83.484 10.584 37.549 6.784 3.402 0.669 0.188 5.767 12.774 1.306 4.671 0.961 0.306 1.015 0.147 0.895 0.192 0.622 0.107 0.817 0.128 0.117 0.096 190.611 3.241 0.244

1572 3.596 18.567 85.176 13.925 17.788 4.751 1.632 0.213 0.318 3.082 7.056 0.754 2.688 0.666 0.113 0.689 0.106 0.603 0.123 0.378 0.06 0.42 0.064 0.076 0.033 5.241 5.623 4.165

1577 1.7 24.451 123.277 1.071 1.669 0.305 0.295 0.009 0.047 0.276 0.421 0.056 0.225 0.043 0.037 0.047 0.007 0.046 0.009 0.031 0.006 0.033 0.004 0.009 0.006 0.804 0.018 0.1

1578 1.775 4.554 78.029 0.434 21.213 0.556 0.245 0.036 0.134 0.228 0.365 0.068 0.286 0.065 0.037 0.077 0.011 0.076 0.018 0.063 0.01 0.076 0.015 0.002 0.033 1.856 0.018 0.024

1579 1.66 30.424 124.14 0.98 3.159 0.386 0.31 0.012 0.015 0.248 0.596 0.086 0.354 0.075 0.03 0.056 0.008 0.055 0.013 0.043 0.008 0.052 0.008 0.011 0.006 3.554 0.071 0.619

1584 2.792 15.573 71.899 2.803 3.376 0.629 0.414 0.067 0.049 0.546 1.082 0.138 0.526 0.095 0.053 0.118 0.016 0.109 0.023 0.068 0.012 0.086 0.017 0.01 0.036 13.349 0.024 0.062

1589 2.007 5.202 79.679 1.225 31.353 1.139 0.33 0.055 0.029 1.141 4.02 0.267 0.947 0.178 0.037 0.209 0.03 0.187 0.04 0.13 0.022 0.15 0.025 0.006 0.036 1.583 0.258 0.17

1590 2.134 24.286 86.951 2.841 2.499 0.734 1.202 0.162 0.048 0.607 1.343 0.161 0.617 0.123 0.03 0.14 0.021 0.127 0.028 0.09 0.015 0.107 0.02 0.042 0.104 0.423 0.058 0.079

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Sample No. Sc V Co Rb Sr Y Zr Nb Cs La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U

9. Interpretation of Chrome Chemistry

257

sample 1525f it is within the range 3.2–13.2. The high Fe2 O3 content in sample 1525f is caused by alteration from chromite to ferrit-chromite, as described in Ulmer (1974), and as such is not considered to a represent crystallization composition.

9. INTERPRETATION OF CHROME CHEMISTRY Chromite analysis has a long history in the interpretation of ophiolitic rocks (Irvine, 1965, 1967; Dick and Bullen, 1984; Kamenetsky et al., 2001; Matveev and Balhaus, 2002) and for Archean sequences (Stowe, 1994; Cotteril, 1969; Chadwick and Crewe, 1986). Chromite is used as it is often the only unaltered mineral and the composition is believed to represent crystallization compositions (Irvine, 1967; Dick and Bullen, 1984; Arai, 1992; Ahmed et al., 2001), particularly for chromite from lava flows, and within massive chromitite, and dunite. The low diffusivity of Cr, Al, Ti and V, within olivine, makes these elements ideal for indicating crystallization compositions (Arai and Yurimoto, 1995). The Cr# indicates the degree of depletion of the mantle source from which the melt was derived, and to a lesser extent, the degree of melt fractionation (Irvine, 1965, 1967; Dick and Bullen, 1984). The TiO2 wt% indicates the degree of depletion of the mantle source and the degree of melt fractionation (Shervais, 1982; Dick and Bullen, 1984; Arai, 1992; Kamenetsky et al., 2001). Vanadium content indicates the degree of depletion of the mantle source, the degree of melt fractionation, and the fO2 during partial melting, melt fractionation and chromite crystallization (Shervais, 1982; Canil, 1999). Fe2 O3 content of spinel has been cited as indicating the fO2 of the mantle source, magmatic system and crystallization environment (Wood and Virgo, 1989; Arai, 1992; Kepezhinskas et al., 1993). Interpretation of chromite compositions, through geochemical modeling of the chromite forming magmatic system, and comparison with both Archean greenstone belts and Phanerozoic SSZ ophiolite complexes allow inferences to magma type and tectonic setting to be made. Power et al. (2000) contest that the use of chrome spinel chemistry for the interpretation of the tectonic setting of dismembered and deformed ophiolitic and other igneous complexes is inherently unreliable. They suggest that changes in chromite composition are caused by sub-solidus equilibrium, serpentinization of the host rock, and metasomatism, all of which result in an increase in Fe and Ti contents. However, with care during analysis, altered chromite can easily be identified. BSE imaging can identify alteration, and scrutinization of the chemical results can reveal which samples are altered (e.g., sample 1525f is obviously ferrit-chromite). The style of chromite mineralization, chromitite grading into dunite and eventually harzburgite, is consistent with formation through reaction between an exotic melt and a harzburgitic mantle (Kelemen, 1990; Kelemen et al., 1990; Zhou et al., 1994, 1996; Oberger et al., 1995; Zhou and Robinson, 1997; Arai, 1997a, 1997b). Cr# is often similar between mantle-hosted and crustal cumulate chromitites (e.g., Dick and Bullen, 1984; Oberger et al., 1995).

258

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

The high Cr# (0.74 to 0.93) of the chromite (Figs. 19 and 20), from the mélange beneath the Dongwanzi ophiolite, is indicative of a very high degree of mantle partial melting, or partial melting of a previously depleted mantle source. In modern and Phanerozoic ophiolitic environments, Cr#s this high are associated with boninitic or arc picrite magmatism in SSZ spreading center and island arc tectonic environments, with slab fluid aided melting of previously depleted mantle that has previously been depleted by melt extraction (Dick and Bullen, 1984; Jan and Windley, 1990; Kamenetsky et al., 2001; Parkinson and Pearce, 1998). However, higher geothermal gradients in the Archean may have produced a melt rich enough in Cr to form high Cr# chromite without previous partial melting of the mantle. High Cr# chromite could also form at a MOR environment after previous extraction of a more depleted melt such as a komatiite. Both modern and Archean komatiites have lower Cr#s, typically 0.5–0.6 (Echeverria, 1980) and 0.66–0.77 (Zhou and Kerrich, 1992) respectively, than those associated with the Dongwanzi ophiolite and Zunhua ophiolitic mélange blocks. However, these Cr#s are higher than those from MOR chromite, less than 0.0–0.7 (Dick and Bullen, 1984; Kamenetsky et al., 2001). It is likely that a melt produced from a source that has previously been depleted by a melt, that was the result of a greater degree of partial melting than MORB, would produce chromite with such high Cr#s. The TiO2 wt% of less than 0.3 wt% (Tables 1 and 2, Figs. 19 and 20) is typical of chromite from SSZ ophiolites (Dick and Bullen, 1984; LeBlanc and Nicholas, 1992) and differs from deposits associated with komatiites, which have greater than 0.3 wt% TiO2 (Zhou and Kerrich, 1992) and those from modern MOR, OIB and LIP settings (Dick and Bullen, 1984; Kamenetsky et al., 2001). A comparison of the Cr# with the TiO2 wt% (Fig. 19) indicates that within a deposit Cr# is roughly constant whilst TiO2 wt% is variable. The variable TiO2 wt% could be the product of alteration, or reflect differing magma Ti contents during chromite crystallization. Chromite from the Dongwanzi and Zunhua fragments have much lower V2 O5 (0.03 to 0.085) than chromite in Phanerozoic SSZ complexes (e.g., Jijal (0.1–0.3; Glass, unpublished); Border Ranges (0.07–0.25; Kusky et al., in preparation); Mayarí-Barcoa (0.07–0.18; Proenza et al., 1999) and komatiite associated chromite deposits; Belingwe (0.17–1.42; Zhou and Kerrich, 1992) and Shurugwi (0.1–0.27; Stowe, 1987). The low V (Tables 1 and 2, Fig. 19) could be caused by: (1) High fO2 during magmatic fractionation and chromite crystallization: High fO2 will cause more vanadium to be in the V4+ and V5+ oxidation states (Shervais, 1982) and behave incompatibly with respect to the crystallizing chromite (Canil, 1999). Low fO2 will cause more V to be in the V3+ oxidation state and enter the chromite lattice more easily. Therefore the low V values associated with the Dongwanzi chromite could be caused by high fO2 in the magmatic system during chromite crystallization. (2) Low fO2 during mantle partial melting: High fO2 will cause more vanadium to be in the V4+ and V5+ oxidation states and behave incompatibly with respect to the mantle during partial melting (Shervais, 1982). Low fO2 will cause more V to be in the V3+ oxidation state. V3+ behaves compatibly with respect to the mantle and is re-

9. Interpretation of Chrome Chemistry

259

Fig. 19. Mineral chemistry plots of TiO2 vs Cr#, V2 O5 vs Cr#, and V2 O5 vs TiO2 for Zunhua and Dongwanzi chromites.

260

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 20. Geochemical plots of TiO2 vs Fe3 + (Fe3 + Cr + Al) and Cr# vs Mg# for Zunhua and Dongwanzi chromites. Mg# is Mg/(Mg + Fe); Cr# is Cr/(Cr + Al).

9. Interpretation of Chrome Chemistry

261

tained in spinel (Canil, 1999). Therefore, low V chromite could be caused by low fO2 during partial melting. In the case of the chromites associated with the Dongwanzi ophiolite and Zunhua mélange this is considered unlikely, as the Cr2 O3 content is so high and V3+ behaves similarly to Cr (Canil, 1997, 1999). Archean komatiite lavas had a similar or higher fO2 than modern oceanic basalts (Canil, 1997). If this represents the mantle source and magmatic system, it is logical to assume that an Archean SSZ spreading center magma would be highly oxidizing and thus vanadium would be dominantly in the V4+ and V5+ oxidation state, and not enter the chromite crystal lattice. (3) Previous mantle partial melting: Chromite associated with Archean komatiitic magmas has high V2 O5 contents (Canil, 1997; Zhou and Kerrich, 1992) and indicates substantial extraction of vanadium from the mantle. Subsequent melts produced are likely to be low in vanadium. Any significant melt extraction under high fO2 will deplete the mantle of vanadium, as more vanadium would have been in the V4+ and V5+ oxidation states and thus incompatible with respect to the mantle (Shervais, 1982; Canil, 1997, 1999). There is an inverse relationship between Cr# and V content (Fig. 19) for the chromites from the Zunhua structural belt. This indicates either chromium occupying the octahedral site, in the chrome spinel lattice, in preference to V3+ , or a different magmatic history, as the higher V values are from disseminated chromite in dunite. Mg and Fe in chromite have also been used to indicate magmatic compositions. Mg and Fe contents may be reliable indicators for rapidly cooled volcanic rocks (Arai, 1992) but in intrusive and mantle deposits Mg and Fe in chromite will undergo subsolidus diffusion with olivine (Irvine, 1965; Roeder et al., 1979; Sack and Ghiorso, 1991). Fig. 20 shows a wide Mg# in relation to the Cr#, indicating extensive re-equilibration of Mg and Fe which precludes the use of these elements as petrogenetic indicators. The Fe2 O3 content of chromite is dependent on the amount of fractionation (Irvine, 1974; Arai, 1992), the oxidation state of the mantle during partial melting, melt evolution (Arai, 1992; Wood and Virgo, 1989; Kepezhinskas et al., 1993), and alteration and metamorphism (Ulmer, 1974; Evans and Frost, 1975). The Fe2 O3 wt% of the chromites analyzed in this study (3.2–13.2) is greater than those reported for chromite from MOR settings (0–4 wt% Fe2 O3 ; Arai, 1992) and most SSZ spreading center settings (e.g., Shetland, 2.5–4 wt% Fe2 O3 , own data; Oman, 3–6 wt% Fe2 O3 ; Auge, 1987) but similar to those from highly oxidizing boninitic and island arc magmas (Irvine, 1974) and Archean komatiites (5–13, Zhou and Kerich, 1992). The high Fe2 O3 content is unlikely to be caused by fractionation as the increase in Fe2 O3 content is not accompanied by an increase in TiO2 wt% (Irvine, 1974; Arai, 1992; Kepezhinskas et al., 1993). Though ferric-ferrous iron ratios may be a reliable indicator for chrome spinel in volcanic rocks (Arai, 1992) serpentinization and metamorphism can lead to an increase in the Fe2 O3 wt% (Ulmer, 1974; Evans and Frost, 1975). This precludes the use of iron as an effective indicator of magma oxygen fugacity in altered ultramafic rocks.

262

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

When compared to Phanerozoic ophiolite complexes the field geology and chromite chemistry show more similarities with a depleted SSZ spreading center than with an island arc or mid-ocean ridge spreading center environment.

10. WHAT DOES THE OPHIOLITIC ORIGIN FOR THE DONGWANZI BELT AND MANTLE FRAGMENTS IN MÉLANGE REVEAL ABOUT ARCHEAN OCEANIC MAGMATIC PROCESSES? The origin and tectonic setting of Archean greenstone belts has been one most hotly debated questions in Precambrian geology for much of the last century. Portions of several Archean greenstone belts have been interpreted to contain dismembered or partial ophiolites (Kusky and Polat, 1999; Abbott, 1996; de Wit et al., 1982, 1987, 1992; Kusky, 1987, 1990, 1991; Kusky and Vearncombe, 1997; Helmstaedt et al., 1986; MacLaughlin and Helmstaedt, 1995; Isachsen et al., 1991; Isachsen and Bowring, 1997; Wilks and Harper, 1997), but none of these contain the complete ophiolite sequence. Several of the Archean greenstone belts that have been interpreted as dismembered Archean ophiolites have three or four of the main magmatic components of a full ophiolite, although all of these have been disputed (Bickle et al., 1994). The Dongwanzi ophiolite and related belts in the North China craton contain a complete, albeit dismembered and metamorphosed ophiolite sequence, including pillow lavas, gabbro, dike complexes, mantle tectonites, and podiform chromitites. The preservation of a complete Archean ophiolite sequence in the North China craton is therefore of great importance for understanding processes of Archean sea-floor spreading, as it is the most complete record of this process known to exist. The Dongwanzi ophiolite has a strongly-deformed and serpentinized mantle section. However, about 60 km to the southwest primary mantle minerals and textures are wellpreserved in many tectonic blocks within the Zunhua mélange in the same structural belt. In this mélange, the lower part of the ophiolite is preferentially preserved, with the main rock types including harzburgite, dunite, podiform chromitites, meta-gabbro, and mafic and ultramafic cumulates. The oceanic mantle rocks show intense serpentinization, consistent with the Central Orogenic belt representing the suture between the East and West blocks of the North China craton (Fig. 2). In a few places pillow lava and sheeted dike complexes are also preserved in the mélange. Some of these have flat REE signatures (Kusky and Li, 2002) and are chemically similar to the Dongwanzi ophiolite. We suggest therefore that the Dongwanzi ophiolite and Zunhua mélange ophiolitic fragments preserved dismembered fragments and ophiolites from the closure of the same ocean basin. But what can the minerals and textures preserved in these ophiolitic fragments tell us about Archean sea floor spreading processes? Harzburgite blocks in the mélange host podiform chromitites with dunite envelopes. The blocks grade up-section into wehrlite, pyroxenite, olivine gabbro (troctolite), and gabbro. Podiform chromitites are commonly hosted by alpine-type or ophiolitic peridotite and are a normal component of ophiolites of different ages. Podiform chromite deposits are

10. What Does the Ophiolitic Origin for the Dongwanzi Belt and Mantle Fragments Reveal?

263

located in the transition zone between layered gabbro and peridotite tectonite, and the lherzolite/harzburgite (asthenosphere-lithosphere) transition in ophiolites (Nicolas and Azri, 1991). They are commonly attributed to oceanic settings such as mid-oceanic ridges, intraoceanic suprasubduction zones (back-arc basins or island arcs) (LeBlanc and Nicholas, 1992; Zhou et al., 1996). Their geological occurrence is closely associated with oceanic spreading processes (Nicolas and Azri, 1991). Late Proterozoic podiform chromitites in ophiolites have been described in several areas, including Ethiopia, Saudi Arabia, Morocco, south China, and Egypt, and Phanerozoic examples are numerous (Li et al., 2002). The oldest relatively intact podiform chromitite previously recognized is that from the Outokumpu ophiolite complex (2.0 Ga), Finland (Vuollo et al., 1995). The Zunhua chromite ores exhibit remarkable similarities to the podiform ores described from the examples mentioned above. Characteristic features of podiform chromites include: (1) lensoidal geometry, typically elongate along foliation although more rarely as discordant pods; (2) chromite bands with rootless to tight folds; (3) unique magmatic textures and structures, such as dunite envelopes for the chromite ores, and nodular, orbicular, and high-temperature shear fabrics of antinodular chromites; and (4) strong plastic deformation associated with harzburgite, with metamorphic foliations and lineations. The origin of the podiform chromitites is attributed to melt-rock reaction, or dynamic magmatism within melt channels in the upper oceanic mantle (Nicolas and Azri, 1991; Zhou et al., 1996). The presence of water in the melt is thought to be important for the formation of podiform chromite (e.g., Edwards et al., 2000). The Zunhua chromitites are typical podiform chromitites as classified by Thayer (1969). The chromites are variably deformed from strongly-stretched to weakly reworked. The nodular and orbicular chromites, apparently first described by Johnston (1936) from the Josephene ophiolite of Northern California, are characteristic of alpine-type peridotites or ophiolitic chromite ores (Nicolas, 1989; Gass et al., 1984; Zhou et al., 1996; Lippard et al., 1986; Peters et al., 1991; Dilek et al., 2000). It is now believed that this type of chromite accumulated below the transition between oceanic crust and mantle based on numerous investigations in ophiolites. Inclusions within chromites, olivine, and orthopyroxene of the host peridotites record high-temperature plastic deformation. The flattening and elongation of chromite parallel to foliation and lineation are intensive high-temperature shear strain. These textures probably record the plastic flow of the upper mantle, now mainly preserved in the core of tectonic blocks. These early lineations defined by deformed magmatic inclusions and the elongation of ore zones are not parallel to later lineations related to the emplacement of the blocks along shear zones, supporting the idea that they represent early mantle-deformation related fabrics. Podiform chromitites are remarkably resilient to later deformation and metamorphism since they are generated at high temperatures (1200–1300 ◦ C) and become very rigid when cooled, thus resisting later shear. These asthenospheric chromite pods are miniature time capsules preserving extraordinary amounts of information about the Archean mantle that we have only begun to tap and understand.

264

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Chromite can be used as a petrogenetic indicator for tectonically uncertain ultramafic massifs. The mineral chemistry of the Zunhua chromitites suggests that they formed in a suprasubduction zone ophiolitic environment. The Cr#, TiO2 wt% and V2 O5 wt% indicate partial melting of a depleted source under high fO2 conditions. Applied to the modern tectonic paradigm these compositions indicate formation in a SSZ setting and fluid aided melting of a previously depleted source. However, due to different geothermal gradients, and possibly magmatic systematics, analogy with modern and Phanerozoic systems may not be directly applicable. Higher heat flow in the Archean could lead to great enough degrees of partial melting to form high Cr# number chromite without fluid aided melting of previously depleted mantle. The Zunhua chromitites occur within dunite envelopes, grading into mantle harzburgite. Such textures form by interaction between a melt and a depleted mantle source. This raises the possibility that mantle and magmatic dynamics were similar to modern systems and that suprasubduction zone environments operated by similar magmatic systems to those of today. Comparison between Zunhua chromite ores and younger examples reveals a surprising similarity in their textures and structures. Podiform chromitites are present almost exclusively in ophiolites, being generated in the uppermost oceanic mantle beneath active spreading ridges above intraoceanic suprasubduction zones (Lago et al., 1982; Leblanc, 1997; Zhou et al., 1996). Coupled with the presence of a full ophiolite sequence in the Dongwanzi complex, the documentation of the Zunhua chromitites provides convincing evidence for the operation of sea-floor spreading and plate tectonics during the Archean before 2.50 Ga. We prefer to ascribe a faster to moderate spreading rate to the formation of the Zunhua podiform chromitites, as podiform chromite is mainly associated with harzburgite-type (HOT) ophiolites (Nicolas and Azri, 1991). Although the field and petrographic observations are consistent with the Neoarchean ophiolites of the Central Orogenic belt preserving relatively hot mantle features, we do not have evidence that this mantle record was any hotter than the present day range of mantle temperatures. However, the hot Archean North China mantle is consistent with some of the higher heat production during the Archean being accommodated by faster creation of oceanic lithosphere from a slightly hotter oceanic asthenosphere.

11. EMPLACEMENT OF ARCHEAN OPHIOLITES AND MANTLE BOUDINS IN THE CENTRAL OROGENIC BELT; IMPLICATIONS FOR THE EVOLUTION OF THE NORTH CHINA CRATON AND ARCHEAN CONTINENTAL GROWTH The Central Orogenic belt is over 1,600 km long, and separates the distinctly different Eastern and Western Blocks of the North China craton. It contains a diverse suite of rocks including high-pressure granulites, strongly deformed metasedimentary and metavolcanic rocks, and a number of different intrusive plutonic suites and dike swarms. The Central Orogenic belt is bounded on the east by a group of sedimentary basins that are filled with flysch-molasse like sequences of graywacke, shale, and conglomerate. We interpret the

11. Emplacement of Archean Ophiolites and Mantle Boudins in the Central Orogenic Belt

265

Central Orogenic belt as the ca. 2.5 Ga suture between the Eastern and Western Blocks of the North China craton. We interpret the mafic and ultramafic blocks in the biotite gneiss matrix to represent a strongly dismembered ophiolite in a metasedimentary and metavolcanic matrix. Relationships are strongly reminiscent of younger ophiolitic mélange terranes, where blocks of ophiolite are preserved in a metasedimentary accretionary prism/trench complex (e.g., Kusky et al., 1997; Kusky and Polat, 1999; Kusky and Young, 1999). The Zunhua ophiolitic blocks in mélange do not preserve an overall younging direction, although a few of the blocks show younging directions toward the west. Similarly, the Dongwanzi ophiolite to the northeast preserves an overall westward-younging sequence. The foliation in the Zunhua structural belt strikes north to northeast, and dips steeply west to northwest. The ophiolitic relicts clearly underwent intense tectonic transposition and amphibolitefacies metamorphism during or after their structural emplacement. The ultramafic rocks of the ophiolite are characterized by intense strain and structural dislocation associated with major thrusting and recumbent folding. As a result they are widely tectonically interleaved with country rock, and have mylonitic margins. Shear zones within gneiss separate different parts of the ophiolite complex. These tectonic slices were dismembered during obduction over an older gneissic terrane to the east. These relationships suggest, although do not require, that the ophiolites were emplaced into the mélange during westward directed subduction, then thrust over the eastern block during closure of the intervening ocean basin (Fig. 21). In this model the contemporaneous arc would be located to the west of the Zunhua structural belt. We interpret a narrow belt of deeply eroded and strongly metamorphosed 2.55–2.50 Ga arc-type TTG plutonic rocks and a greenstone belt in the Wutai-Hengshan-Taihang Mountains to the southwest (Figs. 2 and 21) to represent the remnants of this arc (Li et al., 2000a; Wilde et al., 1998). The ca. 2.50–2.40 Ga Qinglong, Hutuo, and Dengfeng sedimentary sequences and other similar basinal deposits east of the Central Orogenic belt (Figs. 2 and 21) may represent the foreland basin sequence resulting from the collision of the east and west blocks. These basin sequences consist of lower turbidite and upper molasse sequences, with more intense thrusting and folding in the west adjacent to the Central Orogenic belt. The 2.50–2.40 Ga granitoids that intrude the base of the ophiolite and much of the Central Orogenic belt could represent collisional to post-collisional granites formed during crustal thickening during orogenesis. This model also explains the exhumation of ca. 2.50 Ga high-pressure granulites and retrograde eclogites in the Hengshan belt to the west (Kusky and Li, 2003) (Fig. 21). The Hengshan granulite belt is over 700 km long, and is associated with the Neoarchean arc rocks on the western side of the Central Orogenic belt. If the continental margin of the eastern block were deeply subducted beneath the overriding arc of the Western block, it would rebound isostatically soon after collision. This rapid uplift and exhumation would remove most of the arc sequence, now preserved as the eclogites and granulites of the prominent Hengshan belt on the western side of the Central Orogenic belt (Fig. 21). Deepest levels of exhumation are recorded in the north, whereas central sectors of the Central Orogenic belt (Wutai Mountain) preserve lower-grade 2.50 Ga arc plutonic rocks and greenstone sequences. Other evidence for this late-stage exhumation and col-

266

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Fig. 21. Tectonic evolution of the Central Orogenic belt of the North China craton.

lapse of the Central Orogenic belt is provided by a swarm of 2.50–2.40 Ga extensional mafic dikes and flood basalts which occur in a larger area of the Central Orogenic belt (Fig. 21). After 2.40 Ga, an extensional graben system to cratonic basin developed with deposition of marine carbonate-mudstone, represented by the Middle to Upper Paleoproterozoic sequence (2.40–1.90 Ga) of the North China craton (Kusky and Li, 2003). The Dongwanzi ophiolite and Zunhua ophiolitic mélange are located in a previously identified suture zone separating two different cratonic blocks, and there is a parallel high-

Acknowledgements

267

pressure granulite belt (2.5 Ga) almost distributed in the same trend. These two different belts record the subduction and final closure of a former oceanic basin (2.50 Ga). It has been suggested that there were one or two supercontinents in the Neoarchean (Kenoraland; Heaman, 1997). The 2.5 Ga orogeny in the North China craton might record the final assembly of this supercontinent. The Central Orogenic Belt is cored by an internal zone with high-grade metamorphic nappes structurally overlying a lower-grade thrust imbricate zone containing many ophiolitic fragments thrust over a passive margin sequence. A foreland basin sequence has syn-orogenic clastic rocks in the west succeeded eastward by a thick flysch sequence. The overall structural zonation of the orogen is much like younger examples, showing that collisional plate processes similar to those of the Phanerozoic were in operation in the Archean.

ACKNOWLEDGEMENTS This work was supported by the U.S. National Science Foundation grants 02-07886 and 01-25925 (awarded to T. Kusky), China National Natural Science Foundation grant 49832030 (awarded to J.H. Li), Peking University Project 985, and St. Louis University. This work has benefited greatly from discussions with numerous scientists, with special emphasis by Brian Windley, Kevin Burke, Alfred Kröner, Mingo Zhai, Ali Polat, and Kent Condie.

REFERENCES Abbott, D., 1996. Plumes and hotspots as sources of greenstone belts. Lithos 37, 113–127. Ahmed, A.H., Arai, S., Attia, A.K., 2001. Petrological characteristics of podiform chromites and associated peridotites of the Pan African Proterozoic ophiolite complexes of Egypt. Mineralium Deposita 36, 72–84. Anonymous, 1972. Ophiolites. Geotimes 17, 24–25. Arai, S., 1992. Chemistry of chromian spinel in volcanic rocks as guide to potential magma chemistry. Mineralogical Magazine 56, 173–184. Arai, S., 1997a. Origin of podiform chromitites. Journal of Asian Earth Sciences 15, 303–310. Arai, S., 1997b. Control of wall-rock composition on the formation of podiform chromitites as a result of magma/peridotite interaction. Resource Geology 47, 177–187. Arai, S., Yurimoto, H., 1995. Possible sub-arc origin of podiform chromitites. The Island Arc 4, 104–111. Auge, T., 1987. Chromite deposits in the northern Oman ophiolite: Mineralogical constraint. Mineralium Deposita 22, 1–10. Bai, J., Dai, F.Y., 1996. The early Precambrian crustal evolution of China. Journal of Southeast Asian Earth Sciences 13, 205–214. Bai, J., Dai, F.Y., 1998. Archean crust of China. In: Ma, X.Y., Bai, J. (Eds.), Precambrian Crustal Evolution of China. Springer Geological Publishing, Beijing, pp. 15–86.

268

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Bai, J., 1996. Precambrian Crustal Evolution of China. Geological Publishing House, Beijing, p. 165. Bai, J., Wang, R.Z., Guo, J.J., 1992. The Major Geological Events of Early Precambrian and Their Dating in Wutaishan Region. Geological Publishing House, Beijing, pp. 1–60. Berhe, S.M., 1990. Ophiolites in north and east Africa: implications for Proterozoic crustal growth. Journal of the Geological Society of London 147, 41–57. Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Brandl, G., de Wit, M.J., 1997. The Kaapvaal Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 581–607. Canil, D., 1997. Vanadium partitioning and the oxidation state of Archean komatiite magmas. Nature 389, 842–844. Canil, D., 1999. Vanadium portioning between orthopyroxene, spinel and silicate melt and the redox states of mantle source regions for primary magmas. Geochimica et Cosmochimica Acta 63, 557– 572. Chadwick, B., Crewe, M.A., 1986. Chromite in the early Archean Akilia association (ca 3800 m.y.) Ivisartoq region, inner Gothabsfjord, southern Greenland. Economic Geology 81, 184–191. Cotterill, P., 1969. The chromite deposits of Selukwe, Rhodesia. Economic Geology Monograph 4, 154–186. Dann, J.C., 1991. Early Proterozoic ophiolite, central Arizona. Geology 19, 590–593. Dann, J.C., 1997a. Pseudostratigraphy and origin of the Early Proterozoic Payson ophiolite, central Arizona. Bulletin of the Geological Society of America 109, 347–365. Dann, J.C., 1997b. Branching sheeted dikes and seafloor spreading within an Early Proterozoic intraarc basin. Journal of Geophysical Research 102, 24,917–24,929. Dann, J.C., 2004. The 1.73 Ga Payson Ophiolite, Arizona, USA. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 73–93. Davis, G.A., Qian, X.G., Zheng, Y.D., Tong, H.M., Yu, H., Wang, C., Gehrils, G., Shafiquallah, M., Fryxell, J., 1996. Mesozoic deformation and plutonism in the Yunmeng Shan: a metamorphic core complex north of Beijing, China. In: Yin, A., Harrison, T.M. (Eds.), The Tectonic Evolution of Asia. Cambridge Univ. Press, pp. 253–280. de Wit, M.J., Hart, R.A., Martin, A., Abbott, P., 1982. Archean abiogenic and probable biogenic structures associated with mineralized hydrothermal vent systems and regional metasomatism, with implications for greenstone belt studies. Economic Geology 77, 1783–1802. de Wit, M.J., Hart, R.A., Hart, R.J., 1987. The Jamestown ophiolite complex, Barberton mountain belt: a section through 3.5 Ga oceanic crust. Journal of African Earth Science 6, 681–730. de Wit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., de Ronde, C.E.J., Green, R.W.E., Tredoux, M., Peberdy, E., Hart, R.A., 1992. Formation of an Archean continent. Nature 357, 553–562. 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. Dilek, Y., Moores, E., Elthon, D., Nicolas, A. (Eds.), 2000. Ophiolites and Oceanic Crust, New Insights from Field Studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 552. Droop, G.T.R., 1987. A general equation for estimating Fe3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses using stoichiometric criteria. Mineralogical Magazine 51, 431–435. Echeverria, L.M., 1980. Tertiary or Mesozoic komatiites from Gorgona Island, Colombia field relations and geochemistry. Contributions to Mineralogy and Petrology 73 (3), 253–266.

References

269

Edwards, S.J., Pearce, J.A., Freeman, J., 2000. New insights concerning the influence of water during the formation of podiform chromitite. In: Dilek, Y., Moores, E., Elthon, D., Nicolas, A. (Eds.), Ophiolites and Oceanic Crust, New Insights from Field Studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 139–148. Evans, B.W., Frost, B.R., 1975. Chrome-spinels in progressive metamorphism—a preliminary analysis. Geochimica et Cosmochimica Acta 39, 959–972. Fang, L., Friend, C.R.L., Li, Q., Li, S.Z., Liu, W., Powell, D., Thirwall, M.F., Yang, Z.S., Zhang, Q.H., 1998. Geology of the Santunying Area of Eastern Hebei Province. Geological Publishing House, Beijing, p. 134. Gao, S., Rudnick, R., Carlson, R., McDonough, W., Liu, Y.S., 2002. Re-Os evidence for replacement of ancient mantle lithosphere beneath the North China craton. Earth and Planetary Science Letters 6135, 1–15. Gass, I.G., Lippard, S.J., Shelton, A.W., 1984. Ophiolites and Oceanic Lithosphere. In: The Geological Society Special Publications, vol. 13. Blackwell Scientific, Oxford, p. 413. Griffin, W.L., Zhang, A., O’Reilly, S.Y., Ryan, C.G., 1998. Phanerozoic evolution of the lithosphere beneath the Sino-Korean craton, in Mantle Dynamics and Plate Interactions in East Asia, A.G.U. Geodynamics 27, 107–126. Harper, G.D., 1999. Structural styles of hydrothermal discharge in ophiolite/sea floor systems. Reviews in Economic Geology 8, 53–73. Harper, G.D., 1985. A dismembered Archean ophiolite, Wind River Mountains, Wyoming (USA). Ophioliti 10, 297–306. He, T.X., Lin, Q., Fang, Z.R., et al., 1992. The Petrogenesis of Granitic Rocks in Eastern Hebei. Jiling Science and Technology Press, Changchun. pp. 1–4 (in Chinese with English abstract). He, G.P., Lu, L.Z., Ye, H.W., 1991. The Early Precambrian Metamorphic Evolution of the Eastern Hebei and the Southeastern Inner Mongolia. Jilin Univ. Press, Changchun, pp. 1–17 (in Chinese with English abstract). Heaman, L.M., 1997. Global mafic magmatism at 2.45 Ga; remnants of an ancient large igneous province? Geology 25, 299–302. Helmstaedt, H., Padgham, W.A., Brophy, J.A., 1986. Multiple dikes in the lower Kam Group, Yellowknife greenstone belt: Evidence for sea floor spreading? Geology 14, 562–566. Holtzman, B., 2000. Guaging stress from mantle chromitite pods in the Oman ophiolite. 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 Paper 349, 149–158. Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga Podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithosphere mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337. Huson, R., Kusky, T.M., Li, J.H., 2004. Geochemical and petrographic characteristics of the central belt of the Archean Dongwanzi ophiolite complex. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 283–320. Irvine, T.N., 1965. Chromian spinel as a petrogenetic indicator: Part 1, Theory. Canadian Journal of Earth Science 2, 648–671. Irvine, T.N., 1967. Chromian spinel as a petrogenetic indicator: Part 2, Petrologic applications. Canadian Journal of Earth Science 4, 71–103.

270

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Irvine, T.N., 1974. Petrology of the Duke Island ulramafic complex, southeastern Alaska. Geological Society of America Memoir 138, 240. Isachsen, C.I., Bowring, S.A., Padgham, W.A., 1991. U-Pb zircon geochronology of the Yellowknife volcanic belt, N.W.T.: New constraints on the timing and duration of greenstone belt magmatism. Journal of Geology 99, 55–67. Isachsen, C.I., Bowring, S.A., 1997. The Bell Lake Group and Anton Complex—a basement cover sequence beneath the Yellowknife greenstone belt revealed and implicated in greenstone belt formation. Canadian Journal of Earth Science 34, 169–189. Jahn, B.M., Zhang, Z.Q., 1984a. Archean granulite gneisses from Eastern Hebei Province, China: Rare Earth geochemistry and tectonic implications. Contributions to Mineralogy and Petrology 85, 225–243. Jahn, B.M., Zhang, Z.Q., 1984b. Radiometric ages (Rb-Sr, Sm-Nd, U-Pb) and REE geochemistry of Archean granulite gneisses from Eastern Hebei Province, China. In: Kröner, A., Hanson, G., Goodwin, A. (Eds.), Archean Geochemistry. Springer-Verlag, Berlin, pp. 204–244. Jahn, B.M., Auvray, B., Cornichet, J., Bai, Y.L., Shen, Q.H., Liu, D.Y., 1987. 3.5 Ga amphibolites from Eastern Hebei Province, China field occurrence, petrography, Sm-Nd isochron age and REE geochemistry. Precambrian Research 34, 311–346. Jan, M.Q., Windley, B.F., 1990. Chromian spinel-silicate chemistry in ultramafics rocks of the Jijal Complex, Northwest Pakistan. Journal of Petrology 31, 667–715. Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004. Neoproterozoic ophiolites in the Arabian Shield: Field relations and structure. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162. Johnston Jr., W.D., 1936. Nodular, orbicular, and banded chromite in Northern California. Economic Geology 31, 417–427. Kamenetsky, V.S., Crawford, A.J., Meffre, S., 2001. Factors controlling chemistry of magmatic spinel: an empirical study of associated olivine, Cr spinel and melt inclusions from primitive rocks. Journal of Petrology 42. Karson, J.A., 2001. Oceanic crust when Earth was young. Science 292, 1076–1077. Kelemen, P.B., 1990. Reaction between ultramafic rock and fractionating basaltic magma I. Phase relations, the origin of cal-alkaline magma series and the formation of discordant dunite. Journal of Petrology 31, 51–98. Kelemen, P.B., Joyce, D.B., Webster, J.D., Holloway, J.R., 1990. Reaction between ultramafic rock and fractionating basaltic magma II. Experimental investigation of reaction between olivine tholeiite and harzburgite at 1150–1050 and 5 kb. Journal of Petrology 31, 99–134. Kepezhinskas, P.K., Taylor, R.N., Tanaka, H., 1993. Geochemistry of plutonic spinels from the North Kamchatka Arc: comparisons with spinels from other tectonic settings. Mineralogical Magazine 57, 575–589. Kontinen, A., 1987. An Early Proterozoic ophiolite—The Jourma Mafic-Ultramafic Complex, northeastern Finland. Precambrian Research 35, 313–341. Kröner, A., 1985. Ophiolites and the evolution of tectonic boundaries in the late Proterozoic ArabianNubian Shield of north east Africa and Arabia. Precambrian Research 27, 277–300. Kröner, A., Cui, W.Y., Wang, S.Q., et al., 1998. Single zircon ages from high-grade rocks of the Jianping complex, Liaoning province, NE China. Journal of Asia Earth Sciences 16 (5–6), 519– 532. Kröner, A., Wildes, S., Wang, K., Zhao, G.C., 2002. Age and evolution of a late Archean to early Proterozoic upper to lower crustal section in the Wutaishan/Hengshan/Fuping terrain of northern China, A Field Guide. GSA Penrose Conference, Beijing, China, September 2002.

References

271

Kusky, T.M., 1987. Comment on “Multiple dikes in the lower Kam Group, Yellowknife Greenstone Belt: Evidence for Archean sea-floor spreading?” Geology 15, 280–282. Kusky, T.M., 1990. Evidence for Archean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10, 820–841. Kusky, T.M., Li, J.H., 2002. Is the Dongwanzi complex an Archean Ophiolite? Response to Zhai, M., Zhao, G., Zhang, Q. Science 295, A923. Kusky, T.M., Li, J.H., Tucker, R.T., 2001. The Archean Dongwanzi ophiolite complex, North China Craton: 2.505 Billion Year Old Oceanic Crust and Mantle. Science 292, 1142–1145. Kusky, T.M., Li, J.H., 2003. Paleoproterozoic tectonic evolution of the North China Craton. Journal of Asian Earth Sciences 22, 383–397. Kusky, T.M., Polat, A., 1999. Growth of Granite-Greenstone Terranes at Convergent Margins and Stabilization of Archean Cratons. In: Marshak, S., van der Pluijm, B. (Eds.), Special Issue on Tectonics of Continental Interiors. Tectonophysics 305, 43–73. Kusky, T.M., Vearncombe, J., 1997. Structure of Archean Greenstone Belts. In: de Wit, M.J., Ashwal, L.D. (Eds.), Tectonic Evolution of Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 95–128. Kusky, T.M., Bradley, D.C., Haeussler, P.J., Karl, S., 1997. Controls on accretion of flysch and mélange belts at convergent margins: Evidence from the Chugach Bay thrust and Iceworm mélange, Chugach Terrane, Alaska. Tectonics 16, 855–878. Kusky, T.M., Li, J.H., Raharimahefa, T., Carlson, R.W., 2004. Re-Os isotope chemistry and geochronology of chromite from mantle podiform chromites from the Zunhua ophiolitic mélange belt, N. China: Correlation with the Dongwanzi ophiolite. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 275–282. Kusky, T., Young, C., 1999. Emplacement of the Resurrection Peninsula ophiolite in the southern Alaska Forearc During a Ridge-Trench Encounter. Journal of Geophysical Research 104 (B12), 29,025–29,054. Kusky, T., Glass, A., Tucker, R., Harper, G., Bradley, D., Ozdogan, M., in preparation. Structure, Chemistry, Age and Origin of the Border Ranges Ultramafic/Mafic Complex: A Suprasubduction Zone Ophiolite Complex. Journal of Geophysical Research. Lago, B.L., Rabinowicz, M., Nicolas, A., 1982. Podiform chromitite ore bodies: a genetic model. Journal of Petrology 23, 103–125. Leblanc, M., 1997. Chromitite and ultramafic rock compositional zoning through a paleotransform fault Poum, New Caledonia—A reply. Economic Geology 92, 503–504. LeBlanc, M., Nicholas, A., 1992. Ophiolitic chromitites. International Geology Review 34, 653–686. Li, J.H., Qian, X.L., Gu, Y.C., 1998. Outline of Paleoproterozoic tectonic division and plate tectonic evolution of North China Craton. Earth Science 23, 230–235. Li, J.H., Qian, X.L., Huang, X.N., Liu, S.W., 2000a. The tectonic framework of the basement of North China Craton and its implication for the early Precambrian cratonization. Acta Petrologica Sinica 16 (1), 1–10. Li, J.H., Kröner, A., Qian, X.L., O’Brien, P., 2000b. The tectonic evolution of early Precambrian high-pressure granulite belt, North China Craton (NCC). Acta Geological Sinica 274 (2), 246– 256.

272

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Li, J.H., Kusky, T.M., Huang, X., 2002. Neoarchean podiform chromitites and harzburgite tectonite in ophiolitic mélange, North China Craton, Remnants of Archean oceanic mantle. GSA Today 12 (7), 4–11. Lippard, S.J., Shelton, A.W., Gass, I.G., 1986. The ophiolite of northern Oman. Geological Society of London Memoir 11, 178. MacLaughlin, K., Helmstaedt, H., 1995. Geology and geochemistry of an Archean mafic dike complex in the Chan Formation: basis of a revised plate tectonic model of the Yellowknife greenstone belt. Canadian Journal of Earth Science 32, 614–630. Matsumoto, I., Arai, S., 1999. Morphological variations of chromian spinel in dunite and harzburgite from the Sangun Zone, Southwest Japan, as a marker of melt/peridotite reaction. Science Reports of the Kanazawa University 44 (1–2), 11–24. Matveev, S., Balhaus, C., 2002. Role of water in the origin of podiform chromite deposits. Earth and Planetary Science Letters 203, 235–243. Menzies, M.A., Fan, W., Zhang, M., 1993. Paleozoic and Cenozoic lithoprobes and the loss of > 120 km of Archean lithosphere, Sino-Korean craton, China. In: Prichard, H.M., Alabaster, T., Harris, N.B., Neary, C.R. (Eds.), Magmatic Processes and Plate Tectonics. Geological Society Special Publication 76, 71–81. Moores, E.M., 1982. Origin and emplacement of ophiolites. Reviews in Geophysics 20, 735–750. Moores, E.M., 2002. Pre-1 Ga (pre-Rodinian) ophiolites: their tectonic and environmental implications. Geological Society of America Bulletin 114, 80–95. Nicolas, A., 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Boston, p. 367. Nicolas, A., Azri, H.A., 1991. Chromite-rich and chromite-poor ophiolites: the Oman case. In: Peters, T.J., Nicolas, A., Coleman, R.G. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Boston, pp. 261–274. Nicolas, A., Boudier, F., 2000. Large mantle upwellings and related variations in crustal thickness in the Oman ophiolite. 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 Paper 349, 67–73. Oberger, B., Loran, J.P., Girardeau, J., Mercier, J.C.C., Pitragool, S., 1995. Petrogenesis of ultramafic rocks and associated chromitites in the Nan Uttaradit ophiolite, Northern Thailand. Lithos 35, 153–182. 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 of Petrology 39, 1577–1618. Parson, L.M., Murton, B.J., Browning, P. (Eds.), 1992. Ophiolites and Their Modern Oceanic Analogs. Geological Society Special Publication 60, 330. Peltonen, P., Kontinen, A., 2004. The Jormua ophiolite: A mafic-ultramafic complex from an ancient ocean-continent transition zone. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 35–71. Peters, Tj., Nicolas, A., Coleman, R.G., 1991. Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Boston, p. 903. Power, M.R., Pirrie, D., Anderson, J.C., Wheeler, P.D., 2000. Testing the validity of chrome spinel chemistry as a provenance and petrogenetic indicator. Geology 28, 1027–1030. Proenza, J., Gervilla, F., Melgarejo, J.C., Bodinier, J.L., 1999. Al- and Cr-rich chromitites from the Mayarí-Baracoa Ophiolitic Belt (Eastern Cuba): Consequence of interaction between volatilerich melts and peridotites in suprasubduction mantle. Economic Geology 94, 547–566.

References

273

Rasmussen, B., 2000. Filamentous microfossils in a 3.235 million-year-old volcanogenic massive sulphide deposit. Nature 405, 676–679. Reed, C., 2002. Chimneys from an Ancient Ocean. Geotimes 23. Roeder, P.L., Campbell, I.H., Jamieson, E., 1979. A re-evaluation of the olivine-spinel geothermometer. Contributions to Mineralogy and Petrology 68, 325–334. Sack, R.O., Ghiorso, M.S., 1991. Chromian spinels as petrogenetic indicators: thermodynamics and petrological applications. American Mineralogist 76, 827–847. Scott, D.J., St. Onge, M.R., Lucas, S.B., Helmstaedt, H., 1991. Geology and chemistry of the Early Proterozoic Purtuniq ophiolite, Cape Smith Belt, Northern Quebec, Canada. In: Peters, T.J. (Ed.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Boston, pp. 817– 849. Scott, D.J., Helmstaedt, H., Bickle, M.J., 1992. Purtuniq ophiolite, Cape Smith belt, northern Quebec, Canada: A reconstructed section of Early Proterozoic oceanic crust. Geology 20, 173–176. Shen, Q.H., Xu, H.F., Zhang, Z.Q., 1992. The Early Precambrian Granulites in China. Geological Publishing House, Beijing, pp. 134–140 (in Chinese). Shervais, J.W., 1982. Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59, 101–118. Sleep, N.H., Windley, B.F., 1982. Archean plate tectonics: constraints and inferences. Journal of Geology 90, 363–379. Stern, R.J., Johnson, P.R., Kröner, A., Yibas, B., 2004. Neoproterozoic ophiolites of the ArabianNubian Shield. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 95–128. Stowe, C.W., 1987. Chromite deposits of the Shurugwi Greenstone Belt, Zimbabwe. In: Stowe, C.W. (Ed.), Evolution of Chromium Ore Fields. Van Nostrand-Reinhold, New York, pp. 71–88. Stowe, C.W., 1994. Compositions and tectonic setting of deposits through time. Economic Geology 89, 528–546. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.C., 1997. Volcanic Aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 55–90. Thayer, T.P., 1969. Gravity differentiation and magmatic re-emplacement of podiform chromite deposits. In: Magmatic Ore Deposits. In: Economic Geology Monographs, vol. 4, pp. 132–146. Tian, Y.Q., 1991. Geology and Mineralization of the Wutai-Hengshan Greenstone Belt. Shanxi Science and Technology Press, Taiyuan, pp. 137–152 (in Chinese). Ulmer, C.G., 1974. Alteration of chromite during serpentinisation in the Pennsylvania-Maryland district. American Mineralogist 59, 1236–1241. Vuollo, J., Liipo, J., Nykanen, V., Piirainen, T., 1995. An early Proterozoic podiform chromitite in the Outokumpu ophiolite complex, Finland. Economic Geology 90, 445–452. Wan, Y.S., Geng, Y.S., Wu, J.S., 1998. The geochemistry of early Precambrian metabasaltic rocks of North China Craton. In: Cheng, Y.Q. (Ed.), Proceeding of Precambrian Geology of North China Craton. Geological Publishing House, Beijing, pp. 39–59. Wang, Q.C., Zhang, S.Q., 1995. The age of the Hongqiyingzi Group: a further discussion. Regional Geology of China 2, 173–180. Wang, K., Li, J.H., Hao, J., 1997. Late Archean mafic-ultramafic rocks from the Wutaishan, Shanxi Province; a possible ophiolitic mélange. Acta Petrologica Sinica 132, 139–151. Wilde, S.A., Cawood, P.A., Wang, K.Y., Nemchin, A., 1998. SHRIMP U-Pb zircon dating of granites and gneisses in the Taihangshan-Wutaishan area: implications for the timing of crustal growth in the North China craton. Chinese Science Bulletin 43, 144–145.

274

Chapter 7: Origin and Emplacement of Archean Ophiolites of the Central Orogenic Belt

Wilks, M.E., Harper, G.D., 1997. Wind River Range, Wyoming Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 508–516. Wood, B.J., Virgo, D., 1989. Upper mantle oxidation state: Ferric iron contents of lherzolite spinels by 57 Fe Mössbauer spectroscopy and resultant oxygen fugacity. Geochimica et Cosmochimica Acta 53, 1277–1291. Wu, J.S., Geng, Y.S., 1991. Major Geological Events of Early Precambrian in North China Platform. Geological Publishing House, Beijing, pp. 1–11 (in Chinese). Wu, J., Geng, Y.S., Shen, Q.H., 1998. Archean Geology Characteristics and Tectonic Evolution of Sino-Korea Paleocontinent. Geological Publishing House, Beijing, pp. 1–104. Wu, C.H., Zhong, C.T., 1998. The Paleoproterozoic SW-NE collision model for the central North China Craton. Progress of Precambrian Research 21, 28–50 (in Chinese). Xu, Z.G., 1990. Mesozoic volcanism and volcanogenic iron ore deposits in eastern China. Geological Society of America Special Paper 237, 46. Zhang, Q.S., Yang, Z.S., Gao, D.Y., 1991. The Archean High-Grade Metamorphic Geology and Gold Deposits in Jinchangyu Area of Eastern Hebei. Geological Publishing House, Beijing, pp. 1–5 (in Chinese). Zhang, Y.X., Ye, T.S., Yang, H.Q., 1986. The Archean Geology and Banded Iron Formation of Jidong, Hebei Province. Geological Publishing House, Beijing, pp. 1–22 (in Chinese with English abstract). Zhao, G., 2001. Paleoproterozoic assembly of the North China Craton. Geological Magazine 138, 87–91. Zhao, G.C., Wilde, S.A., Cawood, P.A., Sun, M., 2001. Archean blocks and their boundaries in the North China Craton: lithological, geochemical, structural and P-T path constraints and tectonic evolution. Precambrian Research 107, 45–73. Zhao, Z.P., 1993. The Precambrian Geological Evolution of Sino-Korean Paraplatform. Science Publishing House, Beijing, pp. 1–83 (in Chinese). Zheng, Z., O’Reilly, S.Y., Griffin, W.L., Lu, F., Zhang, M., 1998. International Geology Review 40, 471–499. Zhou, M.F., Kerrich, R., 1992. Morphology and composition of chromite in komatiites from the Belingwe Greenstone Belt, Zimbabwe. Canadian Mineralogist 30, 303–317. Zhou, M.F., Robinson, P.T., Bai, W.J., 1994. Formation of podiform chromitites by melt/rock interaction in the upper mantle. Mineralium Deposita 29, 98–101. Zhou, M.F., Robinson, P.T., Malpas, J., Zijin, L., 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, 3–21. Zhou, M.F., Robinson, P.T., 1997. Origin and tectonic environment of podiform chromite deposits. Economic Geology 92, 259–262. Ziegler, A.M., Rees, P.M., Rowley, D.B., Bekker, A., Qing, Li, Hulver, M., 1996. Mesozoic assembly of Asia: constraints from fossil floras, tectonics, and paleomagnetism. In: Yin, A., Harrison, T.M. (Eds.), The Tectonic Evolution of Asia. Cambridge Univ. Press, pp. 371–400.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

275

Chapter 8

RE-OS ISOTOPE CHEMISTRY AND GEOCHRONOLOGY OF CHROMITE FROM MANTLE PODIFORM CHROMITES FROM THE ZUNHUA OPHIOLITIC MÉLANGE BELT, N. CHINA: CORRELATION WITH THE DONGWANZI OPHIOLITE T.M. KUSKYa , J.H. LIb , T. RAHARIMAHEFAa AND R.W. CARLSONc a Department

of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63130, USA b Department of Geology, Peking University, Beijing, China c Department of Terrestrial Magnetism, Carnegie Institute of Washington, Washington, DC, USA

The Zunhua structural belt of the Archean North China craton is a structurally-complex amphibolite-granulite facies terrain characterized by NE-striking, intensely strained gneiss and amphibolites exhibiting tight composite folds and ductile shear zones. It is part of the Central Orogenic Belt that separates the Eastern and Western Blocks of the North China Craton. Various thrust slices, such as TTG gneiss, mafic plutonic rocks, supracrustal sequences, mafic volcanics, banded iron formation, garnet-bearing gneiss, and granites are tectonically intercalated with each other. The 2.5 Ga Dongwanzi ophiolite is one of the largest well-preserved greenstone belts in the northern part of the Zunhua structural belt. More than 1,000 other fragments of gabbro, pillow lava, sheeted dikes, harzburgite, and podiform-chromite bearing dunite occur as tectonic blocks (tens to hundreds of meters long) in a biotite-gneiss and BIF matrix, intruded by tonalite and granodiorite, in the Zunhua structural belt. Blocks in this metamorphosed Archean ophiolitic mélange preserve deeper levels of oceanic mantle than the Dongwanzi ophiolite. Many of the chromite bodies are localized in dunite envelopes within harzburgite tectonite, and have characteristic nodular and orbicular chromite textures, known elsewhere only from ophiolites. The chromites have variable but high chrome numbers (Cr/Cr+Al = 0.74–0.93) and elevated P, also characteristic of suprasubduction zone ophiolites. The high chrome numbers, coupled with TiO2 wt% < 0.2 and V2 O5 wt% < 0.1, indicate high degrees of partial melting from a very depleted mantle source and primitive melt for the chromite. A Re-Os model age from the chromites indicates an age of 2547 ± 10 Ma, showing that they are the same age as the Dongwanzi ophiolite. The ultramafic and ophiolitic blocks in the Zunhua mélange are therefore interpreted as dismembered and strongly deformed parts of the Dongwanzi ophiolite. We suggest the name “Zunhua ophiolitic mélange” for these rocks. The range in initial Os isotopic compositions in the chromites in these ophiolitic blocks is small and DOI: 10.1016/S0166-2635(04)13008-9

276

Chapter 8: Re-Os Isotope Chemistry and Geochronology of Chromite

well within the range seen in modern ophiolites. The chondritic to sub-chondritic initial ratio also is interesting in that it shows more similarity to the values found for abyssal peridotites than OIB’s, pointing to an ocean-ridge rather than plume setting for the initial formation of these peridotites.

1. INTRODUCTION The circa 2.5 Ga Dongwanzi ophiolite and related Zunhua mélange of the North China Craton offer one of the world’s most complete records of Archean oceanic crust and lithosphere (Kusky et al., 2001; Li et al., 2002). The current state of knowledge about the field, structural, geochronological, mineralogical, geochemical, and chromite chemistry of the circa 2.5 Ga Dongwanzi ophiolite and related Zunhua mélange are presented in several papers in this volume (Kusky et al., 2004; Huson et al., 2004; Li et al., 2004; Huang et al., 2004). In this contribution we present Re-Os data on mantle podiform chromite samples collected from the Zunhua mélange and Dongwanzi ophiolite, and show that the two belts are the same age, and have similar chemical characteristics, supporting their correlation. Readers are referred to the other chapters cited above for details of the geology, chemistry, and other aspects of these chromite deposits.

2. RE-OS STUDIES Rhenium and Osmium are siderophilic elements that concentrate in the Earth’s core and mantle during planetary formation. 187Re decays to 187 Os according to the equation 187

Os/188 Osmeasured =

187

Os/188 Osinitial +

187

Re/188 Osmeasured × (eyt − 1),

where y is the decay constant equal to 1.666 × 10−11 years, corresponding to a half-life of 41.60 ± 0.42 × 109 years (Smoliar et al., 1996). The Re-Os system is ideal for tracing the evolution of mafic and ultra-mafic rocks and is good for tracing lithosphere forming events where extraction of a melt volume depletes the initially primitive mantle source in Re and leaves a residue enriched in Os. Both rhenium and osmium prefer metal and sulfide phases. Rhenium is moderately incompatible during mantle melting events, whereas osmium is highly compatible, so mantle melting events (such as associated with the formation of oceanic crust) leave a mantle residue depleted in rhenium relative to osmium. If subducted oceanic lithosphere remains geochemically isolated from the upper mantle, then oceanic crust formation drives isotopic evolution of 187 Os/188 Os of the mantle to become depleted in rhenium with time. Also, Re-Os data from Archean mantle rocks should reveal whether there have been major mantle formation events in the early history of the Earth that led to the generation and stabilization of the lithosphere. The rhenium-osmium composition of the mantle is commonly studied by examining the composition of chromite bodies in ophiolites of various ages. Ophiolitic chromites have a

3. Analytical Methods

277

strong resistance to secondary alteration, have high osmium concentrations, and low Re/Os ratios that limit age corrections and uncertainties. Further, many ophiolitic chromites have crystallized from melt flux gathered over a large domain, minimizing variations in isotopic composition (Tsuru et al., 2000; Walker et al., 2002). The rhenium-osmium isotope system was used as a method of dating samples of the Zunhua chromite bodies. The Re-Os concentration in chromites was measured in twelve samples from the Zunhua mélange and Dongwanzi ophiolite. These samples were chosen because of their extremely fresh appearance and their chromite composition. For this study, only purified chromites derived from the mafic portion of chromite of the ophiolites were examined, and the rocks age ranges from 2.4 Ga to 2.7 Ga. Samples for Re-Os analysis were collected from several of the podiform chromite bodies in the Zunhua Structural Belt at the Maojiachang and Zhuling localities (see chapter by Kusky et al., this volume), and from the cumulate ultramafic section of the Dongwanzi ophiolite. Samples are largely serpentinized, but analysis of the chromite grains has revealed that most of them escaped alteration (Kusky et al., 2004).

3. ANALYTICAL METHODS Chromite rich samples were crushed with an alumina and agate crushing apparatus and ground into a powder, and chromite grains under 250 µm were concentrated using a Frantz magnetic separator. The chromite grains were later washed in distilled water in an ultrasonic bath. The chromite grains were dried carefully and then transferred and kept in plastic sample cans. The chemical separation techniques used in this study for Re-Os analyses followed the “Department of Terrestrial Magnetism” (DTM) extraction procedure. Carius tube equilibration, CCl4 -HBr extraction and also micro-distillation of osmium and anion exchange separation of rhenium were used. The first step is sample weighing, spiking and dissolution. Samples were dissolved and equilibrated with spikes Re-Os-99-4-50 in Carius tubes; the spike concentration was 9.239 × 10−12 m/g. According to the chromite quantity, 180 mg to 650 mg spikes and 100 mg to 200 mg of the finely ground sample powder was put in the tubes. The lower half of a Carius tube was placed in a dewar of methanol-dry ice slush, and we added approximately 2 ml of concentrated HCl and 4 ml of concentrated HNO3 . The tubes were sealed with lab-clamps after they were removed from the dry ice and methanol and heated to 220 ◦ C for 15 h. After cooling and uncapping, the solution was evaporated. Upon opening the tubes with the lab-clamp, 9 ml (3 × 3 ml) of CCl4 (carbon tetrachloride) was added in each sample, this is for the extracting the Os from the A-R. To extract the Os from the CCl4 we used HBr. Three milliliters of CCl4 from the regent bottle was pipetted to the centrifuge tube, capped and shaken for one minute then placed into centrifugal machine. After 5 min we pipetted the CCl4 from the bottom of the tube into Savillex beaker containing 3 ml of 9 N HBr. The tightly capped HBr and CCl4 mixture was then shaken for about one minute,

278

Chapter 8: Re-Os Isotope Chemistry and Geochronology of Chromite

then CCl4 was removed from beneath HBr, and after that the samples were put under heat lamp. In general the Os was oxidized by the addition of a concentrated CrO3 solution, extracted into liquid Br2 and finally purified by micro distillation. For the micro-distillation, the Os was dissolved in one drop of concentrated HBr, and then transferred into the cap of a Savillex conical vial and the HBr was evaporated. 20 µl of distilled 9 N HBr was pipetted into the tip of the conical beaker. 30 µl of CrO3 in 12 N H2 SO4 was added to the sample residue of Os. The beaker was rapidly and tightly closed in the upside down position, and wrapped completely in aluminum foil with a hole at the top of the beaker. This beaker was then placed for at least one and one half-hours on a hotplate having a surface temperature of around 92 ◦ C. CrO3 oxidized the complex OsBr2− 6 to OsO4 and volatile OsO4 was transferred in the gas phase to the conical tip of the beaker where the concentrated HBr reduced it to OsBr2− 6 . After three hours, the Os had been distilled. The beaker was then opened gently and turned back to its normal position. The content of the cap was discarded and the HBr in the beaker was evaporated down to about 1 to 2 µl under the heat lamp. The osmium was then ready for mass spectrometry. The isotopic composition of Re and Os were measured using Mass spectrometry in the DTM lab in Washington, DC. For osmium, the samples were put in Pt filament with 0.5 HBr. Samples were first dried down with a heat lamp and kept loaded in the center of the filament and a nitrogen cold trap was loaded. After that, the filament was heated with 1 amp of current and barium solution was added on top of the dried samples, then dried to a thin layer ready for analysis in the mass spectrometer. For the Os, the filament was heated slowly to a dark red color (showing that the temperature was high enough to download the Os filament), and at the same time the temperature, focus and also the pick signal could to be checked. Os was emitted as OsO− 3 . Os isotopic ratios were calculated relative to 188 Os. The mass spectrometer yielded the isotopic ratios of Os: 192/188, 190/188, 189/188, 187/188, 186/188, and 185/188. Rhenium was loaded and run in a similar way to that for osmium, but without melting the load. Sample preparation for Re analysis included picking the sample in Ba(NO3 )2 solution, and then loading this directly on a platinum filament, and drying it under a heat lamp before mass spectrometry. The Re was emitted as ReO− 4 ions, at slightly lower temperatures than Os.

4. RE-OS RESULTS The age of the chromites and mantle tectonites in the ophiolitic mélange in the North China Craton were studied using in-situ Re-Os analysis of chromites included in podiform chromite. The chromite grains were selected from coarse mineral concentrates from samples that had visible chromite grains, and appropriate chemical compositions for the study (i.e., both Re and Os had to be present). Rhenium and Os concentrations and isotopic ratios for the ophiolitic chromites are given in Table 1. In the 10 samples that have both elements present as detected in mass spectrom-

4. Re-Os Results

Table 1. Re-Os data from the North China Craton Sample Rock type Re (ppb) Os (ppb) 187 Re/ 188 Os O1-10-30 Nodular chromite 0.103 35.4133 0.0139 O1-10-40 Chromite 0.067 34.8094 0.0093 NC-6 Antinodular chromite 0.0522 23.5982 0.0106 NC-20 Antinodular chromite 0.0332 39.4955 0.0092 with envelope Tk-NCD-01-33-a Disseminated chromite 0.0247 5.2018 0.0227 Tk-NCD-01-33-b Banded chromite 0.0366 23.7075 0.0074 O1-10-16 Chromite 0.0786 36.0948 0.0105 NC-13-1 Chromite 0.0872 40.0789 0.0105 NC-19-4 Nodular chromite 0.1778 43.8438 0.0195 Z-3 Folded chromite 0.1735 28.6233 0.0292

279

187 Os/

0.11202 0.11016 0.11161 0.11052

TRD (Ga) 2.40 2.62 2.43 2.57

TBE (Ga) 2.391 2.623 2.430 2.572

0.11131 0.11000 0.11122 0.10969 0.11068 0.11058

2.54 2.63 2.48 2.69 2.61 2.51

2.544 2.634 2.483 2.695 2.614 2.510

188 Os

Fig. 1. Isochron plot of the chromites samples.

etry, the Re varies between 0.0246 ppb and 0.1777 ppb and Os ranges from 5.2018 ppb to 43.8438 ppb. The chromites from the Zunhua and Dongwanzi ophiolites have Archean TRD and TBE ages (2400–2700 Ma and 2300–2700 Ma). All of the samples have 187 Re/188 Os ratios that are much less than the chondritic average of approximately 0.4. The low ratios mean that there is minimal age correction for most samples, even if quite old. The errors were ±0.0009. The mass spectrometry fraction of 187 Os/188 Os was corrected using 192 Os/188 Os = 3.082614 and the measured ratios were corrected for blank contribution, and with an uncertain blank of ±0.00024. The Re weights have uncertainty errors of about ±0.0073 ppb, in contrast of the Os errors with a maximum of ±0.0232 ppb. The isochron diagram in Fig. 1 shows that seven of the samples nearly define an isochron with a slope corresponding to the average TRD age of 2547 ± 10 Ma and an initial 187 Os/188 Os of 0.1107 ± 0.0013. In Table 1 the concentrations of Re and Os are listed in ppb (10−9 g) by weight. The age of each sample was calculated after some correction

280

Chapter 8: Re-Os Isotope Chemistry and Geochronology of Chromite

Fig. 2. Relationship between the Os proportion and the ages of each rock. 187 Os/188 Os was in moles, and it calculated with an error between ±0.00018 and 0.00024.

of the Os blank and Re blank. The Re-Os varies for each type of mineral, indicating that it changes with the structure and texture of the rock itself. For example the nodular chromite (sample NC-19-4) has high concentrations of Osmium and Rhenium (43.8438 ppb and 0.1778 ppb) compared to the disseminated chromite. Some correlations between the two elements (Re and Os) were determined during the analyses. Fig. 2 shows a relationship between the Os isotopic composition and the age of the rocks. The trend line shows the evolution and the variation of Os ratios versus age. The ages of the rocks were calculated using the Os and Re percent included in the rocks themselves. Moreover the dating method using both elements was derived from the rate and concentration of 187 Re and Os isotopic in the samples.

5. DISCUSSION The distribution of TRD ages indicate that the Zunhua chromites and Dongwanzi ophiolites of North China craton formed during the period 2.40–2.7 Ga. The Re-Os ages reported here are in remarkable agreement with the circa 2505 ± 2.2 Ma U/Pb ages reported from gabbroic crustal sections of the Dongwanzi ophiolite (Kusky et al., 2001), supporting the correlation of the Zunhua chromites and Dongwanzi ophiolite. The chromite concentration for each sample changes with the texture of the rock itself, and it affects the percentage of Os and Re in the samples. Disseminated chromite samples (TK-NCD-01-33-a) yielded a very small chromite grain size, with 0.0247 ppb Re and 5.2018 ppb for Os. In this sample Re and Os concentrations were very low because they had other minerals left with the chromite sample, that demands more preparation time than the nodular chromites, which have big nodular chromite that is easy to purify.

6. Conclusion

281

Important variations exist in Re-Os contents and Os isotope ratios in the chromites, because the formation of the ophiolites overlapped the main chromite mineralization. Those variations are observed in our samples. The relationships between isochronal trends of the 187 Os/188 Os versus 187 Re/188 Os were dominated by the variation of Os. Three of the studied rocks plot far from the isochron line. This could result from contamination during the analyses, or the samples may have had too much primary silicate left in the chromite grains. The latter suggestion is favored because during the sample crushing and purification procedure, these samples had some light minerals left in the samples. However, a pseudo isochron trend appears using seven of the samples (Fig. 1). Normally the Re concentration in chromite is around 0.01 to 0.02 ppm (in magmatic rocks), but in this case a very low Re concentration was found, varying from 67 ppt to 1778 ppt (10−12 g). This range in values is characteristic of ophiolites. The chondritic to sub-chondritic initial also is interesting in that it shows more similarity to the values found for abyssal peridotites than OIB’s, pointing to a ocean-ridge rather than plume setting for the initial formation of these peridotites. Moreover, Os is normally sited in trace phases within chromite, not in the chromite structure itself. As both isotopes are siderophiles and the parent and daughter isotopes tend to be concentrated in oceanic crust, which was deformed and metamorphosed, the proportion of Re-Os changed throughout the ophiolite’s genesis until they were emplaced in the North China craton.

6. CONCLUSION Chromite can be used as a petrogenetic indicator for tectonically uncertain ultramafic massifs. The mineral chemistry of the Zunhua chromites suggests that they formed in a suprasubduction zone ophiolitic environment. The Cr#, TiO2 wt% and V2 O5 wt% indicate partial melting of a depleted source under high fO2 conditions. Applied to the modern tectonic paradigm these compositions indicate formation in a SSZ setting and fluid aided melting of a previously depleted source. However, slightly higher heat flow in the Archean could lead to great enough degrees of partial melting to form high Cr# number chromite without fluid aided melting of previously depleted mantle. Mass spectrometry measurements for Re-Os from cumulate chromite and mantle podiform chromite yielded an average 2547 ± 10 Ma TRD ages for the mantle and crustal rocks of the Zunhua mélange and Dongwanzi ophiolite. These ages are virtually indistinguishable from the 2505 ± 4 Ma U/Pb (zircon) age obtained from crustal gabbroic cumulates, supporting the contention that the Dongwanzi ophiolite and Zunhua mélange are related parts of a large Archean ophiolitic terrane. We name this belt the Zunhua ophiolitic mélange, referring to the older oceanic portions of the Zunhua structural belt that were accreted in the oceanic realm, before being intruded by arc-like TTG magmas, and deformed again (see Kusky et al., 2004). The Zunhua chromites occur within dunite envelopes, grading into mantle harzburgite. Such textures form by interaction between a melt and a depleted mantle source. This raises

282

Chapter 8: Re-Os Isotope Chemistry and Geochronology of Chromite

the possibility that mantle and magmatic dynamics were similar to modern systems and that suprasubduction zone environments operated by similar magmatic systems to those of today.

ACKNOWLEDGEMENTS This work was supported by the U.S. National Science Foundation grants 0125925 and 02-07886 (awarded to T. Kusky), China National Science Foundation grant no. 49832030 (awarded to J.H. Li), Peking University Project 985, and St. Louis University. Visits in the field by Alan Huang, Ali Polat, Bob Tucker, Brian Windley, Alfred Kröner, John Meyers, and many other scientists on NSF, NNSF, and Interridge Field trips led to interesting discussions, and are greatly appreciated.

REFERENCES Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga Podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithosphere mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337. Huson, R., Kusky, T.M., Li, J.H., 2004. Geochemical and petrographic characteristics of the central belt of the Archean Dongwanzi ophiolite complex. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 283–320. Kusky, T.M., Li, J.H., Tucker, R.T., 2001. The Archean Dongwanzi ophiolite complex, North China Craton: 2.505 Billion Year Old Oceanic Crust and Mantle. Science 292, 1142–1145. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Li, J.H., Kusky, T.M., Huang, X., 2002. Neoarchean podiform chromites and harzburgite tectonite in ophiolitic mélange, North China Craton, Remnants of Archean oceanic mantle. GSA Today 127, 4–11. Li, J.H., Kusky, T.M., Niu, X.L., Feng, J., Polat, A., 2004. Neoarchean massive sulfide of Wutai Mountain, North China: A black smoker chimney and mound complex within 2.50 Ga-old oceanic crust. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 339–361. Smoliar, M.I., Walker, R.J., Morgan, J.W., 1996. Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099–1102. Tsuru, R.J., Walker, R.J., Kontinen, A., Peltonen, P., Hanski, E., 2000. Re-Os isotopic systematics of the 1.95 Ga Jourma ophiolite complex, northeastern Finland. Chemical Geology 164, 123–141. Walker, R.J., Prichard, H.M., Ishiwatari, A., Pimentel, M., 2002. The Osmium isotopic composition of convecting upper mantle deduced from ophiolitic chromites. Geochimica et Cosmochimica Acta 66, 329–345.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

283

Chapter 9

GEOCHEMICAL AND PETROGRAPHIC CHARACTERISTICS OF THE CENTRAL BELT OF THE ARCHEAN DONGWANZI OPHIOLITE COMPLEX R. HUSONa , T.M. KUSKYa AND J.H. LIb a Department

of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63130, USA b Department of Geology, Peking University, Beijing, China

The Dongwanzi ophiolite of the North China craton (Eastern Hebei province) represents the oldest known complete section of oceanic crust (2505 ± 2 Ma). It consists of three main northeast striking outcrop belts, and numerous related blocks in tectonic mélange. A representative suite of samples focused on the central belt were collected from the ophiolite and related rocks, and were analyzed by X-ray fluorescence to determine its geochemical characteristics. Petrographical studies of thin sections were also conducted to determine mineral assemblages. Major and trace element concentrations are compared with known concentrations from well-studied ophiolites and rocks from tectonic settings to determine tectonic environment of formation. Major element analysis shows samples are subalkalic (in particular, calc-alkaline) to alkalic. Trace element analysis shows enrichment of large ion lithophile elements as well as depletion of high field strength elements relative to midocean ridge basalts. Tectonic settings for ophiolite formation include mid-ocean ridges, suprasubduction zone (including forearc and backarc basins), and arcs. Calc-alkaline geochemical characteristics of oceanic rocks have predominantly been identified in suprasubduction zone settings and their occurrence in the Dongwanzi ophiolite suggests a similar tectonic setting. Trace element signatures are similar to suprasubduction zone ophiolites indicating formation above a subduction zone. 1. GEOLOGIC SETTING The North China, Tarim, and Yangtze platforms make up the bulk of Precambrian rocks in China. The platforms cover an area of about 4 million km2 and occupy about 40% of China (Figs. 1 and 2). The North China Craton is approximately 1.7 million km2 in area and forms a triangular shape covering most of north China, the southeastern part of NE China, Inner Mongolia, Bohai Bay, Northern Korea, and part of the Yellow Sea regions (Ma, 1998). It is divided into the Eastern and Western Blocks, and the central orogenic belt (Li et al., 2000; Zhao et al., 2001; Kusky and Li, 2003). The Western Block and Central Orogenic Belt (COB) are separated by the younger Huashan-Lishi fault in the south and DOI: 10.1016/S0166-2635(04)13009-0

284

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 1. Geologic setting of the North China Craton (after Zhao et al., 1999; Wilde et al., 2002).

by the Datong-Duolun fault in the north. The Central Orogenic Belt and Eastern Block are locally separated by the younger Xinyang-Kaifen-Jianping fault (Fig. 3). Rock formation ages range from 2.7 to 2.5 Ga in the Eastern and Western blocks and 2.5 Ga for the central orogenic belt. Detrial zircons from the Eastern Block have been dated at 3.5–3.8 Ga and are the oldest ages obtained from the North China Craton (Bai et al., 1997; Kröner et al., 1998; Qian and Li, 1999; Li et al., 2000; Zhao et al., 2001). The average lithospheric thickness ranges from 80 km in the east to 160 km in the west (Yuehua and Wang, 1997). The difference in thickness between the blocks is reflected in the lack of a thick mantle root in the Eastern Block. The Zunhua structural belt (ZSB) is located in the northern part of the Central Orogenic Belt, in the eastern part of the Hebei province and covers a 130 × 20 km2 area (Bai and Dai, 1998; Kusky et al., 2001, 2004). Numerous ultramafic boudins have been identified within the ZSB having flat rare earth element to light rare earth element-depleted patterns similar to basalts from suprasubduction and mid-oceanic ridge settings (Li et al., 2002). Field and petrographic analyses revealed typical ophiolitic assemblages including cumulates and pillows, podiform chromite, well-preserved sheeted dikes, pyroxenite, wehrlite, and partly serpentinized harzburgite (Kusky et al., 2001; Li et al., 2002). Tectonic ultramafic blocks in the Zunhua structural belt have been identified in the same belt as the ophiolite and can also be traced into a complex mélange zone found along

2. Field Description and Petrography

285

Fig. 2. Tectonic map of the North China Craton showing divisions into Eastern and Western blocks separated by the Central Orogenic belt and Qinglong foreland basin (after Wilde et al., 2002; Zhao et al., 1999; Ma, 1998; Kusky and Li, 2003).

strike with the ophiolite (Li et al., 2002; Kusky et al., 2004; Huang et al., 2004). Original mantle textures have been overprinted in the Dongwanzi ophiolite and so the tectonic blocks are thought to preserve the deeper sections of the ophiolite (Kusky et al., 2001; Li et al., 2002). Podiform chromites are found within harzburgite tectonite and dunite host rocks and contain nodular and orbicular textures. These types of chromite textures are found exclusively in ophiolites and are thought to have formed during partial melting of the flowing upper mantle (Li et al., 2002). Also podiform chromite can only form where water content of the basaltic melt is high enough to exsolve a water-rich fluid phase (Matveev and Ballhaus, 2002). This suggests that the most likely environment for podiform chromite formation is a suprasubduction zone setting.

2. FIELD DESCRIPTION AND PETROGRAPHY The Dongwanzi Ophiolite Belt is located in the NE part of the Zunhua Structural Belt and has been interpreted to represent a large ophiolitic block within the Zunhua ophiolite mélange (Kusky et al., 2001). The Dongwanzi Ophiolite Belt is about 50 km long and

286

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 3. Distribution of basement rocks in the North China Craton inset shows area of interest (after Zhao et al., 1999, 2001; Kusky and Li, 2003).

between 5 and 10 km wide, see Figs. 4 and 5, and preserves the upper or crustal part of the ophiolite suite, with deeper sections of the ophiolite being preserved in related blocks (Kusky et al., 2001; Li et al., 2002). The ophiolite is broken into three main thrust slices, including the NW Belt, Central Belt, and the SE Belt. In this contribution we describe aspects of all belts, but focus on the Central Belt. All of the belts are metamorphosed to at least greenschist facies and typically amphibolite facies, with conditions approaching granulite in the west. However, primary textures are well preserved in the Central and Eastern Belts, and some primary mineralogy remains. All of the rock descriptions below should be read as preceded by the term meta, but for ease of reading, the term meta is typically omitted.

287

Fig. 4. Location map, with samples.

2. Field Description and Petrography

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 5. Sample location for the central belt.

288

2. Field Description and Petrography

289

2.1. Ultramafics A high temperature shear zone intruded by the Cuizhangzi diorite-tonalite complex marks the base of the ophiolite. Exposed ultramafic rocks along the base of the ophiolite in the southeast part of the central orogenic belt, southeast part of the southeast belt, and in the Zunhua Structural Belt to the southwest, include strongly foliated and lineated dunite and layered harzburgite (Kusky et al., 2001; Li et al., 2002). Aligned pyroxene crystals characterize the peridotites in this section. Strong deformation of serpentinized harzburgite resulted in strongly foliated rock. Harzburgite shows evidence for early high temperature deformation. This unit is interpreted to be part of the lower residual mantle, from which the overlying units were extracted. Harzburgite from the upper central belt has a mineral assemblage (see Fig. 6, NC200113b and Table 1) including orthopyroxene and olivine with minor amounts of clinopyroxene. Orthopyroxene grains are subhedral to anhedral and are distinguished from olivine by the presence of cleavage and green to pink pleochroism. Minor apatite occurs as euhedral hexagonal to long tabular grains, and sphene is present as fractured diamond shaped grains with corroded edges. Fine-grained alteration products including serpentine are also present. Unlike sample NC2001-13b, sample NC2000-17b, has a slightly banded appearance, alternating between layers of faintly aligned hornblende grains and the finer-grained matrix. The mineral assemblage consists of coarse-grained hornblende in a fine-grained matrix of euhedral clinopyroxene, olivine, and opaques. These grains are subhedral to euhedral, show primary twinning, and are zoned. Hornblende grains contain inclusions of apatite, chromite, and some clinopyroxene. Amphibole-bearing pyroxenites (NC2001-13-1) from the central belt include mineral assemblages of primary coarse-grained hornblende and smaller primary titanite grains in a fine-grained altered matrix. Hornblende grains are elongate tabular to equant shaped, show twinning, and are partially altered to actinolite in places, whereas sphene grains are typically diamond-shaped. Two sizes of aligned hornblende grains give the rock a banded or layered appearance. Banding of the hornblendes grades to randomly orientated grains. Clinopyroxene grains are zoned and show oscillatory extinction. Accessory minerals include epidote and chlorite, which suggest greenschist metamorphism. Altered plagioclase feldspar is present as well as apatite. Sphene is present as small, fragmented grains. Apatites form as elongate to characteristic hexagonal shape. The mineral assemblage of a hornblende-bearing pyroxenite from the upper central belt (sample NC2000-16) includes rounded clinopyroxene grains embedded in large poikilitic hornblendes. Larger clinopyroxene grains contain inclusions of apatite and are partially altered to hornblende. Apatite is present and occurs both as euhedral inclusions in clinopyroxene as well as intergrown with the smaller zoned pyroxene. Opaques appear as rounded grains and are also intergrown with clinopyroxene and hornblende. Additional pyroxenites from the central belt (NC2001-13c) show mineral assemblages that include hornblende and clinopyroxene with abundant opaques. Hornblende and clinopyroxene grains are poikilitic and randomly oriented. These grains are zoned and

290

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 6. Ultramafics. Photographs of sample NC2001-13-1: amphibole-bearing pyroxenite: (A) twinned hornblende grains in fine-grained altered matrix. NC2001-13b: harzburgite: (B) orthopyroxene and olivine with primary sphene (center of picture). NC2000-17b: hornblende pyroxenite: (C) twinned hornblende in a fine-grained matrix of euhedral clinopyroxene and opaques; (D) hornblende grains with irregular grain boundaries. Photographs of NC2000-28b: plagioclase-bearing pyroxene hornblendite: (E) hornblende and pyroxene in fine-grained altered matrix. Photographs in cross nicols, bars represent 1 µm.

Sample ID NC2001-2 NC2001-3 NC2001-4:olv-gabbro NC2001-4:hbl-gabbro NC2001-4a NC2001-5 NC2001-5b NC2001-6a:olv-gabbro NC2001-6a:dike NC2001-6b NC2001-7 NC2001-7b NC2001-13-1 NC2001-13-2 NC2001-13b NC2001-13c NC2000-15b NC2000-15c NC2000-16 NC2000-17b NC2000-19 NC2000-19b NC2000-19c

olv 60

cpx 30

25

10 10 20

5

25

5

25 65 55 65

15

20 40 70 40 15 10 15

hbl 25 10 45 50 30 20 30 70 20

5 45

plag 8 (vein) 40 50 (alt) 45/5 (alt) 30 30 30 30 (alt) 40

qtz

bio

serp

apa

alt

opaq

20

2

Other 10

5 35 30

10

5 10

10 10

10 10

3

5

5 5 5

10

3 3

10 25 15

Phlogo: 20 Phlogo: 15

1

5 20

20 5

40 45 35

sph 2 3 5

5 5

15 30 20 25

chl

5

44 (alt)

4

epi

2. Field Description and Petrography

Table 1. Modal estimates for the Dongwanzi Ophiolite, vol%∗

3 10 10 15 2 2

Calcite: 10

30 20 25 (continued on next page)

291

292

Table 1. (Continued) olv

cpx

hbl

20 3

10 10

5 20 3

30

5

70

plag 10 30 45 67 55/5 (vein) 38 42 (alt) 15 35 20

qtz 90 66 10

25 (alt) 55 (alt) 10

5 15

bio

epi

chl

serp

apa

sph

alt

opaq

Other

3 15 15 20

38

3 7

2 3 3

3 5

35 65 3 10

30 15 20 10 30 15

5 5 3 5 5

43 35 20 40

Calcite: 10 Calcite: 20 Calcite: 10

20 Calcite: 10 15

5 1

Calcite: 15

Abbreviations: olv—olivine, cpx—clinopyroxene, hbl—hornblende, plag—plagioclase feldspar, qtz—quartz, bio—biotite, epi—epiodite, chl—chlorite, serp— serpentine, apa—apatite, sph—sphene, alt—alteration product, opaq—opaque. ∗ Modal percentages are visual estimates.

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Sample ID NC2000-20 NC2000-21a NC2000-21b NC2000-22-1 NC2000-22-2 NC2000-23c NC2000-23d NC2000-24a NC2000-24c NC2000-26d NC2000-27c NC2000-27d NC2000-28 NC2000-28b NC2000-28d

2. Field Description and Petrography

293

have simple twins along with kink bands. Opaques include ferro-chromite and predominantly chromite and contain red unidentified minerals (non-isotropic) that are also found throughout some sections. Accessory minerals include apatite (found as large, relatively, euhedral, hexagonal shaped inclusions), sphene and prehnite. Fine-grained material present is plagioclase feldspar that has altered to sericite. Ultramafics from the northern part of the southern belt include plagioclase-bearing pyroxene hornblendites (sample NC2000-28b). This sample shows a faint layering of hornblende. Mineral assemblages include abundant interlocking grains of hornblende. Twinning commonly occurs in hornblende but is not so apparent in clinopyroxene. Biotite shows bent cleavage. Feldspar veins (about 1.5 mm in width) are present and grains show albite and deformation twinning. Apatite is present in trace amounts as inclusions in hornblende. 2.2. Mafic and Ultramafic Cumulates The cumulate layer represents the transition zone between the lower ultramafic cumulates and upper mafic assemblages. The lower part of the sequence consists of orthocumulate pyroxenite, dunite, wehrlite, lherzolite and websterite, and olivine gabbro-layered cumulates. Many layers grade from dunite at the base, through wehrlite and are capped by clinopyroxene (Kusky et al., 2001). Basaltic dikes cut through the cumulates and are similar mineralogically and texturally to dikes in the upper layers. Cumulates in the upper central belt (see Fig. 7, NC2001-2) include mineral assemblages of cumulate olivine cut by younger veins of altered alkali feldspar and epidote with subhedral to euhedral sphene present in veins. Varied grain sizes of olivine result from metamorphic recrystallization and show polygonal mosaic texture. Larger olivine grains are embayed by smaller olivine grains. Clinopyroxene is present bordering the vein.

Fig. 7. Cumulates. Photographs of sample NC2001-2: wehrlite: (A) cumulate olivine. NC2001-7: pyroxenite: (B) phlogopite (phlog) and cumulate diopside. Photomicrographs in cross nicols, bars in photomicrographs represent 1 µm.

294

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Ultramafic cumulates (samples NC2001-7 and NC2001-7b) show mineral assemblages consisting of altered cumulate diopside and large primary phlogopite. Phlogopite results from contact metamorphism with a chlorite-phlogopite-marble. Diopside grains have high birefringence at the rim of the grain, are altered, and have fluid inclusions. These grains also have undulose extinction as a result of high strain. Quartz is present in veins. 2.3. Gabbros The gabbro complex of the ophiolite is up to 5 km thick and grades up from a zone of mixed layered gabbro and ultramafic rocks to one of strongly layered gabbro that is topped by a zone of isotropic gabbro (Kusky et al., 2004). Thicknesses of individual layers vary from centimeter to meter scale and include clinopyroxene and plagioclase-rich layers. Layered gabbros from the lower central belt (see Fig. 8, sample NC2000-19) alternate between fine-grained layers of pyroxene and metamorphic biotite that are separated by layers of metamorphic biotite intergrown with quartz. Biotite and pyroxene layers show a random orientation of grains. Coarse-grained veins of feldspar and quartz are concentrated along faults and fractures. Plagioclase feldspar shows core replacement (sausseriticization) and typically has irregular grain boundaries. Microcline is present and identified by inversion twinning and also shows irregular grain boundaries. Some of the feldspar is altered to muscovite. Biotite has bent cleavages and pyroxene is heavily altered. Alteration is concentrated along faults. In other layered gabbros from the lower central belt (samples 19c and 19b), the mineral assemblage consists of coarse-grained hornblende grains in a matrix of finer grained pyroxenes and feldspars. Hornblende grains show subhedral prismatic habit and are oriented randomly (NC2000-19c-002) grading into layered subhedral elongated grains (NC2000-19c001). Both types of grains show twinning and low-grade metamorphism, altering to biotite and chlorite. Clinopyroxene grains present have rounded edges, are subhedral to anhedral and commonly occur as inclusions within the larger hornblendes or altered feldspars. Most of the plagioclase feldspar is altered and/or shows core replacement. Some opaques are also present in small amounts. Gabbro from the lower central belt consists of altered plagioclase feldspar and clinopyroxene with small amounts of unaltered hornblende and quartz (sample NC0021b). The plagioclase feldspar is almost entirely altered, and shows bent deformation twins. Micrographic textures are also present in the feldspar. Altered clinopyroxene shows coronas of actinolite. Quartz grains have fluid inclusions. Chlorite patches and calcite veins occur throughout. Opaques are present interstitially and contain coronas of hornblende. Gabbro from the southern part of the southern belt (samples NC200022-1 and NC0021b) is slightly coarser grained but very similar to those of the lower central belt. Mineral assemblages include coarse-grained plagioclase and clinopyroxene, both of which are in various states of alteration. Plagioclase grains show deformation twinning, micrographic textures, and have intergrown grain boundaries with quartz. Quartz grains contain fluid inclusions. Both hornblende and clinopyroxene have alteration coronas of biotite and actinolite. Coarse-grained hornblende contains inclusions of apatite and plagioclase with

2. Field Description and Petrography

295

Fig. 8. Gabbros. Photographs of sample NC2000-19: altered layered gabbro: (A) altered layers of fine-grained interstitial material. NC2001-19b: gabbro: (B) altered hornblende in a fine-grained matrix of clinopyroxene. NC2000-19c: pyroxenite gabbro: (C) twinned and aligned hornblende grains with clinopyroxene. NC2000-21b: pyroxene-hornblende gabbro: (D) altered hornblende in matrix of altered feldspar. NC2000-22-1: pyroxene-hornblende gabbro: (E) feldspar with some hornblende grains. NC2000-15b: gabbro: (F) altered gabbro. Photos in cross nicols, bars in photomicrographs represent 1 µm.

296

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

chlorite present as an alteration product. Opaques are present in the altered matrix and, where larger grained, have coronas of biotite. Recrystallized quartz shows irregular grain boundaries and is present with biotite as thin veinlets. Gabbro from the northern part of the southern belt (sample NC2000-15b) includes the mineral assemblage plagioclase, clinopyroxene, hornblende, biotite, and quartz with minor apatite. Plagioclase grains show micrographic texture and secondary twinning (bent deformation twins), and are altered. Clinopyroxene has irregular, rounded, grain boundaries and is partially altered to hornblende. Hornblende is twinned in some places and also altered to chlorite. Biotite shows bent cleavage and opaques occur as rounded grains. 2.4. Sheeted Dikes and Pillows The sheeted dike complex is discontinuous over several kilometers. More than 70% of the dikes exhibit one way chilling on their NE side (Kusky et al., 2001, 2004). Gabbro screens are common throughout the complex and increase in number and thickness downward marking the transition from the dike complex to the fossil magma chamber. In some areas, the gabbro is cut by basaltic-diabase dikes, but in others it cuts through xenoliths of diabase suggesting comagmatic formation. The upper part of ophiolite consists of altered and deformed pillow basalts, pillow breccias, and interpillow sediments (chert and banded iron formations). Many of the pillows are interbedded with more massive flows and cut by sills; however, some well-preserved pillows show typical lower cuspate and upper lobate boundaries that define stratigraphic younging (Kusky et al., 2001). Sheeted dike samples were not available for petrographic description or geochemical analysis. A. Polat (University of Windsor, personal communication) analyzed pillow lavas from the ophiolite. Preliminary geochemical data on these samples was obtained from Polat and is included in the trace element geochemistry and tectonic discrimination diagrams. 2.5. Related Dikes The Dongwanzi ophiolite is cut by at least four dike swarms ranging in composition from mafic dikes, gabbroic and diabase dikes. A cross-cutting granodiorite dike has been dated at about 300 Ma (Kusky et al., 2004). Related dikes in the upper central belt (see Fig. 9, sample NC2001-4) are characterized by large fragmented grains of clinopyroxene that are partially altered to hornblende and biotite. Large grains of fragmented apatite are present as well. Hornblende appears altered, zoned, and tabular. Small grains of sphene form a boundary between coarser grains and a finer grained mixture of biotite, pyroxene, and interstitial material. Abundant biotite and chlorite are present with random orientation. Younger dikes from the upper central belt include sample NC2001-3, a pink felsic dike, that is heavily altered. The mineral assemblage consists of fine-grained interstitial material replacing plagioclase (identified by twinning) and chlorite. Unaltered minerals include

2. Field Description and Petrography

297

Fig. 9. Dikes. Photographs of sample NC2001-4: olivine-gabbro dike: (A) sheared contact between two olivine-gabbro dikes in cross nicols and plane light. NC2001-3: felsic dike: (B) fine-grained interstitial material replacing plagioclase feldspar in cross nicols and plane light. NC2001-5: hornblende gabbro: (C) hornblende and twinned clinopyroxene in fine-grained matrix of altered plagioclase; (D) hornblende and clinopyroxene in fine-grained matrix of altered plagioclase. Photos in cross nicols, bars in photomicrographs represent 1 µm.

hornblende and sphene. Large subhedral grains of hornblende that show twinning and zoning are concentrated in zones. Hornblende is partially altered to actinolite. Fractured sphene grains are found throughout the sample. Opaques are surrounded by a highly birefringent unidentified mineral. The mineral assemblage of the dated plagioclase rich, leucocratic dike (NC2001-5) include large abundant hornblende with apatite inclusions and pyroxene partially altered to hornblende. Small rounded pyroxene grains are mantled by hornblende in places and occur as inclusions within hornblende. Hornblendes are intergrown, randomly oriented and

298

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

show deformation twins (kink bands) as well as simple twinning. Opaques are present and contain inclusions of apatite.

3. GEOCHEMISTRY 3.1. Geochemical Criterion for Tectonic Setting Weathering and metamorphic processes may affect the original major and trace element compositions of ophiolites and other rock suites. Studies of trace elements in ocean floor basalts indicated that certain elements are more affected than others (Pearce and Cann, 1971; Pearce, 1975). Ti, Zr, and Y are insensitive to secondary alteration whereas Sr, Rb, and K are very affected. If the ophiolite formed above a subduction zone, partial melting of the mantle due to dehydration of the subducting plate would enrich it in certain elements and hydrous minerals. However, if the ophiolite formed in a mid ocean ridge setting, it would be expected to be essentially anhydrous and have different geochemical signatures. Geochemical signatures from a suprasubduction zone include high concentrations of K, Rb, Ba, Sr, and Th and low concentrations of Nb, Zr, Ti, Y, and Yb relative to MORB (e.g., Pearce et al., 1984; Shervais, 2001). 3.2. Suprasubduction Zone Ophiolites Most suprasubduction zone ophiolites have geochemical signatures similar to island-arcs and are thought to have formed during the initial stages of subduction prior to development of the volcanic arc (Pearce et al., 1984). Suprasubduction zone ophiolites can form in back arc, as well as fore arc tectonic settings. Back arc suprasubduction zone ophiolites show lithologic and chemical characteristics that are very similar and almost indistinguishable from mid-ocean ridge basalts. Fore arc ophiolites, on the other hand, show chemical signatures that are readily distinct from MORB (Shervais, 2001; Bloomer et al., 1995). These chemical signatures include enrichment of the LILE (Saunders and Tarney, 1984). They are low in high field strength elements and have low Ti/V values. Vanadium and titanium are high field strength elements that are strongly incompatible and immobile. They are both resistant to weathering and metamorphism and are relatively stable during amphibolite and granulite facies metamorphism (Shervais, 1982). MORB basalts are restricted to Ti/V ratios between 20 and 50 (Shervais, 1982) whether they are enriched or not. This field overlaps with that of back arc basins that also have ratios from 20 to 50. Arc tholeiites have Ti/V ratios < 20 with a minor overlap with MORB ratios. Calc-alkaline series have negative slopes implying magnetite fractionation resulting in Ti/V ratios > 15. As mentioned above, back arc basin magmas and MORB ratios are similar geochemically. These discrimination methods were devised using and applied to Phanerozoic examples. However, samples from the Dongwanzi Ophiolite show calc-alkaline characteristics suggesting a SSZ formational environment.

3. Geochemistry

299

3.3. Sample Preparation Major and trace element analysis of 40 representative samples from the Dongwanzi ophiolite were carried out to determine the range and geochemical affinity of the Dongwanzi magmas. Samples selected include dikes, layered gabbros and gabbros, and ultramafics from the central and southern belts. Major elements Na, Mg, Al, Si, P, K, Ca, Ti, Mn, and Fe were analyzed from fused glass discs. Trace element analysis of V, Cr, Co, Ni, Zn, Ga, Rb, Sr, Y, Zr, Nb, Sn, Ba, and Rb was conducted on pressed powder pellets. All samples were prepared and analyzed at the Washington University X-ray fluorescence laboratory. Samples were cut and polished to remove weathered surfaces. After being crushed, the samples were pulverized using a ceramic shatterbox. The powders for major element analysis were combined with flux in a platinum crucible and mixed while melting to ensure a homogeneous end product. The melt was poured into a heated platinum mold and cooled. Powders for trace element analysis were mixed with a binding agent and pressed. Detection limits of analysis are to 1 part per million. 3.4. Geochemical Characteristics of the Dongwanzi Ophiolite Traditionally, volcanic samples are analyzed for geochemical discriminant studies because the compositional fields of various tectonic environments were determined using the chemical range determined from modern volcanic rocks. Inclusion of cumulates and other rock types in discriminant analysis could involve additional processes such as crystal fractionation that can change the composition of the rocks, and hence complicate the interpretation of results. Analyzed samples for the Dongwanzi ophiolite includes dike and pillow basalts (pillow basalt data provided by Polat, in press) represented by solid symbols. Also included in the graphs are gabbros, ultramafics rocks, and cumulates (represented by outlined symbols). These samples are graphed for comparison only and should not be used for determining tectonic setting. However, these samples generally plot in the same fields as the volcanic rocks. 3.5. Major Elements Characteristics The samples were plotted using MgO on the x-axis due to the low SiO2 concentration of many of the mafic and ultramafic members (Figs. 10 and 11). However, the dikes and pillows were plotted on a K2 O vs. silica diagram and TAS to determine the alkalic character of these rocks (Fig. 12). These samples plotted in the medium- to high-K fields suggesting the samples are calc-alkaline. The ultramafics are characterized by 32–49 wt% SiO2 , 0.7–2.5 wt% TiO2 , 4.3– 11.61 wt% MgO, 4.5–17.0 wt% Al2 O3 , and 8.6–21.09 wt% Fe2 O3 and range from 0.66 to 6.23 in total alkali content (Table 2). They are plotted using both MgO and SiO2 for comparison with the dikes and pillows. Examination of the thin section of sample NC200113b shows the sample is heavily altered and has a high amount of apatite explaining the high CaO and low K2 O contents of this sample (sample location, Figs. 4 and 5). Samples

300

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 10. Harker diagrams for samples from the Dongwanzi ophiolite using SiO2 as the x-axis. Low SiO2 samples should not be considered in geochemical trends and are only plotted for comparison.

NC2000-17b, NC2001-13c, and NC2001-5gabbro, from the middle central belt, all show high TiO2 , FeO, and P2 O5 values whereas sample NC2000-24a from the south belt exhibits only high TiO2 . For the most part, samples show similar trends based on location.

3. Geochemistry

301

Fig. 11. Harker diagrams for samples from the Dongwanzi ophiolite using MgO as the x-axis.

Anomalous values (high TiO2 , FeO, P2 O5 , MgO, and CaO) result from proximity to later dikes and/or faults (where samples may have been affected by fluids). The ultramafics as a whole show continuing trends with cumulate and gabbro groups.

302

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 12. (Top) Division of subalkalic rocks using the K2 O vs. silica diagram with fields and nomenclature as defined by Le Maitre et al. (1989) and Rickwood (1989). This diagram is only applicable to pillow and dike samples. Ultramafic, cumulate, and gabbro samples are plotted for comparison only. Pillows and dike samples plot within calc-alkaline fields. (Bottom) Total alkali vs. silica diagram.

3. Geochemistry

303

Table 2. Major element oxides (wt%) and trace element∗ (ppm) analysis of ultramafic samples from the Dongwanzi Ophiolite† SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI

NC28c 49.81 16.98 1.89 6.78 6.90 7.04 4.32 1.91 0.77 0.31 0.13 1.64

NC28b 42.21 12.47 3.11 11.17 11.61 9.83 2.08 1.60 1.28 0.49 0.20 1.37

NC15c 44.66 12.33 2.98 10.73 6.26 7.28 2.78 0.54 1.62 0.13 0.18 7.9

Total

98.48

97.42

97.39

Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

< 6.3 6.2 168.2 19.5 111.7 48.9 < 8.7 23.6 106.4 110.8 164.9 341.3 690.2 0.0 35.2

< 6.9 9.2 229.1 29.3 290.8 31.4 4.4 24.6 174.0 184.3 271.3 517.8 401.1 0.0 60.2

< 6.6 6.0 92.0 26.2 219.9 12.2 < 9.3 18.9 104.9 86.2 367.4 65.1 179.9 0.0 51.7

NC2001-13b 43.39 7.55 2.49 8.95 8.33 22.79 0.61 0.05 1.17 1.68 0.24 0.84 98.09 < 7.2 < 3.0 165.8 22.6 783.0 < 3.6 < 10.8 17.8 107.4 7.1 363.4 < 33.6 10.6 0.0 39.6

NC2001-13c 35.73 11.52 4.26 15.33 8.68 12.93 1.88 1.34 2.20 2.05 0.25 0.82 96.97 < 7.8 3.6 123.2 29.3 1014.8 4.5 < 11.7 23.6 159.6 13.0 694.9 < 30.3 249.7 0.0 86.9

NC17b 37.16 8.97 4.13 14.86 10.70 15.23 1.27 0.95 2.04 1.09 0.22 0.19 96.81

NC2001-13 43.50 13.41 3.10 11.16 7.48 11.93 2.82 1.19 1.54 0.77 0.23 0.77 97.91

< 7.8 < 7.2 2.4 5.9 106.2 144.6 24.8 23.6 660.6 717.7 3.5 6.4 < 11.7 6.1 20.5 22.3 117.6 156.1 59.1 23.2 671.6 444.4 56.8 < 31.5 213.6 86.0 0.0 0.0 69.7 48.8 (continued on next page)

Cumulate samples were taken from only the Central Belt. Cumulates are characterized by 34–50 wt% SiO2 , 0.03–1.8 wt% TiO2 , 8.9–14.13 wt% MgO, 5.0–10.24 wt% Al2 O3 , and 0.99–23.86 wt% Fe2 O3 (Table 3). Cumulates 2x and 2y, lower range of SiO2 , show similar trends to ultramafic samples as well. Low SiO2 and high TiO2 , P2 O5 , and FeO contents closely match ultramafic samples from the middle belt. Samples 7 and 7b have higher SiO2 and show unusually high CaO and MgO, and low K2 O values that are similar to ultramafic sample 13b. Phlogopite results from contact metamorphism with a chloritephlogopite-marble, which explains the high CaO content in these samples. Similarities with 13b suggest these rocks underwent similar alteration. Gabbros are characterized by 48.9–59.6 wt% SiO2 , 0.5–1.43 wt% TiO2 , 4.07–9.6 wt% MgO, 10.51–16.30 wt% Al2 O3 , and 4.54–9.46 wt% Fe2 O3 (Table 4). Gabbro shows trends

304

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Table 2. (Continued) SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI

NC28c 46.99 15.12 2.53 9.10 5.94 9.11 4.46 1.06 1.26 0.75 0.24 1.28

NC28b 32.88 10.93 4.40 15.83 10.84 14.15 1.49 1.16 2.40 2.11 0.21 0.29

NC15c 53.71 16.63 1.77 6.36 5.19 5.04 5.22 1.89 0.70 0.33 0.16 1.65

Total

97.82

96.69

98.65

Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

< 6.6 7.7 194.3 24.1 863.2 8.0 6.1 25.0 141.1 22.2 375.2 < 28.2 137.4 0.0 40.8

< 7.8 < 3.3 74.1 25.3 968.7 5.2 < 11.7 22.4 126.6 21.3 578.1 < 30.6 241.8 0.0 68.9

< 6.0 2.5 73.6 14.7 533.4 45.1 < 8.4 19.7 83.3 97.1 144.9 271.9 630.6 0.0 33.8

NC2001-13b 51.25 4.61 1.61 5.80 19.86 11.73 0.49 0.05 0.16 0.01 0.15 1.89

NC2001-13c 44.10 12.07 2.89 10.38 4.43 8.19 2.81 0.57 2.49 0.34 0.22 8.77

97.61 < 6.3 < 2.4 4.5 5.7 28.0 < 2.7 < 8.4 4.4 62.7 703.2 141.4 4271.1 427.6 0.0 58.7

97.24 < 6.6 10.4 168.8 22.9 398.1 10.1 5.3 22.6 140.6 58.1 316.9 80.3 253.8 0.0 50.6

NC17b 45.62 11.67 2.57 9.24 4.32 10.04 2.80 0.79 1.17 0.12 0.19 9.28 97.81 < 6.6 3.0 71.1 19.3 237.4 17.6 5.6 14.8 93.5 52.1 266.7 52.1 156.3 0.0 44.9

NC2001-13 37.05 4.50 4.59 16.50 10.06 19.31 0.59 0.27 1.51 2.33 0.22 −0.05 96.86 < 8.1 < 3.3 62.9 19.9 443.4 < 4.2 < 12.3 16.7 130.7 48.3 670.5 < 33 48.7 0.0 61.4

∗ Trace element detection limit = 1 ppm. † Major element oxides normalized using Barth-Niggli Cation Norm.

similar to cumulates and ultramafics with the exception of higher SiO2 wt% (ranging from 48 to 59%). Excluding samples 19 through 19c, samples from the middle belt show relatively uniform chemical similarity to samples from the upper and lower belts. Samples 19 through 19c show a slight separation from the other gabbros with higher CaO and MgO values. Sample 19 shows alteration under thin section, which is also evident from the low Na2 O and unusually high K2 O values. The related dikes show varied chemistry that does not appear to be based on location. They are characterized by 48.6–57.4 wt% SiO2 , 0.5–1.3 wt% TiO2 , 2.8–6.6 wt% MgO, 12.8–15.4 wt% Al2 O3 , and 4.92–13.44 wt% Fe2 O3 (Table 5). Younger dikes are characterized by 38.1–52.0 wt% SiO2 , 0.24–1.35 wt% TiO2 , 0.24–4.6 wt% MgO, 18.3–23.6 wt%

3. Geochemistry

305

Table 3. Major element oxides (wt%) and trace element∗ (ppm) analysis of cumulate samples from the Dongwanzi Ophiolite† SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI Total Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

NC2001-7b 50.17 6.28 0.65 2.34 14.13 23.48 0.62 0.17 0.34 0.24 0.10 0.5 99.02 < 6.3 < 2.4 127.9 15.4 345.5 1.8 < 8.7 10.1 141.0 22.5 72.8 34.2 39.0 0.0 6.1

NC2001-7 46.58 10.24 0.74 2.67 12.80 23.97 0.15 0.11 0.31 0.05 0.09 0.8 98.51 < 6.6 2.5 100.3 12.7 354.4 2.3 < 8.7 13.2 90.4 10.3 76.4 < 33.6 63.9 0.0 10.6

NC2001-2x 37.41 7.35 4.03 14.48 10.54 18.60 0.78 0.49 1.88 2.05 0.22 −0.7 97.12 < 8.4 < 3.6 72.1 26.0 519.4 1.9 < 12.9 18.5 151.9 36.4 816.8 < 34.5 83.8 0.0 64.5

NC2001-2y 34.53 5.01 5.19 18.67 8.98 18.58 0.92 0.28 1.74 2.78 0.27 0.35 97.31 < 7.8 1.5 81.8 21.1 614.5 4.9 5.3 22.1 120.8 32.4 616.4 20.6 174.2 0.0 61.4

∗ Trace element detection limit = 1 ppm. † Major element oxides normalized using Barth-Niggli Cation Norm.

Al2 O3 , and 4.69–14.81 wt% Fe2 O3 . Low MgO values are seen in samples NC2001-3 and NC2001-3b. These samples show unusually high Al2 O3 and Na2 O values and fall within the alkalic field and subalkalic fields based on the TAS and K2 O graphs, respectively. Values from the north and south belts appear closely matched in CaO, Al2 O3 , and P2 O5 contents but show no other similar trends. Pillow lavas were analyzed by A. Polat (in preparation) of the University of Windsor, Ontario. They are characterized by 49.0–50.0 wt% SiO2 , 0.40–0.43 wt% TiO2 , 18.7– 18.10 wt% MgO, 6.4–6.6 wt% Al2 O3 , and 9.7–9.8 wt% Fe2 O3 (Table 6) and plot within

306

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Table 4. Major element oxides (wt%) and trace element∗ (ppm) analysis of gabbroic samples from the Dongwanzi Ophiolite† SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI Total Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

NC2001-15b 53.41 16.27 1.70 6.10 5.65 6.85 4.33 1.35 0.71 0.28 0.11 1.56 98.31 < 6.0 4.3 105.9 12.2 1009.7 12.4 5.6 22.2 97.3 89.9 135.7 196.3 1171.8 0.0 33.3

NC19b 50.04 10.51 1.78 6.40 9.64 13.95 2.40 1.21 1.14 0.25 0.16 0.86

NC19c 48.99 11.50 2.06 7.40 9.49 12.01 2.72 1.44 1.43 0.34 0.14 0.91

NC19 52.95 11.59 0.99 3.55 8.86 11.50 0.52 7.27 0.92 0.13 0.09 0.46

NC21b 55.04 15.42 1.72 6.20 5.20 6.42 4.26 1.68 0.68 0.24 0.13 1.51

NC22-1 59.66 15.55 1.40 5.03 4.07 5.48 4.52 1.22 0.55 0.17 0.10 0.94

NC22-2 50.87 16.30 1.99 7.15 5.68 5.49 3.97 0.53 0.99 0.27 0.16 4.42

98.33

98.41

98.84

98.51

98.70

97.80

< 6.6 4.8 68.2 15.3 869.3 18.7 < 9.3 15.2 67.2 143.8 221.9 383.0 765.2 0.0 34.5

< 6.6 4.3 80.5 18.9 816.3 25.2 < 9.3 16.0 78.6 133.6 272.4 308.2 622.2 0.0 45.5

< 6.3 10.7 138.6 19.5 218.1 113.2 17.7 14.9 32.1 22.0 84.5 47.7 616.0 0.0 < 8.4

< 6.3 3.4 88.5 12.5 947.6 12.7 12.8 21.4 89.6 52.2 153.2 156.6 926.0 0.0 28.9

< 6.0 4.5 148.5 13.6 815.1 8.9 7.0 18.3 69.3 34.6 123.3 11.4 584.7 0.0 26.0

< 6.3 3.1 247.6 7.5 741.2 4.1 5.3 20.1 154.3 178.9 161.7 347.9 215.4 0.0 31.8

∗ Trace element detection limit = 1 ppm. † Major element oxides normalized using Barth-Niggli Cation Norm.

alkalic fields on the K2 O diagram. The have high MgO content, characteristic of komatiites, however samples lack spinifex texture. In general, the pillows seem more mafic than dikes or gabbros and plot with the ultramafics on the major element graphs. The AFM diagram is used to distinguish between tholeiitic and calc-alkaline trends (Fig. 13). Ultramafic samples from the ophiolite plot as tholeiites along with the cumulate samples. Cumulate samples that plot closer to the MgO apex are NC2001-7 and NC20017b, both of which are contaminated by contact metamorphism. Pillow lavas have a high MgO content (18.7–18.10 wt%) and plot with the metamorphosed cumulates. The gabbros

3. Geochemistry

307

Table 5. Major element oxides (wt%) and trace element∗ (ppm) analysis of dike samples from the Dongwanzi Ophiolite† SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI Total Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

NC2001-4 48.69 15.48 2.25 8.09 4.95 9.40 2.96 3.73 1.16 0.57 0.21 0.57 98.04 < 6.6 4.8 127.4 16.8 1705.4 45.5 4.0 22.5 103.0 9.2 332.2 < 30.3 877.8 0.0 38.1

NC28 52.88 15.47 1.07 3.85 2.80 7.09 2.51 3.40 0.58 0.26 0.09 8.73

NC28d 57.43 14.49 1.40 5.04 3.80 8.74 4.78 0.71 0.57 0.22 0.11 1.17

NC21 49.63 12.83 2.88 10.36 6.62 7.35 3.20 1.41 1.33 0.14 0.28 2.16

98.73

98.46

98.18

< 5.7 6.4 115.7 15.7 255.5 54.2 12.4 18.6 64.2 48.5 116.1 131.6 543.0 0.0 21.8

< 6.0 < 6.6 4.3 4.0 91.1 84.1 17.1 22.1 1338.9 385.8 8.4 34.3 17.5 3.9 21.0 17.7 50.3 113.6 53.9 47.5 146.9 305.6 203.8 56.0 164.5 308.8 0.0 0.0 16.2 48.0 (continued on next page)

and dikes plot in the alkalic field and the later, younger dikes plot to the left of the alkalic field due to high amounts of alkalics and iron. Major element plots to distinguish between calc-alkaline and tholeiitic basalts include SiO2 vs. FeO(total)/MgO and FeO vs. FeO (total)/MgO (after Miyashiro, 1973). Based on low FeO(tot)/MgO content, the pillow lavas and dikes, as well as the other samples, plot in the calc-alkaline field of the FeO(tot)/MgO vs. SiO2 diagram (Figs. 14 and 15). Samples of younger dikes were excluded in the lower graph to allow trend to be seen. Because of low SiO2 values in the ultramafics, all the samples were plotted on the FeO vs. FeO (total)/MgO diagram. Dike samples from the ophiolite plot through the calc-alkaline field and into the tholeiitic field. Ultramafics remained in the tholeiitic field, and gabbros plotted in the calc-alkaline field. Cumulate samples plotted as alkalics are NC2001-7, NC2001-7b,

308

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Table 5. (Continued) SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI

NC23d 55.08 16.71 0.83 2.97 2.46 5.27 4.16 3.47 0.08 0.45 0.06 7.86

NC21a 72.63 14.53 0.23 0.82 0.37 0.69 5.03 3.97 0.16 0.06 0.01 0.71

Total

99.37

99.21

Sn Nb Zr Y Sr Rb Pb Ga Zn Ni V Cr Ba Ce Co

< 5.7 < 2.1 147.9 14.9 331.8 86.4 7.2 15.3 50.6 36.9 35.0 76.0 304.7 0.0 14.0

< 5.4 3.9 93.9 2.4 441.3 45.1 13.1 16.1 17.2 3.3 18.0 15.2 1775.8 0.0 5.8

NC2001-3 51.81 23.38 1.21 4.35 0.24 8.90 5.44 1.47 0.25 0.12 0.04 1.23

NC2001-3b 52.08 23.66 1.02 3.67 0.48 9.14 5.60 0.99 0.24 0.10 0.04 1.27

98.42

98.28

< 6.0 < 2.4 5.7 1.6 3466.3 23.4 7.2 25.4 17.6 < 8.1 212.0 41.7 1198.1 0.0 7.4

< 6.0 < 2.4 22.9 2.3 4914.6 17.8 5.2 24.0 23.9 4.0 151.0 < 28.5 643.8 0.0 6.4

5plag 38.15 18.33 3.22 11.59 4.62 13.39 0.70 2.80 1.35 1.05 0.21 2.11 97.51 < 7.2 < 3.0 67.5 18.6 1880.8 59.0 4.1 26.0 109.3 < 11.4 454.3 < 30.9 613.8 0.0 40.6

∗ Trace element detection limit = 1 ppm. † Major element oxides normalized using Barth-Niggli Cation Norm.

both of which had high MgO and FeO (total) values. The calc-alkaline character of the samples indicates a non-MORB origin (Wilson, 1989). 3.6. Trace Element Characteristics Pearce and Cann (1971, 1973) and Pearce (1975) devised a method for characterizing basic volcanic rocks by determining alkalic character by using aY/Nb plot, and then plotting the samples on a Ti, Zr, and Y ternary diagram to see if they plotted in the ‘within-plate basalt’ field. If they did not, samples were replotted on a Ti vs. Zr discrimination diagram for highly weathered samples or Ti, Zr, Sr ternary diagram for fresher samples.

3. Geochemistry

309

Table 6. Major element oxides (wt%) and trace element∗ (ppm) analysis of pillow samples from the Dongwanzi Ophiolite SiO2 Al2 O3 Fe2 O3 FeO MgO CaO Na2 O K2 O TiO2 P 2 O5 MnO LOI Total Ni Co Cr Rb Sr Ba Cs V Pb Ga Cu Zn Nb Ta Zr Hf Y Th U

DW02-9 49.11 6.74 9.69 0 19.08 10.04 0.78 0.65 0.39 0.22 0.16 1.9 98.76 602 51.4 2052.6 21 166 94 0.4 104 <5 7 <5 63 3 < 0.5 79 1 10 1.9 0.49

DW02-10 50.03 6.64 9.8 0 18.67 10.28 0.85 0.55 0.43 0.19 0.15 1.65

DW02-9

DW02-10

14 28.9 3.81 16 3.4 0.92 2.75 0.45 2.39 0.46 1.29 0.18 1.2 0.18

14 29.8 4.02 16 3.3 0.91 2.69 0.43 2.19 0.42 1.22 0.17 1.1 0.17

99.24 494 49.3 1710.5 11 187 257 0.1 95 <5 6 10 54 3 < 0.5 83 1 9 1.7 0.38

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

∗ Analyzed by Polat (in press).

The ratio Y/Nb decreases with increasing alkalic character (Pearce and Cann, 1973). Alkalic oceanic island and continental basalts are characterized by ratios of less than one, whereas ocean floor alkalic basalts have ratios of less than two. Ratios greater than three indicate that the rocks are tholeiitic, low potassium tholeiites, island arcs or calc-alkaline basalts. Samples from the central belt of the Dongwanzi ophiolite had a Y/Nb ratio of

310

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 13. An AFM diagram showing the boundary between calc-alkaline and tholeiitic samples from the Dongwanzi Ophiolite. The dashed line represents fields as defined by Kuno (1968) and the solid line represents fields as defined by Irving and Baragar (1971). Pillows and dikes from the Dongwanzi Ophiolite plot within the calc-alkaline field of Kuno.

greater than three with a few exceptions falling into the transition zone (Fig. 16). Younger leucocratic dikes with ratios less than one plotted in the alkalic field. Samples plot near the low-potassium tholeiite and within the calc-alkaline fields of Pearce and Cann (1973). The rock types are fairly uniform between the fields with the exception of the gabbros, which tend to cluster toward the calc-alkaline field (Fig. 17). The gabbros also show higher amounts of Zr, whereas pillows fall outside these fields. Dikes are distributed between ocean island and/or continental basalts and calc-alkaline basalts. The concentration of Ti varies between samples and Y is relatively uniform between rock types. Location of the samples, whether from the north, middle, or south parts of the belt does not appear to be a factor in Ti, Zr, or Y concentration. Overall, samples plotted in the Ti vs. Zr (Fig. 18) discrimination diagram do not appear to fall into any distinct field. Samples of dikes fall into the calc-alkaline and ocean island basalt fields however, pillow lavas do not fall into any fields. When plotted on the Ti, Zr, Sr diagram, samples fall into the calc-alkaline field (Fig. 19). This may be misleading though. The samples that were analyzed are not fresh and so elevated concentrations of Sr are probably a result of element mobility due to weathering. Gabbro samples have higher Sr values whereas all other rock types are distributed uniformly throughout the calc-alkaline field. Another element not readily affected by alteration is chromium. Pearce (1975) devised a Ti vs. Cr model to distinguish between island-arc and ocean floor basalts. Samples from Dongwanzi plot mainly within the island-arc tholeiite field having low concentrations of Cr, pillows however, plot in the ocean floor basalt field (Fig. 20). The Cr-Y discrimination diagram of Pearce (1982) shows the separation of mid-ocean ridge basalts from island arc tholeiites, which include calc-alkaline and alkalic basalts from oceanic arcs. Samples from

3. Geochemistry

311

Fig. 14. FeO(total)/MgO vs. FeO diagram with fields as defined by Miyashiro (1973). The lower graph plots all the samples, excluding the younger dikes. With the exception of two samples, the pillow and dike samples plot in the calc-alkaline fields.

Dongwanzi plot almost entirely within the island arc field except for the pillows, which again, plot outside the island arc field (Fig. 21). The samples from the Dongwanzi ophiolite plot within the island arc tholeiite-back arc basin/MORB field of Shervais (1982). They have Ti/V ratios > 15, however based on this

312

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 15. FeO(total)/MgO vs. SiO2 diagram with fields as defined by Miyashiro (1973). The lower graph plots all the samples, excluding the younger dikes. Pillow lava and dike samples all plot within the calc-alkaline field.

plot (Fig. 22), tectonic environment cannot be determined (ex. MORB ratios are typically between 20 and 50 and so they overlap with those of the island arc and are not distinct). Samples with lower Ti/V ratios (less than 10) are dikes from the middle of the central belt. In general, gabbros seem to have lower Ti and V values, while ultramafics have higher

3. Geochemistry

313

Fig. 16. Determination of alkalic character by Y/Nb ratios with fields as defined by Pearce and Cann (1971, 1973), Pearce (1975). Ratios greater than three suggest tholeiitic and/or calc-alkaline tectonic setting whereas ratios less than one suggest alkalic ocean island and/or continental basalt tectonic setting. With the exception of younger, later dikes, samples from the Dongwanzi ophiolite plot in the transitional to tholeiitic and/or calc-alkaline fields.

Fig. 17. The Ti-Zr-Y discrimination diagram for basalts after Pearce and Cann (1973). Samples from the Dongwanzi ophiolite plot in the ocean island or continental basalt and calc-alkaline basalt field.

314

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 18. The Ti-Zr discrimination diagram with fields as defined by Pearce and Cann (1973). Dike samples plot within the calc-alkaline field whereas pillows have low Ti values and plot just below this field. All other samples are scattered.

Fig. 19. The Ti-Zr-Sr discrimination diagram for basalts after Pearce and Cann (1973). Samples from the Dongwanzi ophiolite plot in the calc-alkaline and low potassium tholeiite field although this may be misleading seen as samples were weathered and high Sr may be due to element mobility.

ones. The cumulate samples are split, having both low and high values and the dikes show a range in Ti values. Beccaluva et al. (1979) proposed two additional discrimination diagrams based on those of Pearce (1975). Ocean floor basalts generally have Ba/Y ratio of less than 4.4 and island

3. Geochemistry

315

Fig. 20. The Ti-Cr discrimination diagram for basalts after Pearce and Cann (1973). Dike samples plot within the island arc field, along with many of the other samples. Pillow lavas though, plotted in the ocean floor basalt field.

Fig. 21. The Cr-Y discrimination diagram for basalts with fields after Pearce (1982). With the exception of the pillow lavas, samples plot within the island arc tholeiite field.

316

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 22. The Ti-V discrimination diagram for basalts with fields after Shervais (1982). Arc tholeiites typically have ratios between 10 and 20, back arc basins have ratios between 20 and 50, and ocean island and alkalic basalts have ratios between 50 and 100. Typical MORB ratios have ratios between 20 and 50 and plot with the back arc basin samples.

arc tholeiites have ratios of greater than 3.9. This diagram is useful in that it has little overlap between the two fields, however, the mobility of Ba during secondary processes limits the use of this diagram for ophiolites. A second plot proposed by Beccaluva et al. (1979), Ti/Cr vs. Ni is more useful. This plot also separates rocks into island arc tholeiites and ocean floor basalts like those of Pearce however, Ni is used because the Ti/Cr ratio is lower in island arc tholeiites at any given Ni content (Beccaluva et al., 1979). Samples from Dongwanzi plot within the island arc tholeiite field of the Ba/Y discrimination diagram although high Ba values may result from element mobility. However, the Ti/Cr vs. Ni diagram supports this where samples except for the pillows, plot in the island arc field. Pillow lavas plot in the ocean floor basalt field on the Ti/Cr vs. Ni graph but have ratios comparable to island arc tholeiites in the Ba/Y plot (Figs. 23 and 24). Samples that plot in the ocean floor basalt field on the Ba/Y plot are altered cumulates and ultramafics. Ratios in the ultramafics range from 3 to 35 and overlap with the gabbros, which range from approximately 30 to 95. The values in the dikes range from about 2.5 to 750 where higher values are most likely due to Ba mobility in secondary processes. In Fig. 24, gabbros have lower Ti values and ultramafics have higher ones, which are consistent with Ti/V plot of Shervais (1982). Dike and pillow samples plot entirely within the back-arc basin field along with the rest of the samples.

4. Conclusions

317

Fig. 23. Determination of alkalic character by Ba/Y ratios with fields as defined by Beccaluva et al. (1979). Ocean floor basalts have ratios lower than 4.4 and island arc tholeiites have ratios greater than 3.8. Little overlap between fields is helpful in determining tectonic setting, however mobility of Ba during secondary processes limits the use of this diagram for ophiolites.

4. CONCLUSIONS The Dongwanzi ophiolite had previously been mapped as a layered intrusion of amphibolite facies. Recognition of sheeted dikes and pillow lavas during regional reconnaissance mapping led to the identification of this belt as the oldest complete ophiolite sequence known. Petrographic analysis conducted on collected samples led to the identification of ultramafic samples, which included hornblende-bearing pyroxenite and pyroxenite, harzburgite, plagioclase-bearing pyroxene hornblendites, and altered ultramafics. Cumulate samples were identified as metamorphic pyroxenite and wehrlite. Gabbroic complexes include layered and altered isotropic gabbros as well as pyroxenite and pyroxenite-hornblende gabbros. Related dikes were identified as hornblende gabbro to gabbroic and hornblendedioritic. Younger dikes range in composition from mafic to felsic to plagioclase-rich. Other associated rocks include chert, banded iron formation, sulfide lenses, and mélange. Younger rocks in the outcrop area include quartzite and altered granites. Geochemical analysis conducted on these samples shows characteristics similar to suprasubduction zone ophiolites. Samples, regardless of rock type, were plotted on graphs that were originally intended for basalts. Therefore ultrabasic to mafic rocks were plotted for comparison and were not regarded when determining tectonic setting. On major ele-

318

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Fig. 24. The Ti/Cr vs. Ni discrimination diagram after Beccaluva et al. (1979). The Ni concentration is lower in island arc tholeiites for any given ratio of Ti/Cr than it is for ocean floor basalts. Dikes plot within the island arc tholeiite field whereas pillows plot as ocean floor basalts.

ment graphs, the dikes and pillow lavas plotted in calc-alkaline field. This was supported, with few exceptions, by the trace element data. Together, all data suggest that the Dongwanzi mafic/ultramafic complex is an Archean ophiolite with affinities to modern suprasubduction zone ophiolites. All samples were exposed to weathering and/or later metamorphism and this was taken into account when determining tectonic setting. Major element analysis also showed that pillows are compositionally similar to komatiites. The absence of complete or dismembered segments of ophiolites from the Archean has led to the theory that plate tectonic processes were significantly different than those in the Phanerozoic. Identification of a complete Archean ophiolite with major and tracel element geochemical signatures similar to modern suprasubduction zone ophiolites at Dongwanzi has important implications for understanding plate tectonic processes in the early earth. Additionally, podiform chromite from the Zunhua structural belt to the southwest provides further evidence for the operation of seafloor spreading and plate tectonic processes before 2.5 Ga. ACKNOWLEDGEMENTS We were assisted in field mapping in the Dongwanzi area by A. Huang, Y.Y. Wu, and Niu Xianglong. Geochemical analysis was conducted at Washington University in

References

319

St. Louis, under the direction of R. Dymek, who is graciously thanked for his efforts. J. Encarnacion assisted with interpretation of petrographic and geochemical features. This work represents part of a M.Sc. thesis by the first author, and was supported by NSF awards to Kusky including (02-07886) and (01-25925), by St. Louis University, and by Peking University (Projects 40242014, 49832030, and Peking University Project 985 to J.H. Li). REFERENCES Bai, J., Dai, F.Y., 1998. Archean crust of China. In: Ma, X.Y., Bai, J. (Eds.), Precambrian Crustal Evolution of China. Springer Geological Publishing, Beijing, pp. 15–86. Bai, J., Wang, H., Yan, Y., 1997. Formation, break-up and amalgamation of the Early Precambrian Continental Crust in China—dynamics and periodicity. Proceedings of the 30th International Geological Congress, Utrecht, The Netherlands. Precambrian Geology and Metamorphic Petrology 17, 89–100. Beccaluva, L., Ohnenstetter, D., Ohnenstetter, M., 1979. Geochemical discrimination between oceanfloor and island-arc tholeiites—application to some ophiolites. Canadian Journal of Earth Science 16, 1874–1882. 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. In: Geophysical Monographs, vol. 88. American Geophysical Union, pp. 1–30. Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga Podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithosphere mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337. Irving, T.N., Baragar, W.R.A., 1971. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences 8, 523–548. Kröner, A., Cui, W.Y., Wang, S.Q., 1998. Single zircon ages from high-grade rocks of the Jianping Complex, Liaoning province, NE China. Journal of Asia Earth Sciences 16 (5–6), 519–532. Kuno, H., 1968. Differentiation of basalt magmas. In: Hess, H.H., Plodervaart, A. (Eds.), Basalts: The Poldervaat Treatise on Tocks of Basaltic Composition, vol. 2. Interscience, New York, pp. 623– 688. Kusky, T.M., Li, J.H., 2003. Paleoproterozoic tectonic evolution of the North China Craton. Journal of Asian Earth Sciences 22, 383–397. Kusky, T.M., Li, J.H., Tucker, R.D., 2001. The Archean Dongwanzi Ophiolite Complex, North China Craton: 2.505-Billion-Year-Old Oceanic Crust and Mantle. Science 292 (5519), 1142–1145. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyer Le Bas, M.J., Sabine, P.A., Schmid, R., Sorensen, H., Streckeisen, A., Woolley, A.R., Zanettin, B., 1989. A Classification of Igneous Rocks and Glossary of Terms. Blackwell, Oxford. Li, J.H., Kusky, T.M., Huang, X., 2002. Archean podiform chromites and mantle tectonites in ophiolitic melange, North China Craton: a record of Early Oceanic Mantle Processes. GSA Today 12 (7), 4–11.

320

Chapter 9: Geochemical and Petrographic Characteristics of the Central Belt

Li, J.H., Kröner, A., Qian, X., O’Brien, P., 2000. The tectonic evolution of the Early Precambrian High-Pressure Granulite Belt, North China Craton. Acta Geologica Sinica 74 (2), 246–258. Ma, X.Y., 1998. Introduction. In: Ma, X.Y., Bai, J. (Eds.), Precambrian Crustal Evolution of China. Springer-Geological Publishing, Beijing, pp. 1–13. Matveev, S., Ballhaus, C., 2002. Role of water in the origin of podiform chromitite development. Earth and Planetary Science Letters 202, 235–243. Miyashiro, A., 1973. The Troodos Ophiolitic complex was probably formed in an Island Arc. Earth and Planetary Science Letters 19, 218–224. Pearce, J.A., 1975. Basalt Geochemistry used to investigate past tectonic environments on Cyprus. Tectonophysics 25 (1–2), 41–67. Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plate boundaries. In: Thorpe, R. (Ed.), Andesites: Orogenic Andesites and Related Rocks. Wiley, Chichester, pp. 525– 548. Pearce, J.A., Cann, J.R., 1971. Ophiolite origin investigated by discriminant analysis using Ti, Zr, and Y. Earth and Planetary Science Letters 12, 339–349. Pearce, J.A., Cann, J.R., 1973. Tectonic setting of basic volcanic rocks determined using trace element analysis. Earth and Planetary Science Letters 19, 290–300. 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: Volcanic and Associated Sedimentary and Tectonic Processes in Modern and Ancient Marginal Basins. Geological Society of London Special Publication 16, 77–94. Rickwood, P.C., 1989. Boundary lines within petrologic diagrams with use oxides of major and minor elements. Lithos 22, 247–263. Saunders, A.D., Tarney, J., 1984. Geochemical characteristics of basaltic volcanism within back-arc basins. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basin Geology: Volcanic and Associated Sedimentary and Tectonic Processes in Modern and Ancient Marginal Basins. Geological Society of London Special Publication 16, 59–76. Shervais, J.W., 2001. Birth, Death, and Resurrection: The life cycle of Supra-Subduction Zone ophiolites. Geochemistry, Geophysics, Geosystems 2. Shervais, J.W., 1982. Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59 (1), 101–118. Qian, X., Li, J.H., 1999. Discovery of Neoarchean unconformity and it’s implication for continental cratonization of North China Craton. Science in China (Series D) 42 (4), 399–407. Wilde, S.A., Zhao, G.C., Sun, M., 2002. Development of the North China Craton during the Late Archean and its amalgamation along a major 1.8 Ga collision zone; including speculations on its position within a global Paleoproterozoic supercontinent. Gondwana Research 5, 85–94. Wilson, M., 1989. Igneous Petrogenesis: A Global Tectonic Approach. Unwin Hyman, London, p. 466. Yuehua, Y., Wang, E., 1997. North China Craton. In: Oxford Monographs on Geology and Geophysics, vol. 35. Oxford Univ. Press, UK, pp. 730–735. Zhao, G.C., Cawood, P.A., Lu, L.Z., 1999. Petrology and P-T history of the Wutai amphibolites: Implications for the tectonic evolution of the Wutai Complex, China. Precambrian Research 93, 181–199. Zhao, G.C., Wilde, S.A., Cawood, P.A., Sun, M., 2001. Archean blocks and their boundaries in the North China Craton: lithological, geochemical, structural and P-T path constraints and tectonic evolution. Precambrian Research 107, 45–73.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

321

Chapter 10

MICROSTRUCTURES OF THE ZUNHUA 2.50 GA PODIFORM CHROMITE, NORTH CHINA CRATON AND IMPLICATIONS FOR THE DEFORMATION AND RHEOLOGY OF THE ARCHEAN OCEANIC LITHOSPHERIC MANTLE XIONGNAN HUANGa , JIANGHAI LIa , T.M. KUSKYb AND ZHENG CHENa a School of

Earth and Space Sciences, Peking University, Beijing 100871, China Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63103, USA

b Department of

The Zunhua Neoarchean podiform chromite, North China, preserves typical ophiolitic magmatic fabrics including nodular and orbicular textures, and magmatic flow structures. The magmatic textures indicate that the Zunhua podiform chromite was formed through five-stages of evolution, with the following time sequence: disseminated chromite, netlike veins, antinodular, orbicular and nodular textures. The evolution of the texture series can be interpreted to result from fast flowing magmatic flowing systems. They result from the vertical accretion of the oceanic mantle. The podiform chromite ores show strong deformation with development of pull-apart structures, banding, folds, and mylonitic foliation. These structures were formed at high temperature in the oceanic mantle during the oceanic ridge spreading as the ores were caught up by plastic flow and sheared transversely. The Zunhua podiform chromite bodies result from active magmatic accretion and strong high-temperature plastic flow, therefore a fast spreading oceanic ridge is suggested for its formation. Silicate mineral inclusions within the chromium spinel and geochemical characteristics of the Zunhua ophiolite support a geological setting in a suprasubduction belt, such as a back-arc basin or fore-arc basin.

1. INTRODUCTION Podiform chromite deposits are distinctive discontinuous concentrations of chromite commonly enveloped by dunite, typically characterized by nodular and orbicular chromite textures, and they may be very irregular in shape, size, and distribution within the host rocks. Podiform chromite deposits are generally associated with harzburgites and only reported in ophiolites and oceanic lithosphere. They are mostly located within a domain including the dunitic transition belt and harzburgitic mantle. Favorable tectonic settings for DOI: 10.1016/S0166-2635(04)13010-7

322

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

podiform chromites include island arcs, back-arc basins, and fast spreading ridges (Johnston, 1936; Thayer, 1964; Leblanc and Nicolas, 1992; Arai and Matsukage, 1998). Because the stable physical and chemical properties of podiform chromites are acquired at high temperatures, podiform chromites sometimes escape the late metamorphism and deformation, and, therefore, have become an important subject of the study for Precambrian ophiolites and mantle characteristics (Ahmed et al., 2001; Li et al., 2002a). The Zunhua podiform chromite (ZPC) occurs within serpentinized harzburgite, enveloped in dunite (also commonly serpentinized) (Li et al., 2002a, 2002b; Huang, 2003). It preserves typical structures and textures of podiform chromite, including nodular and orbicular textures, very similar to the podiform chromites in the Oman ophiolite and the Troodos ophiolite (Matveev and Ballhaus, 2002; Nicolas, 1989). The age of the Zunhua podiform chromite is about 2.50 Ga (Kusky et al., 2004) and is the oldest podiform chromite reported in the world so far. A Re-Os model age of the ZPC is 2547 ± 10 Ma (Kusky et al., 2004). There are not only high-temperature deformation structures such as pull-apart structures and mylonitic structures, but also magmatic fabrics such as nodular and orbicular textures in the Zunhua Neoarchean podiform chromite. This paper presents a detailed study of the microstructures in the Zunhua podiform chromites, and discusses the implications these fabrics have on the characteristics of Archean oceanic mantle.

2. GEOLOGIC AND TECTONIC SETTING The North China craton is divided into two major blocks, the Western Block and the Eastern Block, separated by the Neoarchean Central Orogenic Belt (Fig. 1) (Li et al., 2000a, 2002b). The Western Block is characterized by its khondalite series, a sedimentary formation with platformal or continental margin affinities. The Eastern Block contains a variety of ca. 3.80–2.5 Ga gneissic rocks and greenstone belts. The Central Orogenic Belt is characterized by its orogenic structural styles, including thrust-related subhorizontal foliations and shear zones, recumbent folds. The 700-km-long Hengshan high-pressure granulite belt (2.5 Ga, reactivated at 1.9 Ga) is located along the western side of the Central Orogenic Belt (Li et al., 2000b), while the Qinglong foreland basin and fold-thrust belt is located on the eastern side (Kusky and Li, 2003). In the middle of the Central Orogenic belt, Neoarchean ophiolitic mélanges including Zunhua ophiolitic mélange, Wutai ophiolitic mélange and Liaoxi ophiolitic mélange crop out (Kusky et al., 2001; Kusky and Li, 2002; Li et al., 2002b; Wang et al., 1997; Li et al., 2002c). The Zunhua structural belt is located in the northeastern sector of the Central Orogenic Belt, separated from an Archean granulite-gneiss dome (3.85–2.50 Ga) of the Eastern Block by a major shear zone (Li et al., 2002b). The Zunhua structural belt contains various tectonic slices, including tonalite-trondhjemite-granodiorite gneiss, mafic plutonic rocks, supracrustal sequences and granites, all complexly intercalated with each other, intruded by ca. 2.56–2.5 Ga tonalitic gneiss, then 2.5–2.4 Ga granites (Wu et al., 1998; Kusky et al., 2001). The Zunhua Neoarchean ophiolitic mélange is located in the middle of the Zunhua structural belt, north of the city

2. Geologic and Tectonic Setting

323

Fig. 1. Tectonic divisions of North China Craton (A) and the location of the Zunhua Neoarchean podiform chromite (B).

324

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

of Zunhua, Hebei Province. It extends northeast about 80 km along strike, and is about 20 km wide from west to east. The Zunhua Neoarchean mélange contains an assemblage of typical ophiolitic rock types, including partly serpentinized harzburgite, dunite, pyroxenite, podiform chromite, metagabbro, layered cumulate, pillow lavas, sheeted dykes (metamorphosed to amphibolites), greenschist and other rock types (Kusky et al., 2001; Li et al., 2002b). These rocks occur as lenses and tectonic blocks ranging in size from 10 cm to more than 1000 m, within a strongly deformed matrix of foliated and sheared, finegrained biotite-gneiss and hornblende-gneiss with some layers of amphibolite and banded iron formation (Li et al., 2002b; Huang et al., 2003). The ZPC deposits are mainly located around the Zhuling area north of Zunhua City. The region was an important province for chromium production in China in 1970s, and was being prospected and mined. The former research revealed geological features of these chromite ores including their nodular and orbicular textures. They show a negatively sloped chondrite-normalized PGE pattern, showing U-shaped chondrite normalized REE pattern (Bai et al., 1976, 1993), but no podiform chromite was reported from these deposits until recently (Li et al., 2002a; Kusky and Li, 2003). Detailed petrographic analysis and geochemical study has revealed that these chromite deposits have textures and structures similar to typical podiform chromites, such as podiform chromite in Oman ophiolite and New Caledonia. In addition, ZPC shows similar geochemical characteristics of podiform chromites (Li et al., 2002b; Huang et al., 2003; Huang, 2003; Kusky and Li, 2003).

3. STRUCTURES AND TEXTURES OF THE ZPC 3.1. Magmatic Structures and Textures of the ZPC Chromite ore bodies of the ZPC are variable in size. Most of them are 1–20 m long and 0.5–2 m thick and extend to 10–30 m deep. The length, thickness, and depth of small bodies are all less than 1 m. The larger chromite bodies are rarely longer than 100 m, thicker than 10 m, and reach several tens of meters deep. Along the strike and the inclination, the thickness of the ore bodies is greatly variable. The shape of the ore bodies is also very complicated. Large bodies are commonly present as irregular lenses, and the large lenses are made of many small chromite bodies, which are variable in size and arranged together densely. The shapes of the small bodies are generally nest-like or pod-shaped. On the margin of the chromite pods, irregular chromite veins cut the foliation of harzburgite. In the middle parts of some large pods, layered chromite horizons are present (Fig. 2). We document here four major magmatic textures that are found in the inner parts of the chromite pods of the Zunhua bodies. They are very similar to textures of podiform chromites from Oman and New Caledonia (Peters et al., 1991, Nicolas, 1989; Nicolas and Al Azri, 1991).

3. Structures and Textures of the ZPC

325

Fig. 2. Geological occurrence of the ZPC. (a) Plane occurrence of a podiform chromite body (Maojiachang). (b) Section occurrence of the podiform chromite in Maojiachang. (c) A chromite vein cuts the foliation of the harzburgite around it (Zhuling). The vein is enveloped in dunite, presents massive texture in the middle part and antinodular texture in the outer part. (d, e) Irregular veins of podiform chromite (Zhuling). (f) Layered chromite. Chromite layers show antinodular (occluded) texture, between the chromite layers, disseminated and nodular chromite grains are preserved in the dunite.

3.1.1. Magmatic Layering (Fig. 3a) Chromite and dunite are interlayered. Chromite layers are commonly 1–2 cm thick and show antinodular (occluded) texture. Dunite layers are generally 10–50 cm thick, and most of them are intensely serpentinized. The layers extend less than 10 m. A few of the chromite bodies with disseminated, nodular, and orbicular textures are preserved in the dunite layers. It proves that the podiform chromite has a magmatic cumulate origin, but it had a small magma chamber so that the scale of the cumulate layer structure is rather small.

326

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

Fig. 3. Magmatic structures and textures of the ZPC. (a) Layered structure. Between the chromite layers, a few of chromite with disseminated, nodular, and orbicular textures occur in dunite. (b) Veined structure. Irregular chromite veins with nodular and a few orbicular textures occur in the serpentinized dunite. (c) Size graded cumulate structure. Dense cumulate chromite nodules show sedimentary bedding with size grading. (d) Magma flow structure. Chromite nodules and orbicules form asymmetric fold, suggesting fast flowing magma (turbulent flow).

3.1.2. Veined Structures (Fig. 3b) Irregular chromite veins (1–10 cm wide and 10–100 cm long) with dunitic rim (1–10 cm wide) occur in foliated harzburgite. Chromium spinel is present as densely packed cumulate euhedral crystals or densely packed cumulate nodular chromite. Most of these irregular chromite veins cut the foliation of the harzburgite, proving their intrusive genesis. Veined and layered chromite ores never occur in the same outcrop, but both of them have chromites with nodular texture, suggested that they are associated with different stages of the magmatic evolution process.

3. Structures and Textures of the ZPC

327

3.1.3. Size Graded Cumulate Structures (Fig. 3c) In the inner parts of massive chromite bodies, nodular and orbicular chromites shows sedimentary-style bedding with size-graded layers and dense cumulate layers that form massive ores. There are pointed veins on the margin of the chromite bodies extruding into the dunite envelope, in the inner part of the ores there is always occluded olivine. The nodules and orbicules show patterns of flattening and mutual impression along their contacts with each other. This structure is a result of rapid deposition of chromite nodules or orbicules while they are still plastic. 3.1.4. Magmatic Flow Structure (Fig. 3d) In the inner part of the irregular chromite veins, the long axes of chromite nodules and orbicules are parallel, and chromites and occluded olivines show no hint of deformation. Locally, asymmetric folds composed of chromite nodules are preserved, suggesting fast magma flow (turbulent flow). Five types of magmatic textures are recognized in the chromite bodies of ZPC and listed as follow: (1) Disseminated texture (Fig. 4a). Chromium spinels are randomly scattered in dunite, preserving euhedral-subhedral crystals with a rounded shape. (2) Net-like veined texture (Fig. 4b). Euhedral chromium spinels concentrate locally to form net-like veins. The occluded silicate (mainly olivine) bodies are quite irregular, different from the antinodular texture. (3) Antinodular texture (Fig. 4c). It is also called silica-occluded texture or net mesh texture. Silica, mainly olivine, occurs as nodular shaped bodies scattered in the chromite matrix evenly to form a net-mesh-like texture. Chromium spinel is generally fine grained, euhedral, with mosaic texture. (4) Orbicular texture. Fine grained chromium spinels envelope olivine (serpentinized) and show ball-, ellipsoidal- or irregular ball-shaped chromite masses (made up of several balls) (Figs. 4d–4f). The outer rim of the orbicular texture is smooth, and in the inner part has a zigzag shape with euhedral spinel growing inwardly, or arc-shaped composed of fine grained chromite. (5) Nodular texture (Figs. 4d and 4g). Chromite occurs in the olivine matrix as noduleshapes. Rims of the chromite nodules are rounded and smooth while the inner parts are made up of many euhedral chromium spinels. Skeletal chromite crystalline are present in some of the chromite nodules. 3.2. High-Temperature Deformation with the ZPC The ZPC ore bodies show strong shearing deformation. They are strongly banded, folded, and have well-developed schlieren and mylonitic structures (Fig. 5). Mineral chemical analysis reveal that there is little difference in the chromium spinels between the deformed and undeformed ores (Huang et al., 2003; Huang, 2003; Kusky et al., 2004). Pull-apart structures are the predominant type of microstructure in the banded and schlieren types of chromite ores. Fractures in chromites are perpendicular to the direction of the maximum

328

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

Fig. 4. Magmatic textures of the Zunhua podiform chromite. (a) Disseminated texture. (b) Net veined texture. (c) Antinodular texture. (d) Orbicular and nodular texture occurs on the rim of antinodular chromite. (e) Orbicular texture of chromite. (f) Nodular texture and orbicular texture always coexist. Long axes of chromite orbicules and nodules are parallel, suggesting magma flow. (g) The nodules and orbicules are cumulates that show size grading, and also show patterns of flattening and mutual impression along their contacts with each other. (h) Chromite nodules have accumulated densely to form massive structured ore.

3. Structures and Textures of the ZPC

329

Fig. 5. High-temperature structures of the ZPC. (a, b) Nodular and orbicular chromite showing development of pull-apart texture normal to elongation of nodules, disruption of nodules and orbicules to form disseminated ore. Matrix is serpentinized dunite. Chromite is weakly foliated in part (b). (c, d) Banded structure of deformed chromite ores with flattened antinodular textures in the chromite bands. (e) A fold of chromite bands. (f) Mylonitic structure in chromitite.

extension (close to the direction of foliation), and are filled with olivine (serpentinized), and original olivine around the chromite grains are stretched (Fig. 6a). Formed on grain boundaries and having angular and smooth walls, pull-apart structures result from high-temperature (> 1200 ◦ C) deformation (Holtzman, 2000). They are apparently different from the low-temperature fractures in chromites caused by serpentine expansion. The latter generally cut through the chromite grains and hit the grain

330

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

Fig. 6. Microstructures of the podiform chromitite from the ZPC. (a) Pull-apart structure in massive ore. Fractures are formed along grain boundaries, the walls generally appear angular. (b) Nodules of chromite are disrupted and disseminated in the olivine matrix. (c, d) In the flattened network structure of chromite ore, chromite is pull-apart while aggregations of olivine are flattened and stretched by plastic flow. Long axes of the flattened antinodules are parallel or sub-parallel to foliation direction. (e) Micro-fold of chromite. Axis of the fold is parallel to the foliation and perpendicular to pull apart structure. (d) Mylonitic structure of chromitite.

4. Discussion

331

boundaries, suggesting that the lattice and grain-boundary strengths are not very different at low temperature. In numerical models, at 1200 ◦ C and 175 MPa pressure, the minimum background deviatoric stress in the matrix required to cause pull-apart texture is 127 MPa, at a strain rate of 4.3 × 10−12 s−1 (Holtzman, 2000). Because of the strong shear deformation (pull-apart texture), nodules, orbicules and masses of chromitites are disrupted, being disseminated and forming foliated ores (Fig. 6). In the foliated samples, olivine is plastically deformed to form stretching lineations, and form a flattened network structure with crushed and disseminated grains of chromite (Figs. 6c and 6d). All these microstructures are typical microstructures associated with high temperature (1250– 1300 ◦ C) plastic deformation in the upper mantle section of ophiolites (Nicolas, 1989; Holtzman, 2000). Chromite folds are commonly found in the ZPC ore bodies, with different scales, from more than 1 m to less than 1 cm. They are mainly similar type folds, with flattened limbs and thickened hinges, and axes sub-parallel to the lineation in the harzburgitic rock around them. Locally, rootless folds occur. Studies of the micro-folds in the podiform chromite reveal that they are commonly associated with pull-apart textures, suggesting brittle cracking of chromite at high-temperature, and plastic shearing of olivine causing the plastic deformation of the chromite, especially the folding (Fig. 6e). In the mylonitic structure in chromite, the grain size is fine and can be divided into two groups, ca. 0.5–2 mm and less than 0.2 mm. In chromite bands, pull-apart textures are visible under the microscope (Fig. 6f). Fine grains of chromite, less than 0.2 mm define a lineation. Ratios between the long axes and the short axes of flattened antinodules are consistent, and reveal the minimum strain of the deformed chromite. In undeformed net-mesh-like texture ores, the ratio of axes is close to 1:1. In banded structures, the ratio of axes is arranged from 1.5 to 3. In mylonitic structures, the ratio can be more than 5. This change is consistent to the increase of the strain in the chromite. The foliation and lineation of the banded structure, schlieren structure and mylonitic structure in the chromite are parallel or sub-parallel to the harzburgite that surrounds them, suggesting they were formed during the same tectonic process.

4. DISCUSSION: STRUCTURES AND TEXTURES OF THE ZPC AND THEIR IMPLICATIONS FOR THE EVOLUTION OF THE OCEANIC MANTLE The structures and textures of the ZPC can be divided into two sequences, one is a magmatic structure and texture series, and the other is high-temperature shearing structure and texture series. The first sequence of structures and textures include magmatic structures such as layered structure, veined structure, size grading in cumulate layers, magmatic flow structure, magmatic textures such as disseminated texture, net veined texture, antinodular texture, orbicular and nodular texture. A magmatic textural evolution sequence can be inferred from the detailed study of the magmatic structure and texture series (Fig. 7). The sequence shows that the podiform chromite was originally developed from disseminated grains, then aggregated to form net veins, concentrated more densely into antinodular texture, and then de-

332

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

posited into silicate melts as chromite orbicules and nodules. On the margins of antinodular chromite, chromite nodules and orbicules commonly occur. Locally, disrupted nodules or half-orbicular (arc shaped) chromite is present close to the outer part of the antinodular chromite. It is consistent with the results of chromite concentration experiments (Matveev and Ballhaus, 2002). In the experiments (Matveev and Ballhaus, 2002), the starting mixture was a picritic melt composition synthesized from oxides and doped with excess water and with 2% (Mg,Fe)Cr2O4 oxide component added. Chromite is concentrated into fluid pools, crystallizing and depositing until the density of local chromite concentrations exceeds the density and strength of the underlying and supporting silicate melt sheet. At that point chromite will sink into the melt as droplets. Fluid wetting of the dispersed chromite grains to a sufficient surface tension minimizes disintegration of the nodules while they sink. The antinodular texture, nodular texture and orbicular texture are the results of incomplete mixing between the Cr-Fe-rich oxide melt and Mg-rich silica melt (Ballhaus, 1998; Zhou et al., 2001). In some of our samples, in the chromite nodules, skeletal chromite crystals are preserved and some chromite orbicules have inward growing textures, suggesting that the nodular texture is the result from the condensation of chromite shells of chromite orbicules. In the inner part of massive chromite ores, magma flow structures are preserved, including turbulent flow structures, suggesting fast flow of magma related to the formation in at least of some of the podiform chromite deposits. Careful observation in the field reveals that the chromite ores without magma flow structure are low quality. Therefore, fast flow of magma is essential for large high quality podiform chromite deposits, after the orbicular and nodular textures were formed. The ZPC has typical magmatic textures (such as orbicular and nodular texture) and magmatic structures (such as layered structure, magmatic flow structure), suggesting that they result from the vertical accretion of the oceanic mantle (Nicolas, 1989). The magmatic

Fig. 7. A model for the evolution of the textures of the ZPC. (a) Cr-Fe oxide transmitted from the primary Cr-rich silica or spinel into the fluid-rich melt during the hydrous melting of refractory peridotites, and scattered in the Mg-rich melt and crystallized to form disseminated textures. (b) Cr-Fe oxide melt concentrating to form net veins of chromite. (c) Cr-Fe oxide melt is more coalesced, chromite crystals are free to settle and to accumulate at the bottom of coalesced melt pools, forming layered structure of chromite. Because of the incomplete mixing between the silica rich melt with the Cr-Fe oxide melt, occluded texture (antinodular texture) occurs in the layered structure. (d) In the layered structure of chromite, the local chromite concentration exceeds the density of the underlying and supporting silicate melt sheet, so the chromite sinks into the melt as droplets. (e) Fluid wetting effect sustains the chromite droplets in a round shape, forming orbicules. Orbicules sink and shrink to form chromite nodules. (f) Rapid deposition of chromite nodules or orbicules forming massive chromite ores. (g) Caught up by the strong plastic shearing in the oceanic uppermost mantle, chromite nodules and orbicules are pulled-apart and disrupted. (h) Because of strong shearing deformation (pull-apart texture), nodules, orbicules and masses of chromites are disrupted, becoming disseminated and forming foliated ores. (i) Chromite is strongly sheared to form mylonite, and the chromite deposit body becomes concordant.

4. Discussion

333

334

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

Fig. 8. The evolution model of the structures of the podiform chromitites (modified after Nicolas, 1989). M, Moho; R, Ridge. (A) Nodular and orbicular microstructure. (B) Pull-apart structure. (C) Foliation in chromites. Deformation increases from site A to C.

evolution sequence can be divided into five stages, developed from disseminated texture to nodular texture, and then forming massive ore in the end (Fig. 7). In the massive ores with nodular and orbicular textures, magmatic flow structures are commonly preserved, suggesting that the Zunhua podiform chromite originated in fast flowing magma chambers. This conclusion is consistent with the theory that the formation of Cr-rich podiform chromite requires a fast vertical magmatic accretion in the uppermost oceanic mantle, associated with a medium-fast spreading oceanic ridge (Nicolas, 1989; Nicolas and Al Azri, 1991; Edwards et al., 2000). Volumes of the chromite ore bodies from the ZPC show a great deal of high-temperature shearing structures including pull-apart texture, folding structure, schlieren structure and mylonitic structure, similar to the Oman podiform chromite and New Caledonia podiform chromite (Nicolas and Al Azri, 1991; Cassard et al., 1981). From the dissemination of the chromite nodules and orbicules, to the folding and mylonitization of chromite, a hightemperature deformation structure sequence occurs together with the magmatic structure and texture series (Fig. 7). According to the study of podiform chromitites in ophiolites from different regions (Nicolas, 1989; Edwards et al., 2000), the occurrence of podiform chromitite deposits

5. Conclusions

335

changes with distance away from the spreading center in the upper mantle as the oceanic plates spread. The discordant (with respect to mantle flow foliation) deposits with irregular shapes are emplaced late, and suffer little or remain free from the uppermost mantle plastic deformation. Therefore, the active magmatic accretion structures can be preserved in them. The concordant deposits, such as pod or tabular types of chromite, are tectonically reoriented parallel to the peridotite foliation. Because they are emplaced early during solid-state flow, they suffered a large plastic strain and the primary magmatic structures and textures are reformed to be high-temperature shearing structures, such as pull-apart textures, folding and mylonitic structures. The microstructures of chromitites change from nodular and orbicular to pull-apart, folding, foliated or mylonitic structures, suggesting progressive plastic deformation at high temperatures in the uppermost oceanic mantle as the oceanic blocks move transversely apart. The magmatic sequence and the deformation series of structures and textures suggest that the ZPC were created in the uppermost mantle beneath a fast spreading oceanic ridge, in this environment, active magmatic accretion and strong plastic flow occurred (Fig. 8). Recently, more and more geochemical studies conclude that podiform chromite has characteristics that suggest they formed in forearcs, island arcs, or back-arc basins. These supra-subduction zone settings have water available for the melt, and are favorable tectonic settings for podiform chromite. Podiform chromite commonly has volumes of primary hydrous mineral inclusions, which are perhaps the record of the fluid extracted from the subduction of the oceanic plate. There are not only the structures and textures suggesting faster ocean spreading, but also a great deal of primary hydrous mineral inclusions, such as phlogopite, and hornblende, preserved in the Zunhua podiform chromite. The mafic volcanic rocks in the Zunhua Structural Belt exhibit island-arc and back-arc tholeiite characteristics (Bai et al., 1993). And the mantle peridotite and podiform chromite of the Zunhua Neoarchean ophiolite show island-arc geochemical characteristics (Huang, 2003; Kusky and Li, 2003). Therefore, we prefer a faster spreading back-arc basin for the origination of the Zunhua Neoarchean ophiolite. 5. CONCLUSIONS The ZPC preserves two sequences of structures and textures. One is characterized by nodular textures and orbicular textures, suggesting active vertical magmatic accretion in the uppermost oceanic mantle. The other sequence consists of high-temperature deformation structures including pull-apart structures, banded structures, folds, schlieren structures, and mylonitic structures. These structures reveal the strong plastic flow and transverse movement of the oceanic mantle. The two sequences of structures and textures suggest the Zunhua Neoarchean podiform chromite was formed in a fast spreading oceanic basin. Abundant primary hydrous mineral inclusions are found in the chromium spinel of the ZPC. The geochemical characteristics of the mafic volcanic rocks and the mantle peridotites from the Zunhua ophiolite suggest a suprasubduction zone tectonic setting. Concluding all the restricted conditions above, we infer the Zunhua Neoarchean ophiolite is the relic of a fast spreading suprasubduction zone ridge.

336

Chapter 10: Microstructures of the Zunhua 2.50 Ga Podiform Chromite

ACKNOWLEDGEMENTS This study was supported by grants from the National Natural Science Foundation of China (Grant Nos. 40242014, 49832030), Peking University (Project 985), and the U.S. National Science Foundation (Awards 02-07886 and 01-25925 to T. Kusky). We thank Prof. Geng Yuansheng and Hao Zhiguo, Chinese Academy of Geological Sciences, for comments on this paper. Thanks to Niu Xianglong, Feng Jun, Liu Zhiqiang, Zhang Zhiqiang who participated in part of the field works. REFERENCES Ahmed, A.H., Arai, S., Attia, A.K., 2001. Petrological characteristics of podiform chromitites and associated peridotites of the Pan African Proterozoic ophiolite complexes of Egypt. Mineralium Deposita 36, 72–84. Arai, S., Matsukage, K., 1998. Petrology of a chromitite micropod from Hess Deep, equatorial Pacific: a comparison between abyssal and alpine-type podiform chromitites. Lithos 43 (1), 1–14. Bai, J., Huang, X.G., Dai, F.Y., et al., 1993. Precambrian Crustal Evolution of China. Geological Publishing House, Beijing, p. 223 (in Chinese). Bai, W.J., Zhou, M.F., Hu, X.F., et al., 1993. Mafic-Ultramafic Magmatism and Tectonic Evolution of the North China Craton. Seismological Press, Beijing, pp. 198–210 (in Chinese). Bai, W.J., Wang, X.B., Wang, H.S., et al., 1976. Rock Types and Metallogenic Regularity of the Chromitite Bearing Ultramafic Massif. Geological Publishing House, Beijing, pp. 129–133 (in Chinese). Ballhaus, C., 1998. Origin of podiform chromite deposits by magma mingling. Earth and Planetary Science Letters 156, 185–193. Cassard, D., Nicolas, A., Rabinpvitch, M., et al., 1981. Structural classification of chromite pods in southern New Caledonia. Economic Geology 76, 805–831. Edwards, S.J., Pearce, J.A., Freeman, J., 2000. New insights concerning the influence of water during the formation of podiform chromitite. In: Dilek, Y., et al. (Eds.), Ophioltes and Oceanic Crust: New Insights from Field Studies and the Ocean Drilling Program: Boulder, Colorado. Geological Society of America Special Paper 349, 139–147. Johnston Jr., W.D., 1936. Nodular, orbicular, and banded chromite in Northern California. Economic Geology 31, 417. Huang, X.N., Li, J.H., Chen, Z., et al., 2003. Petrological and structural characteristics of the Zunhua Neoarchaean Ophiolitic Melange: proof of the operation of ancient plate tectonics. Acta Scientiarum Naturalium Universitatis Pekinensis 39 (2), 200–212 (in Chinese with English abstract). Huang, X.N., 2003. Zunhua Neoarchean podiform chromite and its implication for plate tectonics process in North China Craton. Ph.D. thesis. Peking University, Beijing, China (in Chinese with English abstract). Holtzman, B., 2000. Gauging stress from mantle chromitite pods in the Oman ophiolite. In: Dilek, Y., Moores, E., Elthon, D. (Eds.), Ophioltes and Oceanic Crust: New Insights from Field Studies and the Ocean Drilling Program: Boulder, Colorado. Geological Society of America Special Paper 349, 149–158. Kusky, T.M., Li, J.H., Tucker, R.T., 2001. The Dongwanzi Ophiolite: Complete Archean ophiolite with extensive sheeted dike complex, North China craton. Science 292, 1142–1145.

References

337

Kusky, T.M., Li, J.H., 2002. Is the Dongwanzi complex an Archean Ophiolite? (Response to Zhai, M., Zhao, G., and Zhang, Q.). Science 295, 923. Kusky, T.M., Li, J.H., 2003. Paleoproterozoic Tectonic Evolution of the North China Craton. Journal of Asian Earth Science 22, 383–397. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Leblanc, M., Nicolas, A., 1992. Ophiolite chromitites. International Geology Review 34 (7), 653– 686. Li, J.H., Qian, X.L., Huang, X.N., 2000a. The tectonic framework of the basement of North China and its implication for the early Precambrian cratonization. Acta Petrologica Sinica 16 (1), 1–10 (in Chinese with English abstract). Li, J.H., Kröner, A., Qian, X.L., 2000b. The tectonic evolution of early Precambrian high-pressure granulite belt, North China Craton. Acta Geologica Sinica 74 (3), 246–256 (in Chinese with English abstract). Li, J.H., Niu, X.L., Huang, X.N., et al., 2002a. Podiform chromitites: a key to identify the ancient oceanic lithospheric relics. Earth Science Frontiers 9 (4), 235–246. Li, J.H., Kusky, T.M., Huang, X.N., 2002b. Archean podiform chromitites and mantle tectonites in ophiolitic melange, North China Craton: a record of early oceanic mantle processes. GSA Today 12 (7), 4–11. Li, J.H., Niu, X.L., Chen, Z., et al., 2002c. The discovery of podiform chromite in west Liaoning and its implication for plate tectonics. Acta Petroloagica Sinica 18 (2), 187–192 (in Chinese with English abstract). Matveev, S., Ballhaus, C., 2002. Role of water in the origin of podiform chromitite deposits. Earth and Planetary Science Letters 203, 235–243. Nicolas, A., 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Dordrecht. Nicolas, A., Al Azri, H., 1991. Chromite-rich and chromite-poor ophiolites. In: Peters, T.J., Nicolas, A., Coleman, R.G. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Dordrecht, pp. 261–274. Peters, T.j., Nicolas, A., Coleman, R.G. (Eds.), 1991. Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Dordrecht, p. 903. Thayer, T.P., 1964. Principal features and origin of podiform chromite deposits, and some observations on the Guleman-soridag District, Turkey. Economic Geology 59, 1497–1524. Wang, K., Li, J., Hao, J., et al., 1997. Late Archean mafic-ultramfic rocks from the Wutaishan, Shanxi Province: a possible ophiolite melange. Acta Petrologica Sinica 13 (2), 139–151 (in Chinese with English abstract). Wu, J.S., Geng, Y.S., Shen, Q.H., et al., 1998. Archean Geology Characteristics and Tectonic Evolution of Sino-Korea Paleoconitnent. Geological Publishing House, Beijing, pp. 43–157 (in Chinese). Zhou, M.F., Malpas, J., Robinson, P.T., et al., 2001. Crystallization of Podiform Chromitites from Silicate Magmas and the Formation of Nodular Textures. Geology 51 (1), 1–6.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

339

Chapter 11

NEOARCHEAN MASSIVE SULFIDE OF WUTAI MOUNTAIN, NORTH CHINA: A BLACK SMOKER CHIMNEY AND MOUND COMPLEX WITHIN 2.50 GA-OLD OCEANIC CRUST JIANGHAI LIa , TIM KUSKYb , XIANGLONG NIUa , FENG JUNa AND ALI POLATc a Department

of Geology, Peking University, Beijing 100871, China Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63103, USA c Max-Planck-Institut für Chemie, Postfach 3060, D-55020 Mainz, Germany b Department of

The Wutai VMS is one of largest sediment-hosted sulfide deposits in China. It occurs as small lenses, thin sheets, and tabular bodies of massive to layered sulfide, disseminated over a forearc mélange belt. Although they are reworked by late deformation, sulfide deposits formed at different crustal levels still can be identified, including relicts of chimneys, pyritic siliceous exhalite, massive crystallized sulfides, talus of massive sulfides and stockwork zones. The country rock of the Wutai VMS ores show intense silicification and chloritization. Epidosites have been identified within mafic rocks. Under microscopic observations, porous sulfides show a mineralogical zonation around micro-conduits. The colloform textures developed delicate banding and concentric textures. Vuggy cavities are commonly lined by concentric layers consisting of idiomorphic pyrite and silica. The Wutai VMS are spatially associated with a convergent plate boundary, formed in the upper sequence of a former Neoarchean oceanic basin. They have been overthrust by forelandthrust belts following closure of the oceanic basin. Our preliminary studies reveal the presence of black smoker chimneys preserved in the Wutai Mt., suggesting that seafloor black smoker hydrothermal activity at about 2.50 Ga played an important role in the generation and accumulation of Wutai VMS. In addition, the Wutai VMS is quite similar to younger Besshi-type deposits, and it is inferred to have been generated in a forearc setting, later tectonically transported in mélange belts during continental collision. 1. INTRODUCTION Our understanding of the thermal structure and evolution of the ocean crust, hydrothermal circulation of seawater through the crust, genetic process of sulfide deposits, and the development of early life, have been greatly improved since active hydrothermal vents were discovered at many sites on modern spreading centers on mid-ocean DOI: 10.1016/S0166-2635(04)13011-9

340

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

ridges and in backarc basins (Fouquet et al., 1991; Sawkins, 1990; Nisbet and Fowler, 1996; Haymon, 1996; Alt and Teagle, 2000; Herzig and Hannington, 1995; Rona, 2002; Scott, 2002). Therefore, the search for ancient massive sulfide deposits within ophiolites will provide information on hydrothermal activity, and the development of life in ancient oceans. Preserved records of hydrothermal activity associated with oceanic crust was first identified in the Oman and Troodos (Cyprus) ophiolites (Haymon et al., 1984; Little et al., 1999a). Indicators of black smoker activity are commonly preserved as massive sulfides within the upper parts of ophiolites (Boyle et al., 1984; Tunnicliffe et al., 1996; Herrington et al., 1998). At least twenty well-documented examples of fossil hydrothermal vents have been discovered associated with ophiolites and ancient rifts, rifted margins, and immature ocean basin sequences. Most studies of hydrothermal vents have been concentrated in volcanogenic massive sulfides (VMS) of the modern oceanic floor and Phanerozoic ophiolites. The record of black smoker deposits dating back to the Archean is only rarely reported (Vearncombe et al., 1995; Rasmussen, 2000), although VMS are widely distributed in Archean greenstone belts (Herrington et al., 1997; Misca, 2000; Schandal et al., 1990; Polat and Kerrich, 1999). The documentation of Archean black smokers will provide significant information for understanding submarine hydrothermal processes, and provide a window into Archean heat flow, different stages in the development of life, and details about the chemical interaction between the lithosphere, hydrosphere, and atmosphere. The objectives of this study are to highlight the geology of several massive sulfide deposits associated with a Neoarchean ophiolitic mélange (2.50 Ga) of the Wutai Mountain area, North China (Wang et al., 1996). In addition, we report on some vent and chimney structures in the massive sulfide deposits.

2. GEOLOGICAL SETTING The North China craton consists of three main tectonic elements, including the Western Block, Central Orogenic Belt (COB) (2.50 Ga) and Eastern Block (Li et al., 2002; Kusky et al., 2004). To the north a granulite-facies province (1.80–2.50 Ga) is preserved, representing the lower-level crustal tectonites (Fig. 1). From west to east, the COB can be subdivided into a high-pressure granulite belt, a TTG-supracrustal sequence, ophiolitic complexes, and the Qinglong foreland basin (Fig. 1) (Li et al., 2002). These belts record the Neoarchean tectonic assembly and collision between the Western and Eastern Blocks. The research area is located at junction between Wutai and Taihang Mountains, in the central sector of the COB (Fig. 2). Regionally, it is subdivided as a forearc mélange belt. Numerous Neoarchean ophiolitic blocks have been recognized along the belt in the last decade (Li et al., 1990; Bai et al., 1992; Wang et al., 1996; Wilde et al., 1998; Li et al., 2002). The ophiolitic blocks are in thrust contact with foreland basin sequences and older continental margin type rocks (Kusky and Li, 2003). The ophiolitic complex occurs as numerous tectonic blocks and sheets within a mélange matrix. Some large blocks of harzburgite, dunite, pyroxenite, and ultramafic layered complex are recognized

2. Geological Setting

341

Fig. 1. Map showing the main tectonic units of North China Craton and location of ophiolitic complexes and volcanogenic massive sulfide deposits.

(Bai et al., 1993; Wang et al., 1997). All these blocks are intensely sheared and folded together with country rock matrix. The recumbent folds with BIF can be recognized in outcrops. Garnet-staurolite-kyanite can be recognized within the metasediments, indicating amphibolite-facies metamorphism. An important unit associated with the ophiolitic sequences and mélange in the Wutai Mountains are several volcanogenic massive sulfide (VMS) deposits (Fig. 2). At least two large massive sulfide deposits are mined in the working area. Mainly pyrite is mined for sulfuric acid, although chalcopyrite deposits are also present. They are widely developed within a NE-trending Neoarchean structural belt, and can be traced to the interior of Wutai belt (Fig. 2). The massive sulfide deposits and relevant ophiolitic blocks (2.50 Ga) now form an allochthonous unit, overthrust onto foreland basin sediments (2.50–2.40 Ga). 2.1. Geological Relationship of Wutai VMS The Wutai massive sulfide deposits are located approximately 300 km southwest of Beijing. They are associated with a mélange belt (Fig. 2), traced for more than 40 km,

342

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 2. Schematic geological map of the Wutai-Taihang mountains, North China Craton, showing the locations of the volcanogenic massive sulfide deposits, and their association with the Wutai greenstone belt (modified after Yuan, 1988).

within which many ophiolitic blocks are identified. The main rock types in these ophiolitic blocks include pillow lavas, amphibolite, sheeted dike, cumulate mafic and ultramafic rocks

3. Ore Deformation of Wutai VMS

343

(dunites, serpentinites), and serpentinized harzburgite tectonite (Li et al., 1990). Layers of banded iron formation (BIF) are also very common, interleaved with turbiditic sequences. The massive sulfide deposits (MSD) of Wutai are conformably hosted by the BIF (2.50 Ga), which are interpreted as a upper exhalative sequence of the ophiolitic complex. The dismembered ophiolitic complex is thrust over foreland basin deposits. The massive sulfide deposits are intercalated with sedimentary rocks (BIF), and are in structural contact with ophiolitic mélanges. The age of the mafic rocks hosting the ore is 2500 Ma, dated by U-Pb analyses of zircon (Bai et al., 1992). The main VMS of Wutai forms stacked lenses and concordant sheets of massive sulfides within amphibolite, rhyolite, BIF, and metasediment (chert, schist and turbidite). Single bodies are 100–1000 m in strike length. They have highly variable thicknesses, with a maximum thickness of 50–60 m, passing laterally into 0.3 m thick layers. The main ore body belt is elongated NNE and dips 30◦ NW with a down-dip extension of at least 20– 500 m. A normal fault controls the outcrop pattern and distribution of the main deposit. Individual ore bodies commonly consist of a number of lens-shaped to layered irregularshaped masses. Associated stringer ore is uncommon within the Wutai VMS. The Wutai VMS deposits are located near the volcanic-sedimentary rock (mica-quartz schist and gneiss, BIF, chert) interface. They consist of concordant mineralized lenses containing greater than 30% massive sulfide minerals. Lenses or sheets of VMS are intercalated with mafic volcanics, garnet biotite schist, BIF and chert, although they are mainly hosted by the BIF and pyritic chert. VMS deposits may be transitional to volcanicassociated iron formation. The primary ore bodies consist essentially of massive pyrite, and contain varied amounts of chalcopyrite, pyrrhotite and sphalerite, although some ores bodies are composed of chalcopyrite with minor pyrite. Some isolated orebodies of chalcopyrite are also identified. More than ten ore bodies of pyrite and/or chalcopyrite have been investigated in the field, the largest one being more than one kilometer long. The footwall mafic lavas and sediments beneath the deposits show extensive silicification, CO2 metasomatism (carbonate), K-metasomatism (fuchsite with ultramafic rocks), and chloritization (in mafic rocks) during deposition of VMS ores. As the massive sulfide deposits are mainly controlled by normal faults, they are interpreted to have been originally deposited within oceanic grabens.

3. ORE DEFORMATION OF WUTAI VMS 3.1. Structural Deformation with MSD Ores The massive sulfide dominated by pyrite typically occurs as bands or laminae within BIF, and have undergone regional folding, shearing, and flattening together with the host rock (Figs. 3a–3d). Sulfide bands are commonly 0.5–15 cm thick, and sometimes up to 30 cm thick. They are both concordant and discordant with the foliation of host rocks. The bands

344

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 3. Photographs of Wutai massive sulfide ores. (a) Massive sulfide dominated by pyrite occurs as folded and flattened bands within exhalite horizon, coin is about 2 cm in diameter. (b) Massive sulfides are tightly folded within chloritized mafic rocks and exhalite. (c) Sheared massive sulfide exhibiting gneissic texture within BIF. (d) Deformed massive sulfide is discordant with host rock. (e) Sulfide ore minerals precipitated in the pressure shadow regions of porphyroclasts. (f) Syn-sedimentary fold structures are preserved within exhalative layers. (g) Exhalite showing disseminated pyrite crystals within silica matrix. (h) Silica-carbonate exhalite.

4. Mineralogy and Ore Petrology of VMS

345

and laminae often alternated with BIF, and are folded together. Synkinematic mineralization is suggested by ore minerals that have precipitated in the pressure shadow regions of porphyroclasts (Fig. 3e). Gneissic sulfide textures are common in metamorphosed and deformed deposits. Through this second phase of deformation, sulfides locally became concentrated in the hinge zones of folds or occur as tails around silica or garnet blasts. Tight folds of silica layering can be recognized within massive sulfide, indicating intense deformation with ores. The mechanical behavior of the sulfide during folding and flattening is variable and in some cases very unusual. In many cases sulfide bands are folded tightly, with thickening of sulfide layers in the hinge zone. In these cases, the sulfide acted as the weak layers, flowing into the fold hinges. Some pyrite appears to have mechanically penetrated into adjacent layers during folding. The pyrites are typically sheared with the matrix and show brittle fragmentation, within which silica lenses or augen are distributed. As grains of pyrite ore are weakly consolidated, their strength is weak, with pull-apart textures developed along grain boundaries. In other cases grain-size is decreased by mechanical shearing. In addition, garnet porphyroblasts formed within the massive sulfide by progressive metamorphism and deformation; inclusions of pyrite within garnet record rotation of the porphyroblasts. The pillow lava are intensely sheared and metamorphosed into foliated amphibolite. At least two phases of mineralization of sulfide is recognized. In the Wutai VMS, the early ore layers are commonly cut by late veins consisting of quartz, calcite and less abundant pyrite. 3.2. Syn-Sedimentary Structures In the weakly deformed domain, many syn-sedimentary structures are recognized within concordant layers of VMS. Some pyritic BIF exhibit normal-faults on hand-sample scales, with crack-seal veins of carbonate-quartz-euhedral pyrite. BIF with minor veinlets of pyrite-silica, and clasts of pyrite are developed, which show complicated fold patterns confined to individual layers. We suggest that these are a record of syn-sedimentary structures, associated with marine tectonic activity. Micro-olistostromes can be found, with complicated fold patterns of pyrites and disrupted layering in the BIF (Fig. 3f).

4. MINERALOGY AND ORE PETROLOGY OF VMS Although they are reworked by late deformation, the sulfide deposits formed at different crustal levels still can be identified, including clastic sulfides, relicts of chimneys, pyritic siliceous exhalite massive crystallized sulfides, and stockwork zones. The host rock for the sulfide deposit is dominated by BIF, basalt, and minor rhyolite. The sulfide ores commonly grade into horizons of mineralized exhalite. The veinlets are mainly associated with mafic rocks. Locally, several of the large massive sulfide lenses are structurally underlain by discordant, chalcopyrite-bearing pyrite stringer zones, contained in pipes of hydrothermally altered rock. Rocks in the pipes are brecciated to form a chlorite-quartz-pyrite stockworks.

346

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

The lower contacts of the main and the satellite ore bodies are typically gradational into a stringer zone. The stringer zones are interpreted as the locus of hydrothermal fluid discharge (e.g., Thurston et al., 1992). 4.1. Massive Crystallized Sulfides The massive sulfide deposits are formed largely of recrystallized pyrite, with variable amounts of pyrrhotite, bornite, marcasite, chalcopyrite, sphalerite, galena, magnetite, barite, anhydrite, and siderite. Apart from quartz, other gangue mineral present include calcite, chlorite, sericite, and epidote. Specimens of ore from Jinggangku contain pyrite > chalcopyrite > sphalerite > barite, whereas those from Dachuan contain pyrite > sphalerite > galena, or apophyllitite. Quartz, sericite, chlorite, calcite, and anhydrite occur in variable quantities. Three types of ore deposits can be subdivided from those in the region, including massive ore, banded ore, and clastic ore. They are closely associated with siliceous exhalites Massive ores consist mainly of pyrite (up to 80–90%) with minor pyrrhotite and quartz. In podiform ores, pyrite occurs as pods together with layers of BIF within lava. Banded ores consist of pyrite layering interleaved with BIF, chert or gneiss with bands of pyrite. They may accompany quartz-chalcopyrite veinlets. Pyrite crystals can be recognized, and delicate primary banding is typically well preserved. Clastic sulfide ores form thin interlayers on the flanks of the more massive lenses. In disseminated ores, fine-granular pyrites, pyrrhotite and magnetite are distributed along silica layering or foliation of lava. Clasts and debris of sulfides are locally scattered along layering in the BIF and chert. Locally, disseminated laminae are defined by sulfides, and folded together with host BIF. These siliceous metalliferous sediments with bands of pyrite were deposited in a submarine setting. Exhalite sinter vent complex commonly grades laterally into Fe-rich siliceous sediment, and occurs as a finger or arm of the main VMS deposit. They consists of fine-grained massive to laminated silica, and are characterized by disseminated pyrite crystals, carbonates and silicified layers (Figs. 3g and 3h). 4.2. Porous Massive Sulfide: Records and Products of Previous Hydrothermal Fluid Channelways Porous massive pyrite deposits exhibit high porosity, ranging between 5 and 20%. These vugs are commonly irregular and between 0.1 mm and 0.5 cm long. According to their mineralogy, two types of porous massive sulfide ores can be subdivided, including iron sulfide and pyrite-sphalerite ores (Figs. 4a and 4b). They consist of fine-grained pyrite, pyrrhotite, sphalerite, and marcasite, characterized by a honeycomb-like structure on a millimeter scale. Clasts of chloritized amphibolite are very common. Cavities in brecciated rocks are filled with silica. Pyrrhotite and pyrite are enriched around micro-conduits. Some vugs are filled with amorphous silica, idiomorphic pyrite and sphalerite. Irregular patches

4. Mineralogy and Ore Petrology of VMS

347

Fig. 4. Porous massive sulfides of the vent complex. (a) Porous massive sulfides with abundant vugs. (b) Porous massive pyrite with irregular patches of coarse-grained pyrite. (c) Massive Fe-Zn sulfides showing high porosity.

of coarse-grained pyrite within porous massive pyrite are interpreted to mark the locations of former fluid channel-ways within mounds (Fig. 4c).

348

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 5. Polished section photomicrographs of the Wutai-Taihang massive sulfide ores, plane polarized light. (a) A thin rim of pyrite lines the inner wall of a micro-conduit. (b) Concentric conduit lined by millimeter-sized pyrite and sphalerite crystals. (c) Collofrom pyrite structures with cement. (d) Isolated micro-conduit defined by pyrite crystals.

These kinds of structures are commonly interpreted to be formed on the outermost parts of mounds or on the outer walls of vent chimneys (Tivey and Singh, 1997; Hannington and Scott, 1988; Hannington et al., 1998; Knott et al., 1998). The chimney walls are characterized by porous networks because of conductive cooling and fluid mixing process through the wall (Herzig et al., 1993). Some sulfides with chimney-like shape are well preserved within BIF sequence. It suggests that hydrothermal fluid emitted from black smoker chimney on the sea floor played an important role in the formation of the Wutai VMS. Although they have suffered the combined effects of tectonism and metamorphism, many primary textures associated with exhalative process can be recognized in the Wutai VMS deposits under microscopic observations. Voids and colloform textures are very common features within the massive pyrite (Figs. 4 and 5). Some vugs are filled with amorphous silica, chalcopyrite or pyrite with inclusions of other minerals. These tiny crystals grew inward into vugs (Fig. 5a). The voids within iron sulfide are commonly overgrown by

4. Mineralogy and Ore Petrology of VMS

349

euhedral to subhedral pyrite, forming the banding and concentric textures. The vuggy cavities are commonly lined by concentric pyrite layers consisting of idiomorphic pyrite and silica (Fig. 5b). Colloform structures displaying well-developed micrometer-scale growth banding or lamination, interpreted to be micro-conduits (Figs. 5c and 5d). The colloform banding shows complex cross-cutting relationships, suggesting episodic dissolution and precipitation process. The hydrothermal vent textures show concentric patterns of laminated silica and pyrite (Fig. 5b). The pyrite crystals defining the conduit and vent deposition sites are extremely well preserved. Honeycomb textures with pyrite are very common, which in some places are filled by silica (Fig. 5d). The inner walls of the honeycomb-like pyrite are lined with fine-grained euhedral pyrite and chalcopyrite crystals, or delicate dendritic aggregates. Dendritic crystals indicate rapid crystallization due to rapid cooling under unstable conditions (Fouquet et al., 1993). Occasionally, micro-conduits are cored by an assemblage of crystalline pyrite, chalcopyrite and sphalerite (Fig. 5b), indicative of mineral precipitation within a fluid conduit (Brown and McClay, 1998). Late epidote veinlets are also developed. In places where the development of the late foliation and the recrystallization of pyrite is strong, the primary textures are intensely obscured. Similar textures as described above with primary porous massive sulfides have been widely documented with black smoker chimneys and the uppermost part of sulfide mounds and ancient and modern VMS deposits, such as TAG Mid-Atlantic ridge (Rona and Scott, 1993; Brown and McClay, 1998; Hannington et al., 1998; Knott et al., 1998; Butler and Nesbitt, 1999), in the Indian Ocean (Halbach et al., 1998) and those of the Pilbara craton (Vearncombe et al., 1995). The colloform and vuggy pyrite is representative of previous hydrothermal fluid channelways. 4.3. Other Types of Sulfide Associated with Vent Chimneys Talus of massive sulfides are sulfides with rubbly and brecciated structures that are well preserved in the Wutai VMS deposits. The centimeter-sized breccias of iron sulfides are set in and cemented by a fine-grained sulfide-carbonate matrix (Fig. 6a). They are interpreted to be fragments of a submarine talus pile, comparable to sulfide mounds around modern black smoker chimneys on the sea floor. In addition, the pyrite-bearing chert-BIF fragments are cemented within a quartz-sulfide matrix. The pebbles are about 3–6 cm in size. They are suggested to be associated with submarine synsedimentary tectonic activity during mineralization. Massive sulfide with mafic breccias mainly consist of sulfides (pyrite, chalcopyrite, and minor amounts of galena). Small fragments of mafic composition are recognized (Fig. 6b). It is suggested that they formed around hydrothermal vents built on mafic volcanics, closely related to the hydrothermal fragmentation of oceanic crust. Similar rock has been reported from modern submarine black smoker chimneys and mounds. Porous Cu-Fe-Pb sulfides constitute layered sulfide precipitates with silica, forming porous ore deposits. Mafic breccias are commonly distributed within a matrix of sulfide (Fig. 6c). Centimeter-sized vugs are preserved, which are filled with of clusters of apophyllitite crystal that grew inward (Fig. 6c). The lager vugs are up to 4 cm in diameter. A closely

350

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 6. Various massive sulfide and exhalite textures and structures associated with a paleo hydrothermal vent complex. (a) Talus of massive sulfide showing brecciated structures. (b) Massive sulfide with mafic volcanic breccias. (c) Sulfide-bearing garnet-bearing mafic breccia. (d) Porous Cu-Fe-Pb sulfides within exhalite horizon. (e) Carbonate-Fe sulfide chimneys. (f) Cross-sections of the chimney tubes showing concentric zones of carbonate and core sulfide minerals.

5. Stockwork and Alteration of Country Rock

351

relevant rock type is sulfide carbonate (Fig. 6d). With tubular cavities between a honeycomb pyrite network, chalcopyrite occus as infillings to cavities in massive pyrite. Disseminated pyrite crystals are well preserved. They are derived from metamorphic equivalents of altered volcanics and relevant exhalite. They formed near a submarine volcanic center, as it is characterized with intense hydrothermal alteration of mafic rocks. Carbonate-sulfide chimneys are locally well preserved in the VMS deposits. They display typical chimney-like shapes, are about 20 cm high, with diameters of 10–15 cm (Figs. 6e and 6f). Their structures and geometry are consistent with growth in an open space. The chimneys are composed of carbonate, silica and pyrite. Their surfaces are overlain by a brown, metalliferous mud film. The fine-grained pyrite breccias with laminated textures are surrounded by bladed carbonate crystals. Cross-sections of the chimney tubes reveal concentric zones defined by carbonate wall and core sulfide minerals. This type of chimney has been reported from the western Pacific oceanic floor and Eastern African Rift (Wu, 2000; Tiercelin et al., 1993).

5. STOCKWORK AND ALTERATION OF COUNTRY ROCK Locally stockworks underlying the Wutai VMS can be recognized. They occur within mafic to rhyolitic rock, strongly deformed and imbricated by late thrusting and folding. The VMS of Wutai are mainly sheet-like, it is not clear now if this represents an artifact of deformation, or it is a primary feature. 5.1. Brecciated and Veined Altered Mafic Rock Veinlets and disseminated fragments of chalcopyrite-pyrite-quartz occur within altered mafic rocks. The angular mafic breccia and fragments are characterized by veins or cavity infillings of epidote, pyrite, and quartz (Figs. 7a and 7b). They occur in a matrix mainly of pyrite + chalcopyrite together with quartz, or are cemented by quartz and quartz-sulfide veins. Breccias are commonly altered to assemblages of quartz + chlorite + pyrite. Disseminated sulfides are also present in the altered mafic rocks. Pyrite, chalcopyrite-sphalerite, and galena may occur as different veinlets, typically less than 1 cm wide. Some veins are 4–10 cm wide, within which fine-grained euhedral pyrite grains are distributed along fissures. Biotite increases along the margins of sulfide-quartz veins. Brecciated and veined altered mafic rocks can be interpreted as the stockwork feeder veins to the overlying massive sulfide. Dense quartz-pyrite veinlets are preserved, representing hydrothermal feeders to the syngenetic mineralization. The order of mineral appearances in the stockwork is pyrrhotite, pyrite, chalcopyrite, and, finally, galena with younger increments of time. The relative rigidity of the host mafic rock during deformation preserved within it veins of sulfide. These feeder zone sulfides beneath the ore bodies within BIF consist largely of pyrite, chalcopyrite and silica. They are characterized by quartz-pyrite-pyrrhotite-chalcopyrite

352

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 7. Veined and altered country rock. (a) Sulfide vein showing brecciated margins in mafic rocks. (b) Veinlets and disseminated sulfide within altered mafic rocks. (c, d) Stockwork within chert-BIF showing irregular conduits. (e) Epidosites.

5. Stockwork and Alteration of Country Rock

353

mineralization in veinlets, and disseminated sulfide within strongly altered host rocks. Compared with massive sulfides within sheet-like and lensoidal bodies, the content of pyrrhotite and chalcopyrite is higher within veinlets of mafic rocks, and pyrrhotite crystallized earlier than pyrite. In many ophiolites, lenticular iron or Cu-Zn sulfide ore bodies occupy paleo-depressions in the surface of the basalt (Boyle et al., 1984), and are underlain by stockworks of ore minerals. These feeder zones are commonly interpreted to be original conduits for ascending hydrothermal solutions. 5.2. Stockwork with Chert-BIF and Turbiditic Rocks In chert-BIF, irregular conduits and voids are preserved within the rocks (Figs. 7c and 7d), and are commonly lined with fined-grained idiomorphic pyrite and quartz. They can be interpreted as sulfide filling original cavities and fractures in the sediments. The irregular veinlets of sulfide are common in chert, which break the rock into chert breccia. Some cracks are sealed by disseminated pyrite and quartz crystals. They represent shallow-level hydrothermal conduits in submarine sediments. Although they underwent late deformation, mineralized veinlets of sulfide and quartz clearly cut across original bedding. Occasionally, sulfide veinlets that cut across BIF are preserved. Some veinlets of sulfide-quartz within BIF show branches, within which well-crystallized pyrite and quartz are preserved. Within turbiditic rocks, sulfide-quartz breccias are common. Irregular and cross-cutting veinlets (1.5–2 cm wide) of sulfide-quartz are a common texture. Sulfide replaced fragments and brecciated turbiditic rocks. Anhydrite veinlets are also recognized. Biotite or hornblende-rich margins (0.5–5 mm thick) are common with sulfide veinlets, associated with hydrothermal alteration. Garnet occurs with metamorphosed alteration halos. Centimeter sized veins of quartz, pyrite, and epidote occur within veinlets and vesicles, and in fractures in altered turbiditic sediments. Some sulfide veins up to 4 cm wide cut across bedding in turbiditic units, representing deep-level conduits from hydrothermal activity. Compared with massive sulfide, pyrhotite content is increased within the turbiditic stockwork. It mainly occurs along margins of veinlets and crystallized earlier than other sulfides. The host rocks clearly underwent shearing, with garnet showing fined-grained tails of sulfide. 5.3. Alteration and Deformation of Country Rocks The country rock of Wutai VMS ores show early silicification, carbonate, sericite, fuchsite (in chert), phlogopite (in mafic lavas), tourmaline (in mafic lava) and chlorite-graphite alteration. For example, mafic rocks are mainly alterated to fuchsite and phlogopite, rhyolitic rocks are characterized by chlorite-biotite alteration. The sericite-silicification alteration with chert accompanies generation of disseminated pyrite. In BIF, intense silicification and local carbonate are developed. Similar types and patterns of alteration have been documented in many massive sulfide ores of the world (e.g., Gillis and Banerjee, 2000).

354

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Fig. 8. Mineral paragenesis for Wutai-Taihang VMS.

Within mafic rocks underlying VMS of Wutai, epidosites are well preserved. They are characterized by metasomatic replacement of primary igneous minerals by granoblastic assemblages of quartz + epidote + magnetite + sulfides (Fig. 7e). It consists almost entirely of quartz and epidote in subequal proportions, characterized by reaction zones and veinlets of epidote. The accessory minerals with epidosites include chlorite, oxide phases, pyrite, prehnite, and andradite. The protoliths of epidosites are mainly basalts. Epidosites and their transitional assemblages have been identified in the Tonga forearc and many ophiolites, such as Pindos (Greece), Troodos (Cyprus), and Josephine (California) (Valsami and Cann, 1992; Harper, 1999; Banerjee et al., 2000). Epidosites are commonly formed at temperatures of up to 450 ◦ C with high cumulative fluid/rock ratios, and represent hydrothermal fluid conduits (Gillis and Banerjee, 2000). They are supposed to be formed in upflow zones at the bases of ore-forming hydrothermal systems of massive sulfide deposits (Banerjee et al., 2000). Epidosites, vein system, alteration of country rocks, and stockworks mark discharge zones adjacent to sulfide deposits.

6. MINERAL PARAGENESIS FOR WUTAI VMS The mineral assemblage in the Wutai VMS is relatively simple, dominated by pyrite and minor chalcopyrite, marcasite or pyrrhotite, with galena and barite as rare components (Fig. 8). The host rocks of the Wutai MSD is mainly BIF and chert, and less commonly mafic volcanics. In contrast, its stockwork is poorly preserved or not developed. The apparent lack of feeder zones is not a simple artifact of deformation and metamorphism, we

7. Tectonic Setting of the Wutai VMS and Its Comparison of with Other VMS

355

suggest that it is a characteristic, reflecting an affinity of the Wutai VMS with Besshi-type sulfide deposits. Chalcopyrite is commonly distributed around pyrite or in the branched ends of veinlets, suggesting that it was precipitated late in the mineralization sequence. Pyrrhotite is a common mineral with veinlets hosted by BIF and mafic rocks (Fig. 8). From massive sulfide ores to stockworks, the contents of chalcopyrite and pyrrhotite increase markedly, suggesting that chalcopyrite and pyrrhotite were precipiated under conditions of higher temperature. The content of pyrite in sulfide veinlets associated with mafic rocks is higher compared with other host rocks in the Wutai VMS, and chalcopyrite is also precipitated late. It is suggested that the stockwork is clearly related to the deep levels of the hydrothermal system, and the massive sulfide deposit formed at the shallow levels. Similar mineralized characteristics are reported with black smokers of SW Pacific oceanic floor (Fouquet et al., 1993). Sometimes, the increase of pyrrhotite content in the Wutai, sulfide is associated with the occurrence of sphalerite and galena, which may result from alteration close to the center of submarine volcanism. Stratigraphic zoning of the upper pyrite-rich and lower chalcopyrite-rich sections reflect formation of pyritic massive sulfide chimneys and mounds on the sea floor, followed by weak replacement of pyrite by chalcopyrite in the subsurface (Valsami and Cann, 1992).

7. TECTONIC SETTING OF THE WUTAI VMS AND ITS COMPARISON OF WITH OTHER VMS The Wutai VMS deposits occur as small lenses, thin sheets, and tabular bodies of massive to layered sulfide, disseminated over 7 km along the forearc mélange belt. They are located near the tectonic boundary between an island arc complex and a foreland foldthrust belt, and are caught up in a structural belt with oceanic blocks related to a collision (Li et al., 1990; Wang et al., 1996; Kusky and Li, 2003). They are spatially associated with convergent plate boundaries (former forearc setting or oceanic basin), and have been overthrust by foreland-thrust belts following closure of an oceanic basin. Many ultramafic and metagabbroic blocks are identified in the flysch mélange, including serpentinite, gabbro, peridotite, ultramafic complex, harzburgite, dunite, pyroxenite and, mafic sheeted dyke complex (Li et al., 1990; Wang et al., 1996; Li et al., 2002). They show sheared margins and tectonic contacts with surrounding rocks, and are believed to represent tectonically transported fragments of oceanic lithosphere deformed during a continental collision at 2.50 Ga (Li et al., 2002; Kusky and Li, 2003). The amphibolite adjacent to the VMS lavas show mid-ocean ridge basalt and arc-like geochemical signatures, interpreted as oceanic lavas. The geological and tectonic features outlined above strongly support the contention that the Wutai VMS was formed by hydrothermal activity within an oceanic basin. The Wutai VMS is mainly hosted within BIF and mafic rocks, which is similar to those of the Semail ophiolite of Oman (Hannington et al., 1998). It may be associated with volcanic centers built on ophiolitic sequences. The tectonic setting is suggested

356

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

to be a suprasubduction zone ophiolitic belt (forearc basin?). Some volcanic breccia associated with sulfides and BIF have been found within the Wutai VMS sequences, suggest the generation of the Wutai VMS were closely related with eruption of volcanics at sea floor. The sulfide mineralization consists predominantly of iron sulfide (pyrite and pyrrhotite) with lesser chalcopyrite, sphalerite, and magnetite. Other chemical sedimentary rocks (carbonate, iron-oxide beds, chert, pyritic BIF) are also recognized. The feeder alteration system in the footwall to the massive sulfide bodies is locally preserved but rarely exhibited, although host-rocks are commonly altered to a chlorite-quartz assemblage. Gangue minerals include amphibole, chlorite, epidote, carbonate, quartz, sericite, graphite, and tourmaline. Abundant graphite occurs in the sedimentary rock. Mn-rich garnets are present in the metamorphosed exhalative horizon. In addition, it is intensely imbricated by thrusting and associated with a convergent plate boundary. So, it is very similar to Besshi-type VMS deposits on its host rock (thick sequence of terrigenous clastic rocks, BIF), mineral assemblage of sulfides and tectonic setting (forearc, sedimented ridge) (Hutchinson, 1980). Many ophiolites host massive sulfide deposit (MSD), such as Pindos (Fe-Cu) (Greece), Troodos (Fe-Cu), Semail (Fe-Cu-Zn), Karmoy (Fe-Cu-Zn), Zambales, Uralian fold belt, Josephine ophiolite (Gillis and Banerjee, 2000; Zaykov et al., 1996; Alexander and Harper, 1992; Valsami and Cann, 1992). The VMS deposits mentioned above are commonly associated with supra-subduction-zone ophiolites (Sawkins, 1990), and the ore formation has close relationships in time and space with felsic intrusive magmatism. The typical example for these types of deposits is that of the Troodos ophiolite, Cyprus (Sawkins, 1990). Their mineral assemblages are very similar to those forming at mid-ocean ridge and suprasubduction belts (Gillis and Banerjee, 2000). In addition, VMS also occur within some Archean oceanic greenstone belts, such as Abitibi, in the Canadian Superior Province (Thurston et al., 1992; Polat et al., 1998; Kusky and Polat, 1999), where they are interpreted to be associated with island arc. The Wutai VMS is characterized by pyrite-chalcopyrite mainly hosted by BIF, rhyolite and basalt, which is quite similar to that of the Cyprus-type sulfide bodies. There, similar VMS (Fe-Cu-Zn), siliceous metalliferous sediments and underlying discharge zones associated with the ophiolite are well-developed (Haymon, 1983; Haymon et al., 1996). The Wutai VMS is hosted by BIF, turbidite and mafic volcanics, formed essentially synchronously with their enclosing rocks, as indicated by their strata bound occurrence, and olistostrome structures that involve sulfide layers. The ophiolitic blocks represent relevant parts of oceanic lithosphere. It is suggested that the VMS formed in the upper sequence of a former Neoarchean oceanic basin, and the ore deposition was localized on the submarine floor, on top of pillow lavas. There are several large harzburgite blocks associated with the Wutai ophiolite and VMS. These types of mantle tectonite are commonly interpreted to be associated with fastspreading ridges (Nicolas, 1989, 1995). The Wutai VMS is widely scattered over a long belt, and single deposits are small in size. This geological evidence also supports a fastspreading ridge model. In modern fast-spreading ridge settings, the hydrothermal vents are

8. Discussion: Submarine Hydrothermal Model for Generation of the Wutai VMS

357

commonly of small size (< 1000 m3 ), are widely scattered and short-lived (Curewitz and Karson, 1998).

8. DISCUSSION: SUBMARINE HYDROTHERMAL MODEL FOR GENERATION OF THE WUTAI VMS Although the Wutai VMS is reworked by late thrusting and metamorphism associated with the regional orogeny (2.50 and 1.90 Ga), it still can be established through structural analysis that the Wutai VMS was originally underlain by a thick sequence of mafic lava (Fig. 4). Laterally, deposits commonly grade into banded iron formations, chert, carbonate, exhalites, graphite schist, and turbidite. The turbidite beds may be derived from the eroding volcanic arc on its western side. Vertically, the Wutai VMS deposits are interlayered with thin volcanic tuff layers, basalt, and felsic volcanics. Sulfide stringers are mainly found within mafic to ultramafic rocks. Extensive CO2 metasomatism (carbonate), and K-metasomatism (sericite-fuchsite), are recorded by their host rocks. Extensive hydrothermal alteration is intimately related to ore-forming processes during the deposition of the volcanic pile on the sea floor. There are some interlayers of volcanic breccia associated with the Wutai VMS, suggesting that they were generated adjacent to depressions or grabens on the sea floor. Contemporaneous deep fractures cut cross the volcanic piles, now they are marked by syndeformational chloritization (quartz-chlorite, quartz-chlorite-carbonate), pyritization and silicification in the footwall of the hydrothermal system. These paleofractures not only acted as pathways for mineralizing fluid, but also delineated seafloor depressions where sulfide locally precipitated and accumulated (e.g., Calvez and Lescuyer, 1991). The venting of hydrothermal fluids in Wutai belt appear to be closely associated with growth faulting within the ophiolitic crust and overlying chert-BIF sequence. In many ophiolites and their modern counterparts, VMS deposits were mainly deposited in depressions or grabens formed at the spreading axis (Lippard et al., 1986; Scott, 2002). The Wutai VMS was mainly precipitated within marine sediments of BIF or chert on oceanic floor (Tian et al., 1996; Bai et al., 1992). There is a close relationship between pyritic BIF and porous massive sulfides in the Wutai VMS, indicating that both a brinepool model and a black smoker model allow for its precipitation of sulfides over a large area. The metalliferous saline brines are believed to discharge from scattered epigenetic vents. They are mainly sediment-hosted sulfide. The mineral assemblage and host rocks are comparable to hydrothermal mineralization associated with modern back-arc basins, such as the Lau basin (Rona and Scott, 1993). Here it is suggested that the formation of BIF is derived from an oxide-rich facies of a saline brine pool. Although the VMS suffered the strong modifications of deformation and metamorphism, preliminary studies reveal the presence of black smoker chimneys preserved in the Wutai mélange belt. The porous massive sulfides and relevant chimney fragments provide direct evidence for black smokers in the formation of the Wutai VMS. The apparent lack of

358

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

feeder zones may not be simply an artifact of deformation and metamorphism. The sheetlike form of deposits suggest its formation by exhalative or synsedimentary processes. The mineral assemblage, structures, and textures of the Wutai VMS are comparable to those of sulfides associated with modern marine black smokers, especially those of the southwest Pacific ocean (Wu, 2000; Fouquet et al., 1993). The available data reveal that there is a dismembered submarine hydrothermal vent-associated massive sulfide deposit in the North China that formed at about 2.50 Ga. So, a black smoker can be inferred for their genesis. Furthermore, the Wutai VMS is quite similar to Besshi-type, it is inferred to be generated in the setting of forearc, later tectonically transported in mélange belts during continental collision (Wang et al., 1996). Therefore, it is suggested that seafloor black smoker activity plays an important role for generation and accumulation of Wutai VMS. They are derived from venting hydrothermal fluids in areas of black smoker chimneys. On the other hand, the occurrence of abundant BIF and pyritic iron formation represents an oxide-rich facies of a metalliferous saline brine pool, discharged from scattered epigenetic vents (Slack, 1993). Modern examples of this type include the Red Sea, and the Guaymas Ridge in the Gulf of California. Metalliferous sediments are also very common in the vicinity of many ancient VMS (Scott, 2002). Since the discovery of the first black smoker forming at hydrothermal vents on the East Pacific Rise in 1978, a wide variety of sulfide deposits and related hydrothermal activities have been identified on the seafloor of different oceans (Fouquet, 1999; Scott, 2002). They are mainly generated within the setting of mid-ocean rides, back-arc basins (Curewitz and Karson, 1998), although some also form in island arcs and oceanic fracture zones. It is suggested here that the black smoker model can explain many geological features of the Neoarchean Wutai VMS, which will provide new information for the hydrothermal evolution of Archean oceanic basins.

ACKNOWLEDGEMENTS This work was supported by a National Science Foundation of China (Project Nos. 40242014, 49832030), Peking University (Project 985 to J.H. Li), and the U.S. National Science Foundation (Awards 02-07886 and 01-25925 to T. Kusky). We thank Zhang Zhiqiang and Guo Muzi for assistance in the field works.

REFERENCES Alexander, R.J., Harper, G.D., 1992. The Josephine ophiolite: an ancient analogue for slow to intermediate-spreading oceanic ridges. In: Parson, L.M., Murton, B.J., Browning, P. (Eds.), Ophiolites and Their Modern Oceanic Analogues. Geological Society Special Publication 60, 3–38. Alt, J.C., Teagle, A.H., 2000. Hydrothermal alteration and fluid fluxes in ophiolites and oceanic crust. 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: Boulder, Colorado. Geological Society of America Special Paper 349, 273–282.

References

359

Bai, J., Wang, R.Z., Guo, J.J., 1992. The Major Geological Events of Early Precambrian and their Age Dating in Wutai Region. Geological Publishing House, Beijing, pp. 1–53. Bai, W.J., Zhou, M.F., Hu, X.F., Cai, Y.C., Zheng, X.H., 1993. The Tectonic Evolution of North China Block Lithosphere, Mafic to Ultramafic Complex, and Mineralization. Seismological Press, Beijing, pp. 93–154. Banerjee, N.R., Gillis, K.M., Muehlenbachs, K., 2000. Discovery of epidosites in a modern oceanic setting, the Tonga forearc. Geology 28, 151–154. Boyle, J.F., Robertson, A.H.F., Gass, I.G., Lippard, S.J., Shelton, A.W. (Eds.), 1984. Evolving metallogenesis at the Troodos spreading axis. In: Ophiolites and Oceanic Lithosphere. Oxford Geological Society Special Publications, vol. 13. Blackwell Scientific Publications, pp. 169–181. Brown, D., McClay, K.R., 1998. Data report: sulfide textures in the active TAG massive sulfide deposit, 260N, mid-Atlantic ridge. In: Herzig, P.M., Humphris, S.E., Miller, D.J., Zierenberg, R.A. (Eds.), Proceedings of the Ocean Drilling Program. Scientific Results 158, 193–200. Butler, I.B., Nesbitt, R.W., 1999. Trace element distributions in the chalcopyrite wall of a black smoker chimney: insights from laser ablation inductively coupled plasma mass spectromely (LAICP-MS). Earth and Planetary Science Letters 167, 335–345. Calvez, J.Y., Lescuyer, J.L., 1991. Lead isotope geochemistry of various sulfide deposits from the Oman Mountains. In: Peters, T.J., et al. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Dordrecht, pp. 385–397. Curewitz, D., Karson, J.A., 1998. Geological consequences of dike intrusion at Mid-ocean ridge spreading centers. In: Buck, W.R., Delaney, P.T., Karson, J.A., Lagabrielle, Y. (Eds.), Faulting and Magmatism at Mid-Ocean Ridges. Geophysical Monograph 106, 117–136. Fouquet, Y., 1999. Where are the large hydrothermal sulfide deposits in the oceans? In: Cann, J.R., Elderfield, H., Laughton, A. (Eds.), Mid-Ocean Ridges: Dynamics of Processes Associated with Creation of New Ocean Crust. Cambridge Univ. Press, Cambridge, pp. 211–224. Fouquet, Y., Stackelberg, U., Charlou, J.L., Donval, J.P., Foucher, J.P., Erzinger, J., Herzig, P.M., Muhe, R., Wiedicke, M., Soakai, S., Whitechurch, H., 1991. Hydrothermal Activity in the Lau Back-arc Basin: Sulfides and Water Chemistry. Geology 19, 303–306. Fouquet, Y., Stackelberg, U., Charlou, J.L., Erzinger, J., Herzig, P.M., Muhe, R., Wiedicke, M., et al., 1993. Metallogenesis in back-arc environments: the Lau basin example. Economic Geology 88, 2154–2181. Gillis, K.M., Banerjee, N.R., 2000. Hydrothermal alteration patterns in supra-subduction zone ophiolites. 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: Boulder, Colorado. Geological Society of America Special Paper 349, 283–297. Halbach, P., Blum, N., Munch, U., Pluger, W., Garbe-Schonberg, D., Zimmer, M., 1998. Formation and decay of a modern massive sulfide deposit in the India Ocean. Mineralium Deposita 33, 302– 309. Hannington, M.D., Scott, S., 1988. Mineralogy and geochemistry of a hydrothermal silica-sulfidesulfate spire in the caldera of axial seamount, Juna de Fuca ridge. Canadian Mineralogist 26, 603–625. Hannington, M.D., Galley, A.G., Herzig, P.M., Petersen, S., 1998. Comparison of the TAG mound and stockwork complex with Cyprus-type massive sulfide deposits. In: Herzig, P.M., Humphris, S.E., Miller, D.J., Zierengberg, R.A. (Eds.), Proceedings of The Ocean Drilling Program. Scientific Results 158, 389–415. Harper, G.D., 1999. Structural styles of hydrothermal discharge in ophiolite/sea floor systems. Reviews in Economic Geology 8, 53–73.

360

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Haymon, R.M., 1983. Growth history of hydrothermal black smoker chimneys. Nature 301, 695–698. Haymon, R.M., 1996. The response of ridge-crest hydrothermal systems to segmented, episodic magma supply. In: Macleod, C.J., Tyler, P.A., Walker, C.L. (Eds.), Tectonic, Magmatic, Hydrothermal and Biological Segmentation of Mid-Ocean Ridges. Geological Society Special Publication 118, 157–168. Haymon, R.M., Koski, R.A., Abrams, M.J., 1996. Hydrothermal discharge zones beneath massive sulfide deposits mapped in the Oman ophiolite. Geology 17, 531–535. Haymon, R.M., Koski, R.A., Sinclair, C., 1984. Fossils of Hydrothermal vent worms from Cretaceous sulfide ores of the Samail Ophiolite, Oman. Science 223, 1407–1409. Herrington, R.J., Evans, D.M., Buchanan, D.L., 1997. Metallogenic aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35. Clarendon Press, Oxford, pp. 176–219. Herrington, R.J., Maslennikov, V.V., Spiro, B., Zaykov, V.V., Little, C.T.S., 1998. Ancient vent chimney structures in the Silurian massive sulfides of the Urals. In: Mills, R.A., Harrision, K. (Eds.), Modern Ocean Floor Processes and the Geological Record. Geological Society of London Special Publication 148, 241–257. Herzig, P.M., Hannington, M.D., 1995. Polymetallic massive sulfides at the modern seafloor: a review. Ore Geology Review 10, 95–115. Herzig, P.M., Hannington, M.D., Fouquet, Y., Von Stackelberg, U., Petersen, S., 1993. Gold-rich polymetallic sulfides from the Lau back-arc and implications for the geochemistry of gold in sea-floor hydrothermal systems of the southwest Pacific. Economic Geology 88, 2182–2209. Hutchinson, R.W., 1980. Massive base metal sulfide deposits as Guides to tectonic evolution. In: Srangway, D.W. (Ed.), The Continental Crust and Its Mineral Deposits. Geological Association of Canada Special Paper 20, 659–684. Knott, R., Fouquet, Y., Honnorez, J., Petersen, S., Bohn, M., 1998. Petrology of Hydrothermal mineralization: a vertical section through the TAG mound. In: Herzig, P.M., Humphris, S.E., Miller, D.J., Zierenberg, R.A. (Eds.), Proceedings of the Ocean Drilling Program. Scientific Results 158, 5–26. Kusky, T.M., Li, J.H., 2003. Paleoproterozoic tectonic evolution of the North China Craton. Journal of Asian Earth Sciences, in press. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Kusky, T.M., Polat, A., 1999. Growth of granite-greenstone terranes at convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73. Li, J.L., Wang, K.Y., Wang, Q.C., Liu, X.H., Zhao, Z.Y., 1990. Early Proterozoic orogenic belt in Wutaishan area, China. Scientia Geologica Sinica 1, 1–11. Li, J.H., Kusky, T.M., Huang, X., 2002. Neoarchean podiform chromitites and harzburgite tectonite in ophiolitic mélange, North China Craton. Remnants of Archean oceanic mantle. GSA Today 12 (7), 4–11. Lippard, S.J., Shelton, A.W., Gass, I.G., 1986. The Geological Society Memoir, No. 11. Blackwell Scientific Publications, Beccles/London, pp. 127–128. Little, C.T.S., Cann, J.R., Herrington, R.J., Morisseau, M., 1999a. Late Cretaceous hydrothermal vent communities from the Troodos ophiolite, Cyprus. Geology 27, 1027–1030. Misca, K.C., 2000. Understanding Mineral Deposits. Kluwer Academic, Dordrecht, pp. 450–464.

References

361

Nicolas, A., 1989. Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Dordrecht, pp. 281–285. Nicolas, A., 1995. The Mid-Oceanic Ridges. Springer-Verlag, Berlin/Heidelberg, pp. 1–80. Nisbet, E.G., Fowler, M.R., 1996. The hydrothermal imprint on life: did heat-shock proteins, metalloproteins and photosynthesis begin around hydrothermal vents? In: Macleod, C.J., Tyler, P.A., Walker, C.L. (Eds.), Tectonic, Magmatic, Hydrothermal and Biological Segmentation of MidOcean Ridges. Geological Society Special Publication 118, 239–251. Polat, A., Kerrich, R., 1999. Formation of an Archean tectonic mélange in the Schreiber-Hemlo greenstone belt, Superior province, Canada: implications for Archean subduction-accretion process. Tectonics, 733–755. Polat, A., Kerrich, R., Wyman, D.A., 1998. The late Archean Schreiber-Hemlo and White RiverDayohessarah greenstone belts, Superior Province: collages of oceanic plateaus, oceanic arcs, and subduction-accretion complexes. Tectonophysics 263, 249–276. Rasmussen, B., 2000. Filamentous microfossils in a 3235 Ma old volcanogenic massive sulfide deposit. Nature 405, 676–679. Rona, P.A., 2002. Marine minerals for the 21st century. Episodes 25 (1), 2–12. Rona, P.A., Scott, S.D., 1993. A special issue on sea-floor hydrothermal mineralization: new perspectives, preface. Economic Geology 88, 1935–1975. Sawkins, F.J., 1990. Integrated tectonic-genetic model for volcanic-hosted massive sulfide deposits. Geology 18, 1061–1064. Schandal, E.S., Davis, D.W., Krogh, T.E., 1990. Are the alteration halos of massive sulfide deposits syngenetic? Evidence from U-Pb dating of hydrothermal rutile at the Kidd volcanic center, Abitibi subporvince, Canada. Geology 18, 505–508. Scott, S., 2002. Minerals on land, Minerals in sea. Geotimes 47, 1–8. Slack, J.F., 1993. Descriptive and grade-tonnage models for Besshi-type sulfide deposits. In: Kirkham, R.V., Sinclair, W.D., Thorpe, R.I., Duke, J.M. (Eds.), Mineral Deposit Modeling. Geological Association of Canada Special Paper 40, 343–371. Thurston, P.C., Williams, Sutchliffe, R.H., Stott, G.M., 1992. Geology of Ontario. Ontario Geological Survey. Ministry of Northern Development and Mines Special 4 (2), 1093–1103. Tian, Y.Q., Ma, Z.H., Yu, K.R., Liu, Z.H., Peng, Q.M., 1996. The early Pecambrian geology of WutaiHengshan Mts., Shanxi, China. Field Trip Guide T315. In: 30th Int. Geol. Congress, Beijing, China. Geol. Publ. House, Beijing, p. 52. Tiercelin, J., Pflumio, C., Castrec, M., Boulegue, J., Gentle, P., Rolet, J., Coussement, C., Stetter, K.O., Huber, R., Buku, R., Mifundu, W., 1993. Hydrothermal vents in lake Tanganyika, East African rift system. Geology 21, 499–502. Tivey, M.K., Singh, S., 1997. Nondestructive imaging of fragile sea-floor vent deposit samples. Geology 25 (10), 931–934. Tunnicliffe, V., Fowler, C.M.R., Mcarthur, A.G., 1996. Plate tectonic history and hot vent biogeography. In: Macleod, C.J., Tyler, P.A., Walker, C.L. (Eds.), Tectonic, Magmatic, Hydrothermal and Biological Segmentation of Mid-Ocean Ridges. Geological Society Special Publication 118, 225–238. Valsami, E., Cann, J.R., 1992. Mobility of rare earth elements in zones of intense hydrothermal alteration in the Pindos ophiolite, Greece. In: Parson, L.M., Murton, B.J., Browning, P. (Eds.), Ophiolites and Their Modern Oceanic Analogues. Geological Society Special Publication 60, 219–232.

362

Chapter 11: Neoarchean Massive Sulfide of Wutai Mountain, North China

Vearncombe, S., Barley, M.E., Groves, D.I., Mcnaughton, N.J., Mikucki, E.J., Vearncombe, J.R., 1995. 3.26 Ga black smoker-type mineralization in the Strelley belt, Pilbara craton, Western Australia. Journal of the Geological Society of London 152, 587–590. Wang, K.Y., Li, J.L., Hao, J., Li, J.H., Zhou, S.P., 1996. The Wutaishan orogenic belt within the Shanxi province, Northern China: a record of late Archean collision tectonics. Precambrian Research, 95–103. Wang, K., Li, J., Hao, J., et al., 1997. Late Archean mafic-ultramafic rocks from the Wutaishan, Shanxi Province: a possible ophiolite mélange. Acta Petrologica Sinica 13 (2), 139–151 (in Chinese with English abstract). Wilde, S., Cawood, P., Wang, K.Y., Nemchin, A., 1998. SHRIMP U-Pb zircon dating of granites and gneisses in the Taihangshan-Wutaishan area: Implications for the timing of crustal growth in the North China Craton. Chinese Science Bulletin 43, 143–144. Wu, S.Y., 2000. The Hydrothermal Sulphide Resource at Sea Floor of the World. Oceanic Press, Beijing, pp. 1–290. Yuan, G.P., 1988. The geology of Wutai greenstone belt. Shanxi Geology 3, 357–366. Zaykov, V.V., Maslennikov, V.V., Zaykov, E.V., Herrington, R.J., 1996. Hydrothermal activity and segmentation in the Magnitogorsk-West Mugodjarian zone on the margins of the Urals palaeoocean. In: Macleod, C.J., Tyler, P.A., Walker, C.L. (Eds.), Tectonic, Magmatic, Hydrothermal and Biological Segmentation of Mid-Ocean Ridges. Geological Society Special Publication 118, 199–210.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

363

Chapter 12

INFERRED OPHIOLITES IN THE ARCHEAN SLAVE CRATON P.L. CORCORANa , W.U. MUELLERb AND T.M. KUSKYc a Department

of Geology, University of Western Ontario, Canada Geology, University of Quebec at Chicoutimi, Canada c Department of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63103, USA b Department of

Ophiolites have always been difficult to recognize in the Archean rock record because of fragmental preservation and intense tectonism. This chapter reviews three distinct areas in the Slave craton and assesses their potential of containing ophiolite sequences. These include the (1) Yellowknife, (2) Point Lake, and (3) Beaulieu and Cameron River volcanic belts, the best modern analogy of which may be Tethyan-type ophiolites resting structurally over continental crust. The 6 km-thick Chan Formation of the Yellowknife volcanic belt resembles modern ophiolites with tholeiitic massive to pillowed flows, abundant gabbro dykes/sills, interflow sedimentary rocks, and a mafic sheeted dyke swarm. Notwithstanding, the base of this crustal-floored sequence is sheared or stitched by late-tectonic plutons and the dunite-peridotite-gabbro segment is lacking. The inferred base of the Point Lake volcanic belt is composed of mafic mylonite with low-strain domains of gabbro, pyroxenite, dunite, and peridotite. The mafic mylonite is overlain by gabbro, layered gabbro, minor mafic dikes, pillowed flows, massive flows, hyaloclastite, and local chert. A welldefined sheeted dyke swarm is absent although mafic dykes are locally preserved in the Augustus granite. The Beaulieu and Cameron River volcanic belts are spatially associated with mafic dyke swarms that intrude the Sleepy Dragon basement complex. Mafic massive and pillowed flows and sub-volcanic sills are predominant above the sheared basement contact. In the strict sense, these belts or belt segments do not fit an ophiolite definition, but meet the general requirements. Ophiolites can be found in numerous geodynamic settings and complete preservation from tectonized mantle to surficial ocean floor products is highly unlikely, especially for Archean rocks. Therefore, the nature of the basement contacts is particularly significant. If the contacts are tectonic, then parts of the ophiolitic sequences may have been sheared off, which is commonly the case for the maficultramafic intrusive component. However, if the contacts between basement and massive and pillowed flows, dykes and sills is conformable, an ophiolite interpretation is not recommended. DOI: 10.1016/S0166-2635(04)13012-0

364

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

1. INTRODUCTION The possibility that some mafic greenstone belts in the Slave Craton may be Archean ophiolites or other ancient analogs of modern tectonic environments has been a contentious issue (Kusky, 1991a; Bickle et al., 1994; Sylvester et al., 1997). Recent models have compared ophiolites with supra-subduction zone settings, including arcs and backarcs, or rift-related settings, such as mid-ocean ridges. Identifying ophiolite complexes using only geochemistry is problematic, and thus, elucidating the regional geological context, structure, and stratigraphy is crucial (e.g., Anonymous, 1972; Pearce et al., 1984; Pearce and Peate, 1995; Searle and Cox, 1999; Metcalf et al., 2000; Moores, 2002). A succinct overview of Slave greenstone belts and their features is necessary in order to provide a comprehensive case either for or against ophiolites in the Archean rock record. As pointed out by Moores et al. (2000) the “ophiolite conundrum” marks a discrepancy between structural and stratigraphic setting, and geochemical characteristics. These authors argue that the mantle is heterogeneous at all scales and geodynamic locations, and that a distinct geochemical signature for ophiolites is lacking. This has ramifications, especially for Archean terranes, in which volcano-sedimentary sequences are generally incomplete, structures are complex, and Fe-tholeiites and komatiites are abundant (rare to non-existent in modern ophiolites). Must the rocks comprise a complete stratigraphic sequence (i.e., tectonite, dunite-peridotite, gabbro, sheeted dikes, pillows, pelagic sedimentary rocks) to be qualified as an ophiolite? What portion of the succession is necessary in order to be called an ophiolite? Is there a distinction between Phanerozoic and Archean ophiolites? The answers to these questions are dependent on the main topics being considered: (1) physical volcanology, (2) stratigraphy and depositional setting, (3) structural stacking/repetition, and (4) geodynamic context. Ophiolites are generally considered a distinct suite of obducted ocean floor rocks associated with all four topics, with a highly varied geochemical affinity depending on tectonic setting (Sylvester et al., 1997; Dilek et al., 2000). Using the term “ophiolite” for any mafic-dominated sequence containing massive and pillowed basalt flows of tholeiitic affinity, dikes and local pelagic or volcaniclastic deposits is tenuous. Descriptive terms such as volcano-sedimentary or mafic volcanic sequence are better used when the origin of a succession is undetermined. If interpreted correctly, ophiolites are significant in understanding Archean evolutionary systems. The Slave Province, located in the Northwest Territories of Canada, is a 500 × 700 km large Archean Craton (Fig. 1). It is composed of gneisso-plutonic complexes, abundant late tectonic granitoid rocks, extensive sedimentary successions, and numerous linear, Ntrending greenstone belts, together straddling 1.5 b.y. of Early Earth’s history from ca. 2.58–4.03 Ga (Mortensen et al., 1992; van Breemen et al., 1992; Isachsen and Bowring, 1994, 1997; Bowring and Williams, 1998; Davis and Bleeker, 1999). Volcanic belts comprising the Slave Province include mafic to intermediate rocks predominant in the west, and felsic-dominated sequences prevalent in the east (Henderson, 1981; Padgham, 1985; Kusky, 1989; Padgham and Fyson, 1992). The line of demarcation between these two segments is the north-trending, pan-Slave Beniah Lake fault zone (Kusky, 1990; Mueller, 1997), a crustal-scale structure along which late-stage orogenic basins formed

1. Introduction

365

Fig. 1. Lithological map of the Slave Province (SP) in the Northwest Territories (NT), Canada. Locations of the Point Lake (PLVB), Beaulieu River (BRVB), Cameron River (CRVB), and Yellowknife (YVB) volcanic belts are indicated. Note the locations of the ancestral Jackson Lake and Beniah Lake faults. Modified from Mueller and Corcoran (2001).

366

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

(Corcoran et al., 1998, 1999). The Beniah Lake fault zone roughly separates juvenile Archean crust (Hackett River terrane; Kusky, 1989) in the east from volcano-sedimentary successions underlain by continental crust in the west of the Slave Craton the Anton complex (Kusky, 1989). The main period of volcanic belt formation appears to have occurred between ca. 2720 and 2650 Ma with a concentration at ca. 2700–2670 Ma, although 3.1– 3.2 Ga volcanic belts with minor komatiitic basalts have been locally identified (e.g., Winter Lake belt; Hrabi et al., 1995). The striking characteristic of the Slave Province, which differentiates it from many Archean cratons, is the association of sialic basement with craton-cover sequences. Numerous distinct regional- and pan-Slave-scale unconformities are located on topographic basement highs between older gneisso-plutonic suites that include the Sleepy Dragon Complex, Augustus Granite, Beniah Lake Complex, and/or volcanic and sedimentary sequences (Lambert, 1988; Kusky, 1990; Mueller et al., 1998; Corcoran et al., 1998; Mueller and Corcoran, 2001). These unconformable relationships, as well as important structural contacts (Kusky, 1990, 1991b) have major implications for the interpretation of Archean ophiolites.

2. SLAVE CRATON CHARACTERISTICS AND OPHIOLITES IN GREENSTONE BELTS Specific volcanic belts containing possible ancient ophiolite remnants were proposed by Helmstaedt et al. (1986) and Kusky (1989, 1990, 1991a), and include the 2722 ± 1 to 2658 ± 2 Ma (Mortensen et al., 1992; Isachsen and Bowring, 1994, 1997) Yellowknife belt, the 2690 ± 3 to ca. 2660 Ma (Mueller et al., 1998; Northrup et al., 1999) Point Lake belt and the 2686.8 ± 3.2 to 2663 + 7/−5 Ma (Henderson et al., 1987; Mueller et al., 2001) Cameron and Beaulieu River Belts (Figs. 1 and 2), all of which structurally overlie a central basement high (Mueller and Corcoran, 2001) or Central Slave Basement Complex (Bleeker et al., 1999). Debate has centered on whether these structural contacts are “modified unconformities”, original thrusts, or both. U-Pb ages of 3.22 Ga for the Augustus Granite at Point Lake (Northrup et al., 1999), 2.95 Ga for the Beniah Complex at Beniah Lake (Isachsen and Bowring, 1997), 2.8–2.95 Ga for the Sleepy Dragon Complex at Cameron and Beaulieu Rivers (Henderson et al., 1987; Lambert and van Breemen, 1991; Bleeker et al., 1999) and 2.93–3.41 Ga for the Anton Complex near Yellowknife (Isachsen and Bowring, 1997) indicate the presence of an older gneissic basement in all study areas. Of significance is an intervening > 2.8 Ga quartz arenite sequence overlying the > 2.8 Ga basement unconformably, but underlying the 2.72–2.64 Ga greenstone belts (Helmstaedt and Padgham, 1986; Kusky, 1991a; Covello et al., 1988; Stubley, 1989; Jackson, 1996; Pickett and Mueller, 2000). These quartz arenites, representing a pan-Slave time line, support the notion of a consolidated older crust in the western Slave by 3.0 Ga (Kusky, 1989; Mueller et al., 1998). Moores (1982, 2002) suggested a division into Tethyan- and Cordilleran-type ophiolites based on a general geodynamic context. Tethyan-type ophiolites overlie continental crust

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts 367

Fig. 2. Correlative stratigraphy of the Yellowknife, Cameron River, Beaulieu River, and Point Lake volcanic belts in the Slave Province. References for ages: (1) Isachsen et al. (1991), Isachsen and Bowring (1994, 1997), (2) Henderson et al. (1987), (3) unpublished data from W. Mueller (unpublished), (4) Northrup et al. (1999), (5) Mueller et al. (1998). BF, Burwash Formation; SD, Sleepy Dragon Complex; CF, Contwoyto Formation; AG, Augustus Granite. Modified from Mueller and Corcoran (2001). Most contacts between volcanic rocks and basement are faulted, and the original relationships uncertain. See text for explanation.

368

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

and continental margin or platform sedimentary rocks, whereas Cordilleran-type ophiolites are not associated with continental crust nor platformal sedimentary deposits, but tectonic mélanges with high-grade metamorphic rocks are common. Partially preserved Slave Craton analogues are more akin to Tethyan-types, but only if their origin is allochthonous rather than autochthonous. An autochthonous ophiolite sequence overlying basement is possible if crustal ultramafic cumulates, but not the tectonized mantle harzburgite segment are preserved and are part of the ophiolite sequence. Problems of interpretation arise in areas where pillowed and massive flows overlie basement with a depositional contact that has been subsequently transformed into a zone of high strain. These could represent ophiolites, the transition from stretched continental crust to oceanic crust, volcanic rifted margins, or subaqueous continental rift environments. 2.1. Yellowknife Volcanic Belt The Yellowknife volcanic belt, located adjacent to the N-trending ancestral Jackson Lake fault, is divided into the mafic flow-dominated Kam Group, the felsic volcaniclasticdominated Banting Group and the clastic-dominated Duncan Lake Group (Figs. 2 and 3; Helmstaedt and Padgham, 1986). Although the inferred 10–12-km-thick Kam Group contains four distinct formations which include the: (1) Chan, (2) Crestaurum, (3) Townsite, and (4) Yellowknife Bay, the lowermost Chan Formation is the only volcanic sequence inferred to contain remnants of an ophiolite complex (Helmstaedt and Padgham, 1986; Helmstaedt et al., 1986; MacLachlan and Helmstaedt, 1995). The Chan Formation is approximately 6 km thick and consists mainly of tholeiitic mafic flows and intrusions (Henderson and Brown, 1966; Helmstaedt and Padgham, 1986; MacLachlan and Helmstaedt, 1995), with minor ultramafic rocks (Considine et al., 1987). It is characterized by a gabbro-dominated base, overlain by a multiple dike complex and capped by massive and pillowed flows with minor interflow sedimentary rocks (Fig. 4; Helmstaedt et al., 1986). The contacts between dikes, irregular shaped intrusions, and flows are either sharp or gradational, with dikes locally budding into pillows (Helmstaedt et al., 1986). The dikes, which are perpendicular to the general strikes of flows, display a marked sheeted nature, resembling typical ophiolite sheeted dike complexes. The irregular shaped intrusions are characterized by numerous chilled margins, consistent with multiple intrusion (MacLachlan and Helmstaedt, 1995). Although the lower tectonite and peridotite layers are absent in the Chan Formation, the up-section change from gabbro to sheeted dikes to massive and pillowed flows resembles the stratigraphy of more modern ophiolite sequences. Geochemically, the Kam Group consists of tholeiitic basalts to andesites that produce flat to slightly LREE enriched patterns (Jenner et al., 1981), and have been interpreted as ocean floor basalts (Cunningham and Lambert, 1989). More specifically, major and trace elements in the Chan Formation are similar to MORB with flat REE patterns (MacLachlan and Helmstaedt, 1995), albeit Cousens (2000) reported a large number of samples with negative Nb anomalies, and Th and LREE enrichments typical of E-MORB. Based on the geochemical data and the ophiolite-like stratigraphy, the Chan Formation has been

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

369

Fig. 3. Lithological map of the Yellowknife volcanic belt displaying the eastern fault-bound margin and the western unconformity between the Jackson Lake Formation and the underlying Kam Group. Modified from Corcoran and Mueller (2002).

370

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 4. Schematic stratigraphy of the Kam Group, Yellowknife volcanic belt, illustrating multiple and sheeted dikes of the Chan Formation grading upsection into predominantly pillowed flows. Overlying units include variolitic and porphyritic flows and felsic tuffs and cherts that are considered marker horizons. Modified from Helmstaedt et al. (1986). Contact between Chan Formation and older granite is sheared.

interpreted to represent the early rifting stages of a marginal or back-arc basin (Isachsen, 1992; MacLachlan and Helmstaedt, 1995; Cousens, 2000). Although a back-arc basin interpretation is feasible for the Chan Formation, one major discrepancy precludes an ophiolitic origin according to the classic Penrose definition, which is the presence of sialic crust currently underlying the Yellowknife belt. If the contact between 2.93 and 3.41 Ga Anton Complex and the overlying greenstone belt is tectonic as suggested by Kusky (1987), a Tethyan-type ophiolite complex for the Chan Formation is tenable. Although contact relationships remain enigmatic though highly sheared, the

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

371

presence of > 2800 Ma zircons in cherty tuffs of the overlying Crestaurum Formation have been inferred to support a proximal granitoid basement (Isachsen, 1992) during evolution of the volcanic belt. εNd values from the Crestaurum, Townsite and Yellowknife Bay Formations range from −3 to +3.4, thus corroborating the involvement of sialic crust during evolution of the Kam Group (Cousens, 2000). These results support an autochthonous origin, which is inconsistent with the defined ophiolite complexes as described by Moores (1982) that are tectonically emplaced on continental crust. However, if an ophiolite is strictly regarded as a suite of rocks irrespective of the nature of the basal contact, as advocated by Sylvester et al. (1997) and Anonymous (1972), the Chan Formation remains a strong candidate for a remnant Archean ophiolite. In addition, the detailed nature and significance of the shearing at the base of the Kam Group remains to be documented. 2.2. Point Lake Volcanic Belt The Point Lake volcanic belt is located approximately 300 km north of Yellowknife, along the north-trending, crustal-scale Beniah Lake fault (Figs. 1 and 5), and is composed of three major formations: (1) the mafic-dominated Peltier Formation, (2) the dacitic Samandré Formation, and (3) the rhyolitic Beauparlant Formation (Figs. 6 and 7). West of Cyclops Peninsula (Fig. 5) the volcanic belt is represented by a mafic mylonite with low-strain domains of gabbro, pyroxenite, dunite, and peridotite (Kusky, 1991a). The mafic mylonite is overlain by generally less-deformed gabbro, layered gabbro, mafic dikes, pillow lavas, massive flows, hyaloclastite, and chemical sediments, and truncated by an unconformity with overlying conglomerates containing clasts from the greenstone belt, as well as the nearby basement complex. Kusky (1991a) suggested that this sequence represents an ophiolite complex, and correlated the mafic rocks west of Cyclops Peninsula with those to the east, which contain only the upper parts of the proposed ophiolite. Using this correlation, Kusky (1991a) constructed a series of restored structural cross sections, and suggested that the Point Lake greenstone belt was thrust at least 69 km over the quartzofeldspathic gneisses of the Anton terrane to the west. This interpretation is based on stratigraphic features and structural characteristics inherent to a ca. 6 km wide region situated at the boundary between the older Anton and juvenile Contwoyto terranes of Kusky (1989) (Fig. 8). The lithotectonic boundary roughly coincides with the trace of the Beniah Lake fault zone. Important for the present discussion is the observation in the final reconstruction of Kusky (1991a) that the greenstone belt is adjacent to the quartzofeldspathic gneisses of the Anton terrane. East of the mafic mylonite, the most extensive homoclinal sequence in the Peltier Formation is approximately 1.5 km thick (Corcoran, 2000). The basalt-dominated sequence is characterized by pillowed and massive flows, pillow breccia, mafic dikes and sills, and minor hyaloclastite and pelagic shale deposits. Andesitic-dacitic volcaniclastic rocks are locally interstratified with massive flows. Massive parts of basalt flows laterally and vertically become pillowed, are overlain by pillow breccia, and are capped by hyaloclastite and turbitiditic tuffs. Pillowed and massive flows are in places cut by synvolcanic mafic dikes and sills. Dikes, 2–15 m thick, are generally perpendicular to pillow strike. Numer-

372 Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 5. Map of the Point Lake fold-thrust belt, showing relationship of the Point Lake volcanic belt to the gneiss terrane to the west, and the graywacke-turbidite terrane to the east. Modified after Kusky (1991a).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

373

Fig. 6. Lithological map of the Point Lake area illustrating the volcanic facies that comprise the mafic-dominated Peltier Formation. Locations of Fig. 9 and Cyclops Peninsula are indicated. Modified from Corcoran (2000).

374

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 7. Schematic stratigraphy of the Point Lake volcanic belt. The mafic-dominated Peltier Formation contains minor calc-alkaline basaltic and andesitic flows and interflow sediments forming peperite. Locally, mafic dikes and sills are abundant and are associated with pillowed flows. The Peltier, Samandre, and Beauparlant formations are interstratified with the turbiditic Contwoyto Formation, which overlies the Augustus Granite unconformably. Age dates from: (1) Isachsen et al. (1991), Isachsen and Bowring (1994, 1997), (2) Mueller et al. (2001), (3) Northrup et al. (1999). Modified from Corcoran and Dostal (2001). Basal contact of volcanic belt is sheared.

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

375

Fig. 8. Early tectonic division of the Slave Province into an older mature continental fragment in the west (Anton terrane) and a juvenile arc-type terrane in the east (Hackett River terrane; modified from Kusky, 1989).

376

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 9. Characteristics of mafic intrusions in the Peltier Formation, Point Lake belt. (A) Three parallel mafic dikes and an east-west-trending sill cut pillowed flows. Geochemically, the sill, dikes and flows are comparable. (B) Nine chilled margin contacts in one dike are indicative of multiple intrusion during evolution of the volcanic belt. Modified from Corcoran (2000).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

377

ous mm-thick chilled margins, 10–75 cm apart, mark the presence of smaller intrusions within the central parts of most dikes at one locality (Fig. 9). Intrusions of this type are considered synvolcanic based on geochemical similarities and the presence of multiple intrusions, indicating continuous pulses of magma supply (Corcoran, 2000). The facies types and thickness, and lateral and vertical facies changes are characteristic of the subaqueous portions of shield volcanoes or seamounts (Corcoran, 2000). Multiple feeder dikes cutting numerous flow units associated with sills are typical of the central part of a volcanic edifice where construction is initiated (e.g., Easton, 1984). However, these multiple feeder dikes are of limited extent and do not comprise a classical mafic dike swarm. Geochemical data from Corcoran and Dostal (2001) indicate that the rocks of the Peltier Formation include N-MORB with flat to slightly LREE-enriched patterns, tholeiitic LREE-enriched E-MORB with Nb and Ti depletions, and strongly LREE-enriched calc-alkaline compositions with marked Nb and Ti depletions. The geochemistry, volcanology and stratigraphy are consistent with a back-arc basin setting, which is a favorable site for the development of an ophiolite sequence. However, the physical volcanology and the possibility of a tectonically modified basal unconformable contact with the Augustus Granite (supported by the unconformity between the Augustus Granite and Contwoyto Formation) have been used to argue against this succession representing an ophiolite sequence. An autochthonous origin for the greenstone belt is supported by a range of εNd values from −6.3 to +3.06 in the volcanic rocks (Northrup et al., 1999; Corcoran and Dostal, 2001), and the interstratification of the volcanic rocks and turbiditic Contwoyto Formation (Henderson, 1998; Corcoran et al., 1998), which overlies the Augustus granite unconformably (Corcoran et al., 1998). Alternatively, Kusky (1991a, 1992) has suggested that the interlayering of the Contwoyto Formation and volcanic rocks may be related to deposition in forearc basins with volcanic activity or intense structural repetition of the underlying volcanic rocks. Considine (1995) has shown that the volcanic rocks in the Contwoyto Formation are alkaline, and chemically distinct from those in the underlying Peltier Formation. An allochthonous model for the structural setting of the greenstone belt is supported by the observations that (1) mafic dikes in the basement are separated from the greenstone belt by shear zones, (2) there are shear zones everywhere between the greenstone belt and Augustus granite, and (3) kinematic indicators and metamorphic mineral assemblages in the shear zones are consistent with thrusting of a thick hot ophiolitic slab over colder sediments and gneisses (Kusky, 1992). It is necessary to consider, however, that the shearing may have occurred following volcanic belt formation on continental crust. Based on stratigraphy, U-Pb ages, major and trace element chemistry, and isotopic results, Corcoran and Dostal (2001) proposed a tectonic model for the evolution of the Point Lake belt involving eruption of tholeiitic basalts between 2700 and 2680 Ma and calcalkaline basalts and andesites from 2680 to 2660 Ma into a back-arc that developed following extension and rifting of an older remnant arc overlying continental crust (Fig. 10). The work of Kusky (1991a) is consistent with an ophiolite origin for segments of the Point Lake belt west of Cyclops Peninsula, but an ophiolitic stratigraphy represented by the general up-section change from ultramafic rocks, gabbro, sheeted dikes and pillowed

378

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 10. Schematic model illustrating the evolution of the Point Lake volcanic belt. (A) Prior to 2.7 Ga, a volcano-sedimentary arc sequence was accreted onto > 3.0 Ga sialic crust. Subduction ensued after 2.7 Ga. (B) From 2700 to 2680 Ma, hinge roll-back propagated crustal attenuation and subsequent rifting of the older crustal arc sequence. A new arc was constructed and mafic tholeiitic magmas were erupted into the back-arc (Peltier Formation). The continental crust was eroded, detritus was transported via turbidity currents and deposited concomitant with volcanism (interstratified Contwoyto and Peltier Formations). (C) Between 2680 and 2673 Ma, calc-alkaline magmas were erupted through the tholeiitic pile. Mass wasting of the arc resulted in the deposition of felsic-intermediate volcaniclastic material (Samandre and Beauparlant Formations). Modified from Corcoran and Dostal (2001).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

379

flows is notably absent in the east. Possibly, the mafic mylonite and related rocks studied by Kusky (1991a) farther to the west, could represent a dismembered ophiolite that is unrelated to the evolution of the eastern Point Lake belt, and may be part of a 2.9–2.8 Ga event. Alternatively, the rocks to the east of Cyclops Peninsula could preserve upper parts of the ophiolitic stratigraphy, whereas rocks to the west preserve a more complete though highly metamorphosed and dismembered part of the ophiolitic sequence. The main differences between the autochthonous and allochthonous models is whether the volcanic rocks are essentially in place, and erupted through the adjacent granitoid gneisses, or whether the greenstone belt was thrust at least several tens of kilometers to the west from its place of origin, which may still have been adjacent to thinned continental crust (Fig. 11). 2.3. Beaulieu and Cameron River Volcanic Belts The Beaulieu River volcanic belt is preserved along the north-trending Beniah Lake fault in the south-central part of the Slave Province (Figs. 1, 12 and 13) and overlies the granitoid Sleepy Dragon Complex. Kusky (1990) postulated that the Beaulieu River belt and its western counterpart, the Cameron River belt, formed during opening and closing of an Archean ocean, based on the presence of basement complex intruded by a dense swarm of mafic dikes, overlain unconformably by a rift/passive margin sequence, then structurally overlain by the greenstone belts, interpreted as an ophiolite complex. Kusky (1990) suggested that only the upper parts of a dismembered ophiolite were preserved, as these greenstone belts contain pillow lavas, massive and brecciated flows, (abundant) gabbro sills (Fig. 13), feeder dikes, and local dense swarms of diabase dikes with pillow lava screens (Kusky, 1991b). In the Patterson Lake structural complex, a shear zone separates rocks of the greenstone belts from the underlying gneiss complex, and kinematic indicators in the shear zone show westward and upward transport of the Beaulieu River belt, and westward and downward transport of the Cameron River belt (Kusky, 1990). If a regional anticlinorium in the core of the Sleepy Dragon Complex were unfolded, the two greenstone belts would have a similar transport direction and internal stratigraphy, suggesting that they were once contiguous and formed one large sheet thrust toward the west. Lambert et al. (1992) discounted Kusky’s interpretation mainly because isotopic evidence is consistent with mixing of a depleted mantle source and Sleepy Dragon-type sialic crust. In addition, Lambert et al. (1992) argued that a continuous zone of 100% dikes is lacking, and that layered gabbros and ultramafic rocks are minor. Detailed physical volcanology and geochemistry of the central and northern parts of the Beaulieu River belt revealed notable variations in composition and volcanic facies. The central Beaulieu River belt lacks continuous multiple dikes, unlike the 20 km long Step’nduck dike swarm in the southern part of the Beaulieu River belt (Lambert et al., 1992). The central Beaulieu River volcanic sequence features tholeiitic basalts, calcalkaline basalts, andesites and rhyolites, and mafic and felsic autoclastic breccia and felsic volcaniclastic rocks of pyroclastic origin (Figs. 14 and 15). Although pillowed flows comprise up to 50% of the sequence, mafic dikes account for only 5% (Corcoran, 2000). The

380

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 11. (A) Structural map of the western Point Lake area (modified after Kusky, 1991a). (B) Sequentially restored down-plunge projections across the western Point Lake area, showing how after 69 km of slip on thrust faults, the western Point Lake greenstone belt is sitting, with an undefined contact relationship, over older quartzofeldspathic gneisses (modified after Kusky, 1991a).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

381

Fig. 11. (Continued.)

parallel dikes are 30–40 cm wide and 10’s of cm apart, cut pillow breccia and pillowed flows, but also locally propagate into individual pillows and are perpendicular to flow tops (Fig. 16). These features corroborate a synvolcanic origin, but the low density and number of intrusions discount a dike swarm interpretation. Geochemically, the mafic rocks are tholeiitic to calc-alkaline basalts and andesites, with LREE-enriched patterns and marked negative Nb and Ti anomalies. Felsic samples are calc-alkaline rhyolites, LREE-enriched, and also have large negative Nb and Ti anomalies (Corcoran, 2001). The combination of pyroclastic deposits, calc-alkaline basalts, andesite and rhyolite flows, the isotopic results of Lambert et al. (1992), as well as only minor tholeiitic basalts, is more consistent with a continental arc setting rather than an ophiolite origin for these parts of the belts. A possible depositional setting as presented by Corcoran (2001) involves the development of tholeiitic to calc-alkaline seamounts during initial arc formation evolving up-section into a summital calc-alkaline dome-flow complex (Fig. 17). This model, however, does not account for the early structural stacking documented by Kusky (1990).

382

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 12. Regional map of the central and northern parts of the Beaulieu River belt and their stratigraphic relation to the Cameron River belt (CRB), Sleepy Dragon Complex (SDC), and Burwash Formation (BF). Note the location of the Beniah Lake fault. Modified from James and Mortensen (1992).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

Fig. 13. Map of the Cameron River greenstone belt, showing mafic sill complexes, and a shear zone separating the greenstone belt from the underlying Sleepy Dragon gneiss complex (modified from Kusky, 1990). 383

384

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 14. (A) Distribution of mafic and felsic volcanic units in the central part of the Beaulieu River belt. (B) Schematic diagrams illustrating the predominant mafic volcanic facies. (C) Schematic diagram showing the felsic volcanic facies and the upsection transition into pillow breccia and mafic flows. Modified from Corcoran (2000).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

385

Fig. 15. Schematic stratigraphy of the Central Beaulieu River belt with the Sleepy Dragon Complex as inferred basement to the mafic-dominated volcanic belt. A ca. 2.9 Ga quartz arenite is in fault contact with the basement gneiss and overlying mafic flows. There is an up-section increase in felsic volcanic units and mafic fragmental facies. Ages from: (1) Henderson et al. (1987), Lambert and van Breemen (1991), (2) Henderson et al. (1987), Bleeker et al. (1999), (3) Mueller et al. (2001), and (4) Corcoran et al. (1999). From Corcoran (2001).

Similarly, a continental arc setting was proposed for the up to 3–4.2-km-thick Cameron River belt which thins out laterally and is interstratified (or intersheared; Kusky, 1990, 1991a) with the 200-m-thick, NNW-trending Raquette Lake Formation (Fig. 18; Mueller and Corcoran, 2001). Regional and facies mapping indicated an autochthonous basementcover relationship between the Raquette Lake Formation and gneissic Sleepy Dragon Com-

386

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 16. Characteristics of mafic intrusions in the Central Beaulieu River belt. (A) Outcrop sketch of the relationship between mafic facies with crudely bedded pillow breccia overlying a pillowed flow. Feeder dikes that cut pillow breccia have chilled margins and locally propagate into individual pillows. (B) Dike (D) propagating into a pillow (P) in pillow breccia. Arrow points to top. Scale, pencil 14 cm.

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

387

Fig. 17. Depositional setting of the Central Beaulieu River belt. (A) Model illustrating the inferred location of the study area on a seamount based on the thin pillowed and massive flows, abundant stratified pillow breccia and hyaloclastite. (B) Model displaying the felsic facies comprising the uppermost dome-flow complex. (C) Subaqueous depositional setting of the Central Beaulieu River belt illustrating the development of felsic edifices at the top of a seamount. From Corcoran (2001).

388 Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 18. Paleogeographic reconstruction of the Raquette Lake Formation and its lateral interdigitation with volcanic deposits of the Cameron River belt. This transition zone marked the presence of an active continental arc characterized by bimodal volcanism. From Mueller and Corcoran (2001).

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

389

plex, and an abrupt but transitional contact with the overlying Japan Sea-like, turbiditedominated Burwash Formation (Kusky, 1990, 1991b; Mueller and Corcoran, 2001). Similarly, Lambert (1988) showed that the Cameron River mafic volcanic rocks were interlayered with the turbiditic and pelagic sedimentary deposits of the overlying Burwash Formation. Although a lateral interdigitation of the 2.68–2.69 Ga Raquette Lake Formation with the mafic to felsic volcanic rocks of the Cameron River belt is favored (Mueller and Corcoran, 2001), Kusky (1990) initially proposed an intervening décollement and Bleeker (1997a, 1997b) suggested an unconformity as proposed by Henderson (1985). In order to put these arguments into perspective, detailed facies analyses along a critical 7-kmlong segment were conducted (Mueller and Corcoran, 2001). Facies mapping revealed a complex volcano-sedimentary evolution of bimodal mafic-felsic, subaerial-subaqueous volcanic flows and felsic volcaniclastic deposits, mafic-gabbroic and felsic sills (low-angle dikes), and coarse-clastic fan deltas with detritus originating from both the gneissic basement and contemporaneous volcanic activity. The formation passes along strike into the mafic pillow-dominated Cameron River belt, and numerous marker horizons could be traced along strike that support the inference of lateral facies interdigitation (Fig. 6 of Mueller and Corcoran, 2001). As observed by Kusky (1990), the autochthonous basementcover relationship is masked by a zone of increasing high strain toward the NNW and then NNE. However, it remains to be determined if this fault is entirely post-depositional, or if it may have been active and responsible for uplift and erosion of the volcanic terrane to the north that shed detritus into the Raquette Lake and Burwash basins. The mafic sheeted and multiple dike and sill complexes of the Cameron and Beaulieu River belts (Lambert, 1988; Lambert et al., 1992; Kusky, 1991b), and their uncertain relationship to dikes in the underlying Sleepy Dragon Complex remains problematic. The dikes in the basement are interpreted in some models to represent feeder dikes to both the Cameron River and Beaulieu River volcanic belts, and if so, an ophiolitic origin seems questionable. However, since the basement with amphibolite and gabbro dikes are structurally separated from the overlying pillowed flows and dike/sill complexes by sheared contact zones, the greenstone belts may have an allochthonous origin, and it can be argued that these sequences are part of a dismembered and tectonized ocean floor sequence lying on stretched continental crust. The association of pillowed flows with gabbro dikes and sills, interstratified turbiditic and pelagic deposits, and an underlying mafic sheeted dike complex certainly qualifies the sequence as an ophiolite. A dense mafic dike swarm cuts the quartzofeldspathic gneiss of the Sleepy Dragon Complex (Fig. 19), however, the volcanic belt is here separated from the basement complex by a shear zone (Patterson Lake structural complex; Kusky, 1991b). If the dikes represent feeders to a volcanic sequence, it is not the same volcanic succession that is now adjacent to the dike swarm. One possibility is that the dike swarm formed during extension of continental crust near the continent/ocean transition, and that oceanic crust developed outboard of the stretched continental crust. The present Cameron and Beaulieu River greenstone belts may have originated near this stretched continental crust-ocean transition zone, and were later thrust toward the west during closure of this (back arc?) basin.

390 Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 19. Dike swarm in Sleepy Dragon complex near Upper Ross Lake (map modified after Kusky, 1990). These dikes may have fed mafic flows that have since been eroded, with the gneiss complex and overlying sediments then being overthrust by the Cameron River greenstone belt. Alternatively, the dikes may have fed the greenstone belt.

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

391

Fig. 20. Regional map of the Northern Beaulieu River belt illustrating the distribution of mafic and felsic volcanic deposits and their relation to quartz arenite of the Beniah Formation. Note the location of the Beniah Lake fault zone and discrete units of ultramafic rocks throughout the area. Modified from Roach (1990) and Corcoran (2001).

The northern portion of the Beaulieu River belt overlies the 2.9–3.1 Ga (Isachsen and Bowring, 1997) quartz arenite-dominated Beniah Formation unconformably, and inferred plutonic basement is represented by granodiorite-tonalite gneisses of the Beniah Complex (Figs. 20 and 21). Discrete north- to northeast-trending ultramafic units up to 0.5 km wide and 1.8 km long are locally preserved in the Beniah Lake area (Fig. 20). Contacts with basement rocks and the shallow water sequence are nowhere preserved, but detrital chromite from dunite are preserved in fuchsite laminae of the quartz arenite, indicating that the sedimentary sequence is younger. The dunite is characterized by olivine pseudomorphs, chromite grains, and up to 20 cm-thick chromite seams (Fig. 22A). Microprobe data from zoned chromite grains is suggestive of a mixed arc-back-arc origin and except for 4 cores, are inconsistent with MORB (Fig. 22B). Stratigraphically, the ultramafic sequence does not appear to be directly related to the volcanic belt. A U-Pb zircon age of

392

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 21. Schematic stratigraphy of the Northern Beaulieu River belt. The Beniah Complex and ultramafic intrusions are considered basement to the Beniah Formation, which in turn underlies the volcanic belt. The mafic-dominated volcanic sequence contains felsic volcaniclastic units that grade upward into pillowed flows. Mafic intrusions are abundant and cut quartz arenite and mafic flows, but are eroded by younger sedimentary rocks. Ages from: (1) Isachsen and Bowring (1997), and (2) Mueller et al. (2001). From Corcoran (2001).

2672.3 + 8.5/−5.3 Ma, determined from a felsic tuff within the volcanic sequence (Corcoran, 2001), is significantly younger than the age of the detrital chromite-bearing quartz arenite. Parallel basaltic and gabbro intrusions, akin to the dike swarms of Lambert et al. (1992), are common in the northern portion of the Beaulieu River belt. Although the intrusions appear to account for approximately 25% of the volcanic facies in the study

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

393

Fig. 22. Characteristics of ultramafic rocks of the Beniah Lake area. (A) Altered dunite contains chromite seams (Ch) and displays a cumulate texture created by olivine pseudomorphs (O). Scale, coin 2.4 cm diameter. (B) Relationship between Fe3+ # and TiO2 based on microprobed data from chromite grains. The low TiO2 negates an intraplate origin, but the high Fe3+ # (in all but four samples) is inconsistent with MORB. The results are best attributed to an arc/back-arc transitional setting. Diagram after Arai (1992).

394

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

area, distinguishing between mafic intrusions and massive flows in the field was problematic because most contacts between intrusions and other volcanic facies are irregular or unexposed. The dikes are parallel, generally north-trending, and range from 3 to 20 m in width. Several sharp contacts were identified between the dikes, dunite and quartz arenite (Fig. 23A). Younger sedimentary breccia (interpreted as ca. 2.6 Ga) preserved along the Beniah fault zone west of Beniah Lake, was deposited unconformably on a mafic dike, as indicated by an erosional contact. Geochemically, the mafic intrusions can be subdivided into NMORB varieties with (La/Yb)n ratios ranging from 0.58 to 0.91, and EMORB types with (La/Yb)n ratios of 1.83–2.69 and marked Nb depletions on incompatible element diagrams. The REE patterns of intrusion subgroups are markedly comparable with those of mafic flows (Fig. 23B), thus supporting a synvolcanic origin. Abundant dikes associated with predominant massive flows are suggestive of a proximal setting with respect to a fissure or vent (e.g., Easton, 1984; Schmidt and Schminke, 2000) and also may indicate open-ocean spreading systems (Batiza and White, 2000). These dikes are considered feeders to the overlying basaltic pile (Corcoran, 2001). Pillowed flows are less common in the northern portion of the Beaulieu River belt and individual pillowed flow events are locally separated by felsic tuff horizons. Felsic volcaniclastic rocks are represented by planar- and cross-bedded tuff and lapilli tuff, tabular and low-angle to wavy bedded tuff, and planar laminated tuff. Mudstone laminae are common between bedforms, and up-section, lenses of black fine-grained tuff are intercalated with planar-bedded tuff. These black tuff horizons increase toward the contact with an overlying pillowed flow. An upward transition from planar- and cross-bedded lapilli tuff to composite beds of planar-, wavy- and cross-bedded tuff characterized by mudstone laminae, is consistent with a shallow subaqueous setting. The predominance of mafic massive flows is commonly attributed to high eruption rates from fissures (Smith and Cann, 1992; Kennish and Lutz, 1998). The subsurface features of fissures are dike complexes (Walker, 2000) and mafic dikes associated with swarms are consistent with rifting (Carey and Sigurdsson, 1984; Walker, 1993). Mafic flows and abundant mafic dikes associated with shallow water felsic volcaniclastic deposits suggest a subaerial to subaqueous transition. A depositional model proposed for the Northern Beaulieu River belt by Corcoran (2001) is illustrated in Fig. 24. Following eruption of a subaerial felsic edifice, pyroclastic debris was deposited into shallow water and was subjected to wave reworking. Mafic lavas derived from small fissure-fed edifices were being emplaced on the ocean floor concommitant with subaerial eruption. Drowning of the felsic edifice due to subsidence or a rise in water level resulted in the deposition of finer-grained material overlain by pillowed flows. The subaerial to shallow subaqueous setting supports deposition on a continental shelf or arc platform. Mafic samples from the Northern Beaulieu River belt are tholeiitic and resemble NMORB with depleted LREE patterns, and slightly LREE-enriched EMORB with negative Nb and some negative Ti anomalies. εNd values are between +0.39 and +8.26. One felsic tuff sampled is calc-alkaline. A possible tectonic setting for the Northern Beaulieu River belt involves contemporaneous subaerial eruption of felsic arc volcanoes and fissure-

2. Slave Craton Characteristics and Ophiolites in Greenstone Belts

395

Fig. 23. Characteristics of mafic intrusions in the Northern Beaulieu River belt. (A) Mafic dike (D) cuts the Beniah Formation quartz arenite (Qa). Contact is marked by dashed line. Scale, fieldbook 17.5 cm. (B) Chondrite-normalized REE abundances for mafic flows and intrusions. Normalizing values after Haskin et al. (1968).

396 Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 24. Depositional setting of the Northern Beaulieu River belt. A subaerial felsic edifice erupted pyroclastic material into shallow water concomitant with emplacement of subaqueous mafic magmas from fissure-dominated eruptions. Local multiple pillowed basalt flows constructed small seamounts on the ocean floor. Transgression, due to sea level rise and/or subsidence, resulted in drowning of the felsic edifice and deposition of pillowed flows over felsic volcaniclastic deposits. From Corcoran (2001).

3. Discussion

397

dominated eruptions in a back-arc basin. Interstratification of tholeiitic mafic and calcalkaline felsic material occurred along the fringing part of an arc in relatively shallow water.

3. DISCUSSION This review of Slave Province greenstone belts with possible ophiolite affinities leaves many questions unanswered. Numerous greenstone belts fulfill a sensu-lato definition of ophiolites with dismembered segments of thick mafic sequences characterized by highly tectonized mafic-ultramafic rocks at their bases, and abundant dike swarms. The belts that can be related to oceanic crust may indeed be good analogues for modern and Phanerozoic ophiolites. If it is true that some types of ophiolites may form on transitional continental crust and that they do not display a distinct geochemical signature, the field criteria is crucial in defining geodynamic context. In order for a mafic-dominated sequence overlying a sialic basement unconformably to be considered an ophiolite, an ultramafic zone of cumulates should be preserved. Mantle lherzolite or harzburgite tectonite may be preserved beneath a thinned, possibly mylonitic quartzofeldspathic gneiss in this tectonic setting. Subaqueous pillowed flows with intrusions of gabbroic composition resting on gneissic basement, which is common in the Slave craton, do not represent an ophiolite in the classic sense (Anonymous, 1972), but may represent material erupted through transitional continental crust near the junction with true oceanic crust. Alternatively, tectonic contacts between the greenstone belts and quartzofeldspathic gneiss with a thrust component signature and thick dismembered fragments of mafic-ultramafic volcanic and plutonic sequences, may justifiably denote an ophiolite sequence. The question is, what is the significant contribution in interpreting slivers of Archean mafic rocks as ophiolites, especially where felsic volcaniclastic and epiclastic sequences are interstratified with, or are adjacent to the mafic successions? Is it merely a classification scheme or does it yield any insight into tectonic processes? It is possible that different types of ophiolites represent end-members between ophiolites representing full oceanic spreading and those of transitional to highly extended continental crust. An overall geometry of the Slave Province proposed by Kusky (1989) involves an eastern juvenile Hackett River arc, adjacent to an intervening sedimentary-dominated Contwoyto terrane lapping onto older continental crust in the west (Anton and Sleepy Dragon terranes) upon which a subsequent arc-back-arc formed. Although there are numerous modifications on this initial plate tectonic theme, the basic ideas remain valid. Ophiolites possibly formed in both terranes, but this study shows that if some of the segments of these belts can be considered ophiolites, they are preferentially preserved on the continental inboard side of the 2720–2660 Ma pan-arc to back-arc forming Slave event. Early rift tectonics (2720–2700 Ma), attenuated the continental crust and arc-back-arc forming sequences (2700–2660 Ma) subsequently developed. The Point Lake, Beaulieu River and Cameron

398 Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Fig. 25. Geodynamic setting of the Kam Group, Yellowknife belt (A), Point Lake belt (B), Northern Beaulieu River belt (C), Cameron River belt (D), and Central Beaulieu River belt (E) in the Slave Province. The Kam Group and Point Lake belt formed in back-arcs, the Northern Beaulieu River belt formed in an arc to back-arc transition zone, and the Central Beaulieu and Cameron River belts formed on the arc. Modified from Corcoran (2001).

References

399

River belts represent different parts of an arc-back-arc system floored by >2.8 Ga continental crust. The Point Lake Belt represents the back-arc spreading center, the Northern Beaulieu River belt is consistent with the back-arc to arc transition zone, and the Central Beaulieu River and Cameron River belts are best compared with the arc (Fig. 25). The volcanic sequences evolved between 2690 and 2670 Ma, which is consistent with the ca. 15–25 m.y. time span for modern arc-back-arc systems in the western Pacific (Leitch, 1984; Tarney and Saunders, 1984; Wharton et al., 1995). The time span indicates that the arc migrated and evolved into a mature marginal basin with a thick turbidite sequence and rift-related basalts at Point Lake. The arc, possibly due to hinge roll-back, migrated over time, resulting in crustal attenuation, akin to a Japan-Japan Sea-type arc to back-arc setting. The older ages from the Kam Group of the Yellowknife volcanic belt, its interpretation as a continental marginal basin (Helmstaedt and Padgham, 1986; Cousens, 2000), and its N-S trend parallel to the Point Lake, Northern Beaulieu River and Central Beaulieu River belts, is suggestive of development during splitting of an active continental arc prior to the 2.69–2.67 Ga subduction event. The spatial relationship between the Kam Group and the other study areas is consistent with eastward migration of the subduction zone over 30– 50 m.y. (Fig. 25). A parallel series of arcuate arc-back-arc sequences is a classical feature of asymmetrical arc spreading centers in the western Pacific Ocean (e.g., Karig, 1971; Karig et al., 1978; Hathway, 1994; Hawkins, 1995). The western Slave Province preserves the evolution of an ancient rift-arc system from ca. 2720 to 2700 Ma, during which a continental margin attenuated, an arc developed, and a back-arc basin formed. Subsequent volcanism occurred in the eastern part of the older rifted arc, represented by the Central Beaulieu River and Cameron River belts. With continued hinge roll-back, the younger arc split and a second back-arc basin formed (Point Lake belt). The Northern Beaulieu River belt represents the volcanic debris that accumulated in the younger arc to back-arc transition zone. Most of the preserved ophiolite-inferred sequences in the Slave Province are spatially associated with basement, and this strong correlation is observed with many Tethyan ophiolite terranes. From a geodynamic and geochemical perspective, many greenstone belts fall into the ophiolite category (sensu lato definition), but detailed volcanic and sedimentary facies analysis, structural geology, and geochemical studies suggest that many did not form as parts of mid ocean ridges in major ocean basins. These potential ophiolites probably developed during extreme attenuation and stretching of continental crust during the early stages of arc extension and ocean formation.

REFERENCES Anonymous, 1972. Ophiolites. Geotimes 17, 24–25. Arai, S., 1992. Chemistry of chromium spinel in volcanic rocks as a potential guide to magma chemistry. Mineralogical Magazine 56, 173–184.

400

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Batiza, R., White, J.D.L., 2000. Submarine lavas and hyaloclastite. In: Sigurdsson, H., Houghton, B., McNutt, S.R., Rymer, H., Stix, J. (Eds.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 361–381. Bickle, M.J., Niset, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Bleeker, W., Ketchum, J.W.F., Jackson, V.A., Villeneuve, M.E., 1999. The Central Slave Basement Complex, Part I: Its structural topology and autochthonous cover. Canadian Journal of Earth Science 36, 1083–1109. Bleeker, W., Davis, B., Villeneuve, M., 1997a. The Slave Province: evidence for contrasting crustal domains and a complex, multistage tectonic evolution. In: Cook, F., Erdmer, P. (Eds.), SlaveNorthern Cordillera Lithospheric Evolution Transect and Cordilleran Tectonics Workshop Meeting (March 7–9), University of Calgary. Lithoprobe Report 56, 36–37. Bleeker, W., Villeneuve, M., Bethune, K., 1997b. Thematic structural studies in the Slave Province, Northwest Territories; contrasting basement/cover relationships on western and southwestern flanks of the Sleepy Dragon Complex. In: Current Research 1997-C. Geological Survey of Canada, 27–37. Bowring, S.A., Williams, I.S., 1998. Priscoan (4.00–4.03 Ga) orthogneisses from northwestern Canada. Contributions to Mineralogy and Petrology 134, 3–16. Carey, S., Sigurdsson, H.A., 1984. A model of volcanogenic sedimentation in marginal basins. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basin Geology. Geological Society Special Publication 16, 37–58. Considine, P.R., 1995. Geology of an Archean Greenstone Belt at Point Lake, Northwest Territories, Canada. M.Sc. thesis. Old Dominion University, 165 pp., 1 map. Considine, P.R., Kusky, T.M., Isachsen, C., 1987. Ultramafic rocks in the southern portion of the Yellowknife Greenstone Belt. Geological Association of Canada, Summer Field Conference, Yellowknife, Northwest Territories, Aug. 31–Sept. 5, 1987. Abstracts with Programs. Corcoran, P.L., 2000. Recognizing distinct portions of seamounts using volcanic facies analysis: examples form the Archean Slave Province, Northwest Territories, Canada. Precambrian Geology 101, 237–261. Corcoran, P.L., 2001. Physical volcanology, geochemistry, and tectonic evolution of three selected areas in the Point Lake and Beaulieu River volcanic belts, Slave Province, Northwest Territories, Canada. Ph.D. thesis. Dalhousie University, Halifax, Nova Scotia, p. 247. Corcoran, P.L., Dostal, J., 2001. Development of an ancient back-arc basin overlying continental crust: the Archean Peltier Formation, Northwest Territories, Canada. Journal of Geology 109, 329–348. Corcoran, P.L., Mueller, W.U., 2002. The effects of weathering, sorting and source composition in Archean high-relief basins: examples for the Slave Province, Northwest Territories, Canada. In: Altermann, W., Corcoran, P.L. (Eds.), Precambrian Sedimentary Environments: A Modern Approach to Ancient Depositional Systems. IAS Special Volume 33, 183–211. Corcoran, P.L., Mueller, W.U., Chown, E.H., 1998. Climatic and tectonic influences on fan deltas and wave-to tide-controlled shoreface deposits; evidence from the Archean Keskarrah Formation, Slave Province, Canada. Sedimentary Geology 120, 125–152. Corcoran, P.L., Mueller, W.U., Padgham, W., 1999. Influence of tectonism and climate on lithofacies distribution and sandstone and conglomerate composition in Archean Beaulieu Rapids Formation, Northwest Territories, Canada. Precambrian Research 94, 17–204. Cousens, B.L., 2000. Geochemistry of the Archean Kam Group, Yellowknife Greenstone Belt, Slave Province, Canada. Journal of Geology 108, 181–197.

References

401

Covello, L., Roscoe, S.M., Donaldson, J.A., Roach, D., Fyson, W.K., 1988. Archean quartz arenite and ultramafic rocks at Beniah Lake, Slave structural province, N.W.T. In: Current Research Part C. Geological Survey of Canada Paper 88-1C, 223–232. Cunningham, M.P., Lambert, R.S.J., 1989. Petrochemistry of the Yellowknife volcanic suite at Yellowknife, NWT. Canadian Journal of Earth Science 26, 1630–1646. Davis, W.J., Bleeker, W., 1999. Timing of plutonism, deformation, and metamorphism in the Yellowknife Domain, Slave Province, Canada. Canadian Journal of Earth Science 36, 1169–1187. Dilek, Y., Moores, E., Elthon, D., Nicolas, A. (Eds.), 2000. Ophiolites and oceanic crust; new insights from field studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 552. Easton, R.M., 1984. Growth and evolution of volcanic edifices—with implications for Precambrian volcanoes. In: Easton, R.M., Easton, M.G. (Eds.). Geological Association of Canada Short Course Notes 4, 1–40. Haskin, L.A., Haskin, M.A., Frey, F.A., Wildman, T.R., 1968. Relative and absolute terrestrial abundances of the rare earths. In: Ahrens, L. (Ed.), Origin and Distribution of the Elements. Pergamon, Oxford, pp. 889–911. Hathway, B., 1994. Sedimentation and volcanism in an Oligocene-Miocene intra-oceanic arc and fore-arc, southwestern Viti Levu, Fiji. Journal of the Geological Society, London 151, 499–514. Hawkins Jr., J.W., 1995. The geology of the Lau Basin. In: Taylor, B. (Ed.), Back-Arc Basins: Tectonism and Magmatism. Plenum Press, New York, pp. 63–138. Helmstaedt, H., Padgham, W., 1986. A new look at the stratigraphy of the Yellowknife Supergroup at Yellowknife, N.W.T.—implications for the age of gold-bearing shear zones and Archean basin evolution. Canadian Journal of Earth Science 23, 454–475. Helmstaedt, H., Padgham, W.A., Brophy, J.A., 1986. Multiple dikes in Lower Kam Group, Yellowknife greenstone belt: evidence for Archean sea-floor spreading? Geology 14, 562–566. Henderson, J.B., 1981. Archean basin evolution in the Slave Province, Canada. In: Kröner, A. (Ed.), Precambrian Plate Tectonics. Elsevier, Amsterdam, pp. 213–236. Henderson, J.B., 1985. Geology of the Yellowknife-Hearne Lake area, District of Mackenzie: a segment across an Archean basin. Geological Survey of Canada Memoir 414. Henderson, J.B., 1998. Geology of the Keskarrah Bay Area, District of Mackenzie, Northwest Territories. Geological Survey of Canada Bulletin 527, 122. Henderson, J.B., Brown, I.C., 1966. Geology and structure of the Yellowknife greenstone belt, District of Mackenzie. Geological Survey of Canada Bulletin 141. Henderson, J.B., van Breeman, O., Loveridge, W.D., 1987. Some U-Pb zircon ages from Archean basement, supracrustal and intrusive rocks, Yellowknife-Hearne Lake area, District of Mackenzie. In: Radiogenic Age and Isotopic Studies Report 1. Geological Survey of Canada Paper 87-2, 111– 121. Hrabi, R.B., Nelson, M.D., Helmstaedt, H., 1995. Diverse metavolcanic sequences and late polymict conglomerate-associated metasedimetnary rocks in the Winter Lake supracrustal belt, Slave Province, Northwest Territories. In: Current Research, 1995-E. Geological Survey of Canada, 137–148. Isachsen, C.E., 1992. U-Pb zircon geochronology of he Yellowknife volcanic belt and subjacent rocks, N.W.T. Canada: constraints on the timing, duration, and mechanics of greenstone belt formation. Ph.D. thesis. Washington University. Isachsen, C.E., Bowring, S.A., 1994. Evolution of the Slave Craton. Geology 2, 917–920.

402

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Isachsen, C.E., Bowring, S.A., 1997. The Bell Lake group and Anton Complex: a basement-cover sequence beneath the Archean Yellowknife greenstone belt revealed and implicated in greenstone belt formation. Canadian Journal of Earth Science 34, 169–189. Isachsen, C.E., Bowring, S.A., Padgham, W.A., 1991. U-Pb zircon geochronology of the Yellowknife volcanic belt, NWT Canada: new constraints on the timing and duration of greenstone belt magmatism. Journal of Geology 99, 55–67. Jackson, V.A., 1996. Preliminary geology of part of the lower supracrustal succession in the Bell Lake area (85 J/16). Department of Indian Affairs and Northern Development EGS Map 199617. James, D.T., Mortensen, J.K., 1992. An Archean metamorphic core complex in the southern Slave Province; basement-cover structural relations between the Sleepy Dragon Complex and the Yellowknife Supergroup. Canadian Journal of Earth Sciences 29, 2133–2145. Jenner, G.A., Fryer, B.J., McLennan, S.M., 1981. Geochemistry of the Archean Yellowknife Supergroup. Geochimica et Cosmochimica Acta 45, 1111–1129. Karig, D.E., 1971. Origin and development of marginal basins in the Western Pacific. Journal of Geophysical Research 76, 2542–2561. Karig, D.E., Anderson, R.N., Bibee, L.D., 1978. Characteristics of back arc spreading in the Mariana Trough. Journal of Geophysical Research 83, 1213–1226. Kennish, M.J., Lutz, R.A., 1998. Morphology and distribution of lava flows on mid-ocean ridges: a review. Earth Science Reviews 43, 63–90. Kusky, T.M., 1987. Comment on “Multiple dikes in the lower Kam Group, Yellowknife Greenstone Belt: Evidence for the Archean sear-floor spreading?”. Geology 15, 280–282. Kusky, T.M., 1989. Accretion of the Archean Slave Province. Geology 17, 63–67. Kusky, T.M., 1990. Evidence for Archean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991a. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10 (4), 820–841. Kusky, T.M., 1991b. Geology of the Cameron River Greenstone Belt-Sleepy Dragon Metamorphic Complex (parts of NTS areas SNRC 85 I/4 and 85 I/15). Indian Affairs and Northern Development Publication EGS 1994-8, 6 maps with marginal notes. Kusky, T.M., 1992. Relative Timing of deformation and metamorphism at mid-to upper-crustal levels in the Point Lake Orogen, Slave Province, Canada. In: Glover, J.E., Ho, S.E. (Eds.), The Archaean: Terranes, Processes and Metallogeny, Geology Department (Key Centre) & University Extension, Publication No. 22. The University of Western Australia, pp. 59–71. Lambert, M.B., 1988. Cameron River and Beaulieu River volcanic belts of the Archean Yellowknife Supergroup, District of Mackenzie, Northwest Territories. Geological Survey of Canada Bulletin 382, 145. Lambert, M.B., Ernst, R.E., Dudas, F.O.L., 1992. Archean mafic dyke swarms near the Cameron River and Beaulieu River volcanic belts and their implications for tectonic modeling of the Slave Province, Northwest Territory. Canadian Journal of Earth Science 29, 226–2248. Lambert, M.B., van Breemen, O., 1991. U-Pb zircon ages from the Sleepy Dragon Metamorphic Complex and new occurrence of basement within the Meander Lake Plutonic Suite, Slave Province, W.T.T. In: Radiogenic Age and Isotopic Studies: Report 4. Geological Survey of Canada Paper 90-2, 79–84. Leitch, E.C., 1984. Marginal basins of the SW Pacific and the preservation and recognition of their ancient analogues: a review. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basin Geology. Geological Society of London Special Publication 16, 97–108.

References

403

MacLachlan, K., Helmstaedt, H., 1995. Geology and geochemistry of an Archean mafic dyke complex in the Chan Formation: basis for a revised plate-tectonic model of the Yellowknife greenstone belt. Canadian Journal of Earth Science 32, 614–630. Metcalf, R.V., Wallin, E.T., Willse, K.R., Muller, E.R., 2000. Geology and geochemistry of the ophiolitic Trinity Terrane, California evidence of middle Paleozoic depleted supra-subduction zone magmatism in a proto-arc setting. Ophiolites and oceanic crust; new insights from field studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 403–418. Moores, E.M., 1982. Origin and emplacement of ophiolites. Reviews of Geophysics and Space Physics 20, 735–760. Moores, E.M., 2002. Erratum; Pre-1 Ga (pre-Rodinian) ophiolites; their tectonic and environmental implications. Geological Society of America Bulletin 114, 80–95. Moores, E.M., Kellogg, L.H., Dilek, Y., 2000. Tethyan ophiolites, mantle convection, and tectonic “historical contingency” a resolution of the “ophiolite conundrum”; Ophiolites and oceanic crust; new insights from field studies and the Ocean Drilling Program. Geological Society of America Special Paper 349, 3–12. Mortensen, J.K., Henderson, J.B., Jackson, V.A., Padgham, W.A., 1992. U-Pb geochronology of Yellowknife Supergroup felsic volcanic rocks in the Russell Lake and Clan Lake areas, southwestern Slave Province, N.W.T. In: Radiogenic Age and Isotopic Studies 5. Geological Survey of Canada Paper 912, 1–7. Mueller, W.U., 1997. Crustal-scale structures and their influence on basin formation, evolution and dispersal patterns: examples from the Abitibi belt and Slave Province. In: Cook, F., Erdmer, P. (Eds.), Slave-Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop Meeting (March 7–9), University of Calgary. Lithoprobe Report 56, 12. Mueller, W.U., Bowring, S.A., Corcoran, P.L., Pickett, C., 1998. Unconformities, major faults and the evolution of volcano-sedimentary basins on the Slave craton. In: Cook, F., Erdmer, P. (Eds.), Slave-Northern Cordillera Lithospheric Evolution Transect and Cordilleran Tectonics Workshop Meeting (March 6–8), Simon Fraser University. Lithoprobe Report 64, 15–16. Mueller, W.U., Corcoran, P.L., 2001. Volcanic and sedimentary processes operating on a marginal continental arc: evidence from the Archean Raquette Lake Formation, Slave Province, Canada. Sedimentary Geology 140–141, 169–204. Mueller, W.U., Corcoran, P.L., Simard, R.L., Mortensen, J.K., 2001. Continental arc-backarc evolution: the 2650–2710 Ma volcano-sedimentary sequences in the central Slave Province. In: Cook, F., Erdmer, P. (Eds.), Slave-Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop Meeting (Feb. 22–25), Pacific Geoscience Centre. Lithoprobe Report 79, 12–17. Northrup, C.J., Isachsen, C., Bowring, S.A., 1999. Field relations, U-Pb geochronology and SmNd isotope geochemistry of the Point Lake greenstone belt and adjacent gneisses, central Slave craton, N.W.T., Canada. Canadian Journal of Earth Science 36, 1043–1059. Padgham, W.A., 1985. Observations and speculations on supracrustal successions in the Slave Structural Province. In: Ayers, L.D., Thurston, P.C., Card, K.D., Weber, W. (Eds.), Evolution of Archean Supracrustal Sequences. Geological Association of Canada Special Paper 28, 133–152. Padgham, W.A., Fyson, W.K., 1992. The Slave Province: a district Archean craton. Canadian Journal of Earth Science 29, 2072–2086. Pearce, J.A., Harris, N.B., Tindle, A.G., 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology 25, 956–983. Pearce, J.A., Peate, D.W., 1995. Tectonic implications of the composition of volcanic arc magmas. Annual Review of Earth Planetary Science 23, 251–285.

404

Chapter 12: Inferred Ophiolites in the Archean Slave Craton

Pickett, C., Mueller, W.U., 2000. Lithofacies characteristics and petrology of a stale Archean shelf sequence: Slave province, NWT. In: Cook, F., Erdmer, P. (Eds.), Slave-Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop Meeting (Feb. 25–27), University of Calgary. Lithoprobe Report 72, 11–24. Roach, D, 1990. Geology of the Beniah Lake Area NTS 85 P/8. Department of Indian Affairs and Northern Development EGS Map 1990-2, scale 1:50 000. Saunders, A.D., Tarney, J., 1984. Geochemical characteristics of basaltic volcanism within back-arc basins. In: Kokelaar, B.P., Howells, M.F. (Eds.), Marginal Basin Geology. Geological Society of London Special Publication 16, 59–76. Schmidt, R., Schminke, H.U., 2000. Seamounts and island building. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 383–402. Searle, M.P., Cox, J., 1999. Tectonic setting, origin, and obduction of the Oman Ophiolite. Geological Society of America Bulletin 111, 104–122. Smith, D.K., Cann, J.R., 1992. The role of seamount volcanism in crustal construction at the MidAtlantic Ridge (24◦ –34◦ N). Journal of Geophysical Research 97, 1645–1658. Stubley, M., 1989. Geology of the Spencer Lake area; parts of NTS 85 P/1,2. Department of Indian Affairs and Northern Development EGS Map 1989-12, scale 1:50 000. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.C., 1997. Volcanic aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 55–90. van Breemen, O., Davis, W.J., King, J.E., 1992. Temporal distribution of granitoid plutonic rocks in the Archean Slave Province, northwest Canadian Shield. Canadian Journal of Earth Science 20, 2186–2199. Walker, G.P.L., 1993. Basaltic-volcano systems. In: Prichard, H.M., Alabaster, T., Harris, N.B.W., Neary, C.R. (Eds.), Magmatic Processes and Plate Tectonics. Geological Society Special Publication 76, 3–38. Walker, G.P.L., 2000. Basaltic volcanoes and volcanic systems. In: Sigurdsson, H. (Ed.), Encyclopedia of Volcanoes. Academic Press, San Diego, pp. 283–289. Wharton, M.R., Hathway, B., Colley, H., 1995. Volcanism associated with extension in an OligoceneMiocene arc, southwestern Viti Levu, Fiji. In: Smellie, J.L. (Ed.), Volcanism Associated with Extension at Consuming Plate Margins. Geological Society of London Special Publication 81, 95–114.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

405

Chapter 13

3.0 GA OLONDO GREENSTONE BELT IN THE ALDAN SHIELD, E. SIBERIA IGOR S. PUCHTEL Department of the Geophysical Sciences, The University of Chicago, 5734 South Ellis Avenue, Chicago, IL 60637, USA

1. INTRODUCTION Greenstone belts are the best preserved areas in Precambrian shields that provide a window into a distant geological past. One of the most challenging issues in the earth sciences is the reconstruction of the chemical and tectonic evolution of the early Earth. Plate tectonics is the most commonly cited mechanism by which the modern Earth releases its heat. Plate tectonics is responsible for chemical and mass exchange within the mantle and basically for the way our planet looks and evolves today. Whether the Earth was so different in the Archean that plate tectonics did not operate has remained the subject of a long-standing debate (e.g., deWit and Ashwal, 1986, 1997). This question still cannot be answered with a sufficient degree of certainty due to the fact that relicts of bone fide Archean oceanic crust as we understand it have yet to be found. This is mostly because of the contorted and incomplete nature of the record in the greenstone belts none of which can be regarded as an unequivocal Archean ophiolite, and also because of the highly varied ways in which plate tectonic processes can operate. But it appears that we are getting very close indeed. This manuscript is an attempt to address the above problem by studying a 3.0 billion-year-old ophiolite-like association of the Olondo greenstone belt in the Aldan shield. The Olondo greenstone belt is a unique structure for several reasons. First, it is distinguished from the other greenstone belts in the Aldan Shield by an abundance and a great facies diversity of mafic-ultramafic rocks. In this respect, there is no match to this greenstone belt in the Aldan Shield, one of the largest cratonic segments on Earth. Second, the rocks are relatively well preserved both geologically and geochemically compared to other Archean ophiolite-like sequences worldwide, and thus can be regarded as valuable witnesses of the early history of the Earth. Third, the Olondo greenstone belt contains one of the oldest ophiolite-like sequences on the planet. Because of the reasons outlined above, this is arguably the best studied greenstone belt in the Aldan shield. In addition, the age of the Olondo greenstone belt at 3.0 Ga is intermediate between the two most commonly cited periods of global crust-forming activity, namely, 2.7 and 3.4 Ga (Condie, 1995, 1998). Thus, the study of this belt can help fill the gap in our understanding the significance of DOI: 10.1016/S0166-2635(04)13013-2

406

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

the tectonothermal and chemical evolution of the Earth during the time period between the early and late Archean.

2. REGIONAL GEOLOGY The Aldan Shield is the largest basement salient in the Siberian craton. The shield consists of the Aldan granulite-gneiss terrain, and Batomga and Olekma gneiss-greenstone terrains (GGT) to the west and east, respectively (Fig. 1). The geology, stratigraphy and geochronology of the region have been reviewed in numerous publications (Mironiuk et al., 1971; Drugova et al., 1985; Petrov, 1985; Dook et al., 1986, 1989; Smelov, 1989; Puchtel, 1992; Dobretsov et al., 1997; Jahn et al., 1998). The geological framework of the Olekma GGT is defined by the juxtaposition of the Olekma gneiss-migmatite complex and the Subgan granite-greenstone complex, which occur in the proportion 10:1. The Olekma gneiss-migmatite complex comprises tonalitic-trondhjemitic gneisses (TTG),

Fig. 1. Geological sketch map of the central part of the Aldan Shield (modified after Dook et al., 1986, 1989).

3. Geology of the Olondo Greenstone Belt

407

amphibolites, granitic gneisses and migmatites. Amphibolites and tonalite-trondhjemitic gneisses occur as remnants within vast areas of granitic gneisses. The U-Pb zircon ages of tonalite-trondhjemitic gneisses of 3212 ± 8, 3172 ± 24, and 3335 ± 2 Ma (Nutman et al., 1990, 1992; Jahn et al., 1998) and the Sm-Nd isochron age of 3235 ± 174 Ma (Puchtel et al., 1993) are considered to represent the time of emplacement of tonalite-trondhjemitic magmas. The granitic gneisses in the vicinity of the Olekma River range in age between 2.8 and 3.0 Ga, with most of the zircon data clustering around 3.0 Ga (Jahn et al., 1998). Puchtel et al. (1993) obtained a Sm-Nd isochron age of 2835 ± 94 Ma for these rocks, which is close to the U-Pb zircon (SHRIMP) age of 2862 ± 14 Ma (Baadsgaard et al., 1990). These ages reflect the time of high-grade metamorphic event and migmatitization. The Subgan granite-greenstone complex includes sedimentary, volcaniclastic and volcanic rocks of greenstone belts, which have undergone polyphase greenschist to amphibolite facies deformation and metamorphism. The greenstone belts strike N-S and are confined to narrow linear zones (Fig. 1). Within these zones, the rocks of the Subgan complex are folded into narrow elongated troughs, and bounded by the highly schistose to blastomylonitic TTG of the Olekma complex. Various types of granites heal the tectonic contacts between the two complexes. The greenstone belts range in age between 3.2 and 2.7 Ga and can be subdivided into predominantly volcanic and volcano-sedimentary types on the basis of lithology. The Olondo greenstone belt is a typical example of the former group. It is arguably the best exposed and best studied among the greenstone belts in the Olekma GGT.

3. GEOLOGY OF THE OLONDO GREENSTONE BELT The Olondo greenstone belt is located in the central part of the Olekma GGT. It is about 100 km long and has a maximum width of 30 km (Fig. 1). The geology and stratigraphy of the area has been described in a number of publications (e.g., Drugova et al., 1983, 1988; Popov et al., 1990, 1995; Puchtel, 1992; Dobretsov et al., 1992; Puchtel and Zhuravlev, 1993b). The belt occurs as narrow, dislocated and intricately deformed synforms with an approximately N-S trend. The largest, the Olondo synform, has a maximum size of about 30 × 10 km. It has steep limbs and splits up into two branches in its northern portion (Fig. 2). The Olondo synform is characterized by a rather symmetrical distribution of various volcanic rock types with respect to the axis of the V-shaped structure. Contacts between the supracrustal rocks and the surrounding tonalitic-trondhjemitic gneisses are tectonic. Numerous tonalite plutons intruded along the contacts between the TTG and the greenstone sequences. There are also several tonalite massifs in the central part of the synform. These plutons contain xenoliths of metamorphosed mafic-ultramafic to intermediate-felsic volcanic and plutonic rocks. The volcanic sequence of the belt has been subdivided into the lower and upper units (Puchtel and Zhuravlev, 1993b). The boundary between the two is not well defined geologically and is thus not shown on the map. The lower unit is exposed in the eastern and western marginal portions of the belt and has a

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

Fig. 2. Geologic-petrographical map of the Olondo greenstone belt.

408

4. Field Petrology of the Mafic and Ultramafic Rocks

409

maximum thickness of ∼ 500 m. It is composed largely of mafic and ultramafic volcanic and volcaniclastic rocks. Volumetrically, the mafic metavolcanics are predominant. The upper unit occurs in the central part of the belt and consists of intermediate and felsic lavas and subvolcanic rocks with a subordinate amount of volcaniclastic lithologies and minor metasedimentary intercalations. The upper unit is about 600 m thick in total. Several authors opined that the above-mentioned succession of volcanic-sedimentary sequences cannot be regarded as having stratigraphic significance due to the intensive metamorphic reworking and deformations and the presence of numerous intrusive bodies of various compositions (Popov et al., 1995). Moreover, these authors suggest that the upper and lower sequences were spatially and temporary separated and were tectonically juxtaposed during later stages of evolution of the belt. Mafic and ultramafic plutonic rocks constitute a substantial part (> 30%) of the Olondo complex. Ultramafic plutonic rocks are mostly represented by large dunite-peridotite bodies that have clear tectonic relationships with the country rocks. In most places they are confined to the contact between the lower and the upper unit rocks and locally occur within the lower unit lavas. Mafic plutonic rocks are most abundant in the belt. They are chiefly represented by differentiated gabbro sills that are confined to lithological boundaries. The metamorphic grade ranges from epidote-amphibolite to amphibolite facies (T = 550–600 ◦ C, P = 2–5 kbar) and increases slightly towards the margins of the belt (Drugova et al., 1983; Popov et al., 1990). Zircons from samples of felsic volcanics collected in the western part of the belt yielded a crystallization age of 2986 ± 12 Ma, those from the central part, 2998 ± 18 Ma, and zircons from the easternmost samples, 3005 ± 10 Ma (SHRIMP data: Baadsgaard et al., 1990). The greenstone belt is cut by metamorphosed dikes of picrites with a Sm-Nd whole-rock age of 2202 ± 41 Ma (Puchtel and Zhuravlev, 1993a). The eastern branch of the synform (Fig. 2) is best preserved and contains several groups of mafic-ultramafic rocks of the lower unit, including Eastern komatiitic (EKB) and Eastern tholeiitic basalts (ETB), ultramafic plutonic rocks and gabbro sills, and thin horizons of actinolite-chlorite schists first identified as komatiites (here called Western komatiites) in the western branch of the synform by Drugova et al. (1988). Field characteristics and petrology of these rocks are given below. For more detailed information, the reader is referred to publications by Popov et al. (1990), Puchtel (1992), Puchtel and Zhuravlev (1993b).

4. FIELD PETROLOGY OF THE MAFIC AND ULTRAMAFIC ROCKS 4.1. Eastern Komatiitic Basalts Eastern komatiitic basalts (EKB) form a gently dipping volcanic sequence consisting of massive, pillowed and differentiated lava flows 5.5 to 38 m thick which are underlain and overlain by tholeiitic basalts. The differentiated flows consist of the upper pillow, the middle massive, and the lower cumulate zone. The pillow lava zone is composed of closely

410

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

packed almost undeformed fragments a few tens of centimeter to a few meters in size. The fragments are tongue-like, bulbous, spheroidal or pillow-like in shape. Their tops are typically convex, and the bottoms either sag down into interpillow space, or are concave, conforming with the topography of the underlying fragments (Fig. 3). The matrix between the fragments is filled with a fine-grained actinolite-chlorite aggregate. The middle massive zone of the flows has a uniform structure and shows columnar jointing. The lower cumulate zone of the flows, where present, consists of a massive rock enriched in cumulate olivine now completely replaced by chlorite. Some of the thinner flows exhibit easily identified porous, brecciated tops and are composed of pillowed or massive komatiitic basalt throughout. 4.2. Eastern Tholeiitic Basalts (ETB) and Associated Gabbro Tholeiitic basalts, now represented by amphibolite schists, occur as spatially separated volcanic sequences that underlie and overlie the komatiitic basalts. Where primary volcanic structures are preserved, it is possible to identify individual flows that vary in thickness from 5 to 50 m and have an amygdaloidal, massive or pillow structure. The amygdaloidal and massive flows have brecciated flowtops, which are made up of a highly porous, fragmented rock, the product of mechanical shattering of the upper chilled crust during the movement of the lava. In the massive flows, the flowtop breccia is underlain by a finegrained amphibolite, that gives way to a coarser-grained variety. In the amygdaloidal flows, the amygdules decrease in size and abundance towards the bottom, with the lower third usually being massive. Pillow lava flows are composed of severely deformed fragments up to 2 m in size. Associated with the tholeiitic basalts are gabbro intrusions. They occur as concordant bodies, a few meters to several hundred meters thick, and are identified by symmetric chilled zones at the contacts with the country rocks, by relics of gabbro textures and a massive structure. 4.3. Western Komatiites Western komatiites (WK) are now represented by series of concordant actinolite-chlorite schist bodies 0.5–25 m thick, which alternate with mafic lavas and tuffs. Due to their position in the lowermost part of the synform near the contact with TTG, the actinolite-chlorite schists are severely sheared, which obscures the nature of their precursor rock. Because of rather poor exposure, the exact dimensions of the bodies are not well known, but are presumed to be not less than several hundred to several thousand meters along the strike. At the inferred tops of some of the least altered bodies, rocks resembling flowtop breccias of komatiite lavas have been found. Those bodies might have been massive lava flows. Others, which lack evidence of extrusive origin, may represent hypabyssal sills. 4.4. Ultramafic Plutonic Rocks Ultramafic plutonic rocks are represented by lenticular bodies of various size. The best preserved and best studied is the Red Hill massif exposed at the western flank of the eastern

4. Field Petrology of the Mafic and Ultramafic Rocks

411

Fig. 3. Fragments of pillow lavas from the upper part of Eastern komatiitic basalt differentiated flows.

412

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

branch (Fig. 2). It is about 600 × 4000 m in size, and is concordant with the schistosity of the surrounding tholeiitic basalts. The body is cut into blocks by a series of faults and has had a long metamorphic and tectonic history. The western exposed contact is tectonic. The border zone is schistose near the contact and becomes massive a few meters away. The massif is composed of dunites-peridotites with relict igneous textures, and of products of their reworking, represented by metamorphic olivinites. The former occur in the least deformed central portion of the massif. The dunites are made up of 95–97% of equant, closely packed olivine grains 0.5–2 mm in size with a composition of Fo91.2–92.4. The interstices are filled with clinopyroxene and chromite grains. The peridotites consist of 70–90% of elongate to equant euhedral olivine grains 1–8 mm in size, ranging in composition between Fo89.9–90.5. Intercumulus minerals are represented by clinopyroxene and chromite grains. Typical features of the olivines from dunites and peridotites are their red-brown color, the ubiquitous presence of minute secondary mineral inclusions, lack of internal zoning and low Ca and Cr contents. Chromites from both dunites and peridotites contain 7–17% Al2 O3 , 4.5–6.5% MgO, and 44–52% Cr2 O3 , with Cr# = 0.82 ± 0.05 (Cr# = Cr/(Cr + Al)). These chromite compositions plot in the field of boninites and island arc tholeiites, well above the field of MORBs as compiled by Portnyagin et al. (1997). As was also concluded by Dick and Bullen (1984), Cr# of greater than 0.7 are indicative of an arc-related rather than a MORB setting. The olivinites consist of densely packed, bladed, and equant tabular olivine Fo86.6–89.1 grains, 1–15 mm in size, elongated in the direction of the schistosity and showing a distinct cleavage coincident with the schistosity of the rocks and the amphibolites hosting the massif. Chromites from the olivinites are relatively high in Cr2 O3 (39–42%), Al2 O3 (5.7– 12%), and MgO (2.7–4.6%), the contents of which, however, are lower than those in the original cumulate peridotites and dunites.

5. MAJOR AND TRACE ELEMENT GEOCHEMISTRY The chemical compositions of mafic-ultramafic volcanic and intrusive rocks are shown in the primitive mantle normalized plots in Fig. 4, and representative analyses are listed in Table 1. Common features of basalts, komatiites and gabbros include LREE-depletions with (La/Sm)N ranging between 0.54 in WK and 0.96 in EKB, and pronounced positive U and Th anomalies in all rocks but ETB (Nb/U = 12–20 in EKB, WK and gabbro and ∼ 28 in ETB as compared to the chondritic value of ∼ 30). The (Nb/Th)N and (Nb/La)N ratios are on average also lower than chondritic (0.9 ± 0.2 and 0.8 ± 0.1, respectively). 5.1. Eastern Komatiitic Basalts Eastern komatiitic basalts have a rather uniform MgO content of 15 ± 1% in the pillow zones of differentiated flows, which are considered to represent the composition of the emplaced lavas. They are substantially depleted in Al and Y relative to Ca and Ti and in

5. Major and Trace Element Geochemistry

413

Fig. 4. Primitive mantle normalized abundances (Hofmann, 1988) of selected major and trace elements in the Olondo volcanic and plutonic rocks. Major and minor elements were determined by X-ray fluorescence spectrometry, REE by ID-TIMS, rest of trace elements by ICPMS.

Gd relative to Yb (Fig. 4), which led Puchtel and Zhuravlev (1993b) to classify them as belonging to the Al-depleted type of komatiitic rocks of Nesbitt et al. (1979). However, the rocks have pronounced negative Zr and Ti anomalies, which are not observed in pristine plume-related Al-depleted komatiites. 5.2. Western Komatiites Western komatiites show nearly chondritic HREE distribution patterns of an Al-undepleted type komatiite (Fig. 4) and were classified as such by Puchtel and Zhuravlev (1993b). At the same time, these rocks have several chemical features that distinguish them from most Archean komatiites. First, they have about two times lower HFSE abundances at a given MgO content compared to typical plume-related komatiites such as those from the Kostomuksha or Abitibi greenstone belts (Arndt, 1986; Puchtel et al., 1998). Second, they are characterized by elevated Cr abundances (up to 4000 ppm), which correlate positively with MgO. This is in contrast to typical komatiites with > 20% MgO, in which Cr behaves as an incompatible element during lava differentiation (e.g., Arndt, 1986). Third, they have high Al/Ti ratios, about 27 on average, due to a relative Ti-depletion rather than Al-enrichment. Finally, they are enriched in SiO2 (up to 51%) compared to typical komatiites with compa-

414

Table 1. Major and trace element data for mafic and ultramafic rocks from the Olondo greenstone belt

Cr Ni Zr Nb Ta Y Th U La Ce Nd Sm Eu Gd Dy Er Yb

07/16 49.0 0.75 7.72 14.6 0.20 13.9 12.9 0.59 0.11 0.17

Eastern tholeiitic basalt 109/1 507/4 104/1 48.8 48.9 49.2 1.17 0.93 1.10 15.3 15.2 14.6 13.4 14.2 14.5 0.19 0.19 0.19 7.49 7.20 7.38 11.4 11.3 11.0 1.75 1.78 1.55 0.26 0.13 0.33 0.24 0.19 0.23

105/1 49.0 1.27 14.8 14.7 0.18 7.05 10.9 1.70 0.20 0.26

106/2 49.5 1.12 15.2 13.8 0.20 7.09 11.1 1.61 0.16 0.23

West komatiite 8569 8551 46.7 49.6 0.29 0.32 7.12 7.21 12.0 11.5 0.18 0.18 27.1 21.5 6.43 9.26 0.10 0.38 0.02 0.04 0.09 0.09

Gabbro 15/1 15/2 49.5 49.1 0.73 0.70 14.5 14.2 12.9 12.5 0.20 0.21 8.45 8.86 11.7 11.8 1.68 2.27 0.10 0.19 0.16 0.18

86226 46.2 0.67 14.6 14.0 0.21 9.70 12.2 1.98 0.22 0.17

Red Hill massif 86245 86371 40.4 41.0 0.08 0.07 1.24 1.24 10.8 11.2 0.16 0.18 46.5 45.8 0.74 0.47 0.01 0.01 0.01 0.01 0.09 0.09

85358 41.6 0.04 0.76 8.44 0.13 47.4 1.55 0.01 0.01 0.05

1692 1249 1320 1207 349 278 263 249 255 3519 2528 408 463 593 2672 5676 2244 518 370 397 356 137 112 120 108 129 1210 1095 131 155 170 3066 2762 3075 47 45 43 46 62 55 64 65 60 14 15 40 37 36 2.15 2.18 1.99 2.11 3.07 2.56 2.40 2.94 2.32 0.350 0.450 1.36 1.59 1.10 0.148 0.143 0.128 0.134 0.190 0.170 0.158 0.187 0.149 0.019 0.025 0.089 0.099 0.071 15 14 13 14 27 22 24 28 25 6 6 17 16 16 0.349 0.326 0.281 0.325 0.332 0.290 0.268 0.312 0.232 0.091 0.102 0.198 0.248 0.166 0.114 0.107 0.092 0.106 0.108 0.095 0.087 0.102 0.076 0.030 0.033 0.065 0.081 0.054 3.29 2.73 2.64 3.14 3.63 3.54 3.78 3.44 2.78 0.347 0.456 1.84 2.01 1.46 0.126 0.184 0.177 8.89 7.65 7.12 8.31 9.90 9.39 10.2 10.1 8.12 1.23 1.41 5.17 5.42 3.86 0.357 0.458 0.447 6.98 6.38 5.68 6.15 7.70 7.20 8.03 8.47 6.92 1.21 1.24 4.40 4.39 3.39 0.274 0.266 0.290 2.19 2.01 1.80 1.77 2.56 2.28 2.68 2.83 2.35 0.459 0.480 1.51 1.49 1.30 0.088 0.071 0.080 0.805 0.733 0.667 0.706 0.913 0.808 0.965 0.985 0.789 0.124 0.137 0.561 0.524 0.602 0.030 0.017 0.033 2.68 2.47 2.21 2.04 3.34 3.02 3.53 3.86 3.23 0.678 0.716 2.17 2.09 1.92 0.119 0.093 0.101 2.81 2.60 2.31 2.04 4.04 3.60 4.27 4.46 3.79 0.907 0.931 2.63 2.61 2.48 0.149 0.121 0.121 1.61 1.46 1.28 1.15 2.53 2.27 2.64 2.78 2.39 0.611 0.626 1.71 1.69 1.58 0.103 0.096 0.080 1.42 1.30 1.14 1.01 2.39 2.10 2.49 2.68 2.28 0.621 0.634 1.65 1.60 1.51 0.115 0.120 0.083

Major elements in wt%, minor and trace elements in ppm. Analyses recalculated on an anhydrous basis.

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

Eastern komatiitic basalt Sample 07/2 07/6 07/9 SiO2 48.2 49.5 50.7 TiO2 0.78 0.78 0.76 A12 O3 7.83 7.93 7.70 Fe2 O3 14.8 13.6 13.4 MnO 0.21 0.19 0.19 MgO 14.9 14.3 14.0 CaO 12.0 12.4 12.7 Na2 O 0.96 0.93 0.28 K2 O 0.12 0.14 0.14 P2 O5 0.24 0.23 0.23

6. Pb and Nd Isotope Systematics

415

rable MgO content. Therefore, though we continue to apply the term ‘komatiite’ to describe this rock type, we no longer consider them to be true komatiitic lavas. 5.3. Eastern Tholeiitic Basalts and Gabbros Eastern tholeiitic basalts and gabbros both have unfractionated HREE patterns similar to those of Archean tholeiites and are thought to have been related by olivine-plagioclaseclinopyroxene fractionation (Puchtel and Zhuravlev, 1993b). Compared to typical Archean tholeiites (Condie, 1985; Arndt, 1991), however, they are higher in Ni, Co and Cr at a given MgO content and overall are substantially depleted in HFSE. 5.4. Dunites and Peridotites Dunites and peridotites from the Red Hill massif have MgO contents ranging between 43 and 50% and were proposed to represent cumulates derived from the Western komatiite parental magma (Puchtel and Zhuravlev, 1993b). REE patterns vary from slightly depleted to enriched for both LREE and HREE. The abundances of moderately to highly incompatible elements are similar to those in residual ophiolitic peridotites (Jaques et al., 1983; Prinzhofer and Allegre, 1985) and are lower than those in ultramafic cumulates of thick differentiated komatiite lava flows (Arndt and Lesher, 1992).

6. PB AND ND ISOTOPE SYSTEMATICS The Sm-Nd and Pb-Pb data for the Olondo mafic-ultramafic rocks are plotted in Figs. 5 and 6. In the Sm-Nd diagram, the data for the lavas (EKB + ETB) yield an isochron with an age of 3006 ± 84 Ma. This age is in good agreement with the results of ion-probe dating of zircons from different parts of the synform (2986 ± 12 to 3005 ± 10 Ma) and corresponds to the time of the eruption of the lavas. All samples have nearly uniform, positive initial εNd values ranging between +2.5 and +2.8. Cumulate-textured dunites and peridotites of the Red Hill massif exhibit nearly chondritic Sm/Nd ratios; the data yield an isochron with an age of 3003 ± 117 Ma, identical to the emplacement age of the lavas. The initial εNd, however, is somewhat lower, at +1.5 ± 0.10 (Fig. 5). Puchtel and Zhuravlev (1993b) interpreted these lower values as a result of hydrothermal alteration of the rocks, which has taken place nearly simultaneously with their emplacement. In the 207 Pb/204Pb vs 206Pb/204Pb diagram, the data for the mafic lavas also define an isochron with a slope corresponding to an age of 3065 ± 73 Ma, which is identical, within analytical uncertainty, to the Sm-Nd and zircon data. The µ1 , or time-integrated 238 U/204 Pb of the source of the magmas, is 8.76 ± 0.10, which is close to the mantle value of 8.75 at 3.0 Ga (Fig. 6). In the 208 Pb/204Pb vs 206 Pb/204Pb space, the data also define a linear trend, from which the Th/U ratio of the rocks is calculated to be 3.1 ± 0.3. This

416

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

Fig. 5. Sm-Nd isochron diagrams for the Olondo volcanic rocks (top) and Red Hill massif dunites and peridotites (bottom). Analyses were carried out at the Max-Planck Institut für Chemie in Mainz using the technique described by Puchtel and Zhuravlev (1993b).

ratio is lower than that of the contemporary plume-derived mafic-ultramafic lavas from several greenstone belts of 3.3–3.4 (e.g., Dupré and Arndt, 1990; Lahaye and Arndt, 1996; Puchtel et al., 1998, 1999, and references therein). It is noteworthy that modern island arc lavas have Th/U ratios of around 2.5 (e.g., Thirlwall et al., 1996; Peate et al., 1997; Ewart et al., 1998). In Fig. 7, compositions of the Olondo lavas are plotted on the diagram εNd vs Nb/Th, where the Nd isotopic compositions are recalculated relative to the hypothetical MOMO (major orogeny-mantle overturn) evolution line of Stein and Hofmann (1994). As can be

7. Re-Os Isotope Systematics and PGE Geochemistry

417

Fig. 6. Pb-Pb evolution diagrams for the Olondo volcanic rocks. The mantle evolution curve was drawn assuming a single-stage model, µ1 value of 8.75, 4.50 Ga as the age of the Earth, and Canyon Diablo values of Tatsumoto et al. (1973) for the starting Pb isotopic composition. Analyses were carried out at the Max-Planck Institut für Chemie in Mainz using the technique described by Puchtel and Zhuravlev (1993b).

seen from the diagram, Eastern TB plot in the field of the Caribbean oceanic plateau basalts though at the lower end of Nb/Th ratios, while WK are associated with island arc volcanic rocks, and EKB and gabbros have intermediate compositions between the two.

7. RE-OS ISOTOPE SYSTEMATICS AND PGE GEOCHEMISTRY Re-Os isotopic analyses were carried out on three bulk rock samples (Red Hill dunites and a peridotite) and two chromite separates, and the data are plotted on the Re-Os diagram in Fig. 8. All samples analyzed have high Os and very low Re abundances, and, thus, low Re/Os ratios, generally less than 0.1. This implies that the age corrections were minimal and have generally little effect on the measured Os isotopic compositions. All samples

418

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

Fig. 7. (Nb/Th)N vs εNd(CHUR)-εNd(MOMO) data. The parameters of the PREMA reservoir are adapted from Wörner et al. (1986). ACC—average Archean continental crust of Rudnick and Fountain (1995); modern primitive mantle and N-MORB estimates are from Hofmann (1988). Nd isotopic compositions of ACC-contaminated rocks calculated at 2.7–2.5 Ga, and those of Olondo and Kostomuksha lavas at 3.0 and 2.8 Ga, respectively. The sources of the data used are available upon request.

plot close to the 3.0 Ga chondritic reference isochron and have an average initial γ 187 Os (3.0 Ga) value of −0.5 ± 0.5. Since the chromites analyzed are magmatic in origin, and the bulk rocks mostly retain primary magmatic textures and mineralogy, this initial γ 187 Os value is considered to reflect the Os isotopic composition of the mantle source of the rocks, even though they do not form an isochron in the Re-Os diagram. This initial value is within the range of Os isotopic compositions (from −5 to +6) obtained by Tsuru et al. (2000) and Walker et al. (1996, 2001) for Proterozoic and Phanerozoic ophiolites, mostly from suprasubduction zone (SSZ) settings. PGE abundances in the Red Hill dunites and peridotites are plotted as CI-chondrite normalized values in Fig. 9. The patterns for the Olondo peridotites are similar to those for mantle rocks from SSZ-ophiolites (e.g., Rehkämper et al., 1997) and differ substantially from olivine cumulates in the lower parts of differentiated komatiite lava flows (e.g., Puchtel and Humayun 2000, 2001) in having higher Os, Ir and Ru contents and lower Pt and Pd abundances.

8. Tectonic Setting of the Olondo Greenstone Belt

419

Fig. 8. Re-Os diagram for the Red Hill dunites, peridotites and chromite separates. The analyses were performed at the Max-Planck-Institut für Chemie in Mainz using the technique described in detail in Puchtel et al. (2001). The line represents the chondritic reference at 3.0 Ga.

Fig. 9. PGE concentrations in the Red Hill dunites and peridotites normalized to CI-chondrite abundances of Anders and Grevesse (1989). The data were obtained using the Carius tube digestion ICPMS technique (Puchtel and Humayun, 2001).

8. TECTONIC SETTING OF THE OLONDO GREENSTONE BELT Several lines of evidence suggest that the Olondo greenstone belt represents a fragment of an ancient suprasubduction zone ophiolite sequence. 1. Submarine supracrustal sequences of the Olondo greenstone belt have linear rather than areal distribution. This favors their origin in a subduction zone environment and not as part of an Archean oceanic plateau. Unlike Phanerozoic ophiolite complexes, which

420

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

include sheeted dike complex, the Olondo sequence is characterized by the presence of a large amount of mafic and ultramafic sills. This may be indicative of a greater magma supply and faster spreading rates in the Archean, as suggested by, e.g., Saunders et al. (1996). 2. Olondo mafic-ultramafic lavas and gabbros have geochemical signatures typical of boninite-like rocks in Phanerozoic SSZ-ophiolite sequences. These include depletions in HFSE and enrichments in Si and Cr at a given MgO content, which point to a mantle source that experienced a previous history of melt depletion, and enrichments in Th and U, typical of subduction-related magmas (e.g., Taylor et al., 1994; Smellie et al., 1995; Portnyagin et al., 1997, and references therein). These features coupled with the LREE- and Nd-Pb depleted mantle-type characteristics are not observed in oceanic plateau or MORB lavas and are unique to SSZ ophiolite sequences. 3. The presence of high-Mg lavas such as the Western komatiites and Eastern komatiitic basalts with boninite-like geochemical and isotopic signatures indicates greater degrees of mantle melting above a subduction zone induced by fluids from a downgoing slab having lowered the melting point. This process is argued to produce SSZ ophiolite complexes with depleted mantle sequences and thicker ultramafic crustal sections (e.g., Pearce et al., 1984), in contrast to dry and less extensive melting beneath oceanic ridges, which produces undepleted mantle lherzolite and no ultramafic crustal sections. 4. PGE distribution patterns in the Red Hill massif dunites and peridotites are typical of mantle rocks from SSZ environments. They were derived from a source previously depleted in incompatible Pt and Pd. 5. The average initial γ Os of −0.5 ± 0.5 for bulk rock dunite-peridotite samples and chromite separates is regarded as reflecting the Os isotopic composition of the Archean convecting upper mantle in the region. These data allow to trace the evolution of the Os isotopic composition of the convecting upper mantle as far back in the Earth history as the middle Archean.

REFERENCES Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoritic and solar. Geochimica et Cosmochimica Acta 53, 197–214. Arndt, N.T., 1986. Differentiation of komatiite flows. Journal of Petrology 27, 279–301. Arndt, N.T., 1991. High Ni in Archaean tholeiites. Tectonophysics 187, 411–420. Arndt, N.T., Lesher, C.M., 1992. Fractionation of REE’s by olivine and the origin of Kambalda komatiites, Western Australia. Geochimica et Cosmochimica Acta 56, 4191–4204. Baadsgaard, H., Nutman, A.P., Samsonov, A.V., 1990. Geochronology of the Olondo greenstone belt. ICOG-7, 6. Condie, K.C., 1985. Secular variation in the composition of basalts: An index to mantle evolution. Journal of Petrology 26, 545–563. Condie, K.C., 1995. Episodic ages of greenstones: A key to mantle dynamics? Geophysical Research Letters 22, 2215–2218.

References

421

Condie, K.C., 1998. Episodic continental growth and supercontinents: A mantle avalanche connection? Earth and Planetary Science Letters 163, 97–108. deWit, M.J., Ashwal, L.D., 1986. Workshop on tectonic evolution of greenstone belts. LPI, Technical Report 86-10, Houston, TX. deWit, M.J., Ashwal, L.D., 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, p. 809. 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. Dobretsov, N.L., Konnikov, E.G., Dobretsov, N.N., 1992. Precambrian ophiolite belts of southern Siberia, Russia, and their metallogeny. Precambrian Research 58, 427–446. Dobretsov, N.N., Popov, N.V., Smelov, A.P., Bogomolova, L.M., Moskovchenko, N.I., Barton, J.M., 1997. The Aldan-Stanovik Shield. In: deWit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 710–725. Dook, V.L., Kitsul, V.I., Petrov, A.F., 1986. The Early Precambrian of South Yakutia. Nauka, Moscow, p. 280. Dook, V.L., Neymark, L.A., Rudnik, V.A., Kröner, A., 1989. The oldest rocks of the Aldan-Stanovik Shield, Eastern Siberia, USSR. Excursion guide for geological field trip to the Aldan-Stanovik Shield. Leningrad, Mainz, p. 121. Drugova, G.M., Bushmin, S.A., Sochava, A.V., Savel’eva, T.E., Kharitonov, A.L., 1985. Trough complexes of the western part of the Aldan Shield. In: Kuznetsov, V.A. (Ed.), Precambrian trough Structures of the BAM Region and Their Metallogeny. Nauka, Novosibirsk, pp. 102–111. Drugova, G.M., Kharitonov, A.L., Bushmin, S.A., Chukhonin, A.P., Shuleshko, I.K., 1983. Structural-Metamorphic Development of the Olondo Complex (E. Siberia). Nauka, Leningrad, pp. 49–65. Drugova, G.M., Puchtel, I.S., Shustova, L.Y., Berezhnaya, N.G., 1988. The Olondo greenstone belt, Aldan Shield. Izvestiya Akademii Nauk SSSR 8, 40–56. Dupré, B., Arndt, N.T., 1990. Pb isotopic composition of Archean komatiites and sulfides. Chemical Geology 85, 35–56. Ewart, A., Collerson, K.D., Regelous, M., Wendt, J.I., Niu, Y., 1998. Geochemical evolution within the Tonga-Kermadec Lau arc back-arc systems: The role of varying mantle wedge composition in space and time. Journal of Petrology 39, 331–368. Hofmann, A.W., 1988. Chemical differentiation of the Earth: The relationship between mantle, continental crust and oceanic crust. Earth and Planetary Science Letters 90, 297–314. Jahn, B.M., Gruau, G., Capdevila, R., Cornichet, J., Nemchin, A., Pidgeon, R., Rudnik, V.A., 1998. Archaean crustal evolution of the Aldan Shield, Siberia: Geochemical and isotopic constraints. Precambrian Research 91, 333–363. Jaques, A.L., Chappell, B.W., Taylor, S.R., 1983. Geochemistry of cumulus peridotites and gabbros from the Marum ophiolite complex, Northern Papua New Guinea. Contributions to Mineralogy and Petrology 83, 154–164. Lahaye, Y., Arndt, N.T., 1996. Alteration of a komatiite flow from Alexo, Ontario. Journal of Petrology 37, 1261–1284. Mironiuk, E.P., Liubimov, V.K., Magnushevsky, E.L., 1971. Geology of the Western Aldan Shield. Nedra, Moscow, p. 237. Nesbitt, R.W., Sun, S.S., Purvis, A.C., 1979. Komatiites: Geochemistry and genesis. Canadian Mineralogist 17, 165–186.

422

Chapter 13: 3.0 Ga Olondo Greenstone Belt in the Aldan Shield, E. Siberia

Nutman, A.P., Chernyshev, I.V., Baadsgaard, H., Smelov, A.P., 1992. The Aldan Shield of Siberia, USSR: The age of its Archean components and evidence for widespread reworking in the midProterozoic. Precambrian Research 54, 195–210. Nutman, A.P., Chernyshev, I.V., Baadsgaard, H., 1990. The Archean Aldan Shield of Siberia, USSR: The search of its oldest rocks and evidence for reworking in the Mid-Proterozoic. In: Third International Archaean Symposium, p. 27. 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 Special Publication, 77–94. Peate, D.W., Pearce, J.A., Hawkesworth, C.J., Colley, H., Edwards, C.M.H., Hirose, K., 1997. Geochemical variations in Vanuatu arc lavas: The role of subducted material and a variable mantle wedge composition. Journal of Petrology 38, 1331–1358. Petrov, A.F., 1985. Tectonic setting of Precambrian greenstone belts in the Aldan Shield structure. In: Kuznetsov, V.A. (Ed.), Precambrian Trough Structures of the BAM Region and Their Metallogeny. Nauka, Novosibirsk, pp. 90–96. Popov, N.V., Dobretsov, N.N., Smelov, A.P., Bogomolova, L.M., 1995. Tectonics, metamorphism and the problems of evolution of the Olondo greenstone belt. Petrology 3, 73–86. Popov, N.V., Smelov, A.P., Dobretsov, N.N., Bogomolova, L.M., Kartavchenko, V.G., 1990. The Olondo Greenstone Belt. Yakutsk Scientific Center, Yakutsk, p. 172. Portnyagin, M.V., Danyushevsky, L.V., Kamenetsky, V.S., 1997. Coexistence of two distinct mantle sources during formation of ophiolites: A case study of primitive pillow lavas from the lowest part of the volcanic section of the Troodos ophiolite, Cyprus. Contributions to Mineralogy and Petrology 128, 287–301. Prinzhofer, A., Allegre, C.J., 1985. Residual peridotites and mechanisms of partial melting. Earth and Planetary Science Letters 74, 251–265. Puchtel, I.S., 1992. Petrology of mafic-ultramafic rocks and evolution of the crust-mantle system in the early Precambrian of the Olekma gneiss-greenstone terrain, Aldan Shield. Ph.D. thesis. Institute of Ore Deposit Geology, Petrology, Mineralogy and Geochemistry, Moscow, p. 256. Puchtel, I.S., Bogatikov, O.A., Simon, A.K., 1993. The Early Precambrian crust-mantle evolution of the Olekma gneiss-greenstone terrane, Aldan Shield. Petrology 1, 451–473. Puchtel, I.S., Brügmann, G.E., Hofmann, A.W., 2001. 187 Os-enriched domain in an Archaean mantle plume: Evidence from 2.8 Ga komatiites of the Kostomuksha greenstone belt, NW Baltic Shield. Earth and Planetary Science Letters 186, 513–526. Puchtel, I.S., Hofmann, A.W., Mezger, K., Jochum, K.P., Shchipansky, A.A., Samsonov, A.V., 1998. Oceanic plateau model for continental crustal growth in the Archaean: A case study from the Kostomuksha greenstone belt, NW Baltic Shield. Earth and Planetary Science Letters 155, 57– 74. Puchtel, I.S., Hofmann, A.W., Amelin, Y.V., Garbe-Schönberg, C.-D., Samsonov, A.V., Shchipansky, A.A., 1999. Combined mantle plume-island arc model for the formation of the 2.9 Ga SumozeroKenozero greenstone belt, SE Baltic Shield: Isotope and trace element constraints. Geochimica et Cosmochimica Acta 63, 3579–3595. Puchtel, I.S., Humayun, M., 2000. Platinum group elements in Kostomuksha komatiites and basalts: Implications for oceanic crust recycling and core-mantle interaction. Geochimica et Cosmochimica Acta 64, 4227–4242. Puchtel, I.S., Humayun, M., 2001. Platinum group element fractionation in a komatiitic basalt lava lake. Geochimica et Cosmochimica Acta 65, 2979–2993.

References

423

Puchtel, I.S., Zhuravlev, D.Z., 1993a. Nd-isotope systematics and petrogenesis of the early Proterozoic picrites in the Olekma granite-greenstone region. Geochemistry International 30, 37–49. Puchtel, I.S., Zhuravlev, D.Z., 1993b. Petrology of mafic-ultramafic metavolcanics and related rocks from the Olondo greenstone belt, Aldan Shield. Petrology 1, 308–348. Rehkämper, M., Halliday, A.N., Barfod, D., Fitton, J.G., 1997. Platinum-group element abundance patterns in different mantle environments. Science 278, 1595–1598. Rudnick, R.L., Fountain, D.M., 1995. Nature and composition of the continental crust: A lower crustal perspective. Reviews of Geophysics 33, 267–309. Saunders, A.D., Tarney, J., Kerr, A.C., Kent, R.W., 1996. The formation and fate of large oceanic igneous provinces. Lithos 37, 81–95. Smellie, J.L., Stone, P., Evans, J., 1995. Petrogenesis of boninites in the Ordovician Ballantrae Complex ophiolite, southwestern Scotland. Journal of Volcanology and Geothermal Research 69, 323– 342. Smelov, A.P., 1989. Metamorphic Evolution of the Olekma Granite-Greenstone Terrain. Nauka, Novosibirsk, p. 128. Stein, M., Hofmann, A.W., 1994. Mantle plumes and episodic crustal growth. Nature 372, 63–68. Tatsumoto, M., Knight, R.J., Allègre, C.J., 1973. Time differences in the formation of meteorites as determined from the ratio of lead 207 to lead 206. Science 180, 1279–1283. Taylor, R.N., Nesbitt, R.W., Vidal, P., Harmon, R.S., Auvray, B., Croudace, I.W., 1994. Mineralogy, chemistry, and genesis of the boninite series volcanics, Chichijima, Bonin-Islands, Japan. Journal of Petrology 35, 577–617. Thirlwall, M.F., Graham, A.M., Arculus, R.J., Harmon, R.S., Macpherson, C.G., 1996. Resolution of the effects of crustal assimilation, sediment subduction, and fluid transport in island arc magmas: Pb-Sr-Nd-O isotope geochemistry of Grenada, Lesser Antilles. Geochimica et Cosmochimica Acta 60, 4785–4810. Tsuru, A., Walker, R.J., Kontinen, A., Peltonen, P., Hanski, E., 2000. Re-Os isotopic systematics of the 1.95 Ga Jormua Ophiolite Complex, northeastern Finland. Chemical Geology 164, 123–141. Wörner, G., Zindler, A., Staudigel, H., Schmincke, H.-U., 1986. Sr, Nd, and Pb isotope geochemistry of Tertiary and Quaternary alkaline volcanics from West Germany. Earth and Planetary Science Letters 79, 107–119. Walker, R.J., Hanski, E., Vuollo, J., Liipo, J., 1996. The Os isotopic composition of Proterozoic upper mantle: Evidence for chondritic upper mantle from the Outokumpu ophiolite, Finland. Earth and Planetary Science Letters 141, 161–173. Walker, R.J., Prichard, H.M., Ishivatari, A., Pimentel, M., 2001. The osmium isotopic composition of the convecting upper mantle deduced from ophiolite chromites. Geochimica et Cosmochimica Acta 66, 329–345.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Published by Elsevier B.V.

425

Chapter 14

2.8 GA BONINITE-HOSTING PARTIAL SUPRASUBDUCTION ZONE OPHIOLITE SEQUENCES FROM THE NORTH KARELIAN GREENSTONE BELT, NE BALTIC SHIELD, RUSSIA A.A. SHCHIPANSKYa , A.V. SAMSONOVb , E.V. BIBIKOVAc , I.I. BABARINAa , A.N. KONILOVa , K.A. KRYLOVa,1 , A.I. SLABUNOVd AND M.M. BOGINAb a Geological Institute

of RAS, Pyzhevsky per., 7, Moscow, 109017, Russia of Ore Deposits, Petrography, Geochemistry and Mineralogy of RAS, Staromonetny per., 35, Moscow, 109017, Russia c Vernadsky Institute of Geochemistry and Analytical Chemistry of RAS, Kosygin st., 19, Moscow, 17975, Russia d Karelian Research Center of RAS, Institute of Geology, Pushkinskaya st., 11, Petrozavodsk, 185610, Karelia, Russia b Institute

Neoarchean subduction-related assemblages of the North Karelian greenstone belt,, in the NE part of the Baltic Shield, Russia, contain the world’s oldest known boninite series, occurring in at least in two areas of the belt. The first area, referred to here as the Khizovaara structure, shows evidence of a late Archean ocean-island volcanic arc collage formed during two tectonic episodes nearly 2.8 Ga ago. The second area, named the Iringora structure, preserves distinctive features of an ophiolite pseudostratigraphy, including not only gabbro and lava units, but also remnants of a sheeted dike complex. The major and trace element chemistry of the Iringora ophiolitic gabbro, dike and lava units suggests a comagmatic series with a continuous compositional variation from more primitive mafic to strictly boninitic melts. In terms of major- and trace element abundance, the boninite series of the North Karelian greenstone belt is practically indistinguishable from the Group I and II of the Troodos upper pillow lavas. These occurrences strongly suggest that Neoarchean subduction-related processes including boninite-hosting supra-subduction zone ophiolites have not changed substantially over the past 2.8 Ga.

1 Present address: Department of Geological and Environment Sciences, Stanford University, CA 94305-2115,

USA. DOI: 10.1016/S0166-2635(04)13014-4

426

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

1. INTRODUCTION Many ophiolites with arc geochemical signatures are thought to have formed as oceanictype crust above subduction zones. These have been termed suprasubduction zone (SSZ) ophiolites (Pearce et al., 1984). The latter are recognized frequently in Phanerozoic orogenic belts, but seldom if ever in the early Precambrian. Until recently (Kusky et al., 2001) there were no examples of Archean mafic-ultramafic sequences for which one could consider the most likely origin to be oceanic (Bickle et al., 1994). In addition to that, boninitic rocks have also not been reported for the Archean (e.g., Sylvester et al., 1997; see also Polat and Kerrich, 2004). Amongst all rock types, boninites, perhaps, are the only rocks which are almost exclusively associated with ophiolites, but not necessarily vice versa (Cameron et al., 1979). Boninitic rocks, or boninite series (Meijer, 1980) volcanic suites sensu lato, are commonly believed to be unusually sensitive indicators of supra-subduction zone mantle wedge processes (Crawford et al., 1989). They are unique because they have appropriate major element chemical compositions thought to be primary magmas from partial melting of the peridotite mantle wedge above the downgoing lithospheric slab (Fallon and Crawford, 1991). Thus the occurrence of boninite lavas in ancient arc-related assemblages has important tectonic implications. One possible explanation for lack of Archean boninitic volcanism is that higher heat flow in the Archean may have caused subducted oceanic crust to have partially melted, rather than forming dehydrating reactions characteristic of modern subduction zones (e.g., McCulloch, 1993). If so, Archean subduction zones could have been flat and slabs been largely disaggregated in the upper mantle at depths less than 200 km (Martin, 1986; Abbott et al., 1994). Recently, the few occurrences of boninites have been described from the Paleoproterozoic (Poidevin, 1994; Wyman, 1999). Fan and Kerrich (1997) and Kerrich et al. (1998) have reported a 2.7 Ga low-Ti tholeiite association of boninite series akin to ophiolitic basalts of Sun and Nesbitt (1978). Among these examples only the 1.9 Ga Amisk boninite series of the Trans-Hudson Orogen in Canada occurs in an ophiolite-like setting (Wyman, 1999). The older boninite series of the Bogoin belt (2.3 Ga) and of the Abitibi belt (2.7 Ga) are associated with komatiites which are absent from Phanerozoic SSZ ophiolites. This, in turn, suggests a possible mantle plume input in the origin of those occurrences (Poidevin, 1994; Kerrich et al., 1998). This paper reports detailed data for 2.8 Ga boninitic occurrences from the North Karelian greenstone belt (Russia) including the Khizovaara and Iringora Structures (Shchipansky, 1999, 2001). These display geological and geochemical traits similar with some well-known Phanerozoic SSZ ophiolites providing unique insight for understanding crustforming processes from early Earth. 2. GEOLOGY The Archean domain of the Baltic (Fennoscandian) shield traditionally is divided into three distinct parts (Fig. 1). The northern, Lapland-Kola province mainly consists

2. Geology

427

Fig. 1. Main tectonic units in the eastern part of the Baltic shield showing the location of the Khizovaara and Iringora Structures. Abbreviation: NKGB = North Karelian greenstone belts.

428

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

of several previously dispersed Archean crustal terranes that together with the different Paleoproterozoic belts have been involved in a collisional-type orogeny at 2.0 to 1.9 Ga (Bridgwater et al., 2001). A central, NW-trending segment known as the Belomorian mobile belt is occupied by assemblages of gneisses and amphibolites. This part of the Baltic Shield has experienced two major orogenic periods, in the Neoarchean and Paleoproterozoic. The Neoarchean period included several crust-forming events between c. 2.9 and 2.7 Ga which can be interpreted in terms of first subduction-related and later collisional orogeny (e.g., Gaál and Gorbatschev, 1987; Bibikova et al., 1996). In the end of the Paleoproterozoic (c. 1.9–1.8 Ga ago) strong tectonothermal reworking occurred during an event of crustal stacking and thrusting referred to as the Svecofennian orogeny which was caused by other thrusting of Lapland granulite belt onto the Belomorian belt (e.g., Bibikova et al., 1996, 2001; Bogdanova, 1996). Although the Svecofennian high-grade metamorphism and folding affected all of the belt, its major Neoarchean crustal structure reveals that early thrust and fold nappes developed c. 2.74– 2.70 Ga (Miller and Mil’kevich, 1995). In contrast, the Karelian province displays no isotopic evidence for strong Paleoproterozoic reworking (Bibikova et al., 2001). The Karelian craton forms the core of the shield and largely consists of volcanic and sedimentary rocks (greenstones) and granites/gneisses that formed between 3200 and 2600 Ma and were dominantly metamorphosed at low-grade (Lobach-Zhuchenko et al., 1993; Sorjonen-Ward et al., 1997). Local synformal patches of Paleoproterozoic (2.45 to 1.9 Ga) volcano-sedimentary rocks unconformably overlie the Karelian basement. To the southwest of the Archean Karelian craton, the Svecofennian domain represents a large portion of Paleoproterozoic crust developed between 2.0 and 1.75 Ga (Gaál and Gorbatschev, 1987). Although tectonic settings of the Karelian Archean greenstone belts are still a matter of debate, there are some indications for subduction-accretion processes that had operated, at least, since c. 2.9 Ga (e.g., Puchtel et al., 1998, 1999). However, a large involvement of deep mantle-plume derived oceanic plateaus into Archean crustal growth processes remains questionable in a respect of subduction style. New lines of evidence on Archean subduction-related environments come from the North Karelian greenstone belt. 2.1. North Karelian Greenstone Belt The North Karelian greenstone belt (= NKGB for brevity) extends for 300 km along the boundary between the Karelian granite-greenstone terrain and the Belomorian mobile belt (Fig. 1). It is comprised of several greenstone belts isolated each from other by granitoids or unexposed areas. All belts preserve similar lithologies and structural styles. Thus they represent the fragments of a single Neoarchean unimodal greenstone belt (Kozhevnikov, 1992, 2000). U-Pb dating of titanites and rutiles shows that the NKGB is located near the inside boundary of the Svecofennian frontal thrusting directed toward the Karelian craton (Bibikova et al., 2001). Superimposed events include medium- to high-grade metamorphism of the moderate- to high-pressure facies series. However, only minor crustal additions accompanied Paleoproterozoic tectonism within the Karelian-Belomorian border zone. In contrast to the Belomorian mobile belt, here there is no evidence for em-

2. Geology

429

placement of Paleoproterozoic granites or magmatites. Post-Archean igneous activity is restricted to scarce swarms of gabbro-dioritic dikes and slightly dismembered gabbronoritic sills. These are age equivalent to c. 2.45–2.40 Ga mafic intrusions widely distributed throughout both the Belomorian and Karelian parts of the Baltic Shield (Amelin et al., 1995). The entire Belomorian belt strikingly lacks the extensive Paleoproterozoic volcano-sedimentary cover that characterizes substantial parts of the Karelian craton. This cover could be wholly eroded from the Svecofennian mountain belt thus exposing a deeper crustal level that its Karelian neighbour (Bogdanova, 1996). The only remnant of the Paleoproterozoic supracrustal sequence is preserved within the Kukasozero belt that extends along the Karelian-Belomorian border zone. The belt provides important information on the Paleoproterozoic structural development (Babarina, 1998) that has been largely used in this study to assess its overprinting effect onto Neoarchean structural patterns of the NKGB. The NKGB has been the subject of comprehensive field studies that define the basic geological and geochemical characteristics of the belt, interpreted to be a Neoarchean accretionary orogen (Kozhevnikov, 1992, 2000). All unit terms used in this study are modified from Kozhevnikov (2000). 2.2. Khizovaara Structure The Khizovaara greenstone structure is a small but well exposed part of the NKGB. It is made up dominantly of a variety of volcanic rocks bordered on both sides by tonalitic intrusions. All of the Khizovaara rocks have experienced polyphase deformation and lowamphibolite facies metamorphism, but the deformation is heterogeneous and volcanic textures are occasionally preserved, so igneous nomenclature is used. The prefix meta is implicit below. Its structural pattern appears to be an imbricate south-dipping homocline clearly displaying the contrasting lithologic assemblages (Fig. 2). The northern flank essentially consists of a sediment-free mafic volcanic sequence referred to as the Northern assemblage, whereas the southern flank largely displays acid-felsic volcanic and volcanoclastic rocks combined into the Southern assemblage. In addition, local piles of the so-called, upper pillowed tholeiitic basalts overlie unconformably (Thurston and Kozhevnikov, 2000) or are thrust imbricated on top of the acid-felsic volcanogenic assemblage. The tectonostratigraphy of the entire Khizovaara sequence is summarized in Fig. 3. Northern lithotectonic assemblage. This can be subdivided into four volcanogenic units based on differences in outcrop appearance, petrography and geochemistry. The northernmost and likely oldest volcanic unit in the Khizovaara Structure outcrops as a sequence of volcanics and gabbro sills of IAT-like affinity. Rarely preserved pillow textures and the absence of interflow sediments suggest their submarine and probable deep-water origin. Two separate small bodies of highly deformed peridotites are located at the base of the unit (Kozhevnikov, 1992) but their relationships with the lower tholeiites are unknown. Boninite series volcanics structurally and stratigraphically (?) overlie the lower tholeiites. The boundary between the two units is abrupt and is marked by a sudden appearance of primitive low-Ti high-Mg tholeiites. These rocks are described in detail below, because of

430

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 2. Geological map of the Khizovaara Structure (compiled by authors using data of Kozhevnikov, 1992, 2000, and Thurston and Kozhevnikov, 2000).

their paucity in the Archean terrains. Farther south and structurally higher still, yet another mafic volcanic unit consists of ferro-tholeiitic basalts strongly enriched in incompatible trace element abundances compared to the lower volcanic units. Its contacts with both the underlying and overlying units are tectonic and marked by occurrence of the relatively wide (up to 20–30 m) zones of highly deformed rocks unusually rich in garnet. Furthermore, there is no evidence that the intrusive rocks related to the ferro-tholeiitic basalts,

Fig. 3. Tectonostratigraphy, U-Pb zircon ages and main geochemical characteristics of the Khizovaara sequence. Arrows indicate the thrust boundaries. Abbreviations: PM, primitive mantle abundance; N-MORB, normal mid-ocean ridge basalt abundance after Hofmann (1988).

2. Geology

431

432

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 4. NORDSIM ionprobe result for the northern andesite (sample 7/96) from the Khizovaara structure.

i.e., sills or dikes, take place within other volcanic units. It may indicate that originally they formed a separate volcanic pile which was consequently incorporated into a single imbricate stack during later accretionary events. Numerous tonalitic dikes cutting all units of the lower mafic assemblage have yielded an U-Pb discordant age of 2803 ± 35 Ma (Kozhevnikov, 1992). These dikes obviously derive from the northern tonalitic intrusion which shows a similar U-Pb age of 2804 ± 27 Ma. Yet another group of volcanic rocks called the Northern andesites completes the lower lithotectonic assemblage. This includes mostly high-Mg andesite and subordinate andesitedacite metavolcanics clearly distinguished from the Southern counterparts both by their field occurrence and unusual geochemical characteristics. Macroscopically they are amygdaloidal, massive and partly coarse, fragmental acid metavolcanic rocks developed in a subaeral-subaqueous environment (Thurston and Kozhevnikov, 2000). Ionprobe U-Pb dating of a homogeneous magmatic zircon from the Northern andesite yields a concordant age of 2783 ± 10 Ma (Fig. 4). It is interesting that due to deformational and metamorphic transformations the quartz and quartz-plagioclase amygdales broke into roughly cylindrical, pencil-like crystal aggregates thus providing good strain markers. Based on a wide range of strain determination made on the amygdaloidal andesites, Kozhevnikov (2000) has assessed that constrictional strains range between 300% and 680%. Southern lithotectonic assemblage. The contact of the Southern lithotectonic assemblage with the underlying Northern assemblage is largely obscured by subsequent Svecofennian overthrusting events. This assemblage largely consists of both wedge- and slabderived volcanogenic rocks developed in a mature island arc setting. These comprise

2. Geology

433

a variety of andesite-dacite-rhyolite lava flows accompanied by fine- to coarse-grained volcano-sedimentary rocks. A number of semi-simultaneous dacitic and rhyolitic dykes and sills are scattered throughout the effusive facies volcanics. The concordant U-Pb age of 2778 ± 21 Ma for zircons from the Southern felsic lavas defines the time of their emplacement. Volcano-sedimentary rocks dominantly consist of interbedded siliciclastic rocks displaying in some places moderate sorting and graded bedding with aluminaenriched upper parts, all indicative for sediments deposited by turbidity currents (Thurston and Kozhevnikov, 2000). Minor but prominent components of the volcano-sedimentary sequence include several thin horizons of carbonaceous schists and quartz arenites. The latter have recently been studied in detail by Thurston and Kozhevnikov (2000) and interpreted to have formed in a shallow water environment during the late stage of arc volcanism. The upper tholeiite unit is included in the Southern lithotectonic assemblage based only in its positions in the thrust stack. It is composed of pillowed and massive lava flows of tholeiitic basalt and andesite-basalt compositions accompanied by co-magmatic gabbro sills. As opposed to the lower tholeiites, they display flat trace element patterns without negative Nb anomalies and tend to be geochemically similar to N-MORB basalts. Amongst the basalts there are also rare thin bodies of picritic rocks of unclear affinity. The Southern lithotectonic assemblage is flanked to the south by a wide shear zone along the diorite-granodiorite pluton contact. In contrast to the age determination on the felsic volcanics, it displays a somewhat older U-Pb upper-intercept zircon age of 2826 ± 18 Ma indicating a possible contribution of an older crustal component. A few intrusions of coarse-grained gneissose feldspar granites post-date all the above Khizovaara volcanic and igneous suites. Structural development. As shown in Fig. 2, the Khizovaara greenstone sequence comprises steeply dipping and thin-skinned shortened volcanogenic assemblages displaying structural patterns typical of an imbricate fold-thrust wedge (e.g., Kusky and Vearncombe, 1997). Moreover the Southern lithotectonic assemblage was subjected to south-directed overthrusting onto the Northern assemblage. The latter event and subsequent intense shearing along the greenstone margins was obviously due to the Svecofennian stacking indicated by the Khizovaara titanite and rutile U-Pb ages of 1750 Ma (Bibikova et al., 2001). Because of the complex and superimposed Svecofennian deformations accompanied by the relatively high temperature metamorphism, recognition of earlier structural development is problematical. Nevertheless Kozhevnikov (1992) has found some indications of earlier tectonic fabrics and identified three deformation stages for the Archean structural development of the Khizovaara sequence. The earliest D1 stage is found in the Northern lithotectonic assemblage and includes the earliest thrust faulting predating the emplacement of the tonalitic dikes. Second-generation or D2 structures are well exposed in the Southern acid-felsic volcanogenic assemblage. They are dominated by coeval mesoscopic- to the map-scale EW and NE-trending closed folds possibly associated with thrust faults. The D3 stage includes at least four folding events collectively characterized by development of NE to NW-trending upright folds.

434

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

2.3. Boninite Series: Field Occurrence, Petrography and Mineral Chemistry To the author’s knowledge, medium- to high-grade metamorphosed boninites have not formerly been petrologically described in Archean domains, or in younger regions in general. Thus it is of particular interest to describe the unique Khizovaara metaboninitic rocks in detail. The Khizovaara boninite series rocks are exposed as a sequence up to 50 m thick (Fig. 2). Its present thickness obviously is not primary because it is strongly stretched parallel to layering and flattened perpendicular to layering. Kozhevnikov (1992, 2000) estimates that the original thickness of this unit could be at least 4–5 times thicker than the present thickness. The boninite series metavolcanics appear as distinct pistachio-green amphibolite in the field. They include the more abundant low-Ti (TiO2 < 0.5%) high-Mg metatholeiites and the more evolved, strictly metaboninitic rocks, collectively forming a coherent low-Ti volcanic series akin to the high-Ca boninite series of some well-known Phanerozoic counterparts (e.g., Cameron, 1985; Smellie et al., 1995). Notice that it is impossible to distinguish between low-Ti tholeiitic and boninitic amphibolites based on macroscopic field appearance alone. Yet the most evolved members of the boninite series containing up to 60 wt% SiO2 show no clear visual differences with the low-Ti high-Mg amphibolites. Thus, based on limited sampling, these amphibolites were previously described as komatiitic and komatiitic basalt lavas interpreted to have formed in a mafic plateau volcanic setting (Thurston and Kozhevnikov, 2000). However they do not display indications for the presence of amphibole or olivine spinifex textures, typical of volcanic rocks of komatiitic affinity. Only scarce signs of primary volcanic textures exist, including remnants of small pillows and thin-bedded interflow (?) horizons of the same composition, perhaps representing a highly stretched hyaloclastitic material. In addition, there are only a few of stretched-out sheet-like bodies that are distinctly different from other units by their lightgrey color discordant in relationships with the metalavas. They are interpreted, on both the field and geochemical grounds, as remnants of late felsic dikes of boninitic affinity. Thus we emphasize that metaboninitic rocks of dominantly lava facies are preserved within the Khizovaara sequence. The majority of collected samples shows a mineral assemblage of hornblende-chloriteplagioclase ± quartz indicative of medium-grade regional metamorphism. However the samples of boninitic affinity display an unusual mineral assemblage which can be termed as staurolite-bearing amphibolite. The petrography and mineral chemistry of the samples studied by electron probe is summarized in Table 1. As the data clearly suggest, rocks of bulk low-Ti high-Mg tholeiitic compositions are characterized by a plagioclase- and staurolite-free assemblage whereas more evolved rocks exhibit plagioclase-, stauroliteand staurolite-garnet-bearing parageneses. It is significant also that both plagioclase and amphibole compositions are mostly heterogeneous (Table 2). The hornblende (Hbl) and the tschermakitic hornblende (HblTs ) are distinctly distinguished chemically by concentrations of Al2 O3 . Microprobe profiles across plagioclases show their compositional variability from anorthite (PlAn ) in cores to andesine (Pl) in rims. Notice that the plagioclase in

2. Geology

435

Table 1. Mineral assemblages of the boninite series metavolcanics studied by microprobe Sample H-333 H-334 H-326 H-321 H-335 H-332

XRock Fe 0.282 0.312 0.349 0.375 0.388 0.381

Main rock-forming minerals Hbl, Cum, Chl Hbl, Pl, Qtz Hbl, Chl, Pl, Qtz, Cal St, Hbl, Chl, Pl, Qtz St, Cum, (Ged), Hbl, Chl, Pl, Qtz, (Mica) Grt, St, Cum, (Ged, Hbl), Chl, Pl, Qtz, Ky

Opaques

Ilm Rt, Ilm, Prt, Pnt Rt, Prt, Pnt, Cpy

Notes: XRock = molecular ration of Fe/(Fe + Mg). Chemical rock compositions are listed in Table 3. Mineral Fe abbreviations are after Kretz (1983). Mica, K-Ca di-octahedral mica. Minerals in trace constituted are shown in parenthesis.

the paragenesis with staurolite is unzoned anorthite (An0.947). Most staurolite and garnet grains are homogeneous, although a limited amount of retrograde zoning is revealed on some garnet rims compared to cores. The above data on zoning of minerals coupled together with their textural patterns suggests that at least two distinct metamorphic episodes (M1 and M2 , respectively) formed the observed mineral assemblages of the Khizovaara boninite series. Staurolite is commonly thought of as a mineral indicative of metapelitic rocks. Recently the possibility of staurolite equilibriums in amphibolites has been theoretically proven by Arnold et al. (2000). Moscovchenko and Turchenko (1975) first discovered and described unusual staurolite-bearing amphibolites in the Iringora Structure. Then Fed’kin (1975) made detailed equilibrium studies of these rocks and assessed their geothermobarometry. However, until now, the origin and significance of these unusual St-bearing amphibolites remained unclear. Staurolite in the Khizovaara metamorphosed boninites occurs as subhedral grains within a chlorite-amphibole matrix (Fig. 5). It commonly displays resorption relationships with anorthitic plagioclase, and in many examples almost enclosed by the latter (Figs. 5d, e). Fine-grained inclusions of rutile and quartz are also characteristic of the staurolite grains. Apparent equilibrium textures among garnet and staurolite (Fig. 5a), staurolite and rutile, and intergrowths of rutile and gedrite (Fig. 5c) suggest that the all these minerals plus the high-alumina hornblende formed in response to an early metamorphic event (M1 ). Relationships between staurolite and kyanite remain ambiguous because they rarely occur in close proximity (Fig. 5b). However the kyanite is commonly replaced by the M2 -stage chlorite. This in turn suggests that the kyanite may have formed in response to the M1 event. In order to decipher a multi-stage metamorphic history of Khizovaara boninite series it is useful to compare the available data on mineral and whole rock chemistry. For this purpose we have plotted both the whole rock compositions of the boninite series and the chemical compositions of the minerals obtained by microprobe technique on both AFM and CFM ternary diagrams (Fig. 6). The model composition of the tholeiite used in the theoretical approach of Arnold et al. (2000) is also shown for reference. As can be seen

H-321

Mineral n SiO2 TiO2 Al2 O3 Cr2 O3 FeO* MnO MgO CaO Na2 O K2 O ZnO Total

St 3 27.97 0.63 52.12 1.80 11.63 0.32 2.16 0.06 0 0 1.02 97.71

Si AlIV AlVI Ti Cr Fe2+ Mn Mg Ca Na K Zn Cations X

H-335 Hbl 4 46.20 0.55 12.15 0.27 12.41 0.32 12.62 10.87 1.36 0.15 0 96.90

HblTs 3 43.52 0.44 14.74 0.49 12.73 0.22 11.09 10.96 1.34 0.23 0 95.77

Chl 2 26.72 0.05 21.56 0.32 17.75 0.13 21.24 0.06 0 0 0 87.84

8.738 0.068 0.202 1.384 0.039 0.458 0.010 0 0 0.107 14.984

6.744 1.256 0.835 0.060 0.031 1.515 0.040 2.745 1.700 0.385 0.029 0 15.341

6.462 1.538 1.042 0.049 0.058 1.580 0.028 2.455 1.743 0.387 0.044 0 15.385

2.697 1.303 1.262 0.004 0.025 1.498 0.011 3.196 0.007 0 0 0 10.004

2.561 1.422 – 0.009 0 0.020 0 0 0.422 0.569 0.003 0 5.005

2.068 1.931 – 0 0.006 0.006 0.003 0 0.924 0.047 0.004 0 4.989

0.751

0.356

0.392

0.319

0.425

0.947

3.979 –

Pl Rim to Hbl 57.42 0.26 27.05 0.00 0.54 0 0 8.84 6.58 0.04 0 100.74

PlAn Rim to St 44.58 0.01 35.33 0.16 0.14 0.06 0 18.60 0.53 0.06 0 99.48

Hbl 3 46.33 0.42 13.68 0.59 12.48 0.15 12.37 10.88 1.03 0.08 0 98.00

HblTs 4 45.34 0.38 15.74 0.52 12.01 0.18 11.76 10.85 1.15 0.08 0 98.00

Cum Rim to Pl 54.38 0.06 1.00 0 19.46 0.55 20.08 0.78 0.01 0 0 96.32

8.625 0.092 0.141 1.534 0.026 0.481 0.004 0.090 0.002 0 15.033

6.665 1.335 0.984 0.046 0.067 1.501 0.018 2.652 1.678 0.287 0.014 0 15.247

6.509 1.491 1.173 0.041 0.059 1.442 0.022 2.517 1.669 0.320 0.014 0 15.256

7.909 0.091 0.081 0.007 0 2.367 0.068 4.353 0.122 0.003 0 0 15.000

0.761

0.361

0.364

St 5 28.77 0.87 52.14 1.27 13.07 0.22 2.30 0.03 0.33 0.01 0 99.00 4.038 –

Chl 2 25.90 0.16 20.73 0.49 17.53 0.01 20.12 0.08 0.09 0 0 85.11 2.704 1.296 1.254 0.013 0.040 1.530 0.001 3.131 0.009 0.018 0 0 9.997

0.352 0.328 (continued on next page)

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Sample

436

Table 2. Representative electron probe average analyses of minerals of the Khizovaara staurolite-bearing metaboninites

Sample

H-332

Mineral n SiO2 TiO2 Al2 O3 Cr2 O3 FeO* MnO MgO CaO Na2 O K2 O Total

GrtAlm Rim to Qtz 37.84 0 20.49 0.11 32.53 1.71 4.34 2.56 0 0 99.59

Si AlIV AlVI Ti Cr Fe2+ Mn Mg Ca Na K Cations X

St 10 28.32 0.77 51.81 1.20 12.51 0.08 2.88 0.03 0.12 0.02 97.74

Hbl Core 43.00 0.23 16.69 0.33 11.51 0.12 11.34 10.47 1.31 0.26 95.26

Cum Rim to Pl 55.21 0.01 0.99 0.20 19.86 0.22 21.29 0.30 0.11 0.01 98.21

Ged Core 51.26 0.18 5.42 0.25 18.88 0.21 19.68 0.40 0.41 0 96.69

1.933 0 0.007 2.177 0.116 0.518 0.220 0 0 8.001

2.993 0.007 1.928 0.001 0.010 1.888 0.087 0.846 0.272 0 0.002 8.034

4.014 – 8.637 0.082 0.134 1.483 0.009 0.609 0.005 0.033 0.003 15.010

6.359 1.641 1.269 0.026 0.038 1.424 0.015 2.500 1.658 0.375 0.049 15.353

7.868 0.132 0.035 0.001 0.022 2.366 0.027 4.523 0.046 0.031 0.001 15.052

7.425 0.575 0.350 0.020 0.028 2.287 0.025 4.249 0.062 0.114 0 15.136

2.674 1.326 1.473 0.006 0.054 1.104 0.003 3.529 0 0 0.005 9.988

0.808

0.691

0.709

0.363

0.343

0.350

0.238

3.029 –

Grt 8 38.47 0.02 21.10 0.17 29.01 1.32 7.29 3.26 0 0.02 100.66

Chl – 26.80 0.08 22.22 0.68 13.23 0.04 23.72 0 0 0.04 86.80

PlAn Core 45.87 0.12 33.27 0.02 0.32 0.07 0 17.35 1.45 0.03 98.50

Pl 4 56.71 0.03 26.25 0.08 0.27 0.02 0 8.88 6.49 0.05 98.78

Ky 3 37.60 0.04 60.88 0.73 0.23 0.03 0 0.02 0 0 99.52

0.004 0.001 0.013 0.003 0 0.869 0.131 0.002 5.000

2.577 1.406 – 0.001 0.003 0.010 0.001 0 0.432 0.572 0.003 5.005

1.022 – 1.950 0.001 0.016 0.005 0.001 0 0.001 0 0 2.995

0.867

0.429

2.145 1.833 –

2. Geology

Table 2. (Continued)

–

437

Notes: n, numbers of analyses; X = Ca/(Ca + Na + K) for plagioclase and X = Fe/(Fe + Mg) for Fe-Mg minerals. Mineral formulae are calculated on oxygen basis: 14 (chlorite), 12 (garnet), 23.5 (staurolite), 23 (amphiboles), 8 (plagioclase) and 5 (kyanite).

438

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 5. Microphotographs of Khizovaara St-bearing amphibolites (metaboninites) in the samples H-332 (a–c), H-321 (d–e) and H-335 (f) illustrating textural patterns of the first and second stage mineral assemblages. (a) Boxed area marks a direct contact between Grt and St; note also inclusions of rutile (dark grey) and quartz (light grey) in staurolite. (b) Relationship between St and Ky. (c) Textural patterns of M1 assemblage (St-PlAn -Ged) and M2 assemblage (Cum-Chl). (d) Relics of St- PlAn assemblage within the amphibole matrix. (e) 3 times enlarged part of microphotograph (d). (f) Encircled area marks small relics of St-PlAn assemblage. Open nicols. Typical width of microphotograph field is about 1 mm.

2. Geology

439

from Fig. 6a, the model composition (composition A from the Table 2 of Arnold et al., 2000) fits the neighbourhood of the whole rock trend of Khizovaara boninite series. Fig. 6b illustrates a good accordance between the whole rock compositions and the mineral chemistry. The major element composition of the sample H-333 plotted on the molecular normative FeO-MgO-CaO-Al2O3 tetrahedron falls on the same plane that characterizes the chemical compositions of the homogeneous mineral phases. Because the sample H-333 represents a paragenesis of the least ferruginous minerals, in shifting a bulk rock composition towards the M apex the given mineral assemblage will be kept, but mineral phases will be consistently reduced in iron content. The observed textural features indicate the second-stage (M2 ) formation of the Hbl-Cum-Chl assemblage (Fig. 5f). Some St-free samples occupy the same locus on the AFM/CFM diagrams as St-bearing ones (Sample H-326, Table 1, Figs. 6a, c). This can be explained by a complete replacement of staurolite by later mineral phases. The locus of the mineral chemistry and whole-rock composition points on Fig. 6c suggest that a first-stage assemblage included a mineral equilibrium of tschermackitic hornblende, anorthite and some putative Ca-free amphibole (gedride). Hereinafter, these were partly replaced by a second-stage mineral assemblage involving chlorite, hornblende with lower Al contents and more sodic plagioclase (Table 2) as compared with the primary metamorphic minerals. It is of interest that a transformation of the M1 assemblage was accompanied by an emergence of calcite that forms inclusions within the secondary-stage hornblende. This suggests that the bulk alkalinity of the boninite series rocks was almost unmodified. Thus we can infer an isochemical character of the second-stage metamorphism. A series of compatibility diagrams to illustrate the mineral parageneses for St-bearing samples (= strictly boninitic in a chemical composition) of the first- and second-stage metamorphic episodes is shown on Figs. 6d–f. It should be particularly emphasized that the mineral parageneses of sample H-332 represents a boninitic dacite in chemical composition (Fig. 6f). The seven-mineral assemblage (St-Grt-Hbl-HblTs-Pl-Ky-Qtz) is univariant in the NCFMASH system that has been used by Arnold et al. (2000) to constraint thermodynamic fields of staurolite stability in staurolite-bearing amphibolites. It thus appears that across the thin section studied there are slight variations of the mineral chemical compositions that have given rise to a formation of several divariant mineral assemblages. For instance, in the sample H-332 the staurolite is the most rich in magnesium (Table 2). According to a petrogenetic grid of NCFMASH system (see Fig. 3 from Arnold et al., 2000) a divariant assemblage of St-Grt-Hbl-HblTs-Pl-Qtz is stable within the kyanite field in a range of 620– 650 ◦ C at pressure of 6.4–7.8 kbar. To constrain T-conditions under which the observed two-stage mineral assemblages were formed we tried to use a version of the Hbl-Pl geothermometer (Holland and Blundy, 1994). However due to a high rate of anorthitic end-member in the first-stage plagioclase a majority of the analyzed samples shows unrealistically high elevated temperatures. Equilibrium study on the St-free sample H-334 has shown two distinct groups of hornblende having the Al2 O3 contents of 14.6 and 11.6 wt% which are associated with two distinct groups of plagioclase having the anorthite mole fractions of 0.688 and 0.419. Taking into account the pressure determination of about 6 kbar obtained earlier by Fed’kin (1975),

440

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

(a)

(b)

(c) Fig. 6. Series of ternary diagrams illustrating a compatibility exit between major element chemistries of the observed mineral assemblages (± quartz) of and available data on the whole rock compositions. (a) Overall tholeiitic trend of Khizovaara boninite series. (b, c) Compatibility diagrams for the both Pl and St free assemblage (b) and the St-free assemblage (c). (d, f) Phase diagrams for St-bearing assemblages. For detail see explanation in the text. Symbols are: M1 , mineral assemblage; M2 , mineral assemblage.

2. Geology

(d)

(e)

(f) Fig. 6. (Continued.)

441

442

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

these two groups of parageneses correspond to the temperatures of 667 and 579 ◦ C for M1 and M2 metamorphic episodes, respectively. In the St-bearing sample H-321 the compositions of the second-stage paragenesis Hbl + Pl (see Table 2) constraint a T-parameter for the M2 metamorphism as being as high as 598 ◦ C. Only one sample (H-332) is suitable to use as a staurolite-garnet geothermometer. The different thermodynamic models of staurolite-garnet equilibrium show nearly identical values of M1 temperature: 692 ◦ C (Perchuk, 1989), 670 ◦ C (Fed’kin and Aranovich, 1990) and 726 ◦ C (Koch-Müller, 1997). These are in a good accordance with the above results of Hbl-Pl thermometry. It worth noting that the variability in Al content of hornblende mentioned above was detected throughout the analyzed samples. This obviously suggests that the distinct hornblendes, i.e., Hbl and HblTs, were formed subsequently during two metamorphic events that clearly differed in P-T conditions. Nevertheless, as has been argued by Stein and Dietl (2001), an absence of mineral phases to stabilize a hornblende chemical composition precludes using the Al-in-hornblende geobarometer. The same sample H-332 is suitable only to apply the garnet-plagioclase-kyanite-quartz paragenesis (Table 2) for assessing metamorphic pressures by using the model of Newton and Haselton (1981) modified by Koziol and Newton (1989). Allowing for kyanite equilibrium both in the M1 and M2 stages, the compositions of garnet and plagioclase appropriate to these stages yield the pressure values of 7.6 and 6.1 kbar for the M1 and M2 temperatures of 670 and 580 ◦ C, respectively. In summary the above P-T values constraining the two distinct tectonothermal events to have formed the Khizovaara metamorphosed boninite series are in agreement with the available structural and isotope-geochronological data. The first-stage mineral assemblages indicative of Barrovian-type metamorphism were developed in response to a Neoarchean orogenic/accretionary event between 2.8 and 2.7 Ga. The second-stage mineral assemblages of moderate pressure metamorphism were formed in the Paleoproterozoic during the Svecofennian tectonothermal reworking at 1.9–1.75 Ga ago as determined by U-Pb dating of titanites (Bibikova et al., 2001). 2.4. Iringora Structure The Iringora Structure is located about 100 km NW of the Khizovaara Structure. It is composed dominantly of the same lithotectonic assemblage that occurs within the latter, except for the Khizovaara’s Northern andesite units. Here we have found the same boninite series displaying an ophiolite-like pseudostratigraphy unique to the Archean. The Iringora ophiolite-like sequence occurs within a thrust stack gently plunging to the N-NE. This stack preserves several imbricate slices of c. 2.8 Ga mafic volcanic rocks, including both the boninite series and the upper tholeiites of MORB-like affinity, overthrust onto the tectonically juxtaposed complex of arc-derived acid-felsic volcanics and siliciclastic turbidites (Fig. 7). Thus, the above mentioned complex of the arc-trench/forearc system represents an autochthon for the Iringora ophiolite-like sequence. Although the entire Iringora sequence has been intensely deformed and metamorphosed, primary igneous, volcanic and sedimentary features fortunately are locally well preserved. The best

2. Geology

443

Fig. 7. Geological map of the Iringora Structure.

preserved fragment of the ophiolite-like sequence occurs along the north shore of Lake Irinozero where not only gabbroic and lava units, but also remnants of a sheeted dike complex, including half dikes and cusps of dikes, i.e., a transition of dikes into the lava unit, are exposed. Moreover, at the base of the ophiolitic nappe there is a complex of tectonic mélange metamorphosed in the mode of a metamorphic sole. Boninite-bearing ophiolite-like assemblage. The structurally lowermost slice of the Iringora ophiolite-sequence consists of an up to 500 m thick pile of sheeted dikes and well-preserved lava complexes overthrust, in turn, by slices of the ophiolitic gabbro and

444

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

the upper tholeiites (Fig. 8). Owing to the prominent volcanic textures, the lava unit is the most easily identified. Lava unit. Although the thickness of the lava units along the north shore of Lake Irinozero is no more than 100–150 m, the preservation of primary volcanic textures is striking. The lava unit comprises dominantly both pillowed and massive flows. Pillows are a fairly uniform size commonly nearly 40–50 cm in two-dimensional exposures, while locally larger (up to 1 m) as well as smaller pillowed textures occur. Individual massive flows are from 2 to 3 m in thickness. They are visibly outlined by a cumulate zone at the bottom and by autobreccia at the top of each flow. Yet another striking feature of the lava unit is a preservation of mafic pillow breccia and pillow clusters. The breccia consists of fragments of pillow lava in a finer-grained, fragmental, and shard-rich matrix, displaying in some cases the classic pattern of hyaloclastic breccia. In such cases the clasts are commonly between 5 and 20 cm in diameter. This indicates that the cooling units experienced contact with water and were erupted in a submarine environment. Moreover, the whole Iringora lava unit displays no evidence for interflow sediments or visible vesicular textured pillows suggesting that it could have been erupted in a deep-water environment. Aside from the uniquely preserved primary volcanic textures, the rocks of the Iringora lava unit are identical in their petrography and mineral compositions to the Khizovaara boninite series. Dyke unit. Uniformly layered medium- to fine-grained rocks underlying the lava unit are interpreted as a dense swarm of tholeiitic and boninitic dikes, i.e., a sheeted dike complex. The latter differs from the foliated gabbro in dike width (average 40–50 cm) and texture displaying in some places evidence for screens of earlier crust. In the field occurrence the largest portion of this unit appears somewhat different from either the lava unit or the gabbro unit. However, there are a few outcrops within the unit where preserved textural features displaying a sheeted dike affinity do occur. A single outcrop exhibits occurrences of preserved chilled margins indicating the development of dike-within-dike relationships (Fig. 9). Furthermore, the transition from sheeted dikes to volcanics is also preserved in a few localities where the sheeted dikes cut roughly perpendicular the overlying volcanics (Figs. 8b, 10). It is crucial also that rocks of boninitic composition are found both in the lava and in the dike units. Gabbro unit. Medium- to coarse-grained, compositionally layered mafic rocks of the Iringora sequence are interpreted as a gabbro unit, which appears to be the deepest level of ophiolite exposed. The detailed mapping of the Iringora ophiolite-like sequence has shown that the gabbro unit occurs structurally higher than the lava unit and was overthrusted onto the latter. Only a few small bodies of gabbro are exposed along with the dyke unit possibly representing screens of earlier crust. Due to a strong schistosity, primary gabbro textures such as modally graded layers are preserved in only a few localities of low strain. The principal schistosity (S2 ) is sub-parallel to compositional layering and is interpreted

Fig. 8. Detailed geological map of the north shore of Lake Irinozero (a), equal area, lower hemisphere stereographic projections for poles to the D1 and D2 fold axial planes and outcrop sketches (b–d) showing some principal fabric elements of the Iringora ophiolitic sequence.

2. Geology

445

446

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 9. Field photograph of the Iringora sheeted dike complex with asymmetric chilled margins of the same polarity indicating dike-within-dike relationships. The arrow shows the spreading direction. (A) Chilled margins and (B) central parts of the dikes.

as transposition of primary layering into the S2 plane during the main Neoarchean (D2 ) deformation and metamorphism phase. Although there is no direct evidence for a ‘normal’ ophiolitic relationship between the gabbro unit and the presumably overlying sheeted dike complex, it is crucial that several observed dikes were mapped to be intrusive into the gabbro unit in some places. Coupled together with the compositional features of the gabbro unit, this may imply that initially the latter was directly related to the overlying dike and lava units. The available geochemical data on the Iringora gabbro unit suggest without doubt that it has the same parent magmas as both the volcanic and dike unit. The dike unit is underlain by tectonic mélange (Fig. 8). The basal contact represents a décollement zone, or a floor thrust in terms of the geometry of duplex structures (Moores and Twiss, 1995). The gabbro unit occurs structurally higher, together with the upper tholeiites in roof thrusts. This, in turn, may imply that the décollement zone along which the ophiolitic rocks were emplaced may have propagated toward the top of the ophiolitic sequence leading to the development of the imbricate stack during an accretionary event.

2. Geology

447

Fig. 10. Field photograph of the preserved transition zone between the Iringora dike and lava units. It is clearly visible that the dike cuts roughly perpendicular the volcanic pillow. Light patches on the rock exposure surface are outgrowths of lichen.

Tectonic mélange and metamorphic sole. At the base of the ophiolitic nappe there is a sheared chaotic rock unit without discernible stratification, of up to 300 m thickness, interpreted as a tectonic mélange. The tectonic mélange consists of internally fragmented rock bodies containing a mixture of native (i.e., rocks of the boninite series) and exotic (Fe-Ti basalts) blocks in a volcano-sedimentary matrix. In contrast to the Khizovaara sequence, along the entire Iringora sequence the Fe-Ti basalts occur only as ‘facoids’ within the tectonic mélange. Native blocks are represented by strongly recrystallized coarse-grained low-Ti high-Mg amphibolites. Blocks range in size from a few millimeters to, at least, several tens of meters. Because of intense shortening and shearing, blocks tend to be flattened and lens-shaped. In many cases exotic blocks truncate the ophiolitic blocks and vice versa. The propagation of shear zones into the lensoidal blocks resulted in further internal fragmentation, chipping off smaller fragments (Fig. 8d). The matrix is composed of petrographically diverse garnet-, staurolite-, kyanite-, and amphibole-bearing micaschists with lenses of subarkosic sandstone and rarely carbonaceous schist. At the bottom of the tectonic mélange the matrix displays some evidence that it is partially derived from the underlying arc volcanic and volcano-clastic rocks. To the north of the Irinozero lake shore, the matrix grades into less mature red-colored medium- to thick-bedded siliciclastic schists of

448

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

turbiditic affinity. These data suggest that the ophiolitic nappe was thrust onto the accreted assemblage of the mature arc and arc-derived trench or forearc slope turbiditic sediments. The uppermost 100 m of the tectonic mélange is composed of garnet-staurolite-kyaniteplagioclase schists and garnet amphibolites displaying an otherwise coarse-grained porphyroblastic texture relative to both the overlying ophiolitic and underlying volcanoclastic rocks. It should be noted that such unusual coarse-grained rocks containing high P-T minerals occur only along with the footwall of the ophiolitic nappe. These grade rapidly downward into kyanite-free and medium- to fine-grained rocks. Microprobe studies of these rocks have shown that the core compositions of garnet-biotite-hornblende assemblage record a T-parameter as high as 700 ◦ C. Unfortunately we have still not detected in thin section a suitable mineral paragenesis to assess a P-parameter with accuracy. However a geological position of this coarse-grained assemblage that is unusually rich in kyanite and garnet suggests that it could be developed under high-PT conditions (> 7 kbar, at least, and about 700 ◦ C) all indicative of the metamorphic sole of a thick and hot ophiolitic nappe. Upper tholeiites. The field appearance and major- and trace geochemistry of the Iringora upper tholeiites show that this unit is the same as that within the Khizovaara Structure. It is composed of pillowed and massive lava flows of basalt to andesitic basalt of MORB affinity accompanied by co-magmatic gabbro sills. The Iringora upper tholeiites occur tectonically both above the lava unit and the gabbro unit of the SSZ ophiolitic sequence, thus implying that an early thrusting event took place during accretion of an arc-trench system and imbrication of the ophiolitic sequence. Arc-trench assemblage. The southern and western segments of the Iringora Structure are composed predominantly of acid-felsic volcanic and volcanoclastic rocks whereas the NW flank reveals a thick pile of siliciclastic sediments. Farther west and south-west still, two elongated dome-like granodiorite massifs (not shown) are interpreted in terms of major and trace geochemistry as intrusive analogues to the acid-felsic volcanic rocks. Due to large deformational and metamorphic transformations, most parts of the volcanic rocks lost primary textures and are geochemically and mineralogically identifiable as being of calc-alkaline to adakite affinity. The few outcrops of this unit suggest that these originally were eruptive rocks developed as lava flows, tuffs and flow breccias and agglomerates ranging in composition from andesites through dacite to rhyodacites. Among the younger volcanic products, dacitic to rhyodacitic dykes and shallow-level sills are particularly common. Some dykes or small dome-like bodies display silica-rich rhyolitic compositions. In addition, relatively small sill- and dyke-like bodies scattered throughout the arc-derived volcanics are ‘normal’-Ti (approximately 1 wt%) and some are high-Al gabbros. Along the contact with the tholeiite-ophiolite imbricate stack the arc volcanics are interlayered with compositionally similar sandstones. Based both on the field appearance and geochemical characteristics, it seems obvious that the Iringora calc-alkaline volcanics are close analogues of the Khizovaara Southern lithotectonic assemblage. Moreover zircons from a dacite subvolcanic body cutting the Iringora arc volcanics yield a U-Pb discordant age of 2782±9 Ma obtained by conventional method (Fig. 11), i.e., showing within the analytical error the same age as the Khizovaara arc volcanics.

2. Geology

449

Fig. 11. Concordia diagram showing plots of zircon fractions from the Iringora dacite (sample 2772-3).

Sedimentary rocks occupy the NE flank of the Iringora Structure. They are characteristically red-colored contrasting to the light-grey colored arc volcanics. Although in some cases sedimentary rocks are interbedded with the arc volcanics, the bulk of the thick sedimentary pile occurs as a fault-bounded sedimentary domain. In spite of the fact that primary sedimentary structures such as parallel bedding, cross-bedding, and grading are intensely deformed throughout the domain, their turbiditic affinity is clear. These sedimentary rocks consist mainly of thick- to medium bedded greywacke-mudstone pairs and are characteristic of sediments deposited by turbidity currents. Sandstones tend to be more abundant and thicker than siltstone and shale pairs. Locally preserved features such as grading, sole marks and cut-and-fill structures indicates that many, if not all, sediments are proximal turbidites derived from nearby volcanic highlands and deposited in a forearc/trench slope setting. Volumetrically, greywackes of andesitic and rhyodacitic composition prevail, but basaltic greywackes tend to be more abundant in the upper level of the sedimentary pile. Some turbiditic outcrops show occurrences of relatively thin (20–40 m) slices of ampibolites both of ‘normal’-Ti and high-Ti tholeiite compositions which can be interpreted as scraped-off sections of oceanic crust in the Iringora accretionary prism. Geochemically, the ‘normal’-Ti amphibolites are similar to the Khizovaara lower tholeiites, whereas the high-Ti amphibolites are akin to the Khizovaara ferro-tholeiitic basalts. Structural development. The present structural framework of the Iringora greenstones is largely due to an overprinting effect attributed to several subsequent deformational phases developed both in the Neoarchean and Paleoproterozoic. In order to decipher the Neoarchean structural development, as a first step, we must assess the succession and style of Paleoproterozoic refolding. For this purpose, we have used the data previously attained on the Kukasozero belt (Babarina, 1998) which is located about 15 km W of the Iringora Structure and involves a thick sequence of Paleoproterozoic (Karelian) volcano-

450

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

sedimentary rocks. The deformation history of the Paleoproterozoic rocks includes at least two folding stages. During the first stage (D3 ), they were affected by northward-trending recumbent folding of eastward-trending vergence and associated thrust faulting. During the second stage (D4 ) a regional thrust plane formed. This was reworked into NS-NEtrending upright open folds which, during the progressive contractional deformation, developed along with their antiformal planar axis into the closed tight folds or the fault zones. The latter represent a sort of layer-parallel faulting formed as a result of ‘rock squeezing’ between the adjacent synformal counterparts (Fig. 12). With respect to the main Paleoproterozoic deformational patterns, the Neoarchean Iringora sequence occurs within the synformal domain, i.e., within the domain of lower Paleoproterozoic strain. That is the reason that some important primary igneous textures indicating an ophiolitic setting were preserved within the study areas. Two early nappe-style deformational events are documented in the Neoarchean Iringora sequence. Although the largest portion of the Iringora greenstones underwent obviously intense transformation during the D2 and subsequent Paleoproterozoic D3 and D4 deformation phases, nevertheless some unique evidence of the earliest D1 ophiolitic thrusting is preserved along the northern shore of Irinozero Lake (Figs. 8b, c). The major ophiolite nappe boundary is distinctively marked by the tectonic mélange and the metamorphic sole. In addition, the earliest foliations and mineral stretching lineations occasionally appear within the best preserved and primary igneous textured lava unit. The orientation of the D1 fold hinge and axial planes was modified during the subsequent D2 deformational phase crucial in development of the Neoarchean structural pattern. Second-generation structures are dominated by several map-scale ENE to EW trending thrust zones defining a duplex geometry of the entire Iringora ophiolitic sequence. Thrust zones are typically from 0.5 m to few meters thick and are defined by development of mylonitic fabrics indicating intense shearing along thrust boundaries. These imbricate thrusts generate their own minor folds with westward vergence. It should be noted that several thrusts have been identified, but extensive inverted sequences resulting from recumbent folding are not common. These imbricates wrap around Iringora mountain, consisting of relatively thin slices of arc-derived volcanics, tectonic mélange, largely dismembered ophiolitic rocks, upper tholeiites and siliciclastic turbidites. Furthermore, syn-tectonic faultbounded panels of red to pinkish granite were emplaced into the volcanic and volcanoclastic rocks semi-simultaneously with the D2 deformation phase. Compared to the Paleoproterozoic structural development, the Neoarchean phase was accomplished by progressive west-vergent tectonic transportation leading to a regional shortening, crustal thickening and Barrovian-type metamorphism.

3. GEOCHEMISTRY Analytical methods. Chemical studies were undertaken on 15 g aliquots from 400 to 500 g of sample, reduced with a jaw crusher and ground twice in a corundum mill. Major elements were determined by XRF on fused glass pellets using a VRA-20R apparatus at

3. Geochemistry 451

Fig. 12. Summary and sketch of the structural development of the Iringora SSZ ophiolite.

452

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

the United Institute of Ecology, Geophysics, and Mineralogy, Novosibirsk, Russia. All Fe is done as Fe2 O3 . The trace elements were analyzed by inductively coupled plasma spectrometry (ICP-MS) in the Mendeleev Analytical Center, St. Petersburg. Accuracies are estimated as follows: ∼ 2% for major elements, ∼ 5% for trace element. Sm-Nd isotope investigations were carried out at the Max-Planck-Institut für Chemie in Mainz by Dr. I. Puchtel following the technique described by Puchtel et al. (1997). The procedures used for the conventional U-Pb age determination of the zircons were similar to those described by Bibikova et al. (1996). Zircons from the sample 7/96 (the Khizovaara Northern andesite) were studied at the Isotopic Laboratory of Naturhistoriska Riskmusset in Stockholm (the Swedish Royal Museum of Natural History) using the ion microprobe Cameca 1270 NORDSIM. The analytical procedure is described in Whitehouse et al. (1997). The analytical data obtained to constrain the above U-Pb ages both of Khizovaara and Iringora rocks will be published elsewhere. 3.1. Geochemistry of the Boninite Series and Related Mafic Rocks Chemical effects of alteration. All analyzed samples have experienced repeated metamorphism and likely prior submarine hydrothermal alteration. Consequently, no primary minerals were preserved in the NKGB sequences. From the sophisticated point of view one could expect that primary igneous compositions of the analyzed rocks were largely disturbed, as well. Of particular interest is the case of metamorphic rocks with geochemical boninitic affinity because the presence of staurolite and quartz along with these rocks suggests excesses of Al, Si over alkalis. Among other reasons this could probably result from a sea-floor hydrothermal alteration that enriched rocks in Mg and removed alkalis. Thus, a major problem in the chemical characterization of the NKGB metamorphosed boninite series and the interpretation of its petrological evolution is to identify and assess the effects of alteration, and to distinguish them from igneous processes. In order to resolve this problem, as a first step, we sampled rocks from outcrops displaying no evidence for late-stage tectonic fabrics. Then, coherency of the geochemistry was considered in conjunction with detailed petrographic observation for the identification of least altered samples. Based on a large number of studies in Archean greenstone belts, there is a broad consensus that Al, Ti, the HFS elements (Th, Nb, Ta, Zr, Hf, Y), the REE (excepting Eu) and the transitional elements (Cr, Co, Ni, Sc, V) are relatively immobile during, at least, low-grade metamorphic alteration (e.g., Condie et al., 1977; Sun and Nesbitt, 1978; Ludden et al., 1982; Shervais, 1982). In contrast, Ca, Na, K, Rb, Sr, Ba and Cs could be generally mobile. However, most of the alteration studies for Archean metavolcanics have been done on komatiites and related rocks. Poidevin (1994) has shown that the Paleoproterozoic Bogoin boninites affected by high greenschist metamorphism have kept magmatic affinities for most of the major elements (except Na and perhaps K) whereas their trace element abundances are very similar to recent boninites. We tried to assess a possible degree of alteration of the NKGB boninite series using a combination of distinct geochemical and petrological approaches. It is worth noticing that

3. Geochemistry

453

a simple application of Bi-variate diagrams for rocks of boninite series to evaluate possible effects of their post-magmatic alteration faces problems. This can stem from a number of processes involved in their petrogenesis including: melting processes, complex paths of fractional crystallization, different mantle source and metasomatizing components, or some combination of these processes. Despite this complexity, the problem can be resolved if a major- and trace-element dataset is coupled with that from unmetamorphosed possible analogues, as outlined below. First of all, it needs to be ascertained that the NKGB rocks of boninitic affinity, i.e., MgO > 8%, SiO2 > 52% and TiO2 < 0.5% (Le Bas, 2000), are not products of largescale post-magmatic addition of SiO2 (up to 15%) into the high-Mg and less siliceous counterparts. In comparing the low-Ti tholeiitic and strictly boninitic compositions on the mantle-normalized diagrams, both compositional groups show no marked differences in their trace element patterns (Fig. 13). This suggests the preservation of magmatic trace element values rather than their disturbance due to a post-magmatic silicification. Some differences can be seen for the third group, including that the most evolved rocks occur as remnants of felsic dykes. However the pronounced negative Eu anomalies and some more fractionated LREE compared to the less evolved groups are attributable to a strong fractionation of plagioclase and, perhaps, amphibole. It should be noted that all the three compositional groups bear the pronounced positive Sr anomalies. This regularity may be of particular interest taking into account two distinct circumstances. One the one hand, Sr is commonly regarded to be mobile during both hydrothermal and metamorphic alteration processes. One the other hand, a large positive Sr anomaly is one of the crucial features for identification of subduction-related volcanics from recent as well as ancient arcs (e.g., Pearce et al., 1984; McCulloch and Gamble, 1991). The recent study on the determination of trace element partition coefficients between garnet, clinopyroxene and hydrous basaltic liquids mostly pertinent to processes in subduction zone has shown that clinopyroxene-dominated fractionation may produce a positive Sr spike in melt from magma without anomalous Sr contents generated in the spinel lherzolite field (Green et al., 2000). This result provides an independent way to assess effects of alterations for Sr and CaO in the NKGB boninite series. To do this, we need to choose two pairs of element ratios mainly controlled by clinopyroxene fractionation. These are CaO/Al2 O3 and Sr/Ce, because Cpx fractionation forces both the removal of CaO relative to Al2 O3 and the progressive enrichment in Sr relative to the neighbouring MREE. However, it is widely accepted that boninite series compositions are controlled by subsequent crystal/liquid fractionation from a boninite parental magma. The order of crystallization is generally chrome spinel-olivine-both pyroxenes-plagioclaseamphibole-magnetite (e.g., Ohnenstetter and Brown, 1996). If such complexity exists, it is important to isolate more evolved liquids which could be also controlled by crystal fractionation of plagioclase. As shown on the Bi-diagrams (see Figs. 15, 16), the Al2 O3 content increases until Mg# (= Mg/(Mg + Fe2+) ∼ 0.70, while Fe2 O3 decreases, reflecting the late crystallization of plagioclase in this series. In the rocks with Mg# < 0.70 Al2 O3 decreases rapidly, reflecting the crystallization of clinopyroxene together with plagioclase. Thus, the rocks with Mg# > 0.70 or MgO > ∼10 wt% are included in the further proceeding. Finally,

454

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 13. Trace-element spidergrams for NKGB boninites and low-Ti tholeiite lavas normalized to primitive mantle values of Hofmann (1988).

3. Geochemistry

455

Fig. 14. Plot of CaO/Al2 O3 versus Sr/Cepm for selected samples in the NKGB boninite series. See explanation in the text.

the plot CaO/Al2 O3 versus Sr/Cepm demonstrates a good correlation (r = −0.74) indicating that the Sr content was mainly controlled by clinopyroxene fractionation (Fig. 14). The two analyzed layered gabbro samples from the Iringora sequence are a good match for the deduced fractionation trend. This, in turn, suggests that the CaO and Sr contents in the analyzed samples essentially reflect their igneous affinities while CaO could be mobile to some extent in those rocks during the sea-floor alteration of the glass. The relative immobility of CaO in boninites series in such process may be because a large fraction of CaO resides in the unaltered groundmass clinopyroxene, rather than in (albitised) plagioclase laths as in the tholeiites (Meffre et al., 1996). In addition, considering the other evidence pointing toward a boninitic affinity for the NKGB metamorphosed rocks given below, a large post-magmatic alteration for clear resemblance with some unmetamorphozed boninite series in term of most of the major- and trace elements seems incredible. This is not to say that there was no post-magmatic disturbance at all. For example, NaO, K2 O, Ba, Rb, Cs display scatters fields on the plots versus incompatible elements (not shown), that are likely attributable to hydrothermal or metamorphic remobilization. Fractionation history. Distinctive geological and geochemical characteristics of the NKGB mafic volcanic rocks have permitted a subdivision into four suites (Fig. 3): lower tholeiites, boninite series, Fe-Ti tholeiites and upper tholeiites. TiO2 coupled with MgO or Mg# values are particularly attractive for individualization of different rock-types. The considerable geochemical diversity between these volcanic suites in terms of their major and trace element compositions is illustrated in Fig. 15, in which the fields of the three late Paleozoic volcanic suites combined into the Koh SSZ ophiolite sequence, New Caledonia

456

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 15. Mg# variation diagrams for selected major and trace elements showing the three mafic volcanic units from the Khizovaara Northern lithotectonic assemblage. Symbols are as follows: rhombuses, lower tholeiites; white stars and black stars, boninites and associated low-Ti high-Mg basalts, respectively; black circles, Fe-Ti tholeiites. Fields of the Koh SSZ ophiolite (after Meffre et al., 1996) are shown for reference. Abbreviations: KLT = Koh lower tholeiites; KB = Koh boninites; KUT = Koh upper tholeiites. Mg# is calculated as the molecular ratio of Mg/Mg + Fe2+ . Thin dotted lines show the limit boundaries for the IUGS definition of boninites (Le Bas, 2000).

3. Geochemistry

457

Fig. 16. Major and trace element variation in the NKGB boninite series with MgO (on a volatile-free basis) compared with the composition of Troodos upper pillow lavas (field outlined by solid curve). Fields of three geochemical groups (I–III, respectively) of Troodos UPL are also shown by dotted curves. Fields of Troodos UPL are taken from data compiled by Sobolev et al. (1993) and Cameron (1985). The encircled roman numerals are predicted compositions for the distinct geochemical groups of Troodos parental magmas after Sobolev et al. (1993).

458

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

(Meffre et al., 1996) are also shown for reference. It is notable that both the stratigraphic superposition and the geochemical features of the mafic volcanic rocks in the Khizovaara Northern lithotectonic assemblage are principally similar to those of the Koh ophiolite. This study focused on the NKGB boninite series volcanics because it forms the partially preserved ophiolitic succession of the Iringora Structure. On the other hand, rocks of boninitic affinity are extremely rare in the Early Precambrian. Major and trace geochemistry of the Khizovaara and Iringora boninite series is presented in Tables 3 and 4, respectively. Rocks of boninite series from these two localities are almost geochemically identical. Two continuous rock groups, boninitic and low-Ti tholeiite, can be distinguished based on selected element abundances. The volumetrically dominant low-Ti tholeiites fit in a chemically primitive (Mg# = 0.72–0.78) group of mantle melts strongly depleted in incompatible elements. The group identified as strictly boninitic, including high-Mg hypersthene-normative dacites, is characterized by volatile-free recalculated SiO2 = 53.4– 66.2 wt%, MgO = 8.4–13.1 wt%, Mg# = 0.65–0.74, and TiO2 = 0.34–0.48 wt%. The relatively high CaO (up to 12.3 wt%) contents and the high CaO/Al2 O3 values (mostly 0.8–1.1) suggest that they may belong to a high-Ca boninite series (Crawford et al., 1989). The major and trace element geochemical variation of the NKGB boninite series is presented in Fig. 16 where they are compared with the well-known late Cretaceous Upper Pillow Lavas (UPL) of the Troodos ophiolite, Cyprus. It is striking that the compositional range of the 2.8 Ga boninite series volcanics matches the compositional fields of the much younger primitive boninite series lavas. Moreover, in terms of trace element abundance the NKGB boninite series rocks are practically indistinguishable from the geochemical groups I and II (Cameron, 1985) or the groups A and B (Flower and Levine, 1987) of the Troodos UPL (Fig. 17). The sole marked exception is the higher FeO* contents in the NKGB boninite and related volcanic series as compared with the younger counterparts (Figs. 15, 16). Thus we can infer that a fractional history of parental magma to form the whole stem of the NKGB boninite series was very similar to those postulated for the groups I and II which constitute the major portion of the Troodos ophiolite (Cameron, 1985). Indeed, the NKGB boninite series volcanics display high abundances of Cr, Ni and Co which vary from 51 to 79 ppm Co, 187–452 ppm Ni, 324–1680 ppm Cr. The early rapid depletion of Cr indicates that chromite was an important early fractionating mineral (Fig. 15). A rapid decrease in Ni with advancing fractionation and increasing of SiO2 suggests initial crystallization of olivine down to Mg# ∼ 0.70 (Fig. 16). As has been emphasized above, these were followed by crystallization of pyroxenes followed, in turn, by plagioclase at Mg# < 0.70. Note that the crystallization of clinopyroxene before plagioclase in these rocks is likely only in a magma with high SiO2 /Al2 O3 and low CaO/Al2 O3 ratios, such as rocks of boninitic affinity (Natland, 1981). However, plagioclase appears on the liquidus until a MgO value of ∼ 10% in the NKGB evolving magma, whereas the Troodos counterpart shows a more evolved MgO value of 6% (Cameron, 1985). True boninite series or contaminated/altered high-Mg rocks? There is a great ambiguity regarding the use of the term “rocks of boninitic affinity” in Archean geology. Crawford et al. (1989) note that very similar major-element compositions exist between modern highCa boninites and some previously reported Archean rocks of boninitic affinity, but the

Sample SiO2 TiO2 Al2 O3 Fe2 O∗3 MnO MgO CaO Na2 O K2 O P 2 O5 LOI Total Mg# Cr Ni Co Sc V Ba Rb Sr

101/1 UM 41.25 0.25 6.18 11.54 0.16 26.35 4.32 0 0.02 0.10 9.82

X-102 BON 59.33 0.34 9.36 9.21 0.16 8.31 10.04 0.63 0.14 0.12 2.59

X-104 LOTI 45.69 0.47 10.11 13.37 0.20 15.08 12.28 0.70 0.08 0.14 1.95

X-126/1 BON 59.84 0.37 9.67 10.12 0.18 8.22 9.43 0.38 0.08 0.12 1.74

X-128 LOTI 44.41 0.33 9.37 12.10 0.19 18.92 10.23 0.28 0.05 0.12 3.88

X-130 BON 55.12 0.40 11.02 11.17 0.17 10.34 10.01 0.70 0.09 0.13 0.93

H-302 LOTI 44.31 0.38 9.03 12.77 0.18 20.23 7.82 0.29 0.04 0.06 4.94

H-305 LOTI 44.78 0.37 8.86 12.31 0.19 16.67 11.36 0.53 0.07 0.06 4.85

H-319 LOTI 49.99 0.48 11.78 12.45 0.18 13.33 9.60 0.84 0.07 0.07 1.21

H-320 BON 56.68 0.34 9.86 9.26 0.16 8.56 9.98 0.31 0.09 0.05 2.27

H-321 BON 55.81 0.41 11.53 10.68 0.17 8.98 9.98 0.54 0.16 0.05 1.85

99.99 0.84

100.23 0.68

100.07 0.72

100.15 0.65

99.88 0.78

100.08 0.68

100.05 0.79

100.05 0.76

100.0 0.71

100.06 0.68

100.16 0.66

2840 905 89.9 21 142 13.1 1.82 76.1

1680 452 72.2 30.8 130 47.4 2.52 102

764 288 63.2 27.4 101 19.3 0.67 37.5

1003 201 62.6 43.1 265 28.9 2.23 61.3

1610 423 73.8 27 50 7.61 0.98 14.3

10.67 220 64.0 45.8 265 24.5 1.43 45.3

2150 597 74.5 29 213 8.3 0.28 26.4

1780 495 74.3 28.7 217 13.6 0.53 29

629 223 59.6 35.8 185 27.4 0.53 98.4

3. Geochemistry

Table 3. Major element oxide (wt%) and trace element abundances (ppm) for the Khizovaara boninite series

1519 762 415 306 78.7 77.6 41.1 36.3 251 104 45.3 59.2 2.15 2.25 104 82.5 (continued on next page)

459

460

Table 3. (Continued) Sample

La/YbN La/SmN Gd/YbN

X-102 BON 0.670 0.71 20.70 9.73 0.081 0.032 1.030 2.640 0.441 2.440 0.935 0.447 1.200 0.248 1.740 0.390 1.210 1.840 1.090 0.387

X-104 LOTI 0.696 0.73 23.31 9.60 0.115 0.027 1.110 3.100 0.460 2.452 0.909 0.362 1.280 0.249 1.820 0.403 1.110 0.119 1.170 0.187

X-126/1 BON 1.040 0.74 27.70 12.90 0.174 0.040 1.120 2.650 0.465 2.790 1.140 0.401 1.350 0.274 2.160 0.434 1.150 0.214 1.370 0.199

X-128 LOTI 0.620 0.43 16.81 8.40 0.075 0.022 0.776 2.190 0.377 2.050 0.826 0.355 1.040 0.236 1.490 0.350 1.130 0.152 1.140 0.170

X-130 BON 0.910 0.81 32.97 14.1 0.155 0.028 1.150 3.330 0.575 2.850 1.060 0.323 1.520 0.274 2.200 0.513 1.370 0.228 1.440 0.200

H-302 LOTI 0.639 0.46 16.27 8.58 0.098 0.026 0.948 2.460 0.400 2.240 0.799 0.316 1.150 0.235 1.560 0.362 1.040 0.167 1.010 0.144

H-305 LOTI 0.662 0.68 21.35 9.12 0.074 0.016 1.010 2.590 0.388 2.140 0.811 0.320 1.130 0.233 1.510 0.367 1.140 0.149 0.939 0.153

0.525 0.644 0.772

0.638 0.694 0.890

0.640 0.679 0.884

0.552 0.619 0.796

0.459 0.591 0.737

0.539 0.683 0.853

0.634 0.747 0.920

0.726 0.784 0.971

H-319 LOTI 0.742 0.76 27.78 11.30 0.186 0.035 1.370 3.500 0.506 2.730 0.991 0.384 1.380 0.329 2.010 0.445 1.370 0.199 1.190 0.412

H-320 BON 0.787 0.67 20.96 12.40 0.098 0.030 1.330 3.390 0.578 2.620 1.070 0.468 1.310 0.261 1.910 0.404 1.420 0.207 1.270 0.168

H-321 BON 0.754 0.62 19.98 11.00 0.082 0.038 1.040 2.840 0.425 2.380 0.900 0.402 1.270 0.271 1.830 0.433 1.200 0.194 1.220 0.174

0.777 0.707 0.575 0.870 0.783 0.728 0.937 0.834 0.841 (continued on next page)

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Nb Hf Zr Y Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

101/1 UM 0.251 0.250 10.8 3.74 0.052 0.009 0.349 0.882 0.131 0.832 0.341 0.086 0.429 0.109 0.693 0.161 0.421 0.089 0.449 0.525

Sample SiO2 TiO2 Al2 O3 Fe2 O∗3 MnO MgO CaO Na2 O K2 O P 2 O5 LOI

H-323 BON 53.66 0.40 11.86 9.87 0.18 8.95 11.91 0.29 0.06 0.07 2.76

H-324 LOTI 49.37 0.42 10.6 12.27 0.19 14.57 10.88 0.41 0.09 0.05 1.16

H-325 LOTI 48.78 0.42 12.31 11.87 0.20 12.66 10.91 0.46 0.49 0.07 1.84

H-325/1 BON 54.10 0.42 12.19 11.10 0.17 9.21 10.02 0.42 0.42 0.05 2.02

H-326 BON 53.00 0.43 11.92 10.76 0.17 10.13 10.89 0.69 0.17 0.07 1.83

H-332 FEL 65.69 0.33 10.45 10.15 0.16 8.31 3.34 0.61 0.13 0.04 0.79

H-334/1 BON 59.36 0.36 9.92 9.89 0.16 9.04 9.19 0.65 0.07 0.07 1.43

H-334/2 LOTI 49.64 0.42 11.88 12.57 0.18 13.32 9.85 0.68 0.08 0.07 1.33

H-334 LOTI 49.91 0.43 11.21 11.62 0.18 12.92 11.39 0.44 0.09 0.07 1.75

H-333 LOTI 47.13 0.41 10.08 12.93 0.18 16.59 10.48 0.49 0.06 0.05 1.62

H-335 FEL 62.66 0.35 9.32 9.98 0.15 7.95 7.67 0.18 0.16 0.06 1.73

Total Mg#

100.01 0.68

100.01 0.72

100.01 0.71

100.12 0.66

100.0 0.69

100.0 0.66

100.14 0.68

100.02 0.71

100.0 0.72

100.02 0.75

100.21 0.65

Cr Ni Co Sc V Ba Rb Sr

688 221 63.7 34.2 126 36.4 1.26 54.7

1040 346 69.6 34.3 201 28.1 1.47 27

939 223 63.4 42.9 244 103 24.7 73.2

996 215 67.5 52.3 277 342 42.4 85.3

912 219 66.8 47.9 254 23.8 2.66 127

977 226 66.7 48.0 228 45.1 7.38 37.2

919 191 62.3 46.4 223 31.5 0.74 73.5

960 217 64.5 44.9 214 15.5 0.39 43.5

1199 241 64.2 43.8 233 17.2 0.42 76.7

3. Geochemistry

Table 3. (Continued)

1412 1603 325 342 72.2 77.9 40.8 45.7 167 179 8.77 37.6 1.01 6.84 25.5 55.2 (continued on next page)

461

462

Table 3. (Continued) Sample

La/YbN La/SmN Gd/YbN

X-102 BON 0.595 0.66 20.63 9.82 0.084 0.020 0.931 2.560 0.400 2.180 0.804 0.345 1.180 0.254 1.620 0.343 1.040 0.174 1.170 0.157

X-104 LOTI 0.948 0.69 24.03 13.70 0.097 0.021 1.250 3.210 0.524 2.840 0.988 0.407 1.500 0.283 2.100 0.453 1.390 0.204 1.350 0.203

X-126/1 BON 0.986 0.59 21.33 13.10 0.153 0.045 1.160 2.940 0.460 2.450 0.941 0.388 1.310 0.242 1.900 0.450 1.250 0.211 1.360 0.280

X-128 LOTI 0.971 0.89 35.21 14.8 0.122 0.041 1.400 3.580 0.599 2.960 1.070 0.450 1.510 0.298 2.060 0.473 1.560 0.244 1.680 0.206

X-130 BON 1.100 0.74 27.78 11.0 0.137 0.019 1.990 4.820 0.712 3.260 1.120 0.182 1.460 0.267 1.800 0.491 1.450 0.228 1.340 0.221

H-302 LOTI 0.865 0.75 26.53 14.5 0.065 0.008 1.340 3.590 0.540 2.950 1.060 0.372 1.540 0.271 2.200 0.500 1.420 0.216 1.480 0.259

H-305 LOTI 0.877 0.66 18.26 13.3 0.114 0.031 1.310 3.450 0.574 2.910 1.030 0.315 1.430 0.268 1.990 0.451 1.470 0.194 1.380 0.211

H-319 LOTI 0.876 0.76 27.78 13.0 0.083 0.029 1.610 3.620 0.606 2.890 0.959 0.448 1.540 0.277 2.130 0.455 1.340 0.213 1.370 0.201

H-320 BON 0.821 0.70 25.26 12.8 0.121 0.027 1.170 3.020 0.496 2.580 0.983 0.373 1.220 0.225 1.870 0.387 1.300 0.206 1.020 0.176

H-321 BON 0.818 0.68 21.23 13.6 0.113 0.026 1.440 3.510 0.587 2.400 0.987 0.261 1.420 0.274 1.900 0.469 1.390 0.229 1.130 0.177

0.561 0.754 0.789

0.537 0.723 0.937

0.625 0.797 0.898

0.576 0.776 0.778

0.563 0.824 0.726

1.002 1.119 0.880

0.611 0.796 0.841

0.641 0.800 0.873

0.793 1.057 0.908

0.774 0.672 0.967

0.860 0.919 1.016

Abbreviations: UM, ultramafic (cumulate?) rock; LOTI, low-Ti tholeiite; BON, boninite; FEL, felsic rocks of boninite series; LOI, loss on ignition.

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Nb Hf Zr Y Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

101/1 UM 0.810 0.46 15.8 10.7 0.099 0.025 1.080 2.790 0.448 2.430 0.902 0.339 1.270 0.259 1.860 0.416 1.300 0.231 1.300 0.173

3. Geochemistry

463

Table 4. Major element oxide (wt%) and trace element abundances (ppm) for the Iringora SSZ ophiolite sequence Dike unit Gabbro unit Sample Lava unit I-417 I-417/2 I-427 I-427/1 I-429 I-415 I-418/1 I-419/1 I-419 I-429/1 SiO2 49.16 47.54 52.80 49.15 52.83 47.59 49.95 57.52 50.31 50.83 0.42 0.44 0.47 0.47 0.44 0.46 0.33 0.35 0.50 0.42 TiO2 10.65 10.00 13.03 11.41 10.73 11.63 8.64 9.39 13.53 10.35 Al2 O3 12.25 12.77 12.13 12.76 11.60 12.46 11.15 10.51 12.56 11.53 Fe2 O∗3 MnO 0.19 0.19 0.17 0.18 0.15 0.18 0.18 0.17 0.18 0.17 MgO 12.92 14.54 10.35 14.20 12.75 13.68 15.24 12.80 13.00 13.14 CaO 11.37 10.84 7.33 9.14 8.88 11.02 11.45 6.86 7.09 10.09 0.35 0.77 2.44 1.00 0.38 0.24 0.01 0.07 0.86 0.19 Na2 O 0.07 0.05 0.08 0.06 0.11 0.15 0.09 0.10 0.12 0.17 K2 O 0.13 0.09 0.10 0.12 0.12 0.09 0.09 0.08 0.10 0.10 P 2 O5 LOI 2.52 2.85 1.11 1.54 2.10 2.56 2.96 2.21 1.82 3.05 Total Mg#

100.03 100.08 100.01 100.03 100.09 100.06 100.09 100.06 100.07 100.04 0.71 0.73 0.66 0.72 0.72 0.72 0.76 0.74 0.71 0.72

Cr Ni Co Sc V Ba Rb Sr Nb Hf Zr Y Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

705.56 543.17 323.79 672.52 789.42 644.72 972.42 538.78 612.72 708.17 254.40 191.59 186.67 256.84 280.60 215.40 326.12 188.67 164.33 265.90 74.10 72.03 52.56 71.57 70.79 66.37 60.72 50.72 57.85 66.10 37.83 36.32 39.93 38.56 36.26 40.69 29.11 32.63 44.58 33.23 237.95 233.21 338.38 297.40 311.42 251.49 186.10 202.43 296.20 267.17 15.76 26.11 17.68 15.24 34.21 56.94 26.84 11.56 20.48 42.03 0.92 1.03 1.19 0.82 1.93 7.82 2.31 2.63 2.76 5.78 49.30 31.27 62.49 61.22 74.04 101.85 107.84 69.50 170.78 92.96 0.88 0.78 1.01 0.96 0.71 1.01 0.73 1.06 1.08 0.88 0.89 0.72 0.83 1.04 0.71 0.66 0.72 0.75 0.85 0.73 33.92 26.42 28.96 37.91 25.47 24.96 26.67 26.54 28.89 22.06 12.20 12.37 13.87 12.94 10.05 13.94 10.07 12.23 15.41 11.71 0.09 0.07 0.13 0.11 0.11 0.15 0.13 0.17 0.13 0.12 0.04 0.03 0.04 0.04 0.03 0.04 0.06 0.08 0.03 0.03 0.78 1.04 1.16 1.03 1.11 1.37 1.16 1.39 1.34 1.12 2.51 2.96 3.49 3.13 2.87 3.68 3.13 3.55 3.76 3.23 0.39 0.45 0.55 0.53 0.41 0.53 0.41 0.52 0.56 0.54 2.05 2.55 3.42 2.68 2.12 2.77 2.25 2.57 3.33 2.73 0.73 0.79 0.99 0.87 0.68 0.84 0.72 0.82 1.01 0.93 0.35 0.40 0.39 0.32 0.46 0.43 0.36 0.29 0.40 0.34 1.31 1.34 1.62 1.46 1.18 1.38 1.19 1.37 1.60 1.41 0.24 0.23 0.27 0.23 0.22 0.25 0.20 0.27 0.32 0.25 1.77 1.74 2.03 1.90 1.44 1.84 1.51 1.78 2.36 1.75 0.41 0.39 0.45 0.41 0.35 0.46 0.33 0.43 0.51 0.42 1.20 1.19 1.32 1.22 1.02 1.22 0.93 1.26 1.42 1.17 0.17 0.17 0.18 0.16 0.15 0.20 0.12 0.19 0.21 0.19 1.05 1.08 1.24 1.18 0.96 1.28 0.81 1.18 1.48 1.25 0.15 0.14 0.19 0.17 0.14 0.17 0.13 0.19 0.21 0.17

La/YbN La/SmN Gd/YbN

0.501 0.829 1.008

0.650 0.829 1.003

0.631 0.738 1.056

0.589 0.745 1.000

0.781 1.028 0.993

0.722 1.027 0.871

0.967 1.014 1.187

0.795 1.067 0.938

0.611 0.835 0.874

0.605 0.758 0.916

464

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 17. Geochemical compositions of the Iringora and Khizovaara boninite series compared to Troodos UP lavas. Primitive mantle and N-MORB abundances are taken from Hofmann (1988).

3. Geochemistry

465

latter have obviously formed by other ways. Indeed, high-Mg rocks with unusually high SiO2 of up to 58% have been described in the 3.4 Ga Nondweni greenstone belt, South Africa as komatiitic andesites (Riganti and Wilson, 1995) and in some late Archean and early Proterozoic greenstone belts as siliceous high-Mg basalts (SHMB) (Sun et al., 1989). However these are commonly believed to form when ultramafic (komatiitic) magmas have undergone crustal contamination and subsequent fractional crystallization (e.g., Arndt et al., 1997; Sylvester et al., 1997). This leads to the question of whether the NKGB suite of high-Ca boninitic affinity belongs to a true boninite series or it represents an unusual kind of crustally contaminated high-Mg rock of komatiitic affinity. There are several lines of evidence to show that the NKGB suite is not crustally contaminated komatiite-derived magma. Ratios of Ti to the adjacent transitions elements Sc and V in the former are lower than in MORB, komatiites or komatiitic basalts and SHMB and close to the primitive mantle ratios (Fig. 18a). Although Zr/Ti ratios vary between 70 and 140 and tend to be also centered around the primitive mantle ratio (∼ 110), the NKGB suite fields on a plot of Zr vs. TiO2 are far lower than a typical N-MORB and markedly distinct from the field of SHMB (Fig. 18b). Contrary to overall LREE-enriched SHMB, the NKGB boninite series volcanics are relatively depleted in LREE compared to the primitive mantle abundances and range in (La/Yb)N from 0.5 to 1.0 mostly being between 0.6 and 0.8, whereas HREEs show less depleted patterns (Ga/YbN = 0.8–1). There are also variable negative Nb anomalies (av. Nb* /Nb = 0.59) but dominantly positive Zr and Hf peaks (av. Zr* /Zr = 1.14 and av. Hf* /Hf = 1.13), additional distinctive features of many recent boninite series (Hickey and Frey, 1982; Pearce et al., 1999). Mantle-normalized plots (Figs. 19a, b) summarize many of distinctive features shared between the NKGB boninitic rocks and younger high-Ca boninites. Conversely, they display crucial differences as compared with crustally contaminated komatiite-derived counterparts (Fig. 19c). Finally, Sm-Nd isotope investigations were carried out to verify constraints on the source composition that formed the NKGB boninite series volcanics. As can see from Table 5, all the analyzed samples from the Khizovaara Structure bear positive εNd values strongly suggesting that no older felsic crust was involved in their petrogenesis and implying, therefore, that they developed in an intraoceanic setting. In contrast, crustally contaminated komatiite-derived lavas yield almost negative εNd values (e.g., Sun et al., 1989; Riganti and Wilson, 1995; Arndt et al., 1997; Puchtel et al., 1997). However the range of εNd values is wide in the Khizovaara boninite series rocks (from +3.28 to +1.10). A majority of the boninitic samples have initial 143 Nd/144Nd ratios of between 1.36 and 1.10 whereas the most primitive composition of sample X-128 yields εNd value of 3.28. Taking into account the mean ε Nd value of about 3.0 ± 0.5 for a Neoarchean long-term depleted mantle reservoir (e.g., Puchtel et al., 1997, and references therein), it is apparent that the Khizovaara boninites are twice as enriched in a low-radiogenic Nd component. Surprisingly, the same rate of the enrichment relative to the MORB source is typical of the group II Troodos UPL that is believed to have resulted from an addition of small amounts (up to 5%) of LREE-enriched subduction component (Cameron, 1985). Despite few data, the inverse correlation between εNd and a rate of Hf enrichment relative to REE represented by

466

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 18. Ti/V vs. Ti/Sc (A) and TiO2 vs. Zr (B) plots for the Khizovaara (shown by the fields) and the Iringora (shown by rhombus) boninite series. For comparison fields of Troodos UPL also are shown: groups I, II and III from Cameron (1985); groups A, B and C from Flower and Levine (1987). Fields for boninites, komatiites, MORB and SHMB are from the compilation of Poidevin (1994). Dashed lines indicate primitive mantle ratios.

3. Geochemistry

467

Fig. 19. Comparison of mantle-normalized trace-element abundance patterns of North Karelian boninites with recent (a), Phanerozoic and Paleoproterozoic (b) high-Ca boninites and Paleoproterozoic and Archean komatiite-derived crustally contaminated counterparts (c). Data sources are as follows: Mariana Trench (Hickey and Frey, 1982); Troodos ophiolite (Cameron, 1985); Northern Tonga ridge (Sobolev and Danyushevsky, 1994); Koh ophiolite (Meffre et al., 1996); Balantrae ophiolite (Smellie et al., 1995); Birch Lake boninite (Wyman, 1999); Bogoin boninite (Poidevin, 1994); Vetreny belt (Puchtel et al., 1997); Kambalda (Jochum et al., 1990); Nondweni belt (Riganti and Wilson, 1995).

468

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Table 5. Sm-Nd isotopic data for the 2.8 Ga Khizovaara boninite series volcanics Sample X-128 H-325/1 X-130 H-326 H-320 X-126/1

Sm, ppm 0.96 0.72 0.71 0.87 0.79 0.68

Nd, ppm 2.59 1.92 1.82 2.27 2.04 1.72

147 Sm/144 Nd

143 Nd/144 Nd

0.22301 0.22651 0.23718 0.23265 0.23507 0.23857

0.513291 0.513315 0.513455 0.513370 0.513411 0.513468

εNd 3.28 2.48 1.36 1.33 1.26 1.10

Hf −0.338 −0.185 0.040 0.061 −0.055 −0.021

Note: Hf was calculated from Hf, Nd and Yb abundances listed in Table 3.

Fig. 20. εNd –Hf covariations in the Khizovaara boninite series rocks. Hf expression is after Pearce et al. (1999). Symbols are as follows: white stars, evolved (Mg# = 0.65–0.69) boninitic compositions; black star, primitive (Mg# = 0.78) high-Mg low-Ti tholeiitic composition.

Hf (Pearce et al., 1999) occurs (Fig. 20). This implies that a subduction component with a low Nd/Hf ratio was involved in the NKGB boninite series melts. Pearce et al. (1999) have explained the nature of such enrichment by two-stage mantle melting processes. A first stage of melting depletes the mantle source through extraction of melts with a high Nd/Hf ratio. The second stage of melting that produces boninites would then give melts with low Nd/Hf ratio. Origin of the NKGB boninite series. There is a general consensus that the origin of boninites requires a water-saturated high-temperature melting of a mantle wedge that has been depleted by incompatible elements by previous melting episodes (Crawford et al., 1989, and reference therein). Data expressed above strongly suggest that a similarity in

3. Geochemistry

469

major, trace element and isotope chemistry exist when compared the NKGB boninite series with the Cretaceous Troodos ophiolitic lavas. It is vital to note that the Troodos UP lavas are taken as type high-Ca boninite suite (Crawford et al., 1989). The amazing similarities in major- and trace element chemistry along the boninite series of drastically different ages reflect their similar petrogenetic processes. Comparison of fractional histories of both these series strongly suggests that compositionally similar melts in equilibrium with olivine were parental melts for the entire high-Ca boninite series. Cameron (1985) has emphasized that the stem of Troodos UPL boninite series was evolved from the most primitive low-SiO2 end-member. Nevertheless there is no agreement regarding the parental melts for the boninite series. Duncan and Green (1980a, 1980b) have shown that a Troodos UPL potential parental magma with 15–16% MgO could have segregated from a harzburgitic residue at 7–8 kbar and 1360 ◦ C. Other estimates suggest that this might have MgO contents in the range 17–22%, generated in a mantle wedge above a subduction zone at 20–30 kbar and 1380–1550 ◦ C (Sobolev et al., 1993). Comparison of the potential candidates for NKGB boninite series primary melts with the available data on the predicted trace element characteristics for Troodos UPL parental magmas (cf. Sobolev et al., 1993) shows a close approximation to the latter constraint. This enables us to suggest that the MgO contents for the NKGB boninite series primary melts of about 20% are permissible. In order to constraint the P-T conditions under which parental melts were generated to form the NKGB boninite series, its composition coupled together with fields of recent boninite series are plotted on Walker’s ternary projection (Walker et al., 1979) (Fig. 21). In addition, available experimental data on the dry and wet mantle melting are also shown on the diagram. As can be seen from the diagram, under dry conditions the predicted parental melts should be formed at pressures of about 3 Gpa and a potential temperature close to 1550 ◦ C. However it is common knowledge that initial water contents in boninitic parental magmas might lay in a range of 1–3% (e.g., Crawford et al., 1989; Ohnenstetter and Brown, 1996) that must depress to some extent the peridotite solidus. Experimental data of Kushiro (1990) suggest a 0.2 Gpa depression of the solidus for a mantle with 4.4–6.6% H2 O that corresponds to a temperature lowering of nearly 150 ◦ C. This implies that a starting melting of a previously depleted mantle wedge to have formed the NKGB boninite series could be realized at a pressure of nearly 2.0 Gpa and a potential temperature less than 1450 ◦ C. Furthermore the NKGB boninite series parental melts should be having some water excess as compared with the younger counterparts. Indeed, the overall NKGB boninite series trend tends to in progress towards the Qz apex (Fig. 21) implying, therefore, that it could be developed with more H2 O content respectively to the recent ones. This may also indicate that the more evolved portions of the NKGB boninite series were crystallized in the amphibole-in stability field. In summary, all the data expressed above strongly suggest that there is no substantial difference between the generation conditions of the Neoarchean and recent high-Ca boninite series counterparts. The only difference between them is a lack of analogues to the most depleted group III of Troodos UPL in the NKGB counterpart. However, among other reasons, this may be due to a sampling gap of the latter when compared with a huge amount of sampling which has been done in the Troodos ophiolite.

470

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 21. Normative compositions of the North Karelian boninite series in ternary projection olivine-plagioclase-quartz from diopside (Walker et al., 1979). Predicted composition for NKGB parental magmas indicated by the cross. The fields for the Phanerozoic high-Ca boninite series and MORB glasses are shown for reference. The data sources used here are as follows: Cameron (1985), Sobolev et al. (1993) for the Troodos Massif; Fallon and Crawford (1991) and Sobolev and Danyushevsky (1994) for the northern Tonga trench; Hickey and Frey (1982) for the site 458 of DSDP, Mariana trench. The field of MORB glass data (MgO > 9 wt%) is taken from Elton (1989). Compositional trends from the experimental and calculated data on upper mantle melting are shown to constraint P-T generation conditions of the NKGB boninite series. The isobaric compositional trends under dry conditions for 1, 2, 3 Gpa are taken from Hirose and Kushiro (1993). Solid line labeled as W-W is the isobaric compositional trend under the wet conditions (for an H2 O content of 4.4–6.6 wt% at 1.2 Gpa) from Kushiro (1990). Dotted lines are point average compositions for various upper mantle potential temperatures (1580–1280 ◦ C) from McKenzie and Bickle (1988).

3.2. Geochemistry of the Arc-Related Volcanics Most geochemical data on the NKGB acid-felsic volcanics comes from the Khizovaara Structure. Two distinct volcanic rock suites are recognized based both on field appearance and compositional characteristics. These are called the Northern andesite suite and the Southern andesite-dacite-rhyolite suite (Figs. 2, 3). In addition, two distinct plutonic suites, the Northern and the Southern tonalites, surround the Khizovaara greenstones. The Northern andesite suite consists mostly of massive, amygdaloidal, glomeroporphiritic textured andesites. In terms of major element chemistry the Northern andesites are characterized by low Al2 O3 and high Na2 O contents and belong to the tholeiitic magma clan (Thurston and Kozhevnikov, 2000). However, based upon the trace element grounds

3. Geochemistry

471

they define three separate groups being as follows: ‘depleted andesite’, ‘enriched andesite’ and ‘high-Nb andesite’ (Figs. 22, 24). The ‘depleted andesite’ (NAD) group volumetrically prevails. This yields high MgO, Cr and Ni contents and conversely low abundances of LILE, HFSE and REE (Table 6). The trace element patterns show flat to slightly fractionated REE (La/SmN = 0.79–1.9, Gd/YbN = 1.5–1.9) coupled with negative anomalies of Eu, Nb and Ti (Fig. 24). The ‘enriched andesite’ group (NAE) is found along both flanks of the unit. As compared with the above this is depleted in MgO and Cr and conversely enriched in LILE, HFSE and LREE (La/SmN = 3.3–4.1, Gd/YbN = 1.7–2.1) and characterized also by the pronounced negative Nb anomalies. The ‘high-Nb andesite’ (NANb) group occurs only along with the south contact of the unit with the Southern lithotectonic assemblage. Compositionally this group is similar to the enriched andesites. With respect to the latter it is characterized by lower MgO contents, higher REE abundances fractionated in a different manner (La/SmN = 2.8–3.0, Gd/YbN = 1.7–2.1), and enrichment in Ti, Zr and Nb (Figs. 22, 24). Note that such enrichment is accompanied by lower ratios of Ti/Nb, Zr/Nb and Ti/Zr as compared with the ‘enriched andesite’ group (Table 6). Because the trace elements outlined above are commonly treated as practically immobile during low-temperature alteration and metamorphism, their variations might be due to magmatic processes. Most of the major and trace elements reveal similar covariations throughout the three geochemical groups. This suggests similar fractionation histories of their parental melts. Modeling of possible fractionation paths has shown that a liquidus assemblage of plagioclase + pyroxene ± olivine ± amphibole ± Fe-Ti oxides could control a crystal differentiation of parental melts with low H2 O contents at shallow-level depths (< 30–40 km) to reach the observed major element variations. However the trace element patterns cannot be explained by crystal fractionation alone. Indeed, the low concentrations of SiO2 coupled together with the high abundances of siderophylic elements (Table 6) suggest an ultramafic precursor rather than a mafic one for their parental melts. Thus the ‘depleted andesite’ group appears to have formed through a partial melting of previously depleted mantle precursor in a manner similar with that generated boninitic melts. In contrast, the trace element pattern of the ‘enriched andesite’ group (Fig. 24) gives an indication of a LREE enriched metasomatic component involved in a melting of depleted mantle. The ‘high-Nb andesite’ group exhibits a remarkable enrichment in both LREE and HFSE, much as the neighbouring Fe-Ti basalts display. It should be noted also their high Gd/YbN ratios are suggestive of garnet in the mantle residue and, therefore, imply that this might be because they formed at deeper levels (> 70 km) as compared with the other andesitic counterparts. In addition, trends to high MgO, Cr, Ni coupled together with negative Nb spikes (Figs. 22, 24) can be accounted for by interaction between different siliceous slab melts and a subarc mantle wedge. It is amazing that amongst recent counterparts the geochemistry of the Northern andesite suit is best matched to that of the so-called high-Ti series lavas of the Troodos ophiolite (cf. Staudigel et al., 1999). The Northern tonalite suite combines a variety of plutonic rocks that border the Khizovaara greenstones on the north (Fig. 2). It ranges from tonalite to trondhjemite (NT) (Fig. 22) and is accompanied by numerous co-genetic tonalitic dikes (NTD) emplaced into the Northern lithotectonic assemblage. These are characterized by high Al2 O3 , Sr and

472

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 22. Harker compositional diagrams showing selected major and trace element variations for Northern andesite and tonalite suites, and Southern rhyolites of the Khizovaara structure.

3. Geochemistry

473

Fig. 23. Harker compositional diagrams showing selected major and trace element variations for the Southern andesite-dacite and granodiorite suite. Fields labeled by numbers correspond to the compositions of Fig. 23 and are shown for reference.

474 Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Fig. 24. Trace-element spidergrams for Khizovaara acid-felsic rocks normalized to primitive mantle values of Hofmann (1988).

3. Geochemistry

475

Table 6. Mean abundances of major element oxides (wt%) and trace elements (ppm) for the Khizovaara arc-derived volcanic and plutonic rocks Rock type n SiO2 TiO2 Al2 O3 Fe2 O∗3 MnO MgO CaO Na2 O K2 O P 2 O5 Sc Ti V Cr Co Ni Rb Sr Y Zr Nb Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Th U La/YbN La/SmN Gd/YbN Eu/Eu*

NAE 7 59.14 0.83 14.29 10.96 0.16 3.96 5.30 4.51 0.66 0.19 23.5 4582 186 19.3 35.0 39.1 19.63 143 17.60 132.40 5.66 266.30 25.46 52.12 6.19 23.58 4.38 1.18 3.85 0.59 3.07 0.63 1.77 0.27 1.64 0.22 3.25 0.38 5.77 0.93

NAD 9 58.54 0.68 14.35 9.78 0.17 4.73 6.05 5.31 0.19 0.20 22.9 3755 177 56.9 31.8 44.5 1.26 211 15.00 100.29 5.05 39.65 6.59 17.84 2.69 12.09 2.93 0.82 2.76 0.42 2.49 0.52 1.38 0.21 1.30 0.18 2.63 0.29 4.30 0.85

10.51 3.66 1.90 0.88

3.47 1.42 1.71 0.88

NA(Nb) 2 56.57 1.59 14.51 13.12 0.18 3.17 5.82 4.53 0.24 0.26 17.2 9399 148 42.1 47.5 73.5 3.09 284 23.25 203.50 13.35 128.90 32.15 68.00 8.93 35.40 6.91 1.83 5.97 0.83 4.57 0.90 2.36 0.33 2.00 0.29 4.94 0.77 3.39 0.57 10.85 2.94 2.41 0.87

NT 8 70.24 0.35 15.73 3.12 0.12 0.98 3.46 4.24 1.60 0.15 4.87 1944 27.8 24.5 7.18 14.9 34.4 465 5.30 98.85 3.70 421.31 17.61 33.90 4.17 14.47 2.29 0.62 1.64 0.20 0.92 0.17 0.42 0.05 0.37 0.05 2.69 0.24 2.53 0.57

NTD 5 69.11 0.42 16.30 3.41 0.12 1.49 3.67 4.11 1.18 0.19

32.13 4.89 3.67 0.97

32.45 4.83 3.14 0.76

12.5 25.3 2.76 9.56 1.63 0.32 1.01 0.13 0.59 0.12 0.22 0.03 0.26 0.04

SA(Nb) 1 58.27 1.34 17.42 8.87 0.16 3.77 4.59 2.49 2.89 0.20 15.6 7197 72.6 17.4 15.6 20.3 98.60 313 18.90 64.30 11.30 639.00 16.80 37.60 4.79 19.20 4.28 1.13 3.95 0.59 3.25 0.67 1.72 0.37 1.67 0.28 1.46 0.60 2.66 0.48

SA 6 61.16 0.75 17.00 6.66 0.18 3.26 6.21 2.51 2.09 0.20 15.5 4098 135.2 73.9 16.3 33.2 53.99 417 12.38 99.85 4.01 594.33 20.75 45.18 5.75 23.69 4.55 1.22 3.46 0.47 2.37 0.47 1.23 0.19 1.15 0.16 2.71 0.50 3.18 0.41

SD 8 66.98 0.62 16.32 4.77 0.14 2.11 4.25 2.96 1.72 0.14 8.4 2565 45.0 51.6 10.2 31.9 41.73 224 8.13 118.33 5.33 423.67 12.87 22.37 3.14 11.13 2.24 0.68 1.90 0.25 1.49 0.29 0.87 0.13 0.71 0.11 2.97 0.36 3.03 0.44

6.79 12.30 13.67 2.47 2.87 3.77 1.91 2.44 2.37 0.84 0.94 1.04 (continued on next page)

476

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Table 6. (Continued) Rock type SAGm SAGf SDsubv n 2 2 9 SiO2 66.90 73.26 67.27 0.67 0.40 0.47 TiO2 15.39 14.38 16.98 Al2 O3 5.15 2.70 3.65 Fe2 O∗3 MnO 0.13 0.13 0.14 MgO 2.95 1.08 1.71 CaO 3.95 3.83 4.39 2.70 3.40 3.58 Na2 O 2.02 0.72 1.70 K2 O P 2 O5 0.15 0.12 0.12 Sc 8.0 6.2 7.9 Ti 3432 2125 2052 V 45.9 31.0 46.45 Cr 63.9 46.8 38.075 Co 13.5 6.8 10.995 Ni 57.6 33.2 28.55 Rb 53.21 18.07 42.6 Sr 240 264 437.25 Y 9.86 7.62 5.9 Zr 117.97 129.27 89.375 Nb 7.24 6.27 2.9025 Ba 691.36 283.44 379.5 La 13.71 11.50 15 Ce 28.00 20.19 26.75 Pr 3.33 2.57 3.4525 Nd 12.50 8.83 13.025 Sm 2.50 1.65 2.3 Eu 0.85 0.56 0.735 Gd 2.36 1.70 1.9375 Tb 0.36 0.22 0.2335 Dy 1.72 1.29 1.08825 Ho 0.34 0.25 0.22575 Er 0.89 0.34 0.5545 Tm 0.13 0.11 0.086575 Yb 0.72 0.67 0.4755 Lu 0.10 0.09 0.066525 Hf 3.08 2.90 2.3925 Ta 0.53 0.41 0.18475 Th 2.82 2.86 2.53 U 0.63 0.34 0.48425 La/YbN La/SmN Gd/YbN Eu/Eu*

12.96 3.53 2.69 1.06

11.58 4.35 2.05 1.03

21.299759 4.0811771 3.2786683 1.0796298

SRT SRF SRP SDI SGD 7 6 8 1 3 74.62 77.01 76.55 60.01 69.18 0.41 0.38 0.22 0.76 0.38 15.40 15.77 14.34 19.49 16.17 2.09 0.80 1.68 5.91 3.97 0.12 0.12 0.12 0.15 0.14 0.72 0.62 0.99 1.93 1.11 1.94 1.86 1.92 6.07 3.50 2.06 1.82 1.40 3.66 3.27 2.54 1.55 2.73 1.78 2.16 0.10 0.08 0.06 0.23 0.11 6.0 6.5 3.3 8.2 6.6 1817 1800 1051 3990 1961 26.5 41.6 14.9 46.5 40.8 56.0 22.2 24.4 11.4 35.3 5.5 3.3 3.5 12.2 8.5 19.3 10.2 15.4 10.1 22.1 86.10 36.20 67.44 42.80 53.64 151 262 199 487 318 10.72 3.32 7.30 12.00 14.48 99.00 111.67 90.96 88.90 112.47 6.01 3.46 10.00 6.59 6.41 544.50 293.17 456.16 509.00 513.01 36.00 14.76 31.29 18.70 20.31 51.00 28.30 63.59 37.40 38.42 6.04 3.53 6.96 5.54 5.35 20.77 12.80 22.91 22.40 20.12 3.58 2.35 3.34 3.85 3.56 0.75 0.60 0.69 1.35 0.90 2.81 1.87 2.50 3.22 2.96 0.35 0.23 0.30 0.39 0.43 1.64 0.90 1.23 2.03 2.32 0.30 0.16 0.23 0.42 0.46 0.85 0.35 0.63 1.35 1.44 0.13 0.05 0.09 0.21 0.22 0.69 0.30 0.55 1.34 1.49 0.11 0.04 0.08 0.21 0.24 2.57 2.76 2.77 2.02 3.17 0.38 0.23 0.71 0.43 0.72 6.56 2.85 8.47 2.23 4.53 0.87 0.62 1.97 0.18 1.14 24.87 4.31 3.27 0.79

32.56 4.23 4.94 0.92

39.57 9.42 9.79 5.93 3.06 3.70 3.78 1.94 1.70 0.73 1.17 0.87 (continued on next page)

3. Geochemistry

477

Table 6. (Continued) Note: n, number of analyses. Abbreviations: NA = Northern andesite suite, (+ E) ‘enriched’, (+ D) ‘depleted’ and (+ Nb) ‘high-Nb’ groups, respectively; NT and NTD = Northern tonalite and tonalitic dike; S = Southern andesite-dacite suite, (+ Nb) ‘high-Nb’, (+ A) andesite and (+ D) dacites, respectively; SAGm and SAGf = agglomerate textured volcanics, matrix and fragment, respectively; SDsubv = subvolcanic dacites; SRT, SRF and SRP = rhyolites: tuff, flow, and porphyry, respectively; SDI and SGD = Southern tonalite, diorite and granodiorite, respectively.

LREE (La/SmN = 4.6–5.3) contents and, conversely, by very low abundances of HREE (Gd/YbN = 3.6–3.8) and Y all similar to recent adakitic melts. It is widely accepted that adakite magmas result from melting of basaltic material at pressures high enough to stabilize garnet-amphibole ± pyroxene ± plagioclase in the melt residues (e.g., Drummond and Defant, 1990; Martin, 1999, and reference therein). The Southern granodiorite suite includes largely dioritic (SDI) to granodioritic (SGD) plutonic rocks. In contrast to the Northern tonalite, they show apparently less fractionated REE patterns (La/SmN = 3.2–4.1, Gd/YbN = 1.3–2.1) and both Eu and Ti troughs (Fig. 24). Its somewhat older U-Pb zircon age in comparison with the other Khizovaara rocks (Fig. 2) implies that a crustal component could be involved to have formed this suite. The Southern andesite-dacite-rhyolite suite occupies the southern limb of the Khizovaara structure. Based upon both the field appearance and available isotope-geochemical data it seems clear that the Iringora acid-felsic volcanics are close analogues of the Khizovaara Southern andesite-dacite-rhyolite suite. In terms of the major element chemistry their overall compositions make up mostly a calc-alkaline trend indicative of a mature volcanic arc. Amongst the latter high-Al andesite-dacitic compositions prevail. The common occurrence of agglomerate textured andesite-dacitic volcanics strongly suggests their emplacement in a subaeral environment. The Southern andesite includes two distinct geochemical groups referred to here as the ‘Southern andesite’ group (SA) and the ‘Southern high-Nb andesite’ group (SANb). The first group together with the Southern dacites (SD) forms continuous trends on most Harker diagrams (Fig. 23). They are characterized by 0.6–0.9 wt% TiO2 , 2.7–4.8 ppm Nb, low contents of Cr, Ni and Zr (Table 6), fractionated REE patterns (La/SmN = 2.6– 3.2, Gd/YbN = 2.0–2.7) and pronounced Nb and Zr negative anomalies (Fig. 24). Dacites tend to have lower abundances of the major elements and Cr, Ni, V, Sc, Sr, REE while their LREE and HREE patterns are more and variably fractionated (La/SmN = 3.0–4.7, Gd/YbN = 1.2–3.1) in comparison to the andesites. The second group of andesites displays some enrichment in Nb (up to 11 ppm) and TiO2 (up to 1.34 wt%), is rare in the population of sample from the Southern suite. In contrast to the overall Southern andesite-dacite stem it shows moderately fractionated REE patterns (La/SmN = 2.5, Gd/YbN = 1.9) and slight Nb and Zr negative anomalies. These high-Nb andesites being also markedly distinct from the Northern counterparts (cf. CA(Nb) and NA(Nb) in Table 6; Figs. 23, 24) are close in terms of major and trace abundances to the Nb-andesites of recent Western Pacific volcanic arcs (Prouteau et al., 2000).

478

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Subvolcanic dacitic sills and dikes are an integral part of the Southern andesite-daciterhyolite suite. They plot along a compositional trend from high-Al dacite to rhyodacite (Fig. 23) close to fields of both the Southern volcanic counterparts and Northern tonalite. In contrast to the former, the subvolcanic dacites show stronger fractionated REE patterns with lower HREE contents (La/SmN = 3.8–4.3, Gd/YbN = 3.1–3.5) and higher Sr/Y and Sr/Nb (Fig. 24) ratios all indicative of adakitic magmas (e.g., Martin, 1999). Rhyolites occur as small sill-like bodies and tuffaceous horizons. The subvolcanic rhyolites are dominantly porphyritic in texture, with plagioclase and quartz phenocrysts up to few mm long. Compositionally these display a compact field of high siliceous KNa rhyolites (Fig. 22). Their REE patterns are strongly fractionated (La/SmN = 5.5–6.5, Gd/YbN = 3.3–4.5) and characterized by pronounced negative Eu spikes. Most of the HFSE, as well as Sr, show a marked depletion, whereas U, Th and Nb demonstrate a clear enrichment relative to abundances of the neighbouring trace elements (Fig. 24). Trends to lower Al2 O3 , CaO, Na2 O and Sr with increasing silica from 74 to 79 wt% must primarily reflect removal of quartz and plagioclase, which is consistent with the presence of these minerals as phenocrysts. In contrast to the subvolcanic counterparts, the felsic tuffaceous horizons show an abnormal enrichment of Al2 O3 (up to 18 wt%). For individual samples, high abundances of Ni and Cr coupled together with irregular variations of Al2 O3, Ti, Zr, Nb and Th appear to reflect a non-magmatic component, either hydrothermal or sedimentary, was involved in their petrogenesis. Based on the geochemical ground and petrological constraints it should be emphasized that the felsic rocks could not have evolved from any partially melted mafic precursors by fractional crystallization alone. This requires either a different magma source or an addition of enriched siliceous component during the evolution from intermediate to felsic composition. A possible source for such melts could be genetically related with the both suites of adakitic affinity, the Southern andesite-dacite volcanics and the Northern tonalite. Indeed, as has been experimentally shown recently (Prouteau et al., 1999), rhyolitic compositions close to the above might be produced by ∼ 30% crystal fractionation of a rising hydrous adakitic magma. This checks well with evidence for major- and trace-element chemistry of the Southern agglomerate textured volcanics. Compositionally, the matrix of the agglomerate lavas (SAGm) is similar to the average Southern subvolcanic dacite (SDsubv) of adakitic affinity while their fragments (SAGf) tend to be akin to the rhyolitic flow (SRF) (Table 6). In summary, although the Northern and Southern volcanic suites display different geochemical features, all the above compositional varieties could result from partial melting of either the mantle wedge or an oceanic slab, as well as through a combination of these processes. Indeed, it is commonly believed that high-Al andesite-dacite volcanics result from partial melting of hydrated mantle wedge whereas adakitic melts developed through the partial melting of oceanic slab. Origin of Nb-enriched andesites is interpreted in a variety of ways each being explained by interaction between a subarc mantle peridotite and adakitic liquids (Sajona et al., 2000). A similar group of andesitic volcanics has been described in the 2.7 Ga Wawa greenstone belt, Superior Province, Canada by Polat and Kerrich (2001, 2004). It is important that recent adakitic and

4. Conclusions and Geodynamic Implications

479

Nb-enriched andesite occurrences are clearly associated with the subduction of young, hot oceanic lithosphere (e.g., Drummond and Defant, 1990; Drummond et al., 1996; Sajona et al., 2000; Prouteau et al., 2000) that is known to be important in the genesis of recent boninite occurrences as well (Pearce et al., 1992).

4. CONCLUSIONS AND GEODYNAMIC IMPLICATIONS Several lines of evidence suggest that the North Karelian greenstone belt contains remnants of a 2.8 Ga SSZ ophiolite association. The Iringora sequence display partially preserved ophiolite-like pseudostratigraphy including lava, sheeted dike and layered gabbro units, whereas the original ophiolitic stratigraphy of the Khizovaara Structure is obscured by both a carapace of the acid-felsic arc- related volcanics, and later tectonic imbricate events. Remnants of sheeted dikes and their transitions into both the lava and the gabbro units are preserved within the Iringora sequence. Sheeted dikes are one of the straight forward indicators for the interpretation of such extensional magmatic system as an ophiolite. Although sheeted dikes may be developed during rifting of volcanic edifices built even on sialic crustal basement, those of the Iringora sequence, combined with their explicit geochemical features, strongly suggest its SSZ ophiolite nature. This ophiolitic complex has experienced intense structural and high-pressure metamorphic transformations during, at least, two distinct tectonic stages, including Neoarchean (2.8–2.7 Ga ago) and Paleoproterozoic (1.9–1.75 Ga ago). The original layered structure was largely disrupted to an extent that we cannot construct a composite section through the Iringora sequence in order to assess a possible crustal thickness of the ophiolite. Nonetheless, significant information comes from the geochemical and isotope data for the NKGB boninite series and related rocks. Their major and trace element abundances were not largely affected by the post-magmatic alteration processes retaining, therefore, records of the petrogenetic processes appropriate to tectonic implications. We have noted that the 2.8 Ga NKGB boninite series are remarkably similar compositionally to the upper pillow lavas from the Troodos ophiolite, a much younger example of high-Ca boninite series (Crawford et al., 1989). The amazing similarities in major and trace element chemistry along the boninite series of manifestly different ages reflect their similar petrogenic history. This implies, in turn, that a tectonic setting of the North Karelian boninite series volcanism was not substantially different from that of recent analogues. Thrusting of the Iringora ophiolitic sequence onto the arc assemblage resulted from an arc-trench interaction (Figs. 7, 12). This could have occurred in a setting of forearc extension that seems typical for most of Phanerozoic SSZ ophiolite assemblages (e.g., Beccaluva and Serri, 1988; Pearce et al., 1992; Stern and Bloomer, 1992). Recently Shervais (2001) summarized a wealth of geologic evidence on the most-known SSZ ophiolites and concluded “that suprasubduction zone ophiolites display a consistent sequence of events during their formation and evolution that demonstrates that they must form in response to processes that are common to all such ophiolites and are characteristic of their mode of

480

Table 7. Comparison of the geological characteristics of Phanerozoic SSZ ophiolites and the Neoarchean SSZ ophiolite-like sequences of the North Karelian greenstone belt. The stages of the life cycle of SSZ ophiolites and their geological characteristics after Shervais (2001) NKGB inferred

Stage 2: Youth refractory melts, second stage melting

NKGB present

Stage 3: Maturity calc-alkaline, “normal” arc

NKGB present

Stage 4: NKGB Death ridge ridge subduction (?) subduction or directly to obduction

Stage 5: Resurrection obduction onto passive margin (Tethyan) or accretionary uplift (Cordilleran) none

none

accretionary uplift (Cordilleran)

Volcanic rocks

primitive arc tholeiites (basalt to basalticandesite)

present

high-Mg andesites, boninites, tholeiitic ankaramites

present

andesite, dacite, basaltic andesite

present

MORB-like or OIB

Plutonic rocks

layered gabbro, troctolite, dunite

not observed

wehrliteCpxite sill complex

present (Kozhevnikov, pers. com.)

quartz diorite, hornblende diorite, agmatites

present

none

rare granitoids; anatexis of lower plate

present

Metamorphic rocks

hydrothermal alteration of volcanics

inferred

hydrothermal alteration of volcanics

inferred

hydrothermal alteration of volcanics

present

inferred high-grade metamorphic sole

obduction may be cold or hot; new subduction zone may form (accretionary uplift)

hot obduction (?)

Abbreviations: NKGB, North Karelian Greenstone belt; MORB, mid-ocean ridge basalts; OIB, ocean island basalts.

present

NKGB

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Events

Stage 1: Birth initial spreading, hinge rollback

Acknowledgements

481

formation”. Comparison the main geologic characteristics of the Neoarchean North Karelian ophiolite-like sequences with those from the stages of the Shervais ‘Life cycle of SSZ ophiolite’ shows (Table 7) that the most, if not all, of distinctive features of SSZ are either present in the 2.8 Ga NKGB or may by inferred. This implies that a mode of formation and evolution of the Neoarchean SSZ ophiolite-like rocks was very similar to the Phanerozoic SSZ ophiolites. A corollary from this comparison is that a modern subduction style accompanied by a spreading of previously formed oceanic lithosphere operated in the Neoarchean.

ACKNOWLEDGEMENTS This work was supported by the RFBR (grants 990565607, 990564055, 000564295 and 00056401). The underlying field work for this study were mostly funded by the Geological Institute of the Russian Academy of Sciences. The main analytical procedures to date zircons by conventional method were done by T.V. Gracheva, T.I. Kirnozova and V.A. Makarov at the Vernadsky Institute of Geochemistry and Analytical Chemistry of RAS, Moscow. Elena Bibikova is grateful to Dr. S. Claesson from the Swedish Royal Museum of Natural History, Stockholm, for providing access to the ion microprobe NORDSIM technique. The authors would especially like to thank Dr. I. Puchtel for the isotope studies of samples.

REFERENCES Abbott, D.H., Drury, R., Smith, W.H.F., 1994. Flat to steep transition in subduction style. Geology 22, 937–940. Amelin, Yu.V., Heaman, L.M., Semenov, V.S., 1995. U-Pb geochronology of layered mafic intrusions in the eastern Baltic Shield: implication for the timing and duration of Paleoproterozoic continental rifting. Precambrian Research 75, 31–46. Arndt, N.T., Albaréde, F., Nisbet, E.G., 1997. Mafic and Ultramafic magmatism. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 233–254. Arnold, J., Powell, R., Sandiford, M., 2000. Amphibolites with staurolite and other aluminous minerals: calculated mineral equilibria in NCFMASH. Journal of Metamorphic Geology 18, 23–40. Babarina, I.I., 1998. Structural evolution of the Lake Kukas segment, North Karelia collision zone. Geotektonika 3, 80–96. Beccaluva, L., Serri, G., 1988. Boninitic and low-Ti subduction-related lavas from intraoceanic arcback-arc systems and low-Ti ophiolites: a reappraisal of their petrogenesis and original tectonic setting. Tectonophysics 146, 291–315. Bibikova, E.V., Skiöld, T., Bogdanova, S.V., Gorbatchev, R., Slabunov, A., 2001. Titanite-rutile thermochronometry across the boundary between the Archaean Craton in Karelia and the Belomorian Mobile Belt, eastern Baltic Shield. Precambrian Research 105, 315–330.

482

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Bibikova, E.V., Skiöld, T., Bogdanova, S.V., 1996. Age and geodynamic aspects of the oldest rocks in the Precambrian Belomorian belt, Baltic Shield. Geological Society of London Special Publication 112, 58–96. Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Bogdanova, S.V., 1996. High-grade metamorphism of 2.45–2.4 Ga age in mafic intrusions of the Belomorian Belt in the northeastern Baltic Shield. Geological Society of London Special Publication 112, 69–90. Bridgwater, D., Scott, D.J., Balagansky, V.V., Timmerman, M.J., Marker, M., Bushmin, S.A., Alexeyev, N.L., Daly, J.S., 2001. Age and provenance of early Precambrian metasedimentary rocks in the Lapland-Kola Belt, Russia: evidence from Pb and Nd isotopic data. Terra Nova 13, 32–37. Cameron, W.E., 1985. Petrology and origin of primitive lavas from the Troodos ophiolite, Cyprus. Contributions to Mineralogy and Petrology 89, 239–255. Cameron, W.E., Nisbet, E.G., Dietrich, V.J., 1979. Boninites, komatiites and ophiolitic basalts. Nature 280, 550–553. Condie, K.C., Viljoen, M.J., Kable, E.D.J., 1977. Effects of alteration on element distributions in Archean tholeiites from the Barberton greenstone belt, South Africa. Contributions to Mineralogy and Petrology 64, 75–89. Crawford, A.J., Fallon, T.J., Green, D.H., 1989. Classification, petrogenesis and tectonic setting of boninites. In: Crawford, A.J. (Ed.), Boninites. Unwin Hyman, London, pp. 2–44. Drummond, M.S., Defant, M.J., 1990. A model for trondhjemite-tonalite-dacite genesis and crustal growth via slab melting: Archean to modern comparisons. Journal of Geophysical Research 95, 21503–21521. Drumond, M.S., Defant, M.J., Kepezhinskas, P.K., 1996. Petrogenesis of slab-derived trondhjemitetonalite-dacite/adakite magmas. Geological Society of America Special Paper 315, 205–215. Duncan, R.A., Green, D.H., 1980a. The genesis of refractory melts in the formation of oceanic crust. Contributions to Mineralogy and Petrology 96, 326–342. Duncan, R.A., Green, D.H., 1980b. Role of multistage melting in the formation of oceanic crust. Geology 8, 22–26. Elton, D., 1989. Pressure of origin of primary mid-ocean-ridge basalts. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geological Society of London Special Publication 42, 125–136. Fallon, T.J., Crawford, A.J., 1991. The petrogenesis of high-calcium boninite lavas dredged from the northern Tonga ridge. Earth and Planetary Science Letters 102, 375–394. Fan, J., Kerrich, R., 1997. Geochemical characteristics of Al-depleted and undepleted komatiites and HREE-enriched tholeiites, western Abitibi greenstone belt: variable HFSE/REE systematics in a heterogeneous mantle plume. Geochimica et Cosmochimica Acta 61, 4723–4744. Fed’kin, V.V., 1975. Staurolites. Nauka, Moscow, p. 465 (in Russian). Fed’kin, V.V., Aranovich, L.Y., 1990. The Empirical Refinement of the Staurolite-Garnet Geothermometer; Experiment-89. Nauka, Moscow, pp. 41–43. Flower, M.F.L., Levine, H.M., 1987. Petrogenesis of a tholeiite-boninite sequence from Ayios Mamas, Troodos ophiolite: evidence for splitting of a volcanic arc. Contributions to Mineralogy and Petrology 97, 509–524. Gaál, G., Gorbatschev, R., 1987. An outline of the Precambrian evolution of the Baltic Shield. Precambrian Research 35, 15–52.

References

483

Green, T.H., Blundy, J.D., Adam, J., Yaxley, G.M., 2000. SIMS determination of trace element partition coefficients between garnet, clinopyroxene and hydros basaltic liquids at 2–7.5 Gpa and 1080–1200 ◦ C. Lithos 53, 165–187. Hickey, R.L., Frey, F.A., 1982. Geochemical characteristics of boninite series volcanics: Implications for their source. Geochimica et Cosmochimica Acta 46, 2099–2115. Hirose, K., Kushiro, I., 1993. Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters 102, 477–489. Hofmann, A.W., 1988. Chemical differentiation of the Earth: The relationship between mantle continental crust and oceanic crust. Earth and Planetary Science Letters 90, 297–314. Holland, T., Blundy, J., 1994. Non-ideal interactions in calcic amphiboles and their bearing on amphibole-plagioclase thermometry. Contributions to Mineralogy and Petrology 116, 433–447. Jochum, K.P., Arndt, N.T., Hofmann, A.W., 1990. Nb-Th-La in komatiites and basalts: constraints on komatiite petrogenesis and mantle evolution. Earth and Planetary Science Letters 107, 272–289. Kerrich, R., Wyman, D., Fan, J., Bleeker, W., 1998. Boninite series: low Ti-tholeiite associations from the 2.7 Ga Abitibi greenstone belt. Earth and Planetary Science Letters 164, 303–316. Koch-Müller, M., 1997. Experimentally determined Fe-Mg exchange between synthetic staurolite and garnet in the system MgO-FeO-Al2 O3 -SiO2 -H2 O. Lithos 41, 185–212. Kozhevnikov, V.N., 2000. Archean Greenstone Belts of the Karelian Craton as Accretionary Orogens. Karelian Science Center, Petrozavodsk, p. 223 (in Russian). Kozhevnikov, V.N., 1992. Geology and Geochemistry of the Archean North Karelian Greenstone Structures. Karelian Science Center, Petrozavodsk, p. 199 (in Russian). Koziol, A.M., Newton, R.C., 1989. Grossular activity-composition relationship in ternary garnets determined by reversed displaced-equilibrium experiments. Contributions to Mineralogy and Petrology 103, 423–433. Kretz, R., 1983. Symbols for rock-forming minerals. American Mineralogist 68, 277–279. Kushiro, I., 1990. Partial melting of mantle wedge and evolution of island arc crust. Journal of Geophysical Research 95, 15929–15939. Kusky, T.M., Li, J.H., Tucker, R.D., 2001. The Archean Dongwanzi ophiolite complex, North China Craton: 2.505-Billion-Year-Old oceanic crust and mantle. Science 292, 1142–1145. Kusky, T.M., Vearncombe, J.R., 1997. Structural Aspects. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 91–124. Le Bas, M.J., 2000. IUGS reclassification of the high-Mg and picritic volcanic rocks. Journal of Petrology 41, 1467–1470. Lobach-Zhuchenko, S.B., Chekulaev, V.P., Sergeev, S.A., Levchenkov, O.A., Krylov, I.N., 1993. Archaean rocks from southeastern Karelia (Karelian granite-greenstone terrain). Precambrian Research 62, 375–397. Ludden, J., Gelienas, L., Trudel, P., 1982. Archean metavolcanics from the Rouyn-Noranda distinct, Abitibi greenstone belt, Quebec: 2. Mobility of trace elements and petrogenetic constraints. Canadian Journal of Earth Sciences 19, 2276–2287. Martin, H., 1999. Adakitic magmas: modern analogues of Archean granitoids. Lithos 46, 411–429. Martin, H., 1986. Effect of steeper Archean geothermal gradient on geochemistry of subduction zone magmas. Geology 14, 753–756. McKenzie, D., Bickle, M.J., 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology 29, 625–629. McCulloch, M., 1993. The role of subducted slabs in an evolving earth. Earth and Planetary Science Letters 115, 89–100.

484

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

McCulloch, M.T., Gamble, J.A., 1991. Geochemical and geodynamical constraints on subduction zone magmatism. Earth and Planetary Science Letters 102, 358–374. Meffre, S., Aitchison, J.C., Crawford, A.J., 1996. Geochemical evolution and tectonic significance of boninites and tholeiites from the Koh ophiolite, New Caledonia. Tectonics 15, 67–83. Meijer, A., 1980. Primitive arc volcanism and a boninite series: examples from Western Pacific island arc. In: Hayes, D.E. (Ed.), The Tectonic and Geologic Evolution of Southeast Asian Sea and Islands. In: Geophysical Monographs, vol. 23, pp. 269–282. Miller, Yu.V., Mil’kevich, R.I., 1995. Thrust-fold structure of the Belomorian zone and its correlation with the Karelian granite-greenstone region. Geotektonika 6, 80–93. Moores, E.M., Twiss, R.J., 1995. Tectonics. W.H. Freeman and Co., New York, p. 414. Moscovchenko, N.I., Turchenko, S.I., 1975. Metamorphism of Kyanite-Sillimanite Type and Sulphidic Ores (Northern Karelia). Nauka, Leningrad, p. 137 (in Russian). Natland, J.H., 1981. Crystal morphologies and pyroxene compositions in boninites and tholeiitic basalts from Deep Sea Drilling Project Holes 458 and 459B in the Mariana fore-arc region. Initial Reports of the Deep Sea Drilling Project 60, pp. 681–707. Newton, R.C., Haselton, H.T., 1981. Thermodynamics of the garnet-plagioclase-Al2 SiO5 -quartz geobarometer. In: Newton, R.C., Navrotsky, A., Wood, B.J. (Eds.), Thermodynamics of Minerals and Melts. Springer, New York, pp. 131–147. Ohnenstetter, D., Brown, W.L., 1996. Boninites: a review. In: Demaiffe, D. (Ed.), Petrology and Geochemistry of Magmatic Suites of Rocks in the Continental and Oceanic Crust. A Volume Dedicated to Professor Jean Michot, Université Libre de Bruxell. Royal Museum for Central Africa, Tervuren, pp. 307–320. 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.). Geological Society of London Special Publication 16, 77–94. Pearce, J.A., Kempton, P.D., Nowell, G.M., Noble, S.R., 1999. Hf-Nd element and isotope perspective on the nature and provenance of mantle and subduction components in Western Pacific arcbasin systems. Journal of Petrology 40, 1579–1611. Pearce, J.A., van der Laan, S.R., Arculus, R.J., Murton, B.J., Ishii, T., Peate, D.W., Parkinson, I.J., 1992. Boninite and harzburgite from leg 125 (Bonin-Mariana forearc): a case study of magma genesis during the initial stages of subduction. Proceedings of the Ocean Drilling Program, Scientific Results 125, 623–659. Perchuk, L.L., 1989. Intercorrelation of Fe-Mg geothermometers using the Nernst law. Geokhimiya 5, 611–622. Poidevin, J.-L., 1994. Boninite-like rocks from the Palaeoproterozoic greenstone belt of Bogoin, Central African Republic: Geochemistry and petrogenesis. Precambrian Research 68, 97–113. Polat, A., Kerrich, R., 2001. Magnesian andesites, Nb-enriched basalt-andesites, and adakites from late-Archean 2.7 Ga Wawa greenstone belts, Superior Province, Canada: implication for late Archean subduction zone petrogenetic processes. Contributions to Mineralogy and Petrology 141, 36–52. Polat, A., Kerrich, R., 2004. Precambrian arc associations: Boninites, adakites, magnesian andesites, and Nb-enriched basalts. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 567–597. Prouteau, G., Maury, R.C., Sajona, F.G., Cotton, J., Joron, J.-L., 2000. Behavior of niobium, tantalum and other high field strength elements in adakites and related lavas from the Philippines. The Island Arc 9, 487–498.

References

485

Prouteau, G., Scaillet, B., Pichavant, M., Maury, R.C., 1999. Fluid-present melting of ocean crust in subduction zones. Geology 27, 1111–1114. Puchtel, I.S., Haase, K.M., Hofmann, A.W., Chauvel, C., Kulikov, V.S., Garbe-Schönberg, C.-D., Nemchin, A.A., 1997. Petrology and geochemistry of crustally contaminated komatiitic basalts from the Vetreny Belt, southeastern Baltic Shield: Evidence for an early Proterozoic mantle plume beneath rifted Archean continental lithosphere. Geochimica et Cosmochimica Acta 61, 1205– 1222. Puchtel, I.S., Hofmann, A.W., Amelin, Yu.V., Garbe-Schönberg, C.-D., Samsonov, A.V., Shchipansky, A.A., 1999. Combined mantle plume-island arc model for the formation of the 2.9 Ga Sumozero-Kenozero greenstone belt, SE Baltic Shield: Isotope and trace element constraints. Geochimica et Cosmochimica Acta 63, 3579–3595. Puchtel, I.S., Hofmann, A.W., Mezger, K., Shchipansky, A.A., Samsonov, A.V., 1998. Oceanic plateau for continental crustal growth in the Archean: a case study from the Kostomuksha greenstone belt, Nw Baltic Shield. Earth and Planetary Science Letters 155, 57–74. Riganti, A., Wilson, A.H., 1995. Geochemistry of the mafic/ultramafic volcanic associations of the Nondweni greenstone belt, South Africa, and constraints on their petrogenesis. Lithos 34, 235– 252. Sajona, F.G., Maury, R.C., Prouteau, G., Cotton, J., Schiano, P., Bellon, H., Fontaine, L., 2000. Slab melt as metasomatic agent in island arc magma mantle sources, Negros and Batan (Philippines). The Island Arc 9, 472–486. Shchipansky, A.A., Babarina, I.I., Krylov, K.A., Samsonov, A.V., Bogina, M.M., Bibikova, E.V., Slabunov, A.I., 2001. The oldest ophiolites: the late Archean suprasubduction zone complex of the Iringora structure, North Karelian greenstone belt. Doklady Earth Science 377a, 283–287. Shchipansky, A.A., Samsonov, A.V., Bogina, M.M., Slabunov, A.I., Bibikova, E.V., 1999. HighMg, low-Ti quartz amphibolites of the Khizovaara greenstone belt, Northern Karelia: Archean metamorphosed boninites? Doklady Earth Science 365a, 817–820. Shervais, J.W., 2001. Birth, death, and resurrection: The life cycle of suprasubduction zone ophiolites. Geochemistry, Geophysics, Geosystems 2, 2000GS000080. Shervais, J.W., 1982. Ti-V plots and petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59, 101–118. Smellie, J.L., Stone, P., Evans, J., 1995. Petrogenesis of boninites in the Ordovician Ballantrae Complex ophiolite, southwestern Scotland. Journal of Volcanology and Geothermal Research 69, 323– 342. 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. Sobolev, A.V., Portnyagin, M.V., Dmitriev, L.V., Tsamerian, O.P., Danyushevsky, L.V., Kononkova, N.N., Shimizu, N., Robinson, P.T., 1993. Petrology of ultramafic lavas and associated rocks of the Troodos Massif, Cyprus. Petrology 1, 331–361. Sorjonen-Ward, P., Nironen, M., Luukkonen, E., 1997. Greenstone associations in Finland. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 676–698. Staudigel, H., Tauxe, L., Gee, J.S., Bogaard, P., Haspels, J., Kale, G., Meijer, P., Swaak, B., Tuin, M., Van Soest, M.C., Verdurmen, E.A.Th., Zevenhuizen, A., 1999. Geochemistry and intrusive directions in sheeted dikes in the Troodos ophiolite: Implications for Mid-ocean ridge spreading centers. Geochemistry, Geophysics, Geosystems 1, 1999GS000001.

486

Chapter 14: 2.8 Ga Boninite-Hosting Partial Suprasubduction Zone Ophiolite Sequences

Stein, E., Dietl, C., 2001. Hornblende thermobarometry of granitoids from the Central Odenwald (Germany) and their implications for the geotectonic development of the Odenwald. Mineralogy and Petrology 72, 185–207. Stern, R.J., Bloomer, S.H., 1992. Subduction zone infancy: Examples from the Eocen Izu-BoninMariana and Jurassic California arcs. Geological Society of America Bulletin 104, 1621–1636. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.S., 1997. Volcanic aspects. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 55–90. Sun, S.-s., Nesbitt, R.W., 1978. Geochemical regularities and genetic significance of ophiolitic basalts. Geology 6, 689–693. Sun, S.-s., Nesbitt, R.W., McCulloch, M.T., 1989. Geochemistry and petrogenesis of Archaean and early Proterozoic siliceous high-magnesium basalts. In: Crawford, A.J. (Ed.), Boninites. Unwin Hyman, London, pp. 148–173. Thurston, P.S., Kozhevnikov, V.N., 2000. An Archean quartz arenite-andesite association in the eastern Baltic Shield, Russia: implications for assemblage types and shield history. Precambrian Research 101, 313–340. Walker, D., Shibata, T., De Long, S.E., 1979. Abyssal tholeiites from the Oceanographic fracture zone II, phase equilibria and mixing. Contributions to Mineralogy and Petrology 70, 111–125. Whitehouse, M., Claesson, S., Sunde, T., Vestin, J., 1997. Ion microprobe U-Pb zircon geochronology and correlation of Archaean gneisses from the Lewisian Complex of Gruinard Bay, northwestern Scotland. Geochimica et Cosmochimica Acta 61, 4429–4438. Wyman, D.A., 1999. Paleoproterozoic boninites in an ophiolite-like setting, Trans-Hudson orogen, Canada. Geology 27, 455–458.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

487

Chapter 15

THE BELINGWE GREENSTONE BELT: ENSIALIC OR OCEANIC? AXEL HOFMANNa AND TIM KUSKYb a School of

Geosciences, University of the Witwatersrand, P. Bag 3, Wits 2050, South Africa b Department of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63103, USA

The Belingwe belt in Zimbabwe is probably the best preserved late Archaean greenstone belt known. But despite the presence of well exposed rocks of very low metamorphic grade and low strain, the tectonic evolution of the Belingwe belt has been a matter of much controversy, suggesting that greenstone belt geology remains poorly understood. The Belingwe greenstone belt comprises two distinct greenstone successions. The lower, 2.9–2.8 Ga old Mtshingwe Group consists of four stratigraphic units, an intermediate to felsic volcanic and volcanoclastic unit, an ultramafic to mafic lava plain sequence, a conglomerate-shale sedimentary sequence, and a unit of tectonically imbricated sedimentary and volcanic rocks. Although geochronological, geochemical and lithological characteristics are broadly known, the tectonic evolution of the Mtshingwe Group remains a matter of speculation. Controversy surrounds the intensely studied, 2.7 Ga old Ngezi Group, which consists of a thin basal sedimentary sequence, a thick ultramafic to mafic volcanic sequence, and an upper sedimentary succession. The basal unit rests unconformably on up to 3.5 Ga old granitoid gneisses and Mtshingwe Group rocks, consists of fluvial to shallow-marine sedimentary rocks, and is similar to cratonic cover successions. The overlying volcanic unit is a submarine lava plain sequence of massive and pillow basalts with komatiites near the base and andesites near the top. The upper sedimentary unit represents a foreland basin sequence and consists of karstified carbonate ramp limestones overlain by deeper-water turbidite deposits. Autochthonous versus allochthonous models have been proposed for the tectonic evolution of the Ngezi Group. Proponents of the autochthonous model regard the Ngezi Group as a conformable sequence that formed in an ensialic rift setting above a mantle plume. Other workers regard the volcanic sequence as an allochthonous unit of oceanic crust that was obducted onto continental basement. A great number of arguments have been proposed in recent years from structural, sedimentological, and geochemical studies for and against the different models. A critical reappraisal of the various arguments indicate the lack of convincing evidence for an ensialic and autochthonous origin. Arguments for an allochthonous origin are strong, whereas an oceanic origin can DOI: 10.1016/S0166-2635(04)13015-6

488

Chapter 15: The Belingwe Greenstone Belt

only be inferred by assuming that modernistic plate tectonic processes were operating in the late Archaean.

1. INTRODUCTION The tectonic setting of Archaean greenstone belts is controversial. Greenstone belts may have originated from plate tectonic-style horizontal tectonic accretion of oceanic and island arc volcano-sedimentary sequences with continental crustal fragments (Kusky, 1989; Card, 1990; Swager and Griffin, 1990; de Wit et al., 1992; Myers, 1995; Kusky and Vearncombe, 1997; Dirks and Jelsma, 1998a, 2002) or superposition of continental, riftrelated volcanic sequences on already existing segments of differentiated crust (Campbell and Hill, 1988; Bouhallier et al., 1993; Bickle et al., 1994; Jelsma et al., 1996; Hamilton, 1998; Collins et al., 1998). Greenstone belts of the Zimbabwe craton, and the Belingwe belt in particular, feature strongly in this debate. The evolution of the Zimbabwe craton involved a number of stages, each of which was represented by the deposition of a volcano-sedimentary sequence and the emplacement of granitoids (Wilson et al., 1995). The oldest segment of granitoid crust is the 3.6–3.5 Ga Tokwe gneiss complex (Fig. 1) which contains greenstone enclaves of the 3.5 Ga Sebakwian Group. The majority of the volcano-sedimentary sequences formed between 2.9 and 2.6 Ga and include the Lower Bulawayan (2.9–2.8 Ga), Upper Bulawayan (2.7 Ga) and Shamvaian (2.65 Ga) Groups, and granitoids of the Chingezi (2.9–2.8 Ga), Sesombi (2.7 Ga) and Wedza (2.65 Ga) suites (Stagman, 1978; Wilson, 1979; Wilson et al., 1995). Stabilization of the craton was achieved around 2600 Ma ago with the emplacement of large volumes of crustally derived Chilimanzi suite granites (Wilson et al., 1995; Jelsma et al., 1996; Horstwood, 1998; Jelsma and Dirks, 2002), followed by intrusion of the layered igneous complex of the Great Dyke at 2576 Ma (Oberthür et al., 2002). Many workers have interpreted the Zimbabwe craton as vertically accreted crust. According to this model, coherent units of volcano-sedimentary rocks were laid down in rifts on top of older continental basement (ensialic model) and underwent little deformation until the late-stage emplacement of granite-gneiss complexes (Bickle et al., 1975; Blenkinsop et al., 1993; Bickle et al., 1994; Shackleton, 1995; Wilson et al., 1995; Ridley et al., 1997; Horstwood et al., 1999). Vertical accretion of the Zimbabwe craton was first proposed by Macgregor (1951) to explain the geometry of arcuate greenstone belts surrounded by subelliptical granitoid batholiths. The model was supported by the recognition of an unconformity in the Belingwe greenstone belt where a thin succession of Upper Bulawayan Group sedimentary rocks capped by a conspicuous chert horizon (Manjeri Formation) overlies granitoid gneisses and older greenstones (Bickle et al., 1975). The chert horizon is overlain, purportedly conformably, by a several kilometer thick sequence of submarine komatiites and basalts. A similar sequence of basement rocks and older greenstones, chert (with or without underlying sedimentary rocks) and a thick volcanic succession was observed in several greenstone belts in Zimbabwe (Wilson, 1979). As a result, the chert horizon in the Belingwe belt has been regarded as a stratigraphic marker horizon and

1. Introduction

489

Fig. 1. Simplified geological map of the Zimbabwe craton (modified after Kusky and Winsky, 1995) and its tectonic setting in southern Africa (inset).

used to construct a craton-wide stratigraphy based on the Belingwe model (Wilson, 1979; Wilson et al., 1995). The existence of such a layer-cake greenstone stratigraphy, underlain by a basement unconformity, led to the suggestion that continental crust was covered by a vast volcano-sedimentary succession before it was disturbed by vertical processes of granite diapirism. This interpretation ignored the occurrence of shear zones parallel to the stratigraphy, many of which coincide with low-angle truncation planes. Layer-parallel shear zones have been recognized in the past, but they were generally interpreted as insignificant (Blenkinsop et al., 1993), or being the direct result of gravity sliding away from the centers of rising diapiric domes (Stowe, 1984; Jelsma et al., 1993). Recent work has suggested otherwise (Kusky, 1991; Kusky and Kidd, 1992; Kusky and Winsky, 1995; Dirks and van der Merwe, 1997; Dirks and Jelsma, 1998b; Jelsma and Dirks, 2002;

490

Chapter 15: The Belingwe Greenstone Belt

Hofmann et al., 2001a, 2003a). In the Belingwe greenstone belt, Kusky and Kidd (1992) argued that the contact between the Manjeri Formation and the overlying volcanic sequence is sheared. They inferred that the shear zone represents a major décollement and that the volcanic unit represents part of an oceanic plateau that was obducted as a large allochthonous thrust sheet onto granitoid basement and older greenstone sequences. The speculations of Kusky and Kidd (1992) were disputed in the literature (Blenkinsop et al., 1993; Bickle et al., 1994; Kusky and Winsky, 1995; Hunter et al., 1998), but no consensus was achieved. However, clarification of the tectonic evolution of greenstone belts is very important, because this directly reflects on models for the evolution of mantle processes and continental growth through time. The Belingwe greenstone belt is one of the few localities were a conformable relationships between basement gneiss-shallow water shelf succession-submarine ultramafic and mafic volcanic sequence has been described and used as constraints for non-actualistic models on Archaean tectonics and volcanism (e.g., Bickle et al., 1994; Arndt, 1999). In younger orogens on the other hand, similar lithological associations are typically allochthonous (e.g., Dewey and Bird, 1971; Ben-Avraham et al., 1981; St-Onge and Lucas, 1990; Burchfiel et al., 1992; Saleeby, 1992; Kusky et al., 1997; Moores, 2002; Moores et al., 2000) and reflect the operation of plate tectonic processes. In the last few years new sedimentological, structural and geochemical information for the Belingwe greenstone belt has been obtained from different groups. The controversy for an ensialic and autochthonous vs oceanic and allochthonous origin however remains. In this paper we provide a review on the geology of the Belingwe greenstone belt and summarize the evidence for and against an oceanic origin for the Belingwe greenstone belt.

2. GEOLOGICAL SETTING The Belingwe greenstone belt is situated in the southern part of the Zimbabwe Craton, north of the Limpopo belt and east of the Great Dyke (Fig. 1). It contains a well preserved, 2.9–2.65 Ga old volcano-sedimentary sequence which is surrounded by granitoid-gneiss terrains, the 3.5 Ga Shabani gneiss complex to the east and the 2.9 Ga Chingezi gneiss complex to the west (Fig. 2). The greenstone sequence is subdivided into two distinct units, the lower Mtshingwe Group and the upper Ngezi Group (Fig. 3). The Mtshingwe Group is assigned to the Lower Bulawayan Group and is confined to the southeastern and western parts of the belt. It comprises the Hokonui (mafic to felsic volcanic and volcanoclastic rocks), Bend (komatiite, basalt, iron formation), Brooklands (sedimentary rocks, komatiite, basalt) and Koodoovale (shale, conglomerate) Formations (Martin et al., 1993). Orpen (1978) recognized an additional unit, the Bvute Formation (isoclinally folded amphibolites), that occurs between the Chingezi gneiss complex and the Hokonui Formation. This unit is probably the more strongly strained and metamorphosed equivalent of the Hokonui Formation and will not be discussed further. The Ngezi Group is assigned to the Upper Bulawayan Group and comprises the Manjeri (sedimentary rocks), Reliance (komatiite,

2. Geological Setting

Fig. 2. Geological map of the Belingwe greenstone belt (modified after Martin et al., 1993).

491

492

Chapter 15: The Belingwe Greenstone Belt

Fig. 3. Stratigraphic logs of the western and eastern parts of the Belingwe greenstone belt. Important shear zones are shown.

basalt), Zeederbergs (basalt, andesite) and Cheshire (conglomerate, shale, limestone) Formations. It forms the central part of the Belingwe greenstone belt, stretching its entire length (c. 70 km) with a maximum exposure width of 20 km. The Ngezi Group overlies the Shabani gneiss complex to the east and the Mtshingwe Group to the southeast and west. Chilimanzi suite monzogranites intrude the Ngezi Group in the northern and southern parts of the belt (Fig. 2). Most studies of the Belingwe greenstone belt have concentrated on the Ngezi Group, given the good preservation and low metamorphic grade of this stratigraphic unit. Controversy surrounds the tectonic setting of the mafic volcanic sequence of the Reliance and Zeederbergs Formations (oceanic crust vs continental flood basalts). Much less is known for the Mtshingwe Group greenstones and associated granitoids, the geology of which will be addressed only briefly.

3. Shabani Gneiss Complex: 3.5 Ga Basement Gneisses

493

3. SHABANI GNEISS COMPLEX: 3.5 GA BASEMENT GNEISSES Granitoid rocks of the Shabani gneiss complex crop out to the east of the Belingwe belt (Fig. 2). The Shabani gneiss consists of magmatic, banded and massive TTG (tonalitetrondhjemite-granodiorite) gneiss similar in composition and age to the Tokwe gneiss to the east of the Mashaba tonalite (Taylor et al., 1991). Different phases have been dated between 3554 and 3368 Ma (dates summarized in Kusky, 1998 and Jelsma and Dirks, 2002). Together they form the Shabani-Tokwe gneiss complex which forms the mini-craton of the Tokwe segment (Fig. 1; Wilson, 1979, 1990). The inferred extent of this crustal block is considered by Kusky (1998) and Horstwood et al. (1999) to underlie most of the Zimbabwe craton and has been referred to as the Tokwe terrane (3.55–2.95 Ga) or Sebakwe protocraton (pre-3.35 Ga) by these authors. The gneisses have Sm-Nd TDM ages between 3.65 and 3.51 Ga (Taylor et al., 1991; Hunter, 1997; Horstwood, 1998) and low initial Sr ratios (0.700–0.701; Hawkesworth et al., 1975; Moorbath et al., 1977), suggesting that they were derived from primitive magmas. Strongly deformed supracrustal gneisses form meter- to kilometer-scale inclusions in the Shabani gneiss complex. The inclusions include amphibolite-grade mafic and ultramafic rocks, metasedimentary schists, and banded ironstones, and are regarded as greenstone belt remnants of the pre-3.35 Ga Sebakwian Group. U-Pb zircon ages of migmatites in the Tokwe segment have identified two distinct amphibolite facies metamorphic events at c. 3.35 and 2.86 Ga (Nägler et al., 1997; Horstwood et al., 1999), resulting in anatexis and crustal reworking.

4. CHINGEZI SUITE GNEISSES: 2.9 GA ARC GRANITOIDS The 2.9–2.8 Ga Chingezi suite (Wilson et al., 1995) in the Belingwe area comprises the Mashaba and Chingezi tonalite and the Chingezi gneiss complex. Supracrustal rocks of this age form the Mtshingwe Group in the Belingwe belt. The Mashaba tonalite is a weakly foliated, north-south striking elongate intrusion to the east of the belt (Fig. 2). It separates the Shangani from the Tokwe gneisses and has been dated at c. 2950 Ma (DoughertyPage, 1994). The Chingezi gneiss complex crops out to the west of the Belingwe belt. It comprises a suite of magmatic and banded TTG gneisses. A number of weakly foliated plutons, such as the Chingezi tonalite, are intrusive bodies within the Chingezi gneiss complex. Hawkesworth et al. (1979) reported a Rb-Sr whole rock age of 2810 ± 70 Ma for the complex. The Chingezi tonalite is a composite body which has yielded a whole-rock Pb/Pb age of 2874 ± 32 Ma (Taylor et al., 1991) where it intrudes the Mtshingwe Group, providing a minimum age for the greenstone sequence. The Chingezi suite gneisses west of the Belingwe belt show primitive isotopic characteristics and lack significantly older inherited components. This is corroborated by their low initial Sr ratios (0.701–0.702; Hawkesworth et al., 1979), relatively low µl-values (8.1– 8.4; Taylor et al., 1991), and relatively young TDM ages (between 3.32 and 3.12 Ga; Taylor et al., 1991), which suggest that it is unlikely that much older crustal material was involved

494

Chapter 15: The Belingwe Greenstone Belt

in their petrogenesis. This is in contrast with the isotopic characteristics of Chingezi suite gneisses within the Tokwe segment, such as the Mashaba tonalite, that have similar initial Sr ratios (0.701; Hawkesworth et al., 1979), but older TDM ages (3.4 Ga; Hunter, 1997) and contain significantly older inherited zircons (Dougherty-Page, 1994; Hunter, 1997).

5. MTSHINGWE GROUP: 2.9 GA LOWER GREENSTONE SEQUENCE The Mtshingwe Group includes the Hokonui, Bend and Koodoovale Formations in the west and the Brooklands Formation in the southeast. The majority of the rocks have undergone greenschist facies metamorphism. The Hokonui Formation (Fig. 4) comprises andesitic to dacitic lavas, agglomerates and tuffs, and mafic lavas and sills (Orpen et al., 1993). Shale, sandstone and conglomerate consisting of reworked intermediate to felsic volcanic material are frequently intercalated with the volcanic rocks and represent subwave base density current deposits, suggesting a predominantly deep-water environment. A pyroclastic breccia with tonalite blocks up to 10 m in diameter has been reported by Orpen et al. (1993) from one locality and has been interpreted as a vent agglomerate. Our own investigations indicate a tectonic origin for the breccia with the tonalite blocks representing disrupted felsic dykes in an agglomerate matrix. The stratigraphic thickness of the Hokonui Formation is unknown because of folding of the sequence. A U-Pb zircon age of 2904 ± 9 Ma (2950 Ma inheritance) from a felsic clast of a Hokonui volcanic breccia has been obtained by Wilson et al. (1995). The lower contact with the c. 2.9 Ga Chingezi gneiss complex is sheared, or intruded by the 2.87 Ga Chingezi tonalite. The Bend Formation (∼ 2.5 km thick) overlies the Hokonui Formation along a sheared contact (Orpen, 1978) and comprises volcanic units of cumulate- and spinifex-textured komatiite and komatiitic basalt, and pillowed and massive tholeiitic basalt that are bounded by banded ironstone and chert horizons (Martin et al., 1993; Orpen et al., 1993). Ten ironstone horizons (Fig. 4c), typically 2–30 m thick and continuous for several kilometers, have been mapped (Orpen et al., 1993). The ironstones are made up of alternating thin beds of finely parallel-laminated and banded chert and hematite- to magnetite-rich chert. The lithological association suggests that the Bend Formation represents a deep-water lava plain sequence. The ironstone horizons were interpreted to have been deposited during intervals of volcanic quiescence. Furthermore, the continuity of the ironstone-chert horizons has been regarded as evidence “that the Bend Formation has not undergone major internal disruption” (Orpen et al., 1993). However, the ironstones are locally strongly deformed and talc schists are common in the sequence. Judging from the observation of similar, but shear-zone-bounded basalt-chert cycles in the Midlands greenstone belt (Dirks et al., 2002), the Bend volcanic cycles may well represent tectonic cycles, or an intensely imbricated basalt/chert sequence (e.g., Kusky and Vearncombe, 1997; Kusky and Bradley, 1999). The Koodoovale Formation (∼ 1 km thick) overlies banded ironstones of the Bend Formation. According to Orpen et al. (1993), the contact is erosive. The majority of the formation consists of massive conglomerate and poorly exposed shale. Clasts in conglomerate are well rounded, up to 50 cm in diameter and include granite and tonalite, porphyritic

5. Mtshingwe Group: 2.9 Ga Lower Greenstone Sequence

495

Fig. 4. Features of Mtshingwe Group rocks. (a) Dacite clast conglomerate, Hokonui Formation. (b) Turbiditic sandstone and shale, Hokonui Formation. (c) Succession of banded iron formation and intervening volcanic rocks, Bend Formation. (d) Isoclinally folded banded ironstone, Bend Formation. (e) Pillow basalt, Roselyn Member, Brooklands Formation. (f) Sheared conglomerate, Mnene River Member, Brooklands Formation. Scale bar is 10 cm long; lens cap is 5 cm in diameter.

felsite, mafic volcanic rock, spinifex-textured komatiite, tremolite-chlorite schist, dolerite and abundant chert and ironstone clasts, suggesting a granite-greenstone provenance. Bedding is locally discernible by conglomerate layers interbedded with thin to medium, partly normally graded sandstone beds. The facies association suggests a subaquatic, possibly deep-water origin. A felsic agglomerate of rhyolite and dacite clasts in a felsic matrix has

496

Chapter 15: The Belingwe Greenstone Belt

Fig. 5. Geological map of the Brooklands Formation (modified after Nisbet et al., 1993a).

been reported from one locality (Orpen et al., 1993). A dacite clast yielded a U-Pb zircon age of 2831 ± 6 Ma (Wilson et al., 1995). An inherited zircon yielded an age of 2880 Ma. The Brooklands Formation (∼ 4 km thick) in the southeastern part of the belt (Fig. 5) is a volcano-sedimentary sequence that overlies the Shabani gneiss complex along a sheared contact. It is divided into four members, the Ndakosi, Roselyn, Mnene River and Pemba

6. Mtshingwe Group: a Working Model

497

Members (Nisbet et al., 1993a). The sedimentary Ndakosi Member comprises chloritic phyllite, siltstone, quartzite, quartzite breccia, and minor ironstone and chert as well as mylonitic muscovite-chlorite-quartz schist in high-strain zones. The Roselyn Member consists of massive and pillow basalt, komatiitic basalt, minor spinifex-textured komatiite and strongly deformed ultramafic rocks, including chlorite-talc schist and serpentinite. The sedimentary Mnene River Member consists, at its base, of a laterally persistent conglomerate horizon with ultramafic and mafic volcanic, metasandstone and chert clasts (Fig. 4f). The conglomerate is overlain by a thick unit of shale that grades into banded iron formation upsection (Nisbet et al., 1993a). The Pemba Member consists of massive mafic volcanic rocks that are locally pillowed and show spinifex texture. Chert, ironstone and shale occur in subordinate proportions. Contacts between the different lithological units are sharp and sheared, and the contacts between sedimentary and overlying volcanic units appear to be major shear zones. Intrusive ultramafic rock bodies, now serpentinites with relic cumulate textures, are common. The ultramafic rocks range from massive to sheared and are variably carbonatized and silicified. Detrital zircons from a Ndakosi Member quartzite give ages ranging between 3.60 and 3.10 Ga (Hunter, 1997), providing a lower age limit to sedimentation. This is corroborated by Sm-Nd TDM crustal residence ages between 3.64 and 3.16 Ga for fine-grained sedimentary rocks (Hunter, 1997). Basalts from the Hokonui, Bend and Brooklands Formations have similar geochemical signatures (Bolhar, 2001; Bolhar et al., 2003a). Two geochemically distinct groups have been differentiated, a group of unfractionated to depleted basalts ((La/Sm)PM = 0.8–0.9, (Nb/La)PM = 0.7–0.9) and a group of light rare earth element (LREE)-enriched) basalts with distinct negative Nb-Ta-Ti anomalies ((La/Sm)PM = 1.3–2.7, (Nb/La)PM = 0.4–0.6). εNd(t) values for Hokonui Formation basalts range from +0.6 to +1.5. The geochemical signatures of the second group of rocks have been attributed to assimilation of continental crust followed by fractional crystallization (Bolhar et al., 2003a).

6. MTSHINGWE GROUP: A WORKING MODEL The evolution of the Mtshingwe Group is poorly understood. With the currently available data, only a working model can be presented which needs to be tested in the course of further studies. The lithological association and inferred environment of deposition of the Hokonui Formation suggests that it represents a volcanic arc-like sequence that may or may not have been attached to the western margin of the Tokwe segment at 2.9 Ga. This interpretation is corroborated by the age and isotopic signatures (Hawkesworth et al., 1979; Taylor et al., 1991) of the Chingezi gneiss complex that can be regarded as the plutonic root zone of the volcanic arc. The similarity in age between Hokonui Formation volcanoclastic rocks (2904 Ma) and the Chingezi Tonalite (c. 2874 Ma) suggests that plutonism and volcanism may have been coeval. Zircon xenocryst ages in the felsic volcanics are only up to 50 Ma older than the corresponding crystallization ages (Wilson et al., 1995), suggesting the absence of much older continental basement in the arc root.

498

Chapter 15: The Belingwe Greenstone Belt

The compositional contrast and tectonic contact between the Hokonui and Bend Formation suggests an allochthonous origin for the latter. The Bend Formation is a submarine lava-plain sequence and thus bears similarities with an obducted remnant of an oceanic sequence. If one wants to follow this line of reasoning, two different tectonic settings for the Bend Formation could be envisaged prior to obduction onto the Hokonui arc sequence. (1) The Bend Formation represents a back-arc basalt sequence that formed in between the Hokonui volcanic arc and the Tokwe segment and was obducted during arc-continent collision. (2) It represents the substratum of a fore-arc basin. The origin of the Koodoovale Formation cannot be understood without better knowledge on the tectonic setting of the underlying Bend Formation. It may have formed prior to obduction of the Bend Formation, representing, for example, a back-arc basin sedimentary sequence, or it may have formed as a result of the crustal movements that gave rise to juxtaposition of the Hokonui and Bend tectonostratigraphic units. The Brooklands Formation has been interpreted to represent passive-margin (Hunter, 1997) or rift basin deposits (Nisbet et al., 1993a). Both interpretations need to be seriously questioned however, firstly, because of the association of submarine komatiitic lavas and basalts with quartzose sedimentary units, and, secondly, because contacts between the sedimentary and volcanic units are shear zones. Nowhere are volcanic and quartzose sedimentary rocks interbedded with each other. This does not mean that the sedimentary units did not form in a passive margin or rift setting, but such environments cannot be inferred for the Brooklands Formation as a whole. The relationship between the Brooklands Formation and the other stratigraphic units of the Mtshingwe Group to the west of the belt is unknown. The sedimentary and volcanic units of the Brooklands Formation may represent tectonically duplicated slices of Bend and Koodoovale Formation rocks. However, they may also represent different units altogether, unrelated in space and time to the western Mtshingwe Group greenstones. Based on similar detrital zircon age populations (data of Hunter, 1997), Kusky (1998) and Jelsma and Dirks (2002) suggested that the Brooklands Formation is part of a 3.1 to 2.95 Ga sedimentary sequence that can be found in the Buhwa and Shurugwi greenstone belts (Fig. 1; Dodson et al., 1988). The geochemical signatures of enriched mafic volcanic rocks reported from the Hokonui, Bend and Brooklands Formations by Bolhar et al. (2003a) have been attributed to a crustal signal, implying involvement of subcontinental lithospheric mantle or assimilation of continental crust during magma genesis rather than magmatic processes in a subduction zone environment. A nearcontinental tectonic setting for the volcanic rocks would thus be implied. The significance of the geochemical data, which are similar to analytical results obtained for volcanic rocks of the Ngezi Group, will be dealt with in more detail later.

7. NGEZI GROUP: 2.7 GA UPPER GREENSTONE SEQUENCE The Ngezi Group comprises a basal sedimentary unit (Manjeri Formation), overlain by ultramafic and mafic volcanic rocks (Reliance and Zeederbergs Formations, collectively termed Ngezi volcanics) and capped by a sedimentary succession (Cheshire Formation).

7. Ngezi Group: 2.7 Ga Upper Greenstone Sequence

499

The majority of the rocks have undergone greenschist to sub-greenschist facies metamorphism. Amphibolite facies assemblages are confined to contact aureoles around Chilimanzi suite granite intrusions. Controversy surrounds the tectonic setting in which the supracrustal rocks formed. The Ngezi Group has been interpreted by many authors (e.g., Bickle et al., 1975; Martin, 1978; papers in Bickle, 1993; Blenkinsop et al., 1993; Bickle et al., 1994; Hunter et al., 1998; Bolhar et al., 2003a) as an ensialic and autochthonous sequence. In contrast, Kusky and Kidd (1992) and Kusky and Winsky (1995) argued that the Ngezi volcanics originated from an oceanic plateau and were emplaced as a large thrust sheet, the Mberengwa allochthon, onto the Manjeri Formation and underlying continental basement. Further evidence that thrust tectonics affected the Ngezi Group was provided in a series of papers on the Cheshire Formation (Hofmann et al., 2001a, 2001b, 2003a, 2003b). Nevertheless, whether the Ngezi volcanics are oceanic or continental is still a matter of debate (Hofmann and Dirks, 2003). 7.1. Manjeri Formation The Manjeri Formation unconformably overlies granitoid gneisses of the Shabani gneiss complex in the east and Mtshingwe Group greenstones in the south-east and west of the Belingwe greenstone belt (Fig. 2). It has a maximum thickness of 250 m and is absent at several localities except for a continuous ironstone horizon. Hunter et al. (1998) described the Manjeri Formation as consisting of three stratigraphic units. The lower Spring Valley Member consists of fluvial to shallow-water sedimentary rocks, including conglomerate, sandstone, shale, banded ironstone and localized stromatolitic limestone (Fig. 6). The middle Rubweruchena Member comprises alluvial fan to fan delta deposits, mainly conglomerate and sandstone. The top of the sequence is a 5 to 10 m thick horizon (Jimmy Member) of sheared shale, chert and massive sulfide, interpreted as a regional shear zone (Kusky and Winsky, 1995; Hofmann et al., 2003a). Grassineau et al. (2000) reported a wide range (−20% to +17%) of δ 34 S values from Jimmy Member sulfides, which was attributed to biological fractionation, and a sharp peak at 0%, attributed to hydrothermal processes. The Manjeri Formation contains a chlorite-sericite-quartz assemblage, indicating very low metamorphic grade (Martin, 1978; Abell et al., 1985; Bickle et al., 1993). Trace element characteristics and Sm-Nd model ages of Manjeri rocks are consistent with sediment derivation from an early to mid-Archaean terrain of mixed mafic and felsic composition (Hunter et al., 1998). Stromatolitic limestone of the Spring Valley Member yielded a Pb-Pb isochron date of 2706 ± 49 Ma (Bolhar et al., 2002). A comparison of stratigraphic sections from the eastern side of the belt (Fig. 7) shows that the lower Spring Valley Member is laterally relatively continuous and retains a similar thickness. The overlying Rubweruchena Member is discontinuous and shows marked facies changes over a short distance; the section resembles a display of a horst-and-graben depositional setting. Rapid facies changes along strike have been stated as a characteristic feature of the Manjeri Formation (Orpen, 1978). In contrast, the overlying ironstone horizon is laterally continuous and equally thick. Similar relationships can be observed in other parts of the greenstone belt where contacts between the Reliance Formation and underly-

500

Chapter 15: The Belingwe Greenstone Belt

Fig. 6. Features of Ngezi Group rocks. (a) Unconformable contact between foliated tonalite of the Shabani gneiss complex and Manjeri Formation. (b) Typical outcrop appearance of Jimmy Member, Manjeri Formation. Note anastomosing cleavage domains and boudinaged chert layers. (c) Graded bedding in Reliance Formation interflow sedimentary rocks. Accretionary lapilli are common in these beds. (d) Pillow basalt, Zeederbergs Formation. (e) Stromatolitic limestone, carbonate member of the Cheshire Formation. (f) Massive basalt pebble conglomerate, siliciclastic member of the Cheshire Formation. Scale bar is 10 cm long; lens cap is 5 cm in diameter.

7. Ngezi Group: 2.7 Ga Upper Greenstone Sequence

501

Fig. 7. Simplified graphic logs of the Manjeri Formation (after Hofmann et al., 2003a; see Fig. 2 for location of logs). Note the lateral discontinuity of the Spring Valley and Rubweruchena Members and the continuity of the Jimmy Member.

ing rocks (Mtshingwe Group greenstones, granitoids) are occupied by the Jimmy Member ironstone with or without variably thick Manjeri sedimentary rocks below the ironstone. Sulfidic ironstones identical to the Jimmy Member ironstone occur locally within or at the base of the Manjeri Formation. These ironstones have been termed chert tectonite (Kusky and Winsky, 1995) and tectonic ironstones, because they represent silicified and sulfideimpregnated shear zones (Hofmann et al., 2003a). 7.2. Ngezi Volcanics The Reliance Formation (∼ 1 km thick) can be divided into three stratigraphic units (Fig. 8; Nisbet et al., 1977). The lower unit (400–500 m) consists of pillowed and massive tholeiitic and komatiitic basalt with minor graded tuff beds and a differentiated ultramafic sill. The central unit (300 m) comprises a komatiite suite of pillow lavas and lava flows with a few cross-cutting dykes. Olivine is typically serpentinized, except for a few localities where relics of fresh olivine and clinopyroxene occur (Nisbet et al., 1987). The upper unit (200 m) includes poorly exposed komatiitic basalt, tholeiitic basalt and graded lapilli tuff

502

Chapter 15: The Belingwe Greenstone Belt

Fig. 8. Simplified graphic log of the Reliance Formation type section (modified after Nisbet et al., 1977).

and passes upwards into the volcanic rocks of the Zeederbergs Formation. Komatiitic basalt of the Reliance Formation has been dated at 2692 ± 9 Ma (whole-rock Pb/Pb; Chauvel et al., 1993), whilst olivine and spinel concentrates from a komatiite yielded a Re-Os age of 2721 ± 21 Ma (Walker and Nisbet, 2002). Studies by Scholey (1992) and Kusky and Winsky (1995) have shown that the sequence has been dissected by many shear zones and that the lower 200 m of the succession is locally absent. The Zeederbergs Formation (∼ 2.8 km thick; Brake, 1996) consists of pillowed and massive basalt (Fig. 9). Basaltic andesite and andesite are present at two stratigraphic levels in the center and at the top of the sequence. Massive and pillowed lava flows form

7. Ngezi Group: 2.7 Ga Upper Greenstone Sequence

503

Fig. 9. Simplified graphic log of the Zeederbergs Formation (modified after Hofmann et al., 2003b).

504

Chapter 15: The Belingwe Greenstone Belt

randomly intercalated units several meters up to 200 m in thickness. Pillows are typically 0.5–1 m in diameter and exhibit chilled margins and calcite- and quartz-filled vesicles. Pillow breccia commonly fills the space between pillows or forms rare discrete horizons. Locally, massive basalt grades into pillow basalt, indicating that the massive variety represents extrusive lava flows. Intrusive rocks are rare in the Zeederbergs Formation and include narrow (< 2 m), cross-cutting basaltic dykes, dolerite and a few ultramafic sills (Martin, 1978; Brake, 1996). Hyaloclastite forms a minor proportion of the sequence and forms crudely parallel-stratified, continuous horizons, several decimeters up to 10 m thick, intercalated with basalt flows. Rare epiclastic sedimentary rocks include continuous, several decimeter thick beds of shale and normally graded beds, 0.1–1 m thick, of basalt pebble conglomerate, sandstone, and siltstone, showing partial and complete Bouma sequences indicative of turbidity current deposits (Hofmann et al., 2003a, 2003b). No detritus other than basaltic material was observed in the Zeederbergs Formation. Ion microprobe trace element data of glass inclusions in Reliance komatiites are similar to modern mantle plume-related magmas unaffected by crustal contamination (McDonough and Ireland, 1993). A study of glass inclusions in Cr-spinels of Reliance komatiites suggested a 0.8–0.9 wt% H2 O content of the primary komatiitic melt (Shimizu et al., 2001a). Reliance komatiites further show a suprachondritic 187 Os/188 Os ratio interpreted to reflect a mantle source affected by mafic crustal recycling or core-mantle interaction (Walker and Nisbet, 2002). A Sm-Nd and Pb-Pb isotope study by Chauvel et al. (1993) indicated a correlation between model µl and εNd(t) values. This was interpreted to reflect, among others, alteration, fractionation, or assimilation of up to 1% of felsic crust similar in composition to the 3.5 Ga Shabani gneiss. Basalts from the Reliance and Zeederbergs Formations show trace element characteristics similar to Mtshingwe Group basalts (Bolhar, 2001; Bolhar et al., 2003a). Two geochemically distinct types occur, unfractionated to depleted basalts ((La/Sm)PM = 0.7–0.8, (Nb/La)PM = 0.8) and LREE-enriched basalts with negative Nb-Ta-Ti anomalies ((La/Sm)PM = 2.1–2.3, (Nb/La)PM = 0.3–0.5). εNd(t) values range from −0.8 to +2.7. The geochemical signatures of the second group of rocks have been attributed to processes of assimilation of continental crust, followed by fractional crystallization (Bolhar et al., 2003a). Studies by (Shimizu et al., 2001b, 2002) arrived at similar conclusions. 7.3. Cheshire Formation The sedimentary Cheshire Formation (∼ 1.3 km thick) is the uppermost stratigraphic unit of the Belingwe greenstone belt and rests along a sharp and sheared contact on the Zeederbergs Formation. Geochronological data, summarized by Bolhar et al. (2002), suggest deposition at c. 2650 Ma. The Cheshire Formation has been subdivided into two stratigraphic units (Hofmann et al., 2001a), a carbonate member that consists of limestone, shale and local limestone breccia, and a siliciclastic member of conglomerate and shale (Fig. 10). Mafic volcanic rocks are absent in the Cheshire Formation. The carbonate member is restricted to two tectonically duplicated and shear zone-bounded units in the western part of the Cheshire outcrop. It is interpreted as a shallow, wave- and storm-dominated, open

7. Ngezi Group: 2.7 Ga Upper Greenstone Sequence

505

Fig. 10. (a) Geological map of the central part of the Cheshire Formation. The strata are steeply dipping to subvertical. (b) Simplified graphic logs of the western and eastern parts of the Cheshire Formation, measured along the Ngezi River. (c) Detailed map of the western Zeederbergs-Cheshire Formation contact and simplified stratigraphic log (modified after Hofmann et al., 2001b).

506

Chapter 15: The Belingwe Greenstone Belt

marine sequence that formed in an eastward-deepening carbonate ramp setting (Hofmann et al., 2001a, 2004). The limestone sequence is capped by a karst breccia horizon which, in turn, is overlain by deeper marine siliciclastic deposits, indicating subaerial emergence and subsequent drowning of the carbonate platform. The siliciclastic member consists of a succession of deep-water conglomerate and shale with a quartzose sandstone horizon and ironstone at the top. A petrographic study of Cheshire Formation conglomerate has shown that the clastic detritus was predominantly derived from erosion of a terrain similar, or laterally equivalent, to the Zeederbergs Formation (Hofmann et al., 2001a). Felsic detritus derived from a granitoid-gneiss terrain is rare except for the uppermost part of the sedimentary sequence. A geochemical provenance study (Hofmann et al., 2003b) of Cheshire shales indicated that the lower part of the sedimentary sequence was derived from the erosion of LREEenriched, mafic volcanic rocks geochemically similar to enriched basalts of the Zeederbergs Formation. Towards the upper part of the Cheshire Formation, progressively more (ultra)mafic and LREE-depleted volcanic rocks, geochemically similar to Reliance Formation komatiite and depleted basalt of the Zeederbergs Formation, were eroded in the source area. Unroofing of a large tract of granitoid rocks took place after deposition of c. 1000 m of Cheshire sediments, as indicated by sedimentological and geochemical data.

8. DEFORMATION OF THE NGEZI GROUP The earliest deformation event that affected the Ngezi Group is a thin-skinned thrusting event (D1 ) that resulted in the formation of layer-parallel to low-angle shear zones (Kusky and Winsky, 1995). Syntectonic sulfide mineralization and silicification of rocks adjacent to the thrust faults gave rise to the formation of rocks similar to banded iron formation, termed chert tectonite by Kusky and Winsky (1995) and tectonic ironstones by Hofmann et al. (2003a). Shearing along ironstone horizons is indicated by anastomosing foliation domains, folding, boudinage and mylonitic fabrics, and, on a regional scale, by truncation of bedding and/or foliation, an anastomosing geometry of the horizons and duplication/juxtaposition of lithostratigraphic units (Hofmann et al., 2001a). A prominent silicified and sulfide-impregnated shear zone is the Jimmy Member at the contact between the Manjeri and Reliance Formation. In outcrop, the Jimmy Member is dominated by highlyweathered, highly-deformed gossaneous banded ironstone and brecciated banded ironstone (Kusky and Winsky, 1995; Hunter, 1997; Hofmann et al., 2003a). Asymmetrical folds and axial planar quartz veins, isoclinal and rootless folds, and boudinage of chert layers are common. Layer-parallel shearing is indicated by anastomosing cleavage domains bounding asymmetrical, boudinaged chert fragments. In the subsurface, the ironstone is dominated by massive sulfide (pyrite, minor pyrrhotite) and chert, intercalated with chloritic shale, graphitic and calcareous mudstone and vein quartz. Zones of high strain as indicated by anastomosing cleavage domains, mylonitic fabrics and intensely folded carbonaceous horizons showing an axial planar cleavage are common (Hunter, 1997).

8. Deformation of the Ngezi Group

507

Detailed structural studies of the Manjeri/Reliance contact have been completed in a number of places along the eastern and western sides of the greenstone belt (Kusky and Winsky, 1995). Since this contact represents one of the most critical relationships upon which general models for the tectonic setting of greenstone belts and the petrotectonic environment of komatiites are based (e.g., Nisbet et al., 1987; Kusky and Kidd, 1992; Blenkinsop et al., 1993; Kusky and Polat, 1999), a review of contact relationships is provided here. Deformation of the Cheshire Formation will be treated separately. 8.1. Contact Relationships at Manjeri Type Section At the Manjeri Formation type section (Unconformity National Monument, Figs. 2 and 11) the Spring Valley Member consists of several-deepening upward sequences (Blenkinsop et al., 1993) of sandstone-mudstone-jaspilite, with a total thickness of 25 m (Fig. 12). These are succeeded by the Rubweruchena Member, comprising about 50 m of sandstone-argillite couplets, capped by 10 m of siltstone and shale (Hunter et al., 1998). Shale is overlain by about 5 m of intensely deformed and disaggregated chert with Fe-rich gossaneous argillite of the Jimmy Member. The lower parts of the Manjeri Formation at this location show little strain (Fig. 12). Deformation is marked only by steep dips and a weak northwest-striking cleavage (S3 of Hofmann et al., 2001a), oriented about 25◦ –30◦ counterclockwise from bedding (Blenkinsop et al., 1993; Kusky and Winsky, 1995). The siltstone and chert/argillite of the Jimmy Member are markedly more strained (Fig. 12). The siltstone is homogenized by cataclastic flow and significantly weathered, because of the former presence of sulfides, but the main fabric is discernible as a composite/planar S-C type of shear zone fabric. S-surfaces are defined by lozenge-shaped chert fragments separated by thin wispy layers of argillite, whereas C-surfaces are defined by through-going shear planes that bend the chert lozenges into sigmoidal forms (Fig. 13). The size of the chert lozenges decreases from a few centimeters to under a millimeter across the 5-m width of the Jimmy Member to the faulted contact with the base of the Reliance Formation, until they are no longer recognizable on the macroscale at the contact. This grain size reduction may be a reflection of a strain gradient, with smaller lozenges reflecting greater disruption and higher strain, in a way analogous to grain size reduction in granitic mylonites (e.g., Vernon et al., 1983; Hanmer and Passchier, 1991; Passchier and Trouw, 1996). The argillite layers do not define regular beds in the outcrop but form wispy layers bounding more coherent packages of chert at several different scales (Fig. 13). At the smallest macroscale, argillite wisps, typically 1–3 mm thick, bound individual chert lozenges and merge at the tails of the lozenges. At a slightly larger scale, other argillite wisps enclose packages of chert lozenges within homogeneously deformed nappe-like structures that are wrapped around fold closures and ramped over fault bends. On a slightly larger scale, most of the C surfaces in the chert unit are marked by discontinuous Fe-rich argillite stringers enclosing small porphyroclasts of chert exhibiting asymmetric fabric trajectories around their margins. At the largest scale observable on the outcrop, Fe-rich argillite layers form through-going shear surfaces separating the chert into approximately 1-m-thick

508

Chapter 15: The Belingwe Greenstone Belt

Fig. 11. Map of the eastern side of the Belingwe greenstone belt, showing locations of the Unconformity National Monument, Rupemba Peak, and Rubweruchena. Modified after Kusky and Winsky (1995).

nappe-like packages of internally deformed chert/argillite (Fig. 13). As the scale of deformation increases, the Fe content of the argillite increases, with the biggest, through-going shear surfaces being the most Fe-rich, which is indicative for the introduction of (now weathered) sulfides along shear zones. Kusky and Winsky (1995) used the asymmetric folds, fault ramps, and fault bend folds in the chert tectonite as kinematic indicators, showing an overall sense of dextral shear

8. Deformation of the Ngezi Group 509

Fig. 12. Map of the stream section (including the Jimmy Member) at Unconformity National Monument. Modified after Kusky and Winsky (1995).

510

Chapter 15: The Belingwe Greenstone Belt

Fig. 13. Contact between Jimmy Member chert tectonite/tectonic ironstone and Reliance Formation ultramafic schist and mylonite (after Kusky and Winsky, 1995). Note the abundance of nappe-style folds, shear (C) surfaces and tectonic dismemberment of chert layers.

for the chert tectonite. Much of the deformation of the top of the Manjeri Formation was related to relative northward movement of the Reliance and Zeederbergs Formations (the Mberengwa allochthon of Kusky and Kidd, 1992) over the Manjeri Formation. The contact between the Jimmy Member and the Reliance Formation is marked by a 3-cm-wide zone of very fine-grained, highly weathered phyllonite (Fig. 13). It may be the ultimate product of cataclastic deformation (fault gouge) or it could represent a more ductile product and be a hydrated and weathered ultramylonite. The basal Reliance Formation also exhibits strong deformation. The lower meter is an extremely fine-grained ultramafic schist that contains a composite S-C planar fabric that is, in places, mylonitic (Fig. 14). The sense of shear is the same as in the underlying chert-tectonite, that is west side to the northwest with little plunge. The succeeding 15 m of the Reliance Formation are very poorly exposed rocks ranging from ultramafic schist, to mylonite, to phyllonite. These ultramafic mylonites have large (several centimeters) porphyroclasts and in contrast to the lowermost meter of the Reliance Formation preserve a steep mineral lineation and shallow intersections between the S and C planes and a maximum fabric asymmetry (Fig. 14) on steep planes. Thus movement in this portion of the shear zone was relatively downdip, indicating that the southwest side moved down (and toward the west) relative to the northeast side. Consequently, Kusky and Winsky (1995) suggested that the shear zone had a

8. Deformation of the Ngezi Group

511

Fig. 14. Field photographs of deformed chert nappes in Jimmy Member tectonic ironstone, slab of chert-tectonite in tectonic ironstone, and ultramafic mylonite showing C-S composite planar fabric.

protracted or polyphase history, in which early, relatively high-temperature fabrics in the ultramafic mylonites record downdip movements toward the west or southwest, (present coordinates) and younger, lower-grade fabrics in the ultramafic schists and chert tectonite record present-day dextral slip movement, with the upper plate (Mberengwa allochthon) moving toward the north. Above the lower 17 m of the Reliance Formation, the lavas are essentially unstrained and consist of a series of ultramafic komatiitic pillows, flows, and tuffs. 8.2. Contact Relationships on Mbalabala Road West of Zvishavane The lower parts of the Reliance Formation are well-exposed (but highly weathered) on the main road leading west towards Mbalabala out of Zvishavane (Fig. 11). Here the ultramafic rocks are phyllonitic to mylonitic in a 200 m wide shear zone that also includes blocks of less deformed komatiite and one large block of well-bedded quartzite and phyllite of the Manjeri Formation (Kusky and Winsky, 1995). The phyllonites are strongly cleaved fine-grained rocks that typically display a strong crenulation cleavage, whereas the mylonites contain porphyroclasts and porphyroblasts and preserve an S-C asymmetry indicat-

512

Chapter 15: The Belingwe Greenstone Belt

Fig. 15. Photomicrographs of serpentinized ultramafic ultramylonites with σ -style porphyroclasts.

ing down dip movement (upper plate toward the west). Asymmetric σ tails (e.g., Passchier and Trouw, 1996), σ -shaped foliation trajectories, and obliquity of mica fish observed in the field also indicate transport of the upper plate (Mberengwa allochthon) downdip toward the west (present coordinates). Because of the lack of strong mineral lineations, interpretation of other shear sense indicators from this location is ambiguous. However, the shallow intersection between C and S surfaces, and the steep mineral lineation preserved elsewhere along the belt margin, suggest relative dip slip motion along the fault at this locality (Kusky and Winsky, 1995). Fig. 15 illustrates some of the microtextural relationships preserved in the ultramafic mylonites. The mylonitic fabric is extremely strong and is defined by the parallel alignment of talc and serpentine, defining an ultramafic ultramylonitic fabric. Numerous strongly asymmetric olivine and pyroxene porphyroblasts are preserved within the ultramylonitic foliation, with maximum asymmetry on steeply dipping surfaces perpendicular to foliation planes. In Fig. 15 porphyroblasts are transected by a curved internal foliation that is continuous with the external foliation, defining small microfolds at the porphyroblast/matrix intersections. In these examples, the external foliation is parallel or nearly parallel to the shear zone boundaries, leading us to suggest that the porphyroblasts have rotated relative to the external foliation, forming the curved inclusion trails. This corresponds to sinistral shear in Fig. 15, and relative motion whereby the upper plate (west side) has moved down and toward the west relative to the lower plate containing the basement gneiss and

8. Deformation of the Ngezi Group

513

overlying shallow water sedimentary sequence (Kusky and Winsky, 1995). The tails on the porphyroblasts in the Mbalabala ultramafic mylonite are unusually long relative to the sizes of the porphyroblasts, indicating exceptionally high strains. Also consistent with exceptionally high strains is the differentiated layering visible in Fig. 15. 8.3. Contact Relationships Near Rupemba Peak The Rupemba Peak area preserves one of the thickest sections (approximately 600 m, possibly structurally repeated) of Manjeri Formation, especially of the carbonate facies (Figs. 2, 11, and 16; Kusky and Winsky, 1995). Jimmy Member gossaneous ironstones around Rupemba Peak are, like elsewhere in the Belingwe belt, high-strain zones that preserve dismembered isoclinal and asymmetric folds (Fig. 17) C-S mylonitic banding, and asymmetric tails indicative of noncoaxial strains (Kusky and Winsky, 1995). These high shear strains contrast markedly with the relatively little strained rocks on either side of the shear zone. On Rupemba Peak, duplication of the carbonate facies along an ironstonedecorated thrust defines a hanging wall anticline above a footwall cutoff (Kusky and Winsky, 1995). Like elsewhere in the Belingwe belt, the ironstone is interpreted to mark the site of a structural detachment (Fig. 16), an interpretation consistent with the duplication of the carbonate facies at this location. 1.5 km northeast of Rupemba Peak, along the railroad line, the argillite facies are cut out by faulting. Limestones are intensely deformed adjacent to the contact and locally show mylonitic fabrics. Sheared carbonates grade upward into 5 to 10 m of mylonitic talc schists similar to that at the Unconformity National Monument and then pass into rocks of the Reliance Formation which preserve good mafic and ultramafic primary textures. 8.4. Summary of Manjeri-Reliance Contact Relationships On the scale of the entire greenstone belt, the Manjeri-Reliance contact appears quasiconformable, but even on this scale it is apparent that the Manjeri Formation shows marked thickness variations from approximately 600 m near Rupemba Peak, to 0 m just north and south of Rupemba Peak, to 250 m near Zvishavane, then back to 0 m north of Zvishavane, and are generally much thinner on the west side of the belt (Figs. 7 and 11; Kusky and Winsky, 1995; Hofmann et al., 2003a). Only part of the above described variation can be attributed to late plutonism; the rest must be caused by either variations in initial sedimentary thicknesses, or by tectonism structurally omitting parts of the Manjeri Formation. The Manjeri Formation consists of three main facies: a shallow water siliciclastic shelf sequence, locally capped by limestone (Spring Valley Member); a fan-delta to turbiditic sandstone-shale sequence indicating drowning of the shelf (Rubweruchena Member); and a thin chert/argillite ironstone within the high-strain zone (Jimmy Member). The thickness variations are not caused by deposition on irregular topography as claimed by Blenkinsop et al. (1993) and Bickle et al. (1993, 1994), because the shallow water sandstones are more continuous relative to overlying carbonates and turbidites. The shear zone marking the Manjeri-Reliance contact is in most cases parallel to stratification in the underlying

514

Chapter 15: The Belingwe Greenstone Belt

Fig. 16. Simplified map of Rupemba peak, showing tectonically repeated tectonic ironstone of the Jimmy Member. Modified after Kusky and Winsky (1995).

rocks, but in several locations structural ramps cut out part of the Manjeri Formation, placing deformed ultramafic rocks of the Reliance Formation into tectonic contact with the shallow water limestones and/or sandstones (Kusky and Winsky, 1995). For instance, on Rupemba Peak (Fig. 16), parts of the carbonate shelf sequence are repeated along contractional faults, and just northeast of Rupemba, ultramafic mylonites are juxtaposed with intensely deformed carbonates. Although more difficult to discern, the basal Reliance Formation appears to be cut out by faulting along hanging wall ramps or late faults oriented at a low angle to the ManjeriReliance contact (Scholey, 1992; Kusky and Winsky, 1995). One of the most intriguing relationships is the typical association of having thick, ductily sheared serpentinite present at the base of the Reliance Formation where the entire Manjeri Formation is present but having thick serpentinite absent where parts of the Manjeri Formation are cut out by faults

8. Deformation of the Ngezi Group

515

Fig. 17. Views of tectonic ironstone from Rupemba Peak. (a) View from top of peak showing tectonic ironstone at top of Manjeri Formations- ridges in background are quartzites defining the eastern edge of the greenstone belt. (b) View of fine laminations in tectonic ironstone. Note pinching of layers, intercalation of different lenses, and isoclinal folds with strongly attenuated limbs. (c) Tectonic ironstone showing isoclinal folds with strongly attenuated limbs. (d) Late asymmetric folds in tectonic ironstone.

(as mapped by Martin, 1978). Although this could be caused by the fortuitous juxtaposition of footwall and hanging wall ramps, it more likely indicates that two generations of subparallel faults are present along the contact (Kusky and Winsky, 1995). An early generation of shearing parallels the contact and forms a thick shear zone with ductile fabrics, including ultramafic mylonites, schists, and phyllonites in the hanging wall and tectonic disruption/ transposition in the footwall chert tectonite. A second generation (but perhaps related to the same tectonic event) of brittle faults cuts the early shear zone at a low angle juxtaposing different units in the footwall and hanging wall and superimposing brittle fabrics (thin fault zones, fault gouge) upon the earlier mylonitic, schistose, and phyllonitic fabrics. Differences in the movement directions of early and late faults results in a presently steep down dip solution for kinematic indicators in the ductile high-strain zones and shallow, presently strike-slip solutions for kinematic indicators in the late brittle faults. Variable amounts of motion on these late faults juxtapose rocks from within or above the early

516

Chapter 15: The Belingwe Greenstone Belt

high-strain zone in the Reliance Formation with any of the three members of the Manjeri Formation, depending on the amount of motion and the angle between the early and late faults. 8.5. Deformation of the Cheshire Formation In the Cheshire Formation, D1 affected poorly consolidated sediments and is recorded in bedding-parallel ductile shear zones, boudins, folds, block-in-matrix structures, and tectonic ironstones (Hofmann et al., 2001a). D1 shear zones separate the Ngezi volcanic sequence from the overlying Cheshire sedimentary rocks and occur between the carbonate and siliciclastic members. Stratigraphic units were locally duplicated along the D1 thrust faults, including a tectonic slice of Zeederbergs Formation volcanic rocks (Fig. 10) that was emplaced onto carbonates along a chaotic unit similar to a tectonic mélange. Blockin-matrix structures as defined by inclusions of various lithologies wrapped around by a plastically deformed argillaceous matrix indicate that deformation affected unconsolidated to semiconsolidated sediment (Fig. 18). Kinematic indicators suggest that bedding-parallel shearing and stratigraphic duplication resulted from northwestward directed tectonic transport (Hofmann et al., 2001a). Subsequent deformation of the Belingwe belt included tight upright folding of the greenstone succession during D2 , forming the distinctive southeast-striking synclinal structure (Fig. 2), cross-folding along an east-west axis (D3 ) and dextral strike-slip faulting along the Mtshingwe fault (D4 ). The D2 and D3 events are of higher metamorphic grade, gave rise to a penetrative cleavage (S2 , S3 ), and took place after consolidation and lithification of the Cheshire sediments. D2 may be an event related to D1 insofar as continued shortening in a late stage of horizontal tectonic deformation combined with a possible switch to an east-west oriented stress field gave rise to the synclinal folding of the belt. D3 and D4 are unrelated to the tectonic evolution of the Belingwe belt and indicate orogenic processes that affected a large section of the already consolidated Zimbabwe craton adjacent to the Limpopo mobile belt (Hofmann et al., 2001a; Kusky, 1998).

9. SYN- AND POST-TECTONIC INTRUSIONS Several dolerite sills, up to 80 m in thickness and greater than 10 km in lateral extent, are unfolded with the sedimentary succession of the Cheshire Formation (Hofmann et al., 2001b), suggesting that emplacement took place prior to synclinal folding of the belt. The dolerite sills have depleted trace element patterns and lack negative Nb-Ta-Ti anomalies ((La/Sm)PM = 0.65–0.71, (Nb/La)PM = 0.85–1.03; Bolhar, 2001). Dolerite sills also occur in the Zeederbergs Formation. Differentiated ultramafic sills, ranging from dunite to gabbro and up to 90 m in thickness, are locally present in the Reliance and Zeederbergs Formations (Fig. 8; Nisbet et al., 1977; Martin, 1978; Brake, 1996). Limited data (Scholey, 1992)

10. Ngezi Volcanics—Autochthonous or Allochthonous?

517

Fig. 18. Features of Cheshire Formation mélange zone. (a) Block of shale (center-right) in sheared muddy matrix. (b) Blocks of limestone (showing aragonite pseudomorph fans) in a matrix of sheared and disrupted mudstone and siltstone.

indicate that the mafic rocks intrusive into the Reliance and Zeederbergs Formations are geochemically very similar to those in the Cheshire Formation. The granitoid-greenstone terrain of the Belingwe area was intruded by dunite-peridotite bodies belonging to the Mashaba ultramafic suite. To the east of the Belingwe greenstone belt are the Shabani and Vukwe complexes, in the west are the Gurumba-Tumba, Vanguard and Ingolubi complexes (Fig. 2). Most of the ultramafic bodies intruded into the gneiss terrain or the Lower Greenstones. The Ingolubi ultramafic complex intruded the Manjeri and Reliance Formation prior to folding of the greenstone belt. The Mashaba-Chibi mafic dyke swarm is a suite of east-west trending dykes that cut the gneiss terrains both to the east and west of the belt. The dykes also cut the Shabani ultramafic complex. Together with the ultramafic complexes, they have been interpreted as the feeder dykes and magma chambers for the Ngezi volcanics (Wilson, 1990; Nisbet et al., 1993b). It should be stressed in the context of this paper that the ultramafic complexes are differentiated, igneous intrusions and not tectonically emplaced ultramafic bodies. Voluminous intrusions of Chilimanzi suite monzogranites was the last major igneous event prior to stabilization of the Zimbabwe craton. In the Belingwe belt, Chilimanzi granites that intrude into the greenstone sequence are represented by the Chibi granite in the south and the Shabani granite in the north (Fig. 2). Intrusion took place after synclinal folding of the belt. Metamorphic aureoles, locally reaching amphibolite facies, developed in the country rock. A Chilimanzi suite granite to the northeast of the Belingwe belt has yielded a U-Pb zircon age of 2634 ± 17 (Horstwood et al., 1999). Intrusion of the layered igneous complex of the Great Dyke and its satellites, the East dyke and Umvimeela dyke, took place after stabilization of the craton 2576 Ma ago (Oberthür et al., 2002).

10. NGEZI VOLCANICS—AUTOCHTHONOUS OR ALLOCHTHONOUS? Deformation of the Jimmy Member ironstone has been attributed to accommodation of strain related to layer-parallel slip during synclinal folding of the greenstone sequence

518

Chapter 15: The Belingwe Greenstone Belt

(Blenkinsop et al., 1993; Bickle et al., 1994) or thrusting of the Reliance Formation onto then Manjeri Formation (Kusky and Kidd, 1992; Kusky and Winsky, 1995). However, shear sense indicators do not support deformation during synclinal folding. Fold asymmetries, fault ramps, and fault bend folds in the ironstone and S-C fabrics in associated mylonitic schists from the eastern side of the belt suggest west- to northward-directed tectonic transport (Kusky and Winsky, 1995). This shear sense is contrary to what would be expected for flexural slip related to folding. Early deformation of the Cheshire Formation was shown to be unrelated to synclinal folding of the belt (Hofmann et al., 2001a). Tectonic movement directions along D1 thrust faults in the Cheshire Formation on the western side of the belt are unidirectional towards the northwest, identical to the sense of shear in the Jimmy Member. We thus favor the interpretation that the Jimmy Member ironstone represents a regional shear zone along which the overlying volcanic and sedimentary units were thrust towards the northwest, negating an autochthonous origin. Some researchers considered it unlikely that a thrust contact would have remained at the same stratigraphic level along the sulfidic ironstone of the Jimmy Member (Blenkinsop et al., 1993). However, they did not take the possibility into account that the sulfidic horizon itself formed as a result of shearing, nor did they recognize the footwall and hangingwall ramps exposed along the thrust system (Kusky and Winsky, 1995). Hofmann et al. (2001a, 2003a) have shown that silicified and sulfide-impregnated shear zones (tectonic ironstones) can form from a variety of host rocks. This means that the Jimmy Member is not a stratigraphic marker horizon, but only marks the location of the basal shear zone of the overlying allochthon. The magnitude of displacement along the shear zone cannot be determined without knowledge of the root zone of the thrust. The abrupt lithological change across the shear zone from quartzose, shallow-water deposits to deep-water pillow basalts lacking terrigeneous sedimentary rocks may suggest that displacement was substantial. Kusky and Kidd (1992), using the geometry of the synclinal structure, estimated a minimum displacement of 28 km along the detachment.

11. MANJERI FORMATION—RIFT, FORELAND, OR CRATONIC COVER SEQUENCE? The Manjeri Formation has been regarded by many workers as a continental rift sequence (Blenkinsop et al., 1993; Bickle et al., 1994; Hunter et al., 1998). Crustal stretching has been attributed to the ascent of a mantle plume that, following rift sedimentation, gave rise to the extrusion of continental flood basalts of the Ngezi volcanics (Hunter et al., 1998). The rift basin interpretation has been based on the heterogeneity of the sedimentary rocks, abrupt lateral facies and thickness changes, and local sediment sources. Dirks et al. (1999) argued that the sedimentary sequence is unusually thin for a rift sequence, while Hunter et al. (1998) proposed uplift associated with the mantle plume and subsequent volcanism as the cause for the reduced sediment thickness. Nevertheless, abrupt facies changes and local sediment sources are not in any way a diagnostic feature of rift basins. Hofmann et al.

12. Ngezi Volcanics—Ensialic or Oceanic?

519

(2003a) have shown that the local presence of sulfidic shear zones within the Manjeri Formation suggests that rapid lateral facies and thickness changes are not primary, sedimentary phenomena, but may be related to tectonic dismemberment of the Manjeri Formation during thrusting of the overlying rock units. The reinterpretation of the Jimmy Member as a sulfidic shear zone rather than a banded iron formation is also consistent with “vertical facies variations” (Fig. 7), because the sudden superposition of alluvial to shallow-water deposits by a tabular horizon of banded iron formation in an active extensional basin consisting of numerous small-scale horst-and-graben structures would be difficult to explain sedimentologically. Kusky and Kidd (1992) and Kusky (1998) suggested that the lower, shallow-marine part of the Manjeri Formation is a remnant of a regionally extensive passive margin sequence deposited on the southeastern edge of the proto-Zimbabwe craton in an interval of 2.83 to 2.70 Ga. Support for this interpretation is provided by Pb-Pb ages of ∼ 2.82 Ga from stromatolitic limestones of a Manjeri-type sedimentary unit that unconformably overlies granitoid basement in the Masvingo greenstone belt (Fig. 1; Moorbath et al., 1987; Collerson et al., 2002). The middle, fluvial to deeper-water part of the Manjeri Formation (Rubweruchena Member) was interpreted by Kusky and Kidd (1992) and Kusky (1998) as a foreland sequence that formed in front of the allochthonous thrust sheet of the Ngezi volcanics. Hunter et al. (1998) argued that the Rubweruchena Member is too thin to represent a foreland sequence and that it lacks detritus from the allochthonous volcanic sequence. However, these authors did not consider that the units suggested to be parts of a dismembered passive margin sequence preserve a wedge-shaped thickness distribution, thickening towards the southeast. Regional stratigraphic relationships suggest that the Spring Valley and Rubwerruchena Members of the Manjeri Formation forms a southeast thickening sedimentary wedge that prograded onto the Tokwe terrane (e.g., temporally and lithologically similar units are much thicker (up to 4 km) in the Buhwa and Mweza belts to the south, and preserve more deeper-water facies; Kusky, 1998), in a manner analogous to the Ocoee-Chilhowee and correlative Sauk Sequence shallow-water progradational sequence of the Appalachians, and similar sequences in other mountain belts (Sloss, 1963; Hatcher, 1989). The progradation could have been driven by sedimentary or tectonic driven flexural loading of the southeastern margin of the craton, but most evidence points to the latter cause. The top of the Manjeri-type units represents a regional detachment surface, upon which allochthonous units of the southern greenstones were emplaced. Loading of the passive margin by these thrust sheets would have induced flexural subsidence and produced a foreland basin that migrated onto the Tokwe terrane (e.g., Bradley and Kusky, 1986).

12. NGEZI VOLCANICS—ENSIALIC OR OCEANIC? Whether mafic volcanic rock units in greenstone belts, and the Ngezi volcanics of the Belingwe belt in particular, represent ophiolite-like sequences that originated in oceanic

520

Chapter 15: The Belingwe Greenstone Belt

environments and were obducted onto continental crust or whether they represent continental flood basalt (CFB)-like, autochthonous sequences that extruded onto continental crust is an ongoing debate. Below we summarize the available evidence for and against ensialic and enzymatic origins for the volcanic sequence of the Ngezi Group. 12.1. Depositional Setting Continental flood basalts are characterized by the subaerial eruption of large volumes of lava onto emergent continental crust (see papers in Mahoney and Coffin, 1997). Many continental flood basalts, such as the Cretaceous-Tertiary Deccan traps, have significant quantities of fluvial or subaerial sediments interbedded with the lava flows (e.g., Subba Rao, 1988). A shallow subaqueous to subaerial origin has been inferred for the Ngezi volcanics (Blenkinsop et al., 1993; Bickle et al., 1994) in favor of a continental setting. This interpretation has been based on the occurrence of (1) vesicles in pillow lavas, which was suggested to indicate eruption in moderate to shallow water depths, (2) accretionary lapilli in tuffs, suggesting subaerial phreatoplinian eruption, and (3) locally developed cross-bedding in intercalated sedimentary rocks. However, none of these features give an unambiguous indication for shallow-water conditions. The vesicularity of pillow basalts is not a reliable indicator of water depth, because it is also a function of volatile content of the magma (Fisher and Schmincke, 1984). Accretionary lapilli have been described from graded beds within thin sedimentary horizons that are intercalated with massive and pillow basalts of the Reliance Formation (Nisbet et al., 1993c). Our own observations indicate that the graded beds show sedimentary features reminiscent of turbidites. Lapilli occur at the base or scattered within thin to thick, normally graded to massive beds with basal load casts and flame structures. Other circular, lapilli-like features in komatiites of the Reliance Formation are more likely ocelli (Kusky and Winsky, 1995) related to liquid immiscibility between comingling komatiitic and tholeiitic melts. Current-ripple and undulating lamination occurs in the upper parts of some beds; small-scale cross-cross-bedding has been observed at only one locality. However, the local presence of cross-bedding does not preclude a deepwater environment (Pickering et al., 1986). Many similar lapilli beds occur in the Barberton greenstone belt, where they have been regarded as reworked deposits of a shallow-water (Lowe, 1999) or deep-water environment (Stanistreet et al., 1981). Nevertheless, accretionary lapilli, although now found as reworked material in subaquatic deposits, are generally thought to form in phreatomagmatic eruption columns and ash clouds, suggesting shallow-water to subaerial volcanism somewhere during Reliance times. However, accretionary lapilli, both in the Belingwe and Barberton (Lowe, 1999) greenstone belts, are only associated with ultramafic volcanism. This relationship would suggest local subaerial conditions only during times of komatiite volcanism. Because this relationship has not yet been explained, we think more work needs to be done before accretionary lapilli can be used as palaeoenvironmental indicators. Accretionary lapilli are absent in the Zeederbergs Formation. Instead, turbidites are common in the sedimentary horizons intercalated with the volcanic rocks (see above). In summary, we interpret the Ngezi volcanics as a lava-plain sequence (cf. Dimroth et al., 1982) that extruded in a subaqueous setting well below wave

12. Ngezi Volcanics—Ensialic or Oceanic?

521

base, as indicated by pillow basalts and intercalated turbidite deposits. No evidence for subaerial extrusion has been observed. Time-equivalent Continental Flood Basalts (CFB’s) (Ventersdorp Group, South Africa; Burke et al., 1985a, 1985b; Burke et al., 1986; Nelson et al., 1992), and CFB’s in general, are dominated by subaerial lava flows. 12.2. Continent-Derived Influx A continental evolutionary setting has been suggested by many authors on the basis of the observation of local quartz sand in a drill core from the Reliance Formation (e.g., Blenkinsop et al., 1993). However, the core was never properly logged or described, and has since been lost (A. Martin, personal communication, 1999), and no evidence for quartzose sediment has ever been reported from outcrop. Nisbet et al. (1993c) stated that the drill core “shows thin beds of quartz grit intermingled with tuffs of the basal Reliance Formation, implying that some clastic sediment was deposited after the onset of volcanic activity”. Tectonic intermingling of Manjeri sandstones with Reliance tuffs along the Manjeri-Reliance shear zone may explain the observation. Without a published description, photograph, or reproducible observation of this purported sandstone, we cannot even rule out the possibility that the quartz may have been a disaggregated vein. A geochemical study on the other hand indicated that shale of the Zeederbergs Formation does not contain extrabasinal detritus, suggesting an environment removed from the influence of continental crust (Hofmann et al., 2003b). 12.3. Geochemistry Two groups of geochemically distinct volcanic rocks occur in the Ngezi volcanics, a group of unfractionated rocks and a group of LREE-enriched rocks with negative anomalies for Nb, Ta, Ti and P, and low εNd values. The geochemical signatures of the second group of rocks have been attributed to processes of assimilation of continental crust, followed by fractional crystallization (Bolhar et al., 2003a). Geochemical evidence for crustal contamination has been used by many authors to imply an ensialic origin for the Belingwe volcanics and an autochthonous origin for the Upper Greenstones (Chauvel et al., 1993). If we accept that the geochemical signatures are indeed a result of contamination by continental crust rather than related to other processes, such as the presence of an enriched mantle source or source contamination, we think that an ensialic origin as CFB’s cannot be inferred from the geochemical evidence for several reasons (Hofmann and Dirks, 2003). (1) Archaean CFB’s, exemplified by the 2.7 Ga Ventersdorp and 2.9 Ga Dominion Groups in South Africa and the 2.7 Ga Fortescue Group in Australia (Burke et al., 1985a, 1985b; Nelson et al., 1992), are geochemically very similar to the enriched Belingwe volcanic rocks. However, they completely lack unfractionated basalts (Fig. 19), such as typify the Zeederbergs Formation of the Belingwe belt, and are thus geochemically distinct. The enriched geochemical signatures of the Archaean CFB’s are attributed to melting of an enriched mantle source (sub-continental lithospheric mantle) rather

522

Chapter 15: The Belingwe Greenstone Belt

(a)

(b) Fig. 19. Primitive-mantle-normalized trace element diagrams for (a) Zeederbergs Formation basalts and basaltic andesites (Bolhar, 2001) and (b) Ventersdorp basalts (Nelson et al., 1992).

than crustal contamination processes (references in Nelson et al., 1992), although a geochemical distinction may hardly be possible. (2) Many volcanic greenstone sequences show evidence that they formed in a nearcontinental setting, as exemplified, for example, by gneiss bodies engulfed in mafic lava in the Mafic Formation of the Midlands greenstone belt, Zimbabwe (Horstwood et al., 1999; Dirks et al., 2002). However, as in many modern tectonic settings, no clear-cut distinction can be made between a continental or oceanic origin, because modern day oceans develop by thinning and rifting of continental crust (see discussion in Sylvester et al., 1997). As such, contamination of basalts produced in infant oceans and back-arc basins by continental crust can be expected. Ophiolitic suites produced

12. Ngezi Volcanics—Ensialic or Oceanic?

523

near thinned continental crust are also described from the Slave Province of Canada (Corcoran et al., 2004) and in the Proterozoic Jourma ophiolite of Scandinavia (Peltonen and Kontinen, 2004). (3) Evidence for crustal contamination has been stated to be inconsistent with an oceanic plateau and mid-ocean ridge setting. However, some oceanic plateau-basalts (Kerguelen Plateau; Ingle et al., 2002) and some mid-ocean ridge basalts (South Atlantic Ridge; Kamenetsky et al., 2001) show evidence for crustal contamination. In addition, many ophiolites show geochemical signatures different from mid-ocean ridge basalts (Moores, 2002; Kusky et al., 2004). Geochemical signatures of apparent crustal contamination are common in younger suprasubduction zone ophiolites, and it is not surprising that allochthonous Archean mafic/ultramafic sequences show similar patterns. They could result from contamination by stretched continental crust during initial basin opening, or they could result from petrogenetic processes in the suprasubduction zone environment (e.g., Moores, 2002). The problem is that most ophiolites are produced in suprasubduction zone environments, and most continental crust is also produced in subduction zone environments, so geochemical patterns between the two are very similar (e.g., Parman and Grove, 2004). 12.4. Zircon Xenocrysts and Lower Crustal Garnets Frequently cited evidence for a continental setting of the Ngezi volcanics is the presence of old inherited zircons in the Belingwe belt and other greenstone belts in Zimbabwe (Hunter et al., 1998; Bolhar et al., 2003a). However, inherited zircons have never been described from the Ngezi volcanics. Xenocrysts were observed in the 200 Ma older Mtshingwe Group (Wilson et al., 1995), which is unrelated to the Ngezi Group (see above). A similar line of reasoning in favor for a continental setting has been used by Blenkinsop et al. (1993) who stated that siliciclastic sedimentary rocks are intercalated with Zeederbergs-type mafic volcanic rocks in the Shamva greenstone belt 400 km to the north of the Belingwe belt. It is obvious that observations from different greenstone sequences with different ages or tectonic settings (the Shamva greenstone belt is interpreted as part of a continental arc sequence; Kusky, 1998) cannot be used to infer a continental origin for the Ngezi volcanics. Shimizu et al. (2002) observed garnet xenocrysts in a komatiite of the Ngezi volcanics. They stated that major element compositions and trace element ratios are consistent with derivation from lower crust. However, garnet-bearing rocks are found in both subcontinental and suboceanic lithospheric mantle and subducted slabs. The uncertainty of the significance of the interpretation of this observation is expounded by the possibility that the Ngezi Group lavas were derived from melting of either stretched continental crust or in a suprasubduction zone setting. We thus want to stress the point that even if old xenocrystic zircons or other crust-derived material would be present in the Ngezi volcanics, this would not imply an autochthonous setting. Neither would it rule out an infant or back-arc oceanic setting for the Ngezi volcanics in the light of the presence of Archaean zircons in rocks

524

Chapter 15: The Belingwe Greenstone Belt

from the Proterozoic Jormua ophiolite (Peltonen et al., 2003), and old xenocrystic zircons in the Paleozoic Josephine ophiolite of California (Harper et al., 1994). 12.5. Crustal Magma Chambers and Feeder Dykes Possibly the strongest evidence for an extensional continental setting of the Ngezi volcanics is the presence of ultramafic intrusions and mafic dykes within the granitoid gneiss terrain, such as the Shabani ultramafic complex and the east-west trending Mashaba-Chibi dyke (Fig. 2). These and similar intrusions have generally been regarded as broadly contemporaneous with Ngezi volcanism and interpreted as the crustal magma chambers and feeder dykes for the Ngezi volcanics (e.g., Prendergast, 2003). However, the timing of emplacement of these units is unknown. Although mostly found in the gneiss terrain and Mtshingwe Group, the Ingolubi ultramafic complex (Fig. 2) is intrusive into the Manjeri and Reliance Formation, but is folded together with the host rocks, suggesting that it formed after deformation along the Manjeri-Reliance contact and after Reliance volcanism, but prior to D2 synclinal folding of the belt. Higher up in the stratigraphy of the Ngezi Group, sills of ultramafic rocks can be found although dolerite sills become more common. It is important to note that folded dolerite sills occur in the Cheshire Formation (Fig. 10; Martin, 1978; Hofmann et al., 2001b). If the dolerite sills, ultramafic sills and cross-cutting ultramafic intrusions are all part of the same igneous event, which is supported by limited geochemical data, then this magmatism was unrelated to volcanism that resulted in the Ngezi volcanics. If this is the case, an event of mafic magmatism affected the Belingwe belt area shortly after or late syn-D1 -thrusting, but prior to D2 -synclinal folding. The Shabani ultramafic complex, which represents a differentiated ultramafic sill of dunite, peridotite, pyroxenite, and gabbro, strikes parallel to, but has a much shallower dip than, the Ngezi Group rocks (Laubscher, 1963). This relationship indicates that intrusion took place when the Ngezi Group rocks were already folded to some degree (∼ 20◦ according to Laubscher, 1963). Intrusion of the Shabani ultramafic complex is thus early syn-D2 . The Mashaba-Chibi dykes cross-cut the Shabani ultramafic complex (Laubscher, 1963) and are thus even younger and cannot represent feeders for the Ngezi volcanics. A large ultramafic intrusion in the Midlands greenstone belt that is very similar to those in the Belingwe belt forms protrusions that intruded along thrust faults in the late phase of a D1 shortening and tectonic stacking event (Dirks et al., 2002). Similarly, deformed ultramafic intrusions in the Masvingo belt preferentially form sheets that intruded along the (thrust) contact between Manjeri-type sedimentary rocks (and underlying granitoid gneisses) and overlying mafic volcanic rocks similar to the Reliance and Zeederbergs Formation of the Belingwe belt. It thus appears that ultramafic to mafic intrusive magmatism affected the Zimbabwe craton in a late stage of, or following, thrust deformation. How this event fits into the larger geodynamic scenario of the Zimbabwe craton is as yet unclear. It is clear, however, that the intrusive magmatism postdates emplacement of the Ngezi volcanics and cannot be used as evidence for an ensialic origin. This is a very similar relationship to the Kalgoorlie area of the Yilgarn craton where an event of mafic intrusive magmatism postdates greenstone belt volcanism by 20 to 30 Ma (Bateman et al., 2001).

13. Cheshire Formation—a Flexural Foreland Basin?

525

12.6. Other Evidence Several other lines of evidence have been put forward for an ensialic and autochthonous origin of the Ngezi volcanics. For instance, Re-Os systematics of chromites from ultramafic complexes intrusive into the Tokwe segment provide geochronological and geochemical evidence for successive growth of lithosphere beneath this crustal block since 3.8 Ga (Nägler et al., 1997). Bolhar et al. (2003b) suggested that since some of these rocks are from near Belingwe, the origin should be the same. They fail to note that the rocks analyzed by Nägler et al. (1997) are largely 3.5 and 2.9 billion years old, not recognizing the complexity of the tectonics of the region, and not recognizing that others have suggested that the Ngezi Group volcanic rocks in the Belingwe belt were likely thrust over the older craton (Kusky, 1998). The Re-Os data are, therefore, irrelevant to rock units that are allochthonous with respect to the old continental nucleus. Stromatolites in the Belingwe belt have distinctive Pb isotopes derived from a longlived source with substantially higher 238 U/204 Pb than mantle (Bolhar et al., 2002). The Pb isotopic characteristics of the stromatolites may reflect seawater composition (B. Kamber, personal communication, 2003) or reequilibration with diagenetic fluid or groundwater (Bolhar et al., 2002). Nevertheless, it has been argued (Bolhar et al., 2003b) that the Pb isotopes require derivation from the old gneisses of the Zimbabwe craton, favoring an ensialic origin of the Ngezi Group. It is evident that the stromatolites of the Manjeri Formation should show old crustal signatures, because the Manjeri Formation unconformably overlies 3.5 Ga granitoid basement. Stromatolites of the Cheshire Formation are wedged in between Zeederbergs mafic volcanic rocks. The Pb isotopes probably reflect seawater composition, because diagenetic fluids would show a strong signature from the adjacent volcanic sequence. Thus, the isotopic signatures indicate exposed continental basement during deposition of the Cheshire Formation. However, the Cheshire Formation is genetically unrelated to the underlying volcanic sequence. Instead, it represents a foreland basin sequence that formed during the thrusting event which resulted in basement unroofing as discussed by Hofmann et al. (2003b). Foreland basins characteristically show sources including the exposed allochthon, uplifted and exhumed basement rocks in high thrust sheets, and the underlying basement terrane exposed in normal fault scarps on the outer trench slope. This model not only explains but predicts basement signatures in the stromatolites. The ensialic model, which assumes that the whole of the continental basement was draped by an autochthonous volcanic sequence, fails to do so.

13. CHESHIRE FORMATION—A FLEXURAL FORELAND BASIN? The Cheshire Formation was interpreted to represent a foreland-type sedimentary basin that formed in front of a fold-and-thrust belt situated in the southeast and contemporaneously with NW-directed thrusting of the underlying volcanic sequence (Kusky and Winsky, 1995; Hofmann et al., 2001a, 2001b). The term foreland basin was used in the sense of a

526

Chapter 15: The Belingwe Greenstone Belt

flexural basin that developed in front of a thrust belt or thrust sheet, with subsidence related to thrust loading. Some of the evidence for a foreland basin include (Hofmann et al., 2001b): (a) The asymmetric facies and thickness distribution with deeper water sediments and thicker strata towards the thrust source. (b) Incorporation of the basin fill into the thrust stack during or shortly after its formation, as indicated by duplication of strata. (c) Clastic material was derived from erosion of Zeederbergs-like mafic volcanics which have been identified to form a thrust sheet in the sequence. (d) The succession of a drowned carbonate platform sequence overlain by deep water turbidite deposits is common to foreland basins. The subaerial exposure of the platform carbonates prior to drowning might be a consequence of flexural arching of the lithosphere some distance away from the tectonic load. Similar relationships are common in younger foreland basin sequences (e.g., Bradley and Kusky, 1986). Facies relationships indicate that basal sedimentary units, i.e., shallow-water carbonates in the west and deep-water siliciclastic rocks in the east, may have formed synchronously and were juxtaposed during thrusting. Part of the sequence was carried piggy-back on top of an active thrust, possibly during or shortly after deposition (Hofmann et al., 2001a, 2001b). The original dimension of the basin is difficult to establish from the current outcrop configuration. However, judging from the vertical and lateral facies distribution, a minimum of several tens of kilometers east-west and greater than 100 km north-south is a first approximation. Difficulties with a foreland basin interpretation arise from the absence of evidence for the thrust load which led to flexure of the lithosphere and provided accommodation space. The uppermost preserved part of the Cheshire Formation, which crops out in the central part of the Belingwe belt, contains D1 shear zones, suggesting the presence of a thrust sheet which was emplaced on top of the Cheshire Formation and which has since been eroded (Hofmann et al., 2001a). In greenstone belts close to the Belingwe belt (Masvingo, Filabusi, Shangani, Gweru; Fig. 1), mafic volcanic rocks similar to the Zeederbergs Formation are overlain by a sedimentary succession that is lithologically very similar to the Cheshire Formation. In all four belts the sedimentary sequence has been mapped to be overlain again by a mafic volcanic sequence (Wilson, 1964; Harrison, 1969; Cheshire et al., 1980; Baglow, 1998), possibly along a thrust fault. Flexural loading may also have been resulted from thrusting and crustal thickening in the Northern Marginal Zone of the Limpopo mobile belt to the southeast of the Belingwe belt, although this event probably took place at a later stage of the evolution of the Zimbabwe craton at ∼ 2.62–2.58 Ga (Mkweli et al., 1995; Jelsma and Dirks, 2002).

14. IMPLICATIONS FOR REGIONAL STRATIGRAPHIC MODELS The stratigraphic division of the Belingwe greenstone belt, and of the Zimbabwe Craton in general, has gone through several iterations. Early workers recognized several different gneissic units of the Rhodesian basement complex and adopted a tripartite stratigraphic division for the greenstone belts, dividing them into a 3.5 Ga Sebakwian Group, a 2.9–2.7 Ga

14. Implications for Regional Stratigraphic Models

527

Bulawayan Group, and an unconformably overlying sedimentary Shamvaian Group (Macgregor, 1947, 1951). Most subsequent workers (Stagman, 1978; Wilson, 1979; Foster et al., 1986) divided the Bulawayan into the Lower Bulawayan Group (or Lower Greenstones) and Upper Bulawayan Group (or Upper Greenstones). The Belingwe greenstone belt was divided into the Mtshingwe Group, which has been correlated with the Lower Bulawayan Group, and the Ngezi Group, which has been correlated with and served as the regional type section of the Upper Bulawayan Group for the entire craton (e.g., Wilson, 1979; Bickle, 1993). Despite some early hints that the Zimbabwe craton may be composed of a number of distinct terranes (Stowe, 1971; Wilson, 1979; Kusky, 1991), much work on the Zimbabwe craton has been geared toward making lithostratigraphic correlations between different greenstone belts, widely interpreted as collapsed ensialic rift basins, and attempting to link them all to a single supergroup style nomenclature (e.g., Bickle, 1993; Blenkinsop et al., 1993, 1997; Wilson et al., 1995; Prendergast, 2003). As summarized above for the Belingwe greenstone belt, a regional shear zone exists between submarine volcanic rocks and underlying granitoid basement and its sedimentary cover sequence, providing evidence that the volcanic rocks are allochthonous. Because the Belingwe greenstone belt has been used as the type section for the Upper Greenstone succession of Zimbabwe (Wilson, 1979; Wilson et al., 1995), and as a model for a greenstone belt overlying continental crust (Bickle, 1993; Bickle et al., 1994), the recognition of significant structural breaks within this belt has important implications for stratigraphic and tectonic interpretation of the Zimbabwe craton as a whole, and for granite-greenstone terranes worldwide. If distinct rock units everywhere rest above a regional shear zone, then they are in structural contact with underlying rocks, and may be allochthonous. Older stratigraphic divisions that lump the Manjeri Formation with structurally overlying allochthonous units are no longer valid—the sedimentary Manjeri Formation rests unconformably on older gneisses and Lower Greenstones, but the overlying volcanic Reliance and Zeederbergs Formations are everywhere separated from the Manjeri Formation and the underlying unconformity by a shear zone. Therefore, these rocks cannot be grouped into the same stratigraphic unit (see Kusky and Kidd, 1992; Kusky and Winsky, 1995; Kusky, 1998; Hofmann et al., 2003a). Kusky and Kidd (1992) suggested to break the stratigraphic units of the Belingwe belt into the basement gneiss complex, the Mtshingwe Group (which appears to contain significant structural and stratigraphic breaks), the Manjeri Formation, the Ngezi Group (consisting of the Reliance and Zeederbergs Formations), and the Cheshire Formation (which may best be assigned to the Shamvaian Group). Further understanding of the tectonic evolution of the Zimbabwe craton hinges upon recognition of the structural complexities of the greenstone belts. While deformed remnants of some lithotectonic units such as shelf sequences, volcanic belts, and gneiss complexes may be correlatable in a broad sense, the recognition that cherty horizons previously used as stratigraphic markers are tectonic high strain zones that jump across stratigraphic units invalidates this approach (Hofmann et al., 2003a). Broad similarities in age, tectonic association, and structural setting may be correlated for tectonic analysis (e.g., Kusky, 1998), but layer-cake stratigraphic modeling should be avoided.

528

Chapter 15: The Belingwe Greenstone Belt

15. DISCUSSION AND CONCLUSION We have shown that the volcanic sequence of the Ngezi Group (Mberengwa allochthon of Kusky and Kidd, 1992) is allochthonous with respect to the continental basement and the sedimentary sequence of the Manjeri Formation that unconformably overlies the basement. It is evident that allochthoneity in itself does not preclude an ensialic origin for the volcanic rocks. However, we have further shown that evidence presented so far for an ensialic origin of the Ngezi volcanics as continental flood basalts is not very strong. Geochemical signatures of crustal contamination, including the presence of crustal xenocrysts, has generally been regarded as strong evidence for a continental setting of a volcanic sequence as continental flood basalts. However, as shown for the Belingwe belt, Archaean continental flood basalts are distinct in their environment of deposition and in their geochemical signatures. Evidence for crustal contamination can only indicate a tectonic setting in the vicinity of continental crust, continental fragments and subcontinental lithosphere and does not preclude settings as diverse as infant oceans, back-arc basins and oceanic plateaus. Furthermore, crustal signatures in a mantle-derived magma, such as LREE enrichment coupled with negative Nb-Ta anomalies, can be related to either contamination by continental crust or subduction-related metasomatic processes. There are currently no unequivocal geochemical criteria to distinguish between these two processes in the Archaean. Although assimilation and fractional crystallization processes may explain the widely observed enriched patterns in Archaean volcanic rocks best or may be more easily modeled, this clearly does not preclude processes in Archaean suprasubduction zone environments. Bolhar et al. (2003a) observed remarkably similar geochemical signatures in the mafic rocks of each volcanic stratigraphic unit from both the Lower Greenstones (Hokonui, Bend, and Brooklands Formations) and Upper Greenstones (Reliance and Zeederbergs Formation), so that essentially identical petrogenetic processes are implied. Hofmann and Dirks (2003) stressed the point that the similar geochemical signatures are in contrast to the different lithological and tectonostratigraphic attributes of individual stratigraphic units. For example, the 2.90 Ga Hokonui Formation is lithologically similar to an island-arc sequence because of the abundance of andesitic to dacitic lavas, pyroclastic and epiclastic rocks, and only minor amounts of mafic volcanic rocks that are intruded by probable synvolcanic granitoid plutons. The 2.69 Ga Zeederbergs Formation is a submarine lava plain sequence of basalts and basaltic andesites, similar to the extrusive sequence of oceanic crust or oceanic plateaux. How can such discrepancies between field relationships and geochemistry be resolved? Proponents of the ensialic model would have to argue that incompatible element depleted mantle plume magmas ascended beneath continental crust at different times during a 200 Ma time period, were contaminated by crustal material of similar composition and in a similar way and were extruded to form CFB sequences that are unlike modern and Archaean CFBs. We have shown above that available sedimentological, structural and stratigraphic data are all inconsistent with this model. Other models that may explain the observed geological and geochemical features of the Belingwe greenstone belt better may include the following: (1) All volcanic rocks of the Belingwe belt formed in a suprasub-

15. Discussion and Conclusion

529

duction zone setting. However, this model is difficult to reconcile with the presence of komatiites and incompatible element depleted to unfractionated tholeiites that are generally regarded to be mantle plume-derived (e.g., Campbell et al., 1989; Storey et al., 1991; McDonough and Ireland, 1993; Arndt et al., 1997), although some authors have proposed plume-arc interaction for the origin of similar greenstone belt volcanic rocks in Canada (e.g., Hollings et al., 1999). (2) Trace element geochemistry is unable to distinguish between different tectonic settings in the Belingwe belt stratigraphy, because of the extrusion of geochemically similar magmas in different environments. This would indicate a strong mantle imprint on magma chemistry, and a geochemically heterogeneous mantle plume, consisting of depleted and enriched components, could be envisaged. (3) The geochemical signatures are only superficially similar and indicate both subduction processes and interaction of mantle plume magmas with fragments of continental crust, continental lithosphere and subduction-modified lithosphere that operated at different times in oceanic but mostly near-continental and near-arc environments. It has long been known that geochemistry alone does not allow a clear cut distinction between different tectonic setting, and this is particularly so for the Archaean, emphasizing the importance of field observation to unravel the tectonic evolution of greenstone belts. Submarine mafic volcanic successions that rest along a thrusted contact on continental crust in younger orogens are generally regarded as remnants of some form of obducted oceanic crust (e.g., Dewey and Bird, 1971; Ben-Avraham et al., 1981; St-Onge and Lucas, 1990; Burchfiel et al., 1992; Saleeby, 1992; Kusky et al., 1997). The question is if this analogy is sufficient evidence for an oceanic origin of the Ngezi volcanics. The lithotectonic assemblage of the Mberengwa allochthon contrasts strongly with that of younger rift, arc, and continental flood basalt sequences, but is very similar to that found in the upper portions of younger ophiolites, oceanic plateaus, seamounts, and crust produced where plumes and ridges are coincident such as in Iceland. In Phanerozoic oceans, the great thickness of the lava and crustal sections distinguishes oceanic plateautype lithosphere from normal oceanic lithosphere (e.g., Burke, 1988), but a similar distinction is more complicated in ancient Archean sequences. First, it must be determined that the reconstructed stratigraphy does not contain hidden shear zones and stratal repetition (e.g., Kusky and Vearncombe, 1997), a task not yet adequately addressed in Belingwe. Second, the possibility of slightly higher mantle temperatures in the Archean makes it possible that Archean oceanic lithosphere had a thicker crustal section than Phanerozoic oceanic lithosphere, and may have resembled younger oceanic plateaus (Sleep and Windley, 1982; Moores, 1986, 2002; Burke, 1988; Kusky and Polat, 1999). Ultramafic rocks within these sections of Archean oceanic plateaus or thick oceanic crust are likely to be komatiitic flows reflecting the slightly higher mantle temperatures, as opposed to mantle ultramafics, which would have been too deep to be scraped off at subduction zones (Hoffman and Ranalli, 1988; Kusky and Polat, 1999). It is clear that the Mberengwa allochthon does not represent a complete Phanerozoic-style ophiolite, which is not to be expected to be found in the Archaean anyway. The Ngezi volcanics consist of a c. 4 km thick extrusive volcanic section that is bounded below by a major thrust fault. A similar section occurs in several greenstone belts in the southern part of the Zimbabwe carton (e.g., Masvingo, Filabusi;

530

Chapter 15: The Belingwe Greenstone Belt

Fig. 1), where it is in sheared contact with underlying granitoid gneisses or older greenstones that are locally unconformably overlain by Manjeri-type sedimentary rocks. These southern greenstones exhibit stratigraphic and structural characteristics reminiscent of the upper part of thick oceanic crust, suggesting that they may represent a large thrust sheet that was obducted onto the Tokwe terrane at ∼ 2.7 Ga. Kusky (1998) suggested that the southern greenstones are distributed in a zone confined to 100 km from a line of passivemargin-type sediments extending from the Matsitama to Mutare greenstone belts (Fig. 1), named the Umtali line. It is possible that this represents the place where an ocean or backarc basin opened between 2.9 and 2.8 Ga, and closed at 2.7 Ga, and forms the root zone from which the southern greenstones were obducted. Future work to unravel the geological evolution of the Belingwe greenstone belt will thus have to focus on the geology of the adjacent greenstone belts, none of which has yet been subjected to any detailed structural, sedimentological or geochemical study.

ACKNOWLEDGEMENTS Work by AH in the Belingwe belt was funded by DAAD, University of Mainz, Stichting Schürmannfonds (1997–2001/13), and German Science Foundation grant Ho 2507/1– 1/2. We thank Tony and Charlotte Rauch of Manjeri Ranches for hospitality in the field. T. Kusky acknowledges support from NSF grant 9304647, and from Boston University and St. Louis University. Pam Winsky is thanked for assistance in the field, and employees of the Nilton in Zvishavane are thanked for unforgettable service.

REFERENCES Abell, P.I., McClory, J., Martin, A., Nisbet, E.G., 1985. Archaean stromatolites from the Ngesi Group, Belingwe Greenstone Belt, Zimbabwe preservation and stable isotopes—preliminary results. Precambrian Research 27, 357–383. Arndt, N.T., 1999. Why was flood volcanism on submerged continental platforms so common in the Precambrian? Precambrian Research 97, 155–164. Arndt, N.T., Albarede, F., Nisbet, E.G., 1997. Mafic and ultramafic magmatism. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 233–254. Baglow, N., 1998. The geology of the Filabusi greenstone belt and surrounding granitic terrane. Zimbabwe Geological Survey Bulletin 91, 263. Bateman, R., Costa, S., Swe, T., Lambert, D., 2001. Archaean mafic magmatism in the Kalgoorlie area of the Yilgarn Craton, Western Australia: a geochemical and Nd isotopic study of the petrogenetic and tectonic evolution of a greenstone belt. Precambrian Research 108, 75–112. Ben-Avraham, Z., Nur, A., Jones, D., Cox, A., 1981. Continental accretion: from oceanic plateaus to allochtonous terranes. Science 213, 47–54. Bickle, M.J., Nisbet, E.G. (Eds.), 1993. The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2. Balkema, Rotterdam, p. 239.

References

531

Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archaean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Bickle, M.J., Martin, A., Nisbet, E.G., 1975. Basaltic and peridotitic komatiites, stromatolites and a basal unconformity in the Belingwe greenstone belt, Rhodesia. Earth and Planetary Science Letters 27, 155–162. Bickle, M.J., Orpen, J.L., Nisbet, E.G., Martin, A., 1993. Structure and metamorphism of the Belingwe Greenstone Belt and adjacent granite-gneiss terrain: The tectonic evolution of an Archaean craton. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 39–68. Blenkinsop, T.G., Fedo, C.M., Bickle, M.J., Eriksson, K.A., Martin, A., Nisbet, E.G., Wilson, J.F., 1993. Ensialic origin for the Ngezi group, Belingwe greenstone belt, Zimbabwe. Geology 21, 1135–1138. Blenkinsop, T.G., Martin, A., Jelsma, H.A., Vinyu, M.L., 1997. The Zimbabwe Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, pp. 567–580. Bolhar, R., 2001. Archaean mafic magmatism and crust formation in the eastern Pilbara craton (Western Australia) and Belingwe greenstone belt (Zimbabwe). Ph.D. thesis. University of Melbourne. Bolhar, R., Hofmann, A., Woodhead, J.M., Hergt, J.M., Dirks, P.H.G.M., 2002. Pb- and Nd-isotope systematics of stromatolitic limestones from the 2.7 Ga Ngezi Group of the Belingwe Greenstone Belt: constraints on provenance, deposition and deformation. Precambrian Research 114, 277– 294. Bolhar, R., Woodhead, J.D., Hergt, J.M., 2003a. Continental setting inferred for emplacement of the 2.9–2.7 Ga Belingwe Greenstone Belt, Zimbabwe. Geology 31, 295–298. Bolhar, R., Woodhead, J.D., Hergt, J.M., 2003b. Reply to Hofmann, A. and Dirks, P., Continental setting inferred for emplacement of the 2.9–2.7 Ga Belingwe Greenstone Belt, Zimbabwe. Geology 31, e31. Bouhallier, H., Choukroune, P., Ballèvre, M., 1993. Diapirism, bulk homogeneous shortening and transcurrent shearing in the Archaean Dharwar craton, the Holenarsipur area, southern India. Precambrian Research 63, 43–58. Bradley, D.C., Kusky, T.M., 1986. Geologic methods of estimating convergence rates during arccontinent collision. Journal of Geology 94, 667–681. Brake, C., Tholeiitic magmatism in the Belingwe greenstone belt, Zimbabwe. Unpublished Ph.D. thesis. University of Edinburgh, p. 184. Burchfiel, B.C., Cowan, D.S., Davis, G.A., 1992. Tectonic overview of the Cordilleran orogen in the western United States. In: Burchfiel, C.B., Lipman, P.W., Zoaback, M.I. (Eds.), The Geology of North America, The Cordilleran Orogen. Conterminous U.S. Geological Society of America C-3, pp. 407–478. Burke, K., 1988. Tectonic evolution of the Caribbean. Annual Review of Earth and Planetary Sciences 16, 201–230. Burke, K., Kidd, W.S.F., Kusky, T.M., 1986. Archean foreland basin tectonics in the Witwatersrand, South Africa. Tectonics 5 (3), 439–456. Burke, K., Kidd, W.S.F., Kusky, T.M., 1985a. The Pongola Structure of southeastern Africa: the world’s oldest recognized well-preserved rift? Journal of Geodynamics 2 (1), 35–50. Burke, K., Kidd, W.S.F., Kusky, T.M., 1985b. Is the Ventersdorp rift system of southern Africa related to a continental collision between the Kaapvaal and Zimbabwe cratons at 2.64 Ga ago? Tectonophysics 11, 1–24.

532

Chapter 15: The Belingwe Greenstone Belt

Campbell, I.H., Hill, R.I., 1988. A two-stage model for the formation of the granite-greenstone terrain of the Kalgoorlie-Norseman area, Western Australia. Earth and Planetary Science Letters 90, 11– 25. Campbell, I.H., Griffiths, R.W., Hill, R.I., 1989. Melting in an Archaean mantle plume: head it’s basalt, tail it’s komatiites. Nature 339, 697–699. Card, K.D., 1990. A review of the Superior Province of the Canadian Shield, a product of Archaean accretion. Precambrian Research 48, 99–156. Chauvel, C., Dupré, B., Arndt, N.T., 1993. Pb and Nd isotopic correlation in Belingwe komatiites and basalts. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 167–174. Cheshire, P.E., Leach, A., Milner, S.A., 1980. The geology of the country between Gwelo and Redcliff. Zimbabwe Geological Survey Bulletin 86, 300. Collerson, K.D., Kamber, B.S., Schoenberg, R., 2002. Applications of accurate, high-precision Pb isotope ratio measurement by multi-collector ICP-MS. Chemical Geology 188, 65–83. Collins, W.J., Van Kranendonk, M.J., Teyssier, C., 1998. Partial convective overturn of Archaean crust in the east Pilbara Craton, Western Australia: driving mechanisms and tectonic implications. Journal of Structural Geology 20, 1405–1424. Corcoran, P.L., Mueller, W.U., Kusky, T.M., 2004. Inferred ophiolites in the Archean Slave Craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 363–404. Dewey, J.F., Bird, J., 1971. Origin and emplacement of the ophiolite suite: Appalachian ophiolites. In: Bird, J.M. (Ed.), Newfoundland, in Plate Tectonics. Journal of Geophysical Research 76, 3179–3206. de Wit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., de Ronde, C.E.J., Green, R.W.E., Tredoux, M., Perbedy, E., Hart, R.A., 1992. Formation of an Archaean continent. Nature 357, 553–562. Dimroth, E.I., Laszlo, I., Rocheleau, M., Goulet, N., 1982. Evolution of the south-central part of the Archaean Abitibi Belt, Quebec; Part I: Stratigraphy and paleogeographic model. Canadian Journal of Earth Sciences 19, 1729–1758. Dirks, P.H.G.M., van der Merwe, J., 1997. Early duplexing in an Archaean greenstone sequence and its control on gold mineralisation. Journal of African Earth Sciences 24, 603–620. Dirks, P.H.G.M., Jelsma, H.A., 1998a. Horizontal accretion and stabilization of the Archean Zimbabwe Craton. Geology 26, 11–14. Dirks, P.H.G.M., Jelsma, H.A., 1998b. Silicic layer-parallel shear zones in a Zimbabwean Greenstone sequence; horizontal accretion preceding doming. Gondwana Research 1, 177–194. Dirks, P.H.G.M., Jelsma, H.A., 2002. Crust-mantle decoupling and the growth of the Archaean Zimbabwe craton. Journal of African Earth Sciences 34, 157–166. Dirks, P.H.G.M., Jelsma, H.A., Hofmann, A., 1999. Comment to Hunter, M.A., Bickle, M.J., Nisbet, E.G., Martin, A., Chapman, H.J. Continental extensional setting for the Archean Belingwe greenstone belt, Zimbabwe. Geology 27, 667. Dirks, P.H.G.M., Jelsma, H.A., Hofmann, A., 2002. Thrust-related accretion of an Archaean greenstone belt in the Midlands of Zimbabwe. Journal of Structural Geology 24, 1707–1727. Dodson, M.H., Compston, W., Williams, I.S., Wilson, J.F., 1988. A search for ancient detrital zircons in Zimbabwean sediments. Journal of the Geological Society of London 145, 977–983. Dougherty-Page, J.S., 1994. The evolution of the Archaean continental crust of northern Zimbabwe. Ph.D. thesis. The Open University, Milton Keynes. Fisher, R.V., Schmincke, H.U., 1984. Pyroclastic Rocks. Springer-Verlag, Berlin.

References

533

Foster, R.P., Mann, A.G., Stowe, C.W., Wilson, J.F., 1986. Archaean Gold Mineralization in Zimbabwe. In: Anhaeusser, C.R., Maske, S. (Eds.), Mineral Deposits of Southern Africa 1. Geological Society of South Africa, pp. 43–112. Grassineau, N.V., Nisbet, E.G., Bickle, M.J., Fowler, C.M.R., Lowry, D., Mattey, D.P., Abell, P., Martin, A., 2000. Antiquity of the biological sulfur cycle: evidence from sulfur and carbon isotopes in 2700 million-year-old rocks of the Belingwe Belt, Zimbabwe. Proc. R. Soc. London B 268, 113–119. Hamilton, W.B., 1998. Archean magmatism and deformation were not products of plate tectonics. Precambrian Research 91, 143–179. Hanmer, S., Passchier, C.W., 1991. Shear sense indicators: a review. Geological Survey of Canada Paper 90, 17–72. Harper, G.D., Saleeby, J.B., Heizler, M., 1994. Formation and emplacement of the Josephine ophiolite and the age of the Nevadan orogeny in the Klamath Mountains, California-Oregon: U/Pb zircon and 40 Ar/39 Ar geochronology. Journal of Petrology 18, 487–520. Harrison, N.M., 1969. The geology of the country around Fort Rixon and Shangani. Rhodesia Geological Survey Bulletin 61, 152. Hatcher, R.D., 1989. Tectonic synthesis of the U.S. Appalachians. In: Hatcher, R.D., Thomas, W.A., Viele, G.W. (Eds.), The Appalachian-Ouchita Orogen in the United States. The Geology of North America F-2, DNAG, GSA, Denver, pp. 511–536. Hawkesworth, C.J., Moorbath, S., O’Nions, R.K., Wilson, J.F., 1975. Age relationships between greenstone belts and “granites” in the Rhodesian Archaean craton. Earth and Planetary Science Letters 25, 251–262. Hawkesworth, C.J., Bickle, M.J., Gledhill, A.R., Wilson, J.F., Orpen, J.L., 1979. A 2.9 b.y. event in the Rhodesian Archaean. Earth and Planetary Science Letters 43, 285–287. Hoffman, P.F., Ranalli, G., 1988. Oceanic flake tectonics. Geophysical Research Letters 15, 1077– 1080. Hofmann, A., Dirks, P.H.G.M., Jelsma, H.A., 2001a. Horizontal tectonic deformation geometries in a late Archaean sedimentary sequence, Belingwe greenstone belt, Zimbabwe. Tectonics 20, 909–932. Hofmann, A., Dirks, P.H.G.M., Jelsma, H.A., 2001b. Late Archaean foreland basin deposits, Belingwe greenstone belt, Zimbabwe. Sedimentary Geology 141–142, 131–168. Hofmann, A., Dirks, P.H.G.M., 2003. Comment to Bolhar, R., Woodhead, J., Hergt, J., Continental setting inferred for emplacement of the 2.9–2.7 Ga Belingwe Greenstone Belt, Zimbabwe. Geology 31, e30–e31. Hofmann, A., Dirks, P.H.G.M., Jelsma, H.A., Matura, N., 2003a. A tectonic origin for ironstone horizons in the Zimbabwe craton and their significance for greenstone belt geology. Journal of the Geological Society of London 160, 83–97. Hofmann, A., Bolhar, R., Dirks, P.H.G.M., Jelsma, H.A., 2003b. The geochemistry of Archaean shales derived from a mafic volcanic sequence, Belingwe greenstone belt, Zimbabwe: provenance, source area unroofing and submarine vs subaerial weathering. Geochimica et Cosmochimica Acta 67, 421–440. Hofmann, A., Dirks, P.H.G.M., Jelsma, H.A., 2004. Shallowing-upward carbonate cycles in the Belingwe greenstone belt, Zimbabwe: a record of Archaean sea-level oscillations. Journal of Sedimentary Research 74, 64–81. Hollings, P., Wyman, D., Kerrich, R., 1999. Komatiite-basalt-rhyolite volcanic associations in Northern Superior Province greenstone belts: significance of plume-arc interaction in the generation of the proto continental Superior Province. Lithos 46, 137–161.

534

Chapter 15: The Belingwe Greenstone Belt

Horstwood, M.S.A., 1998. Stratigraphy, geochemistry and zircon geochronology of the Midlands greenstone belt, Zimbabwe. Unpublished Ph.D. thesis. University of Southampton. Horstwood, M.S.A., Nesbitt, R.W., Noble, S.R., Wilson, J.F., 1999. U-Pb zircon evidence for an extensive early Archaean craton in Zimbabwe: A reassessment of the timing of craton formation, stabilization, and growth. Geology 27, 707–710. Hunter, M., 1997. The tectonic setting of the Belingwe Greenstone Belt, Zimbabwe. Unpublished Ph.D. thesis. University of Cambridge. Hunter, M.A., Bickle, M.J., Nisbet, E.G., Martin, A., Chapman, H.J., 1998. Continental extensional setting for the Archean Belingwe Greenstone Belt, Zimbabwe. Geology 26, 883–886. Ingle, S., Weis, D., Scoates, J.S., Frey, F.A., 2002. Relationship between the early Kerguelen plume and continental flood basalts of the paleo-Eastern Gondwanan margins. Earth and Planetary Science Letters 197, 35–50. Jelsma, H.A., van der Beek, P.A., Vinyu, M.L., 1993. Tectonic evolution of the Bindura-Shamva greenstone belt (northern Zimbabwe): progressive deformation around diapiric batholiths. Journal of Structural Geology 15, 163–176. Jelsma, H.A., Vinyu, M.L., Valbracht, P.J., Davies, G.R., Wijbrans, J.R., Verdurmen, E.A.T., 1996. Constraints on Archaean crustal evolution of the Zimbabwe craton: a U-Pb zircon, Sm-Nd and Pb-Pb whole-rock isotope study. Contributions to Mineralogy and Petrology 124, 55–70. Jelsma, H.A., Dirks, P.H.G.M., 2002. Neoarchaean tectonic evolution of the Zimbabwe Craton. In: Fowler, C.M.R., Ebinger, C., Hawkesworth, C.J. (Eds.), The Early Earth: Physical, Chemical, and Biological Development. Geological Society of London Special Publication 199, 183–211. Kamenetsky, V.S., Maas, R., Sushchevskaya, N.M., Norman, M.D., Cartwright, I., Peyve, A.A., 2001. Remnants of Gondwanan continental lithosphere in oceanic upper mantle: Evidence from the South Atlantic Ridge. Geology 29, 243–246. Kusky, T.M., 1989. Accretion of the Archean Slave Province. Geology 17, 63–67. Kusky, T.M., 1991. Are the Upper greenstones at Belingwe, Zimbabwe, allochthonous? In: 22nd Lunar and Planetary Science Conference, March 18–22, Johnson Space Center, Houston, Texas, Proceedings Volume, pp. 769–770. Kusky, T.M., 1998. Tectonic setting and terrane accretion of the Archean Zimbabwe craton. Geology 26, 163–166. Kusky, T.M., Bradley, D.C., 1999. Kinematics of mélange fabrics: Examples and Applications from the McHugh Complex, Kenai Peninsula, Alaska. Journal of Structural Geology 21 (12), 1773– 1796. Kusky, T.M., Kidd, W.S.F., 1992. Remnants of an Archaean oceanic plateau, Belingwe Greenstone Belt, Zimbabwe. Geology 20, 43–46. Kusky, T.M., Polat, A., 1999. Growth of Granite-Greenstone Terranes at Convergent Margins and Stabilization of Archean Cratons. Tectonophysics 305, 43–73. Kusky, T.M., Vearncombe, J., 1997. Structural Aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 95–128. Kusky, T.M., Winsky, P.A., 1995. Structural relationships along a greenstone/shallow water shelf contact, Belingwe greenstone belt, Zimbabwe. Tectonics 14, 448–471. Kusky, T.M., van der Pluijm, B., Condie, K., Coney, P., 1997. Tectonic setting and terrane accretion in Precambrian orogens. Tectonophysics 265, 1–179. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274.

References

535

Laubscher, D.H., 1963. The origin and occurrence of chrysotile asbestos and associated rocks in the Shabani and Mashaba areas. Ph.D. thesis. University of the Witwatersrand, Johannesburg, South Africa. Lowe, D.R., 1999. Shallow-water sedimentation of accretionary lapilli-bearing strata of the Msauli Chert: Evidence of explosive hydro-magmatic komatiitic volcanism. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic Evolution of the Barberton Greenstone Belt, South Africa. Geological Society of America Special Paper 329, 213–232. Macgregor, A.M., 1947. An Outline of the Geological History of Southern Rhodesia. Southern Rhodesia Geological Survey Bulletin, 38. Macgregor, A.M., 1951. Some milestones in the Precambrian of Southern Rhodesia. Proceedings of the Geological Society of South Africa 54, 27–71. Mahoney, J.J., Coffin, M.F. (Eds.), 1997. Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism. In: Geophysical Monographs, vol. 100, p. 438. Martin, A., 1978. The Geology of the Belingwe-Shabani Shist Belt. Rhodesia Geological Survey Bulletin 83, 213. Martin, A., Nisbet, E.G., Bickle, M.J., Orpen, J.L., 1993. Rock units and stratigraphy of the Belingwe Greenstone Belt: The complexity of the tectonic setting. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 39–68. McDonough, W.F., Ireland, T.R., 1993. Intraplate origin of komatiites inferred from trace elements in glass inclusions. Nature 365, 432–434. Mkweli, S., Kamber, B.S., Berger, M., 1995. A westward continuation of the Zimbabwe craton— Northern marginal zone tectonic break and new age constraints on the timing of the thrusting. Journal of the Geological Society of London 152, 77–83. Moorbath, S.M., Wilson, J.F., Goodwin, R., Humm, M., 1977. Further Rb-Sr age and isotope data on early and late Archaean rocks from the Rhodesian craton. Precambrian Research 5, 229–239. Moorbath, S.M., Taylor, P.N., Orpen, J.L., Treloar, P., Wilson, J.F., 1987. First direct radiometric dating of Archaean stromatolitic limestone. Nature 326, 865–867. Moores, E., 1986. The Proterozoic ophiolite problem, continental emergence, and the Venus connection. Science 234, 65–68. Moores, E.M., 2002. Pre 1-Ga (pre-Rodinian) ophiolites: Their tectonic and environmental implications. Geological Society of America Bulletin 114, 80–95. Moores, E.M., Kellog, L.H., Dilek, Y., 2000. Tethyan ophiolites, mantle convection, and tectonic “historical contingency”: a resolution of the “ophiolite conundrum”. Geological Society of America Special Paper 349, 3–12. Myers, J.S., 1995. The generation and assembly of an Archaean super continent: evidence from the Yilgarn craton, western Australia. In: Coward, M.P., Ries, A.C. (Eds.), Early Precambrian Processes. Geological Society Special Publication 95, 143–154. Nägler, T.F., Kramers, J.D., Kamber, B.S., Frei, R., Prendergast, M.D.A., 1997. Growth of subcontinental lithospheric mantle beneath Zimbabwe started at or before 3.8 Ga: Re-Os study on chromites. Geology 25, 983–986. Nelson, D.R., Trendall, A.F., de Laeter, J.R., Grobler, N.J., Fletcher, I.R., 1992. A comparative study of the geochemical and isotopic systematics of late Archaean flood basalts from the Pilbara and Kaapvaal Cratons. Precambrian Research 54, 231–256. Nisbet, E.G., Bickle, M.J., Martin, A., 1977. The mafic and ultramafic lavas of the Belingwe Greenstone Belt, S. Rhodesia. Journal of Petrology 18, 521–566.

536

Chapter 15: The Belingwe Greenstone Belt

Nisbet, E.G., Bickle, M.J., Martin, A., Orpen, J.L., 1993a. Sedimentology of the Brooklands Formation, Zimbabwe: Development of an Archaean greenstone belt in a rifted graben. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 87–120. Nisbet, E.G., Bickle, M.J., Orpen, J.L., Martin, A., 1993b. Controls on the formation of the Belingwe Greenstone Belt, Zimbabwe. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 215–223. Nisbet, E.G., Martin, A., Bickle, M.J., Orpen, J.L., 1993c. The Ngezi Group: Komatiites, basalts and stromatolites on continental crust. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 121–165. Nisbet, E.G., Arndt, N.T., Bickle, M.J., Cameron, W.E., Chauvel, C., Cheadle, M., Hegner, E., Martin, A., Renner, R., Roedder, E., 1987. Uniquely fresh 2.7 Ga old komatiites from the Belingwe Greenstone Belt, Zimbabwe. Geology 15, 1147–1150. Oberthür, T., Davis, D.W., Blenkinsop, T.G., Höhndorf, A., 2002. Precise U-Pb mineral ages, Rb-Sr and Sm-Nd systematics for the Great Dyke, Zimbabwe—constraints on late Archean events in the Zimbabwe craton and Limpopo belt. Precambrian Research 113, 293–305. Orpen, J.L., 1978. The geology of the southwestern part of the Belingwe Greenstone Belt and adjacent country—The Belingwe Peak area. Unpublished Ph.D. thesis. University of Rhodesia. Orpen, J.L., Martin, A., Bickle, M.J., Nisbet, E.G., 1993. The Mtshingwe Group in the west: Andesites, basalts, komatiites and sediments of the Hokonui, Bend and Koodoovale Formations. In: Bickle, M.J., Nisbet, E.G. (Eds.), The Geology of the Belingwe Greenstone Belt, Zimbabwe. In: Geological Society of Zimbabwe Special Publications, vol. 2, pp. 69–86. Parman, S.W., Grove, T.L., 2004. Petrology and geochemistry of Barberton komatiites and basaltic komatiites: evidence of Archean fore-arc magmatism. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 539–565. Passchier, C.W., Trouw, R.A.J., 1996. Microtectonics. Springer-Verlag, Berlin, p. 289. Peltonen, P., Kontinen, A., 2004. The Jormua ophiolite: A mafic-ultramafic complex from an ancient ocean-continent transition zone. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 35–71. Peltonen, P., Mänttäri, I., Huhma, H., Kontinen, A., 2003. Archean zircons from the mantle: The Jormua ophiolite revisited. Geology 31, 645–648. Pickering, K.T., Stow, D.A.V., Watson, M., Hiscott, R.N., 1986. Deep-water facies, processes and models: a review and classification scheme for modern and ancient sediments. Earth Science Reviews 23, 74–174. Prendergast, M.D., 2003. The nickeliferous late Archean Reliance komatiitic event in the Zimbabwe craton—magmatic architecture, physical volcanology, and ore genesis. Economic Geology 98, 865–891. Ridley, J.R., Vearncombe, J.R., Jelsma, H.A., 1997. Relations between greenstone belts and associated granitoids. In: de Wit, M.J., Ashwal, L.D. (Eds.), Tectonic Evolution of Greenstone Belts. Oxford Univ. Press, pp. 376–397. Saleeby, J.B., 1992. Petrotectonic and paleogeographic settings of U.S. Cordilleran ophiolites. In: Burchfiel, B.C., Lipman, P.W., Zoaback, M.I. (Eds.), The Cordilleran Orogen. Conterminous U.S. Geological Society of America G-3, pp. 653–683.

References

537

Scholey, S.P., 1992. The geology and geochemistry of the Ngezi Group volcanics, Belingwe Greenstone Belt, Zimbabwe. Unpublished Ph.D. thesis. University of Southampton. Shackleton, R.M., 1995. Tectonic evolution of greenstone belts. In: Coward, M.P., Ries, A.C. (Eds.), Early Precambrian Processes. Geological Society Special Publication 95, 53–65. Shimizu, K., Komiya, T., Hirose, K., Shimizu, N., Maruyama, S., 2001a. Cr-spinel, an excellent micro-container for retaining primitive melts—implications for a hydrous plume origin for komatiites. Earth and Planetary Science Letters 189, 177–188. Shimizu, K., Nakamura, E., Maruyama, S., 2001b. Petrological and geochemical study of komatiite, komatiitic basalt and basalt in Belingwe greenstone belt, Zimbabwe—evidence for crustal contamination. In: Cassidy, K.F., et al. (Eds.), 4th International Archaean Symposium 2001, Extended Abstracts. AGSO–Geoscience Australia, Record 2001/37, pp. 188–189. Shimizu, K., Nakamura, E., Maruyama, S., Kobayashi, K., 2002. The evolution of the Belingwe greenstone belt, Zimbabwe. Geochimica et Cosmochimica Acta 66, A709. Sleep, N., Windley, F., 1982. Archean flake tectonics: constraints and inferences. Journal of Geology 90, 363–379. Sloss, L.L., 1963. Sequences in the cratonic interior of North America. Geological Society of America Bulletin 74, 93–113. Stagman, J.G., 1978. An Outline of the Geological History of Rhodesia. Rhodesia Geological Survey Bulletin 80. Stanistreet, I.G., de Wit, M.J., Fripp, R.E.P., 1981. Do graded units of accretionary spheroids in the Barberton Greenstone Belt indicate Archaean deep water environment? Nature 293, 280–284. St-Onge, M.R., Lucas, S.B., 1990. Evolution of the Cape Smith Belt: Early Proterozoic continental underthrusting, ophiolite obduction, and thin-skinned folding. In: Lewry, J.F., Stauffer, M.R. (Eds.), The Early Proterozoic Trans-Hudson Orogen of North America. Geological Association of Canada Special Paper 37, 313–351. Storey, M., Mahoney, J.J., Kroenke, L.W., Saunders, A.D., 1991. Are oceanic plateaus sites of komatiite formation? Geology 19, 376–379. Stowe, C.W., 1971. Summary of the tectonic development of the Rhodesian Archean craton. Special Publication of the Geological Society of Australia 3, 377–383. Stowe, C.W., 1984. The early Archaean Selukwe nappe, Zimbabwe. In: Kröner, A., Greiling, R. (Eds.), Precambrian Tectonics Illustrated. Nägele und Obermiller, Stuttgart, pp. 41–56. Subba Rao, K.V.B. (Ed.), 1988. Deccan Flood Basalts. Geological Society of India, Memoir 10. Swager, C., Griffin, T.J., 1990. An early thrust duplex in the Kalgoorlie-Kambalda greenstone belt, Eastern Goldfields Province, Western Australia. Precambrian Research 48, 63–73. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.C., 1997. Volcanic Aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 55–90. Taylor, P.N., Kramers, J.D., Moorbath, S., Wilson, J.F., Orpen, J.L., Martin, A., 1991. Pb/Pb, SmNd and Rb-Sr geochronology in the Archaean Craton of Zimbabwe. Chemical Geology (Isotope Geoscience Section) 87, 175–196. Vernon, R.H., Williams, V.A., Dárcy, W.F., 1983. Grain-size reduction and foliation development in a deformed granitoid batholith. Tectonophysics 92, 123–145. Walker, R.J., Nisbet, E., 2002. 187 Os isotopic constraints on Archean mantle dynamics. Geochimica et Cosmochimica Acta 66, 3317–3325. Wilson, J.F., 1964. The geology of the country around Fort Victoria. Southern Rhodesia Geological Survey Bulletin 58, 147.

538

Chapter 15: The Belingwe Greenstone Belt

Wilson, J.F., 1979. A preliminary reappraisal of the Rhodesian basement complex. In: Anhaeusser, C.R., Foster, R.P., Stretton, T. (Eds.), A Symposium on Mineral Deposits and Transportation and Deposition of Metals. Geological Society of South Africa Special Publication 5, 1–23. Wilson, J.F., 1990. A craton and its cracks: some of the behaviour of the Zimbabwe block from the late Archaean to the Mesozoic in response to horizontal movements, and the significance of some of its mafic dyke fracture patterns. Journal of African Earth Sciences 10, 483–501. Wilson, J.F., Nesbitt, R.W., Fanning, C.M., 1995. Zircon geochronology of Archaean felsic sequences in the Zimbabwe craton: a revision of greenstone stratigraphy and a model for crustal growth. In: Coward, M.P., Ries, A.C. (Eds.), Early Precambrian Processes. Geological Society Special Publication 95, 109–126.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

539

Chapter 16

PETROLOGY AND GEOCHEMISTRY OF BARBERTON KOMATIITES AND BASALTIC KOMATIITES: EVIDENCE OF ARCHEAN FORE-ARC MAGMATISM S.W. PARMAN AND T.L. GROVE Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

1. INTRODUCTION The Barberton Greenstone Belt (BGB) is one of several mid- to late-Archean greenstone belts that lie along the eastern margin of the Kaapvaal craton (Brandl and de Wit, 1997). With an age of 3.49–3.46 Ga (Lopezmartinez et al., 1992), the BGB is among the oldest of the Kaapvaal Craton’s greenstone belts and is part of the nucleus around which the Late Archean greenstone belts to the north (e.g., Murchison and Giyani) and to the south (e.g., Nondweni and Commondale) were attached (Fig. 1). It was in the BGB that komatiites were first recognized as high-MgO magmas (Viljoen and Viljoen, 1969a, 1969c). Subsequently, they were found in many greenstone belts throughout the world, establishing komatiites as a widespread and peculiar aspect of Precambrian magmatism (Arndt and Brooks, 1980). In Barberton, as in most greenstone belts, komatiites make up a fairly small proportion of the total exposed rocks; ∼ 10% (de Wit and Ashwal, 1997). Despite their small volume, komatiites have received broad scientific attention because their unusual compositions are the premiere, and possibly singular, evidence that the Archean mantle was hotter than it is today (Herzberg, 1995; Nisbet et al., 1993; Walter, 1998). Barberton komatiites in particular are central to many geophysical and geochemical models of mantle evolution, because they are among the oldest komatiites known (3.49 Ga) and their compositions suggest the highest mantle temperatures (Herzberg, 1995; Walter, 1998). Due to this, they have been mapped multiple times (Dann, 2000; de Wit et al., 1987; Lowe and Byerly, 1999; Viljoen and Viljoen, 1969c; Williams and Furnell, 1979) and have been sampled repeatedly by geochemists and petrologists (to the great detriment of the outcrops in the type section, it may be noted). The nature of the melting process that gave rise to such unusual magmas has been the subject of diverse speculation. Initially, some very non-uniformitarian processes were proposed, including leaks from persistent magma oceans (Nisbet and Walker, 1982) and melts produced by meteorite impact (Green, 1972). As evidence has mounted for the operation of some form of plate tectonics in the Archean, komatiite enthusiasts have focused on DOI: 10.1016/S0166-2635(04)13016-8

540

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Fig. 1. Regional map of Kaapvaal Craton (after Schmitz, 2001).

linking komatiites to the major modern melting environments: mid-ocean ridges (MOR), subduction zones (SZ), or plumes. The plume model has gained the widest acceptance (Arndt et al., 1997; Herzberg, 1995; Walter, 1998). In this theory, komatiites are proposed to be generated by thermal plumes of mantle rising from a thermal boundary layer within the Earth, the same process that

1. Introduction

541

is thought to have generated the Hawaiian island chain and the Ongtong-Java oceanic plateau. The thermal boundary layer is usually ascribed to either the core-mantle boundary or the 660 km seismic discontinuity. The unusual composition of komatiites is explained by higher temperatures in the Archean mantle, or at the thermal boundary layer. The higher potential temperature of the Archean plumes would cause them to begin melting at greater depth and to a greater extent than modern plumes. Both higher pressure and higher degree of melting would contribute to the high MgO of komatiites (Herzberg, 1992, 1995; Walter, 1998). Other aspects of komatiite chemistry are attributed to the presence of pyrope and/or majorite garnet in the source of komatiites (Walter, 1998; Xie et al., 1995). These minerals are only stable at pressures above 2.6 GPa, and so their geochemical signatures would corroborate a high pressure origin for komatiites. This model requires the Archean mantle to have been 200 to 400 degrees hotter than the present mantle (Herzberg, 1995; Nisbet et al., 1993). While the plume model for komatiite production is the most widely cited, SZ models have persisted. Early on, geochemical similarities between komatiites and modern mafic arc-volcanics were noted (Brooks and Hart, 1974). It was also realized that if komatiites were produced by wet melting, then the high mantle temperatures required by anhydrous melting models could be substantially reduced (Allegre, 1982; Mysen and Boettcher, 1975). Petrographic, petrologic and melt inclusion studies have provided support for the presence of substantial dissolved H2 O contents in some komatiite magmas (Parman et al., 1997; Stone et al., 1997), though this is not universally accepted (Arndt et al., 1998). Presently, SZs are the predominant site of hydrous volcanism, while plume magmas are relatively anhydrous. Thus high H2 O contents favor a SZ origin for komatiites, though hydrous plume models have been proposed (Asahara et al., 1998; Kawamoto et al., 1996). At the time much of the initial geologic and geochemical work on komatiites was being done (mid-1970s), boninites were just beginning to be recognized as a rare but widespread form of highly magnesian, hydrous arc volcanism, though their existence had been known for some time (Crawford et al., 1989). In contrast to the uncertainty surrounding the melting process and tectonic setting of komatiites, the origin of boninites was identified rather quickly. They are produced by hydrous melting of depleted mantle in the fore-arcs, and less frequently, back-arc basins of subduction zones (Crawford, 1989). Presently they form a significant portion of the Marianas fore-arc crust and other fore-arcs in western Pacific SZs, and are also present in a few back-arc spreading centers (Falloon and Crawford, 1991). Boninites are often present in ophiolites (Crawford, 1989). This has lead to the realization that many ophiolites represent fore-arc or back-arc oceanic crust, rather than MOR-type oceanic crust. Researchers have noted a strong similarity between the petrography and geochemistry of boninites and basaltic komatiites (Cameron et al., 1979; Redman and Keays, 1985; Sun et al., 1989b), as well as with unusually high-SiO2 komatiites in Nondweni (Wilson and Versfeld, 1994). Basaltic komatiites are ubiquitous in komatiite sequences. Some can be explained as being the products of fractional crystallization of their associated komatiites, but others have major and trace element compositions that preclude this, and must

542

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

have been generated by their own set of P-T melting conditions. The major and trace element compositions of the boninites and basaltic komatiites largely overlap, and some samples are nearly identical (Parman et al., 2001). Nevertheless, differences in TE and isotope systematics have been used to argue against a genetic relationship (Sun et al., 1989b). Studies have now found boninites sensu stricto conformably interlayered with komatiites, most notably in the greenstone belts of the Superior Province (Kerrich et al., 1998). They propose that the interlayering of komatiite and boninite lavas occurred when a plume was located near a subduction zone. The magmas from the plume (komatiites) overflowed the intervening trench and became interlayered with boninites that were simultaneously erupting in the fore-arc (Kerrich et al., 1998). This plume-arc scenario has been extended to other komatiite bearing greenstone belts (Hollings et al., 1999). The plume-arc scenario seems to us physically unlikely. Perhaps one odd occurrence might be explained as chance, but its apparent repetition is difficult to accept. Either mantle upwellings and downwellings had an unusual attraction for each other in the Archean, or another explanation must be sought. In this paper we propose that komatiites and their associated lavas are the Archean equivalents of boninite sequences and that the similarity between basaltic komatiites and boninites is not fortuitous. Thus the komatiite-basaltic komatiite sequences were produced by a single subduction related event, and no coincidental juxtapositions are required. Our approach inherently assumes that plate tectonics operated in a fashion that, if not exactly the same as today, is at least recognizable geochemically and structurally. Such an actualistic approach has its obvious limits. The modern mantle is not producing komatiites, as far as we know, and so something must have been different in the Archean. Our discussion will focus on the komatiites and their related basaltic komatiites from the Komati and Hoogenoeg formations. These two formations form a continuous stratigraphic section and have been the main focus of our research, though reference will also be made to komatiites in the BGB’s smaller and less well preserved komatiite bearing sequences: Sandspruit, Theespruit, Mendon and Weltevreden. In the end we will put the Barberton data in the context of the global komatiite data set.

2. FIELD GEOLOGY 2.1. Geologic Context The Barberton Greenstone belt (BGB) consists of a number of tectonic blocks sutured together and folded by tectonic processes (Fig. 2), and appears to be an example of a fold and thrust belt (de Wit, 1982; de Wit et al., 1987). The different blocks contain both igneous and sedimentary components and record a > 400 m.y. geologic history (Brandl and de Wit, 1997; de Wit et al., 1987; Lowe, 1999a). At least two periods of arc-related volcanism are recorded by felsic intrusive and extrusive layers. Some of the sedimentary sequences have also been interpreted as produced in a convergent margin (Lowe, 1999a), and the suturing of the various blocks is generally ascribed to one or more collisional

2. Field Geology

543

Fig. 2. Geologic map of western Barberton greenstone belt (Lowe, 1999a).

events. Thus the overall history of the BGB appears to be a familiar one of crustal growth occurring through arc-magmatism and continental collision. The presence of the ultra-mafic units within this convergent setting does not require that they were produced by the subduction zone itself. Because a wide variety of terranes are processed through subduction zones, plausible origins for the komatiites include nearly all tectonic settings: (1) They could represent sections of MOR crust obducted onto a continental mass during convergence, similar to some models for the formation of the Oman ophiolite; (2) they could represent SZ oceanic crust either from a fore-arc or back-arc basin. This is how most ophiolites are thought to form; or (3) they could represent an ocean island or large igneous province (LIP) such as Ongtong-Java, scraped off the subducting crust onto the over-riding plate.

544

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

The BGB contains a number of komatiite bearing sequences. In order from lowest to highest structural level they are the Sandspruit, Theespruit, Komati, Kromberg, Mendon, and Weltevreden (Fig. 2). Reviews of their geology can be found elsewhere (Brandl and de Wit, 1997; Lowe, 1999a). Though more dating is necessary, these appear to be distinct magmatic events, rather than a single repeated section. An interesting similarity between the ultramafic units is that they occur at the bottom of their respective structural blocks and stratigraphic reconstructions, and therefore seem to be the magmatic events that initiate each crust-building process (Lowe, 1999a). 3. BRIEF DESCRIPTION OF KOMATI FM. The Komati Fm. lies above the Theespruit Fm., and is separated from it by the Komati fault. While there is still much work to be done on the timing of events, it appears that at least part of the Komati and Theespruit are coeval, and that the former was thrust over the latter by a collisional event. The Komati Fm. is separated into an Upper and Lower part. The Lower Komati Fm. is distinguishable by the presence of olivine spinifex komatiites, which are interlayered with massive serpentinite and basaltic komatiites. Some of the basaltic komatiites are pillowed, indicating eruption underwater (Dann, 2000; Viljoen and Viljoen, 1969b), though an intrusive origin has also been forwarded (de Wit et al., 1987). The Upper Komati largely consists of pillowed basaltic komatiite, with a small percentage of massive and pyroxene spinifex komatiite. It is overall, noticeably less mafic than the Lower Komati Fm. The stratigraphic thickness of the Lower and Upper Komati is ∼ 1.76 km and ∼ 1.27 km, respectively (Dann, 2000). The Komati Fm. contains one felsic tuff. A single zircon from this unit has yielded an age of 3.481 ± 2 Ga (Bowring, unpublished data). Less mafic still is the overlying Hoogenoeg Fm., with a thickness of ∼ 3 km (Dann, 2000). It mostly contains basalt with abundant pillow structures, and a moderate proportion of basaltic komatiite. The boundary between the Komati and the Hoogenoeg is the Middle Marker. Initially interpreted as a chert horizon, it is now thought to be a pyroclastic layer that has been heavily silicified by the metamorphism that has affected all of the BGB (Lowe, 1999b). Thus the Komati and Hoogenoeg Fms. comprise a single, continuously active, or nearly so, igneous period. The reader is referred to the recent remapping of the Komati Fm. by Dann (Dann, 2000, 2001) for a more detailed stratigraphic description. 4. PETROLOGY 4.1. Petrography All of the units in the Komati and Hooggenoeg Fm. have experienced extensive greenschist facies metamorphism with the lowest structural levels approaching the greenschistamphibolite transition (Cloete, 1994). The original mineralogy of the high-MgO komatiites

4. Petrology

545

Fig. 3. Back-scattered electron image of preserved augite in Barberton komatiite. Nearly the entire core to rim profile is preserved. Image is ∼ 300 microns across. Lighter color of rims indicates higher iron content, typical of igneous zoning. Dark grey overgrowth on augite is metamorphic tremolite. White phase is magnetite. Black matrix material is a crypto-crystalline mixture of chlorite and serpentine, and is interpreted to have been glass, prior to metamorphism. In olivines, serpentine filled cracks that cut relict crystals cause Fe enrichment in the olivine (observable the back-scattered image as a brightening of the adjacent olivine). No such Fe enrichment is associated with the cracks in the augite.

was dominated by olivine, which was the first phase to crystallize along with minor proportions of chrome-rich spinel. In massive textured units, the olivines are equant with typical diameters of a centimeter. In spinifex units, the olivines range from small elongate cm sized crystals near the chill margins, to spectacular meter long blades in the interiors of thicker (7 m) units. The most mafic cores are Fo94 (Parman et al., 1997). The olivines have largely been altered to an assemblage of serpentine and magnetite. Where metamorphic veins cut the olivine relicts, a distinct iron enrichment is seen in the crystals, clearly indicating the compositions of the crystals have diffusively exchanged with metamorphic fluids during their long 3.5 Ga history (Parman et al., 1997). The spinels are overgrown with various amounts of magnetite. Sub-calcic augite was the third and last phase to crystallize. In massive units they occur as equant to elongate interstitial crystals. In the olivine-spinifex units, they range from elongate to thin lathes that fill the polygonal spaces between the olivine blades. A series of metamorphic reactions affect the augites. Initially, they alter to tremolite, which in turn reacts to an assemblage of chlorite, serpentine and magnetite (Fig. 3). Unlike the olivines, no evidence of any diffusional effects are seen in the relict augites (Parman et al., 1997). Pyroxene spinifex komatiites are characterized by decimeter long sheaves of augite needles. As described above, the augite metamorphoses to an assemblage of tremolite, chlorite, serpentine and magnetite. One unit in the Upper Komati Fm. is notable for the presence of pigeonite cores inside the augite relicts (Parman et al., 1997). This unit appears to be the base of a pyroxene-spinifex unit, though the contact is obscured. The basalts and basaltic komatiites are altered to assemblages of amphibole, chlorite, serpentine and magnetite. Unlike the komatiites, the original textures have largely been

546

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

overprinted by the metamorphism and we have not discovered any unaltered minerals. The lack of preserved crystal textures suggests these units were relatively aphyric or microcrystalline. No glass is preserved in any of the units. The homogeneity of the crypto-crystalline serpentine-chlorite intergrowths that replaced the mesostasis in the komatiites suggests that it was glass, though it is possible that it was micro-crystalline. Exceptions to the above general description exist. Heavy to complete silicification of the komatiites is seen in a few locations, as is talc-carbonate alteration (Dann, 2000). Interesting vesicular komatiites, and a number of wehrlite dykes are also present (Dann, 2001). 4.2. Compositional Effects of Metamorphism Most samples from the Komati Fm. contain no igneous phases. Even the least altered samples rarely contain more than 5% fresh minerals, though cm size patches do preserve up to ∼ 70% of the original mineralogy (Fig. 3). The lack of unaltered samples is a fundamental barrier to deciphering the geochemistry of the samples. It is entirely possible that no samples retain their original, igneous composition. That being said, petrologists and geochemists have taken hope from the observation that samples of aphyric and spinifex komatiites (thought to be unaffected by crystal accumulation) plot near an olivine fractionation trend for many elements (Fig. 4a). This is what is expected for such olivine-rich magmas, with the higher MgO samples representing the more primitive magmas, and lower MgO samples being the result of fractionation. Unfortunately, this trend is also close to a serpentinization trend in which the higher MgO samples are the result of element loss during the transformation of the rock to tremolite, chlorite and eventually, serpentine. This second explanation of the chemical trends seems to be favored by the observation that the metamorphic H2 O contents of the samples increases with MgO content (Fig. 4b). Yet the MgO-H2O correlation can also reasonably be expected if the trend was predominantly due to olivine fractionation. During metamorphic hydration, the higher MgO samples have the potential to produce more serpentine, and so will incorporate higher metamorphic H2 O contents than less mafic samples. In this case, one might expect to see some scatter in MgO vs. H2 O contents between the olivine fractionation line and the fully serpentinized trend. The degree of serpentinization in the samples appears quite variable in thin section. Yet little scatter is seen. What then can be said of komatiite alteration? That at least part of the compositional trends in the samples is due to olivine fractionation seems inescapable. The massive komatiites and the basal cumulates of spinifex bearing units are prima facie evidence of extensive olivine fractionation. Yet many of the chill margin samples have Mg#’s (MgO/MgO + FeO∗ ) too high to be in equilibrium with the most forsteritic olivine cores (Fo94), clearly providing evidence for Mg enrichment through serpentinization (de Wit et al., 1987; Nisbet et al., 1993). We are thus forced into adopting a vague yet optimistic stance that the komatiite compositions are close enough to the original igneous compositions that the broad features of their chemistry (high MgO, low TiO2 , high CaO/Al2 O3 ) are indicative of the melting process that produced them and can be used in petrologic estimates of melting

4. Petrology

547

Fig. 4. Compositional effects of metamorphism seen in Barberton komatiites. (a) Ternary projection of chill and aphyric komatiites from the Komati Fm. (data from Smith and Erlank, 1982, 1980; reduction method of Grove, 1993). Arrows show possible origin of compositional variations. The entire trend could be metamorphic (light grey arrow), with the sample furthest from serpentine representing the least altered composition. Alternatively, the trend could mostly be olivine fractionation (dark grey line), with metamorphism causing only a slight rotation of the trend (dark grey arrow). In this case, the sample closest to olivine (and serpentine) would represent the least fractionated melt composition. (b) Correlation between metamorphic H2 O content of samples and their composition (projected onto cpx-ol join). The most mafic samples (lowest cpx/(cpx + ol)) have the highest H2 O content. Arrows show the two possible metamorphic processes described above.

conditions, but that we should be wary of trying to split hairs with such compromised data. Recent experiments crystallizing a least-altered Barberton komatiite composition have reproduced the compositions of the olivine and augite preserved in the rocks, and thus pro-

548

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

vide some support for the compositions being close to igneous (Parman et al., 1997). The similarity between komatiite compositions from greenstone belts around the world also suggests the chemical trends are predominantly igneous, rather than metamorphic in origin. Trace element compositions have also been affected by metamorphism. Both positive and negative Eu anomalies (relative to the other REE) are present in many samples, and Sr isotopes do not yield the igneous age of the samples (Jahn and Sun, 1979). While HREE patterns are fairly similar from sample to sample, LREE and LILE concentrations and patterns vary widely. This has been interpreted as the mobilization of LILE by metamorphic fluids (Lahaye et al., 1995; Lecuyer et al., 1994). Yet variations in LILE can be caused by many pre-metamorphism processes such as crustal contamination and fluid input to the source of melting (e.g., subduction zones). In the following section we take a different approach, and use the TE composition of the preserved igneous augite to estimate the TE composition of the melt from which it grew and to estimate the compositional effects of metamorphism.

5. AUGITE COMPOSITION 5.1. Major Element Composition The augite relicts are perhaps the only unaltered material in the Barberton komatiites, and as such, their compositions provide the clearest record of the emplacement conditions and pre-metamorphosis magma compositions of the BK units. Towards this end, we have analyzed both the major and trace element compositions of BK augites throughout the Komati Fm. (Parman et al., 1997). Major element partitioning is sensitive to pressure and temperature conditions, as well as bulk composition, and is used to provide constraints on the emplacement conditions of BK. Trace element partitioning is relatively insensitive to P-T conditions as well as post emplacement crystal fractionation/accumulation, and is used to infer source composition as well as the post-emplacement effects of metamorphism. The compositions of augite relicts from the olivine spinifex units in the lower Komati Fm. are shown in Fig. 5. The crystals show typical igneous zonation, with high Mg#, high wollastonite (Wo) cores, and low Mg#, low Wo rims. Diffusional re-equilibration with the surrounding matrix during metamorphism would produce high Mg#, high Wo rims, and opposite of the observed zonation. The cores represent the first augite to grow from the melt, after approximately 30 wt% olivine fractionation. They display a limited range of Wo contents, between 0.39 and 0.42, and were used by Parman et al. to estimate the emplacement conditions and H2 O contents (Parman et al., 1997). Augite compositions produced experimentally from a komatiite bulk composition are shown as fields in Fig. 5. The natural augite cores fall within the field of augites produced with 6 wt% H2 O dissolved in the melt at 2 kb pressure. Augites produced in anhydrous experiments at 1 atm have substantially lower Wo, with no compositional overlap with the

5. Augite Composition

(a)

(b)

Fig. 5. Major element composition of augite in Barberton komatiites and in experiments on a least altered Barberton komatiite composition (Parman et al., 1997). (a) Molar Mg# (Mg# = MgO/(MgO + FeO)) versus wollastonite (Wo) content of augite. Dark grey ovals are compositions of augite cores from 10 Barberton samples. Arrows show typical core to rim zoning profiles. All data points (squares) for sample B95-11 are shown to illustrate distribution of data (field for core composition is unfilled). Compositional fields (light grey) of experimentally produced augite are also shown. Augite in anhydrous experiments has distinctly lower Wo contents than those crystallized from hydrous komatiite melt. The cores all fall within the 6 wt% H2 O field. (b) Al content (per 6 oxygens) versus Mg# of augite. Augites in the interior of komatiite units (open circles) have high Mg#’s and relatively low Al. The augite cores (highest Mg# points) fall along the equilibrium curve defined by the compositions of augites produced the experiments (both wet and dry). Filled diamonds show the composition of augites that crystallized near chill margins of units. Due to disequilibrium partitioning during rapid cooling, they are distinctly Al rich compared to the augite from the interior. 549

550

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

natural samples. Augites produced at 1 kb with 3 wt% H2 O (not shown, to prevent figure clutter) mostly fall below the natural augite field, with a small overlap at the low Wo end of the natural range. Anhydrous crystallization under higher pressures would lower the Wo content of the augites (mainly due to the rise in the crystallization temperature) and so cannot reconcile the dry experiments with the natural augites. Disequilibrium crystallization would shift all partition coefficients towards a value of unity and so would also lower the Wo content of the experiments. Furthermore, under fast growth conditions, the augite crystals incorporate high concentrations of Al. This effect is seen in augite crystals that formed within a few cm of the chill margins of units, but is not present in the augite elsewhere in the units (Fig. 5b). The augite compositions shown in Fig. 5 are all from the interior of BK units and have low Al contents. Corroborating evidence that the BK melts contained high H2 O contents is provided by the near absence of pigeonite in BK units (Parman et al., 1997). In the anhydrous experiments, pigeonite appears before augite in the crystallization sequence, and so one would expect it to be an abundant phase in BK samples. In the hydrous experiments, pigeonite nucleation is suppressed, and crystallizes after augite, consistent with the BK petrography. Disequilibrium during rapid crystal growth can also suppress pigeonite nucleation relative to augite, but as discussed above, such conditions only occurred near the margins of units. Pigeonite has been discovered in one sample from the lower Komati Fm. (Dann, personal communication) and one sample from the upper Komati Fm. (Parman et al., 1997). We infer that these units had lower initial H2 O contents or had degassed to a greater extent than the units that lack pigeonite. The presence of H2 O in the melt may also help explain the unusual olivine and spinifex textures in komatiites. H2 O greatly enhances diffusion rates and lowers nucleation densities, both of which encourage the growth of large crystals. The effects of H2 O on crystal growth have been used to explain the spinifex-like olivine textures in the Rhum layered intrusion (Donaldson, 1974). The scatter in the Wo content of the experimental and natural augites is the primary source of uncertainty in estimates of dissolved H2 O contents. Augites produced in experiments with 0, 3, and 6 wt% dissolved H2 O have average Wo contents of 0.338, 0.379, and 0.402, respectively. All three values have a 2-sigma standard deviation of ±0.015. Linear regression through these data yield the expression H2 O = −31.1 + 91.3Wo (R2 = 0.974). The range of Wo contents in the natural augite cores are 0.390 to 0.420, closely corresponding to the range in the 6 wt% H2 O experiments. Using the expression for H2 O, H2 O contents at the time of augite crystallization are 4.6–7.3 ±1.5 wt% H2 O. The large scatter in the experimentally produced augites is an indication of a failure of the experiments to reach equilibrium. The disequilibrium is caused by the short run times (6 h) required to retain H2 O throughout the duration of the experiment (Parman et al., 1997). Prior to the appearance of augite in BK melts, 30 wt% olivine crystallizes, raising the H2 O content of the melt. Correcting for olivine fractionation yields estimates of H2 O content at time of emplacement for BK of 3.2–5.1 wt% H2 O. Two points about these estimates should be emphasized. Our estimates are based on the amount of H2 O present at the time of augite crystallization. H2 O lost by devolatilization are not accounted for by our method

6. Geochemistry

551

and therefore our estimates of pre-emplacement H2 O contents are minima. Second, the H2 O versus Wo relationship is actually curved, not linear (with only three data points we do not feel justified using a polynomial fit). Estimated H2 O contents above 6 wt% (4.2 wt% after olivine correction) should be treated with caution. 5.2. Trace Element Composition We have analyzed the preserved augites for their REE concentrations using the Cameca 3f ion probe at the Woods Hole Oceanographic Institute (Parman et al., 2003). The REE patterns of the augites display significant variations (Fig. 6). The variations correlate with the bulk composition of the samples in which the augite were found. Those augites with the highest La/Sm occur in the samples with the highest bulk La/Sm (Fig. 6a), suggesting the high La/Sm was a compositional feature of the melt from which the crystals grew, and was not introduced by post-emplacement metamorphosis. Using experimentally determined REE partition coefficients, the compositions of the melts in equilibrium with the augites have been calculated. These calculated melts agree with the bulk rock compositions for all REE except Eu, indicating that the REE have not been significantly mobilized in the Barberton samples. Both Sr and Eu show clear evidence of metamorphic mobilization. Sr/Nd of calculated melts are significantly higher than Sr/Nd of the bulk samples (Fig. 6b), indicating loss of Sr. The cause of the LREE enrichments in the BK melts are discussed in section 5.2.

6. GEOCHEMISTRY 6.1. Major Elements The Komati and Hoogenoeg Fms. contain a variety of MgO-rich magmas (Fig. 7). In this paper, we will only distinguish between komatiites (MgO > 18 wt%) and basaltic komatiites (MgO < 18 wt%). The lower MgO, SiO2 basaltic komatiites could also be categorized as low-Ti tholeiites, sometimes referred to as back-arc tholeiites. Overall, the various rock types in the Komati and Hoogenoeg Fms. form a continuous compositional trend, which we will refer to as the komatiite series or as komatiitic magmas. A number of researchers (Cameron et al., 1979; Sun et al., 1989b) have noted the major element similarity between modern boninites and ancient basaltic komatiites. Most notably, both have high SiO2 for a given MgO content, low TiO2 , and high MgO contents. Significant overlap between boninites and Barberton basaltic komatiites is seen for all elements (Fig. 7). The high MgO end of the boninite samples overlaps the low MgO end of the komatiite compositions. From this it can be seen that komatiites share the high SiO2 (for their high MgO) and low TiO2 character of boninites and basaltic komatiites. There are some differences between the boninites and komatiitic magmas. The least amount of overlap is seen for FeO and TiO2 , with most boninites having lower concentrations of both elements (Fig. 7). CaO and Al2 O3 contents largely overlap, but CaO contents

552

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Fig. 6. Trace element (TE) composition of augite in Barberton komatiites compared to bulk sample TE compositions (Parman, unpublished data). (a) La/Sm ratio of augites (open diamonds) and in bulk samples (filled black circles) increases in the samples from left to right. The compositions of melts in equilibrium with the augite were calculated using cpx/melt partition coefficients (Hart and Dunn, 1993). They encompass the bulk sample compositions, indicating that metamorphism did not significantly affect La or Sm concentrations. (b) Sr/Nd ratios for augites, bulk samples, and calculated melts (symbols as in top panel). No correlation is seen between the augite and bulk samples. The calculated melts have higher Sr/Nd than the whole-rock analyses, indicating bulk loss of Sr from the sample.

of the komatiitic samples mostly fall on the high side of the boninite field, while Al2 O3 contents fall on the low end of the field. Sun (Sun et al., 1989b) has used these differences to argue that the similarity between boninites and basaltic komatiites is coincidental and largely due to crustal contamination of more mafic, less siliceous komatiite magmas. Estimated average continental crust does plot at the end of both the boninite and basaltic

6. Geochemistry

553

Fig. 7. Major element composition of komatiites (filled circles) and basaltic komatiites (open squares) from the Komati and Hoogenoeg Fms. Composition of boninites (and associated low-Ti tholeites) indicated by grey fields. Komatiite data from Dann (unpublished data) and Smith and Erlank (1982, 1980). Boninite data from Bloomer and Hawkins (1987), Brown and Jenner (1989), Cameron (1989), Crawford et al. (1989), Falloon et al. (1989), Hickey and Frey (1982), Upadhyay (1982). Basaltic komatiites, and to a lesser extent komatiites, overlap the boninite field in all plots. Least amount of overlap is for TiO2 and FeO.

554

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Fig. 8. CaO/Al2 O3 versus Al2 O3 for komatiites and basaltic komatiites from the Komati and Hoogenoeg Fms. Same symbols and data as in Fig. 7. Average composition of continental crust (Rudnick and Fountain, 1995) indicated by CC. The correspondence between the boninite trend and the komatiitic trend is remarkable.

komatiite trends, but the boninites erupted through thinned oceanic crust and so cannot have been contaminated. One of the most unusual compositional features of the komatiitic magmas is their high CaO/Al2 O3 contents. While CaO/Al2 O3 in MORB and OIB rarely varies from the mantle value of 0.86, komatiites have values of 1–3.5 (Fig. 8). Basaltic komatiites range from 0.6 to 3.0. The high CaO/Al2 O3 ratios correlate with low Al2 O3 , demonstrating that it is a depletion in Al2 O3 that produces the high ratios. This has led to the distinction between Al-depleted komatiites (CaO/Al2 O3 > 1) and Al-undepleted komatiites (CaO/Al2 O3 ∼ 1). Both types are present in the Komati Fm. Progressive metamorphism of the komatiitic rocks decreases CaO contents (Parman et al., 1997). So some of the scatter towards lower CaO/Al2 O3 ratios, especially at the low Al2 O3 contents, is probably due to metamorphism, and is not representative of the magmas. Boninites are one of the few modern magmas that commonly have CaO/Al2 O3 values above 0.86, with values ranging between 0.2 for low-Ca boninites up to 2 for high-Ca boninites. The correlation between CaO/Al2 O3 and Al2 O3 is the same as is seen in the komatiitic series (Fig. 8). In komatiites, the correlation is ascribed to garnet in the source region, which would preferentially sequester Al2 O3 over CaO. Yet this explanation is unlikely to apply to boninites, which are produced at relatively low pressures (1–1.5 GPa), far out of the stability field of garnet in the mantle, especially given the depleted nature of the boninite source region (Crawford et al., 1989).

6. Geochemistry

555

6.2. Trace Elements As one would expect for such mafic magmas, trace element (TE) concentrations in the Komati and Hoogenoeg samples are low (Fig. 9). Chondrite-normalized spidergram patterns are variable, especially in the large-ion lithophile elements (LILE = Th, U, La) and high field-strength elements (HFSE = Nb, Zr, Hf, Ti). The heavy rare earth elements (HREE) show fairly consistent patterns with [Gd/Yb]N (chondrite normalized) values near 1.4. Light rare earth element (LREE) patterns show greater variability, with [La/Sm]N ranging from 0.7 to 1.1 in the komatiites, and up to 2.5 in the basaltic komatiites. Depletions in the HFSE (Ti, Hf, and Zr) are present in both komatiite and basaltic komatiite samples. Nb depletions are not apparent in the komatiites, though they are present in a few of the basaltic komatiites. A few of the basaltic komatiite samples have [Gd/Yb]N between 1 and 1.1, and have slightly U-shaped patterns (Fig. 10). One komatiite sample, B94-4, also has very slightly U-shaped pattern ([La/Sm]N = 1.1; [Gd/Yb]N = 0.9). On top of that, it has peculiar positive Zr and Hf anomalies (Fig. 10). While this is the only sample with such a composition found in the Komati Fm., similar komatiites have been found in a number of other komatiite localities (Jochum et al., 1991), and so we believe its composition to be significant. There are both similarities and differences between the TE systematics of the komatiitic magmas and boninites. Like the Barberton samples, boninites have variable LREE, with [La/Sm]N values both above and below unity, though values > 1 are much more common in boninites. HFSE depletions are also a commonality, though Nb depletions are much more prevalent and more extreme in boninites. Also, some boninites have positive Zr anomalies, like that seen in sample B94-4. The most noticeable difference is in the HREE, where boninites predominantly have [Gd/Yb]N ratios < 1. A diagnostic compositional feature of boninites is the correlation between the HFSE depletions and LREE enrichments (Hickey and Frey, 1982). Samples with high La/Sm have low Ti/Zr ratios, while low La/Sm samples have high, MORB-like ratios (Fig. 11). This is interpreted as mixing between the depleted mantle (low La/Sm) of the boninite source, and the LREE enriched hydrous fluid that fluxes the mantle and causes melting (Hickey and Frey, 1982). Precisely the same pattern is seen in the Komati and Hoogenoeg samples. Like the CaO/Al2 O3 vs. Al2 O3 correlation, crustal contamination of low La/Sm could explain much of the La/Sm vs. Ti/Zr trend in the komatiitic rocks, but not in the boninites. Though in this case, the highest La/Sm basaltic komatiites plot beyond the mixing line between continental crust and komatiite. MORB and OIB have Ti/Zr ratios between 100 and 110, with little correlation to La/Sm (Fig. 11). Thus they show little relationship to either the komatiitic samples or boninites. The Nb-Th systematics of the Barberton samples provides further evidence of link between komatiites and boninites. In the komatiite series, Nb contents are low (0.5–2 ppm). Nb/Th decreases from 13 to 9 with increasing Nb, indicating that Th contents rise faster than Nb concentrations (Fig. 12). The komatiite trend is overlapped and extended to higher Nb (7 ppm) and lower Nb/Th (= 7) by the basaltic komatiites. Three basaltic samples define a lower Nb/Th trend with values between 2 and 3. The same pattern is seen in boninites

556

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Fig. 9. Spidergram plots for Barberton komatiites (top), Barberton basaltic komatiites (middle), and modern boninites (bottom). Data for komatiites from Dann (unpublished data). Boninite data from same references as in Fig. 7. Elements arranged in order of compatibility (Sun and McDonough, 1989a) and normalized to CI chondrite (McDonough and Sun, 1995). The unusual komatiite sample B94-4 is highlighted in grey. Low La/Sm boninite is also highlighted in grey to point out that many boninites are not LREE enriched and do not have U-shaped REE patterns. See text for a further discussion of the data.

6. Geochemistry

557

Fig. 10. Chondrite-normalized spidergram for Barberton basaltic komatiites (filled circles) with boninite-like TE patterns (same normalization as in Fig. 9). Boninite compositions (open circles) shown for comparison. References for samples are: HOL2 (Kerr et al., 1999), 2981 and 46-1 (Hickey and Frey, 1982), 5092 (Jahn and Sun, 1979), S-1C, C-9.8, BC-3-12 (Dann, unpublished data).

(Fig. 12). High-Ca boninites from North Tonga, Cyprus, and Bonin overlap the high Nb/Th komatiite trend and extend it to even higher Nb (20 ppm) and lower Nb/Th (= 6). The low Nb/Th trend is overlapped by both high and low-Ca boninites. The similarity between the komatiitic and boninite samples is remarkable, especially the presence of both the high and low Nb/Th trends.

558

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Fig. 11. Ti/Zr versus La/Sm in Barberton komatiites and basaltic komatiites, and Phanerozoic boninites. Same data and symbols as in Figs. 7 and 8. Both komatiitic magmas and boninites show decreasing Ti/Zr with increased La/Sm. MORB and OIB show fairly constant Ti/Zr of 100–110.

Continental crust plots near the high Nb end of both the komatiite and boninite trends, and so the trends could be the result of contamination. But crustal contamination cannot explain the presence of two separate Nb/Th trends. Also, to produce the basaltic komatiites with > 5 ppm Nb would require > 50% contamination by continental crust (12 ppm Nb), which is not permitted by the high MgO contents of the samples. Estimated slab fluids have a variety of Nb concentrations (0.1–59 ppm, Fig. 12). Nb/Th ratios are less variable (1–5). At the high Nb end of the spectrum, the fluids overlap the composition of continental crust. The fluids with low Nb concentrations could mix with the komatiites to produce the low Nb/Th trend without substantially increasing Nb concentrations. Samples with higher Nb could be produced by mixing with fluids with high Nb such as estimated by Grove et al. (2002). Slab melts would have Nb > 20 and Nb/Th = 10–100 (Rudnick and Fountain, 1995), and so are precluded from playing a major role in the formation of either Barberton komatiitic magmas or modern boninites. The HFSE depletions in komatiites have been attributed to fractionation of majorite garnet (Xie et al., 1993). While majorite/melt HFSE partition coefficients are high (Ohtani et al., 1989), the values for Ti and Zr are not different enough to produce the variability seen in the Ti/Zr ratios (Fig. 11). Majorite fractionation also would not produce two trends in Nb/Th (Fig. 12), and in general it would be a spectacular coincidence that majorite fractionation would produce HFSE systematics that precisely mimic boninite systematics, for which majorite involvement is precluded.

6. Geochemistry

559

Fig. 12. Nb/Th versus Nb in Barberton komatiites and basaltic komatiites, and Phanerozoic boninites. Two distinct trends are seen in both the komatiitic magmas and boninites; one with Nb/Th > 5 and one with Nb/Th < 5. There is no overlap of the komatiitic magmas with MORB or OIB. Same data and symbols as in Figs. 7 and 8. Compositional field for MORB and OIB show by grey region. Boninites shown as grey triangles. Grey bordered boxes show various estimates of hydrous slab fluid TE compositions: A (Ayers, 1998), MG (McCulloch and Gamble, 1991), SN (Stolper and Newman, 1994), G et al. (Grove et al., 2002). Nb composition of SN estimated based on its similarity with Grove et al. (2002) estimate.

6.3. Isotopes Barberton komatiite samples have fairly uniform ε Nd (t) values of +1.9 ± 0.7 (Chavagnac and Bowring, unpublished data). Lower values (−3.5 to +0.2) have been reported for some Barberton basaltic komatiites (Jochum et al., 1991). Similar relationships are seen in many komatiite suites (Jochum et al., 1991). Nd isotopic systematics vary from place to place in boninites. Cape Vogel boninites are quite similar to komatiites in having a fairly limited range of ε Nd (t) (+3.6 to +5.7) over a range of Sm/Nd, whereas Bonin boninites have a wide range of ε Nd (t) values (−8 to +5) that correlate with Sm/Nd (Cameron et al., 1983). Victoria boninites show both trends. This has been interpreted as the result of two slab fluids entering the boninite source (Nelson et al., 1984). One derived from sediments with a very low ε Nd (t) and another from either the basaltic crust or mantle with a slightly positive ε Nd (t). The consistently positive epsilon values in Barberton komatiites suggest little or no involvement of sediments, whereas the basaltic komatiites appear to require some amount of sediment derived fluids. The presence of two slab fluids with different TE compositions may explain the two trends seen in Nb/Th (Fig. 12). At this point we do not have the isotopic composition of the samples in Fig. 12 and so cannot tell if there is any correlation between Nb/Th and ε Nd (t).

560

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

The depleted isotopic signature of the komatiite source region has been used to argue against high H2 O contents on the grounds that any H2 O present in the mantle would be carried away by the melting that produced the LREE depletion (Arndt et al., 1998). Yet, subduction magmas (including boninites) commonly have both high H2 O contents and positive ε Nd (t). The former is produced by the input of hydrous slab fluids just prior to melting. If the fluids have positive ε Nd (t), as is the case in some boninites, they will not affect the Nd isotopes of the komatiite source. And even if the LREE are enriched in the source, the melting occurs too soon after fluid addition (fast enough to preserve U-Th disequilibrium) for radiogenic ingrowth of Nd to lower ε Nd (t). Thus the presence of high H2 O and positive ε Nd (t) is not only consistent with our model, it is required by it.

7. DISCUSSION In light of (1) the enormous major and trace element similarity between the komatiite series and boninites, (2) the petrologic evidence for high magmatic H2 O contents in komatiites, and (3) the structural and field evidence that the Barberton greenstone belt was produced by subduction processes, we propose the Komati and Hoogenoeg magmas were produced in an Archean subduction zone. Modern boninites are produced primarily in fore-arcs, early in the history of a subduction zone (Stern and Bloomer, 1992). Though back-arc boninites are also observed (Falloon and Crawford, 1991). So while a fore-arc setting for the komatiites seems most likely, production in a back-arc is also possible. Modern boninites are produced in rare situations when both high temperatures and high H2 O contents are present at shallow levels in sub-arc mantle. They are volumetrically a minor part of present-day volcanism. Thus we would expect komatiites to make up only a small fraction of Archean magmatism, which is in fact what is observed (de Wit and Ashwal, 1997). This is in contrast to plume theories, where high degree melting of the mantle in superhot plumes would be expected to produce great volumes of komatiite magma, similar to continental flood basalts or oceanic plateaus. Boninites are present in many ophiolites, and so one can consider the KomatiHoogenoeg Fms. as the crustal section of a dismembered Archean ophiolite. Our proposal is separate from that of de Wit, who suggested that the Komati Fm. was part of a complete Steinmann trinity named the Jamestown ophiolite (de Wit et al., 1987). This was based largely on structural interpretation. While the presence or absence of a mantle section in Barberton is beyond the scope and purpose of this paper, we do believe the interpretation of the Komati Fm. as sheeted dikes is incorrect (de Wit et al., 1987; see also de Wit, 2004). Like many previous researchers in Barberton, we interpret the Komati and Hoogenoeg Fms. as a series of subaqueous lava flows with minor contemporaneous, near-surface intrusions. To our knowledge all boninites are erupted subaqueously, and so our physical interpretation is entirely consistent with our chemical interpretation. Being produced by subduction in rather unusual tectonic circumstances, the pertinence of komatiites to global Archean mantle temperature or composition is tenuous. The majority of the TE in komatiites originate in the subducted slab (Grove et al., 2002;

8. Conclusion

561

Stolper and Newman, 1994), while the temperatures at which the komatiites were produced would have largely been controlled by the H2 O content and the pattern of induced mantle corner flow. The disappearance of komatiites from boninite/basaltic komatiite bearing sequences in the early Proterozoic does indicate a secular change in the melting conditions in subduction zones. It is reasonable to attribute this to mantle cooling, whether directly through a decrease in sub-arc mantle temperatures or indirectly through a decrease in subduction rates, a decrease of H2 O input from slabs, or an increase in the strength of the lithosphere. In this context, late Proterozoic ophiolites such as the in the Cape Smith fold belt (Hynes and Francis, 1982) are particularly interesting to the study of secular change, and may record an intermediate stage between Archean komatiite magmatism and Phanerozoic boninite magmatism. The question of whether crustal contamination or slab fluids produced the TE trends in the komatiitic magmas is unresolved. This is not surprising as distinguishing between the two processes is a long standing problem in the geochemistry of modern subduction magmas. Why should it be any different in the Archean? The problem is that the continental crust is produced in subduction zones, and so its TE composition is that of subduction fluids; high LILE, low HFSE. In addition, sediments from the crust are subducted with the downgoing slab and so the similarity between continental crust and slab fluids is further reinforced. The coincidence of the boninite and komatiite trends in Ti/Zr, La/Sm, Nb/Th, and the high MgO contents of the magmas argues intuitively against crustal contamination, as it would seem an amazing coincidence that crustal contamination of plume magmas in the Archean would produce the exact same TE systematics as hydrous melting of depleted mantle in the Phanerozoic. Admittedly, this is a circumstantial argument. Perhaps the best evidence is the presence of low Nb, low Nb/Th samples. More data on the basaltic komatiites is needed to constrain this line of reasoning. 8. CONCLUSION (1) The major element composition of preserved augites indicates high magmatic H2 O contents in Barberton komatiites. (2) Komatiites and basaltic komatiites in the Komati-Hoogenoeg Fms. display remarkably similar geochemical systematics to modern boninites. (3) The source region of modern boninites is more depleted than the komatiitic magmas. (4) Komati-Hoogenoeg magmas were produced by hydrous melting in an Archean subduction zone. (5) The Komati-Hoogenoeg Fms. are the crustal section of a dismembered Archean ophiolite. REFERENCES Allegre, C.J., 1982. Genesis of Archaean komatiites in a wet ultramafic subducted plate. In: Arndt, N.T., Nisbet, E.G. (Eds.), Komatiites, pp. 495–500.

562

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Arndt, N., Brooks, C., 1980. Komatiites. Geology 8 (3), 155–156. Arndt, N., Ginibre, C., Chauvel, C., Albarede, F., Cheadle, M., Herzberg, C., Jenner, G., Lahaye, Y., 1998. Were komatiites wet? Geology 26 (8), 739–742. Arndt, N.T., Kerr, A.C., Tarney, J., 1997. Dynamic melting in plume heads: The formation of Gorgona komatiites and basalts. Earth and Planetary Science Letters 146 (1–2), 289–301. Asahara, Y., Ohtani, E., Suzuki, A., 1998. Melting relations of hydrous and dry mantle compositions and the genesis of komatiites. Geophysical Research Letters 25 (12), 2201–2204. Ayers, J., 1998. Trace element modeling of aqueous fluid; peridotite interaction in the mantle wedge of subduction zones. Contributions to Mineralogy and Petrology 132 (4), 390–404. 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 (3), 361–377. Brandl, G., de Wit, M.J., 1997. The Kaapvaal Craton, South Africa. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 581– 607. Brooks, C., Hart, S.R., 1974. On the Significance of Komatiite. Geology 2 (2), 107–110. Brown, A.V., Jenner, G.A., 1989. Geological setting, petrology and chemistry of Cambrian boninite and low-Ti tholeiite lavas in western Tasmania. In: Crawford, A.J. (Ed.), Boninites, pp. 232–263. Cameron, W.E., 1989. Contrasting boninite-tholeiite associations from New Caledonia. In: Crawford, A.J. (Ed.), Boninites, pp. 314–338. Cameron, W.E., McCulloch, M.T., Walker, D.A., 1983. Boninite petrogenesis; chemical and Nd-Sr isotopic constraints. Earth and Planetary Science Letters 65 (1), 75–89. Cameron, W.E., Nisbet, E.G., Dietrich, V.J., 1979. Boninites, komatiites and ophiolitic basalts. Nature 280 (5723), 550–553. Cloete, M., 1994. Aspects of volcanism and metamorphism of the Onverwacht Group lavas in the south-western portion of the Barberton Greenstone Belt. Ph.D. thesis. University of Witswatersrand, Johannesburg, South Africa. Crawford, A.J., Falloon, T.J., Green, D.H., 1989. Classification, petrogenesis and tectonic setting of boninites. In: Crawford, A.J. (Ed.), Boninites, pp. 1–49. Crawford, A.J. (Ed.), 1989. Boninites, p. 465. Dann, J.C., 2000. The 3.5 Ga Komati Formation, Barberton Greenstone Belt, South Africa, Part I: New maps and magmatic architecture. South African Journal of Geology 103 (1), 47–68. Dann, J.C., 2001. Vesicular komatiites, 3.5-Ga Komati Formation, Barberton Greenstone Belt, South Africa: inflation of submarine lavas and origin of spinifex zones. Bulletin of Volcanology 63 (7), 462–481. de Wit, M.J., 1982. Gliding and Overthrust Nappe Tectonics in the Barberton-Greenstone Belt. Journal of Structural Geology 4 (2), 117. de Wit, M.J., 2004. Archean Greenstone Belts do contain fragments of ophiolites. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 599–613. de Wit, M.J., Ashwal, L.D. (Eds.), 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, p. 809. de Wit, M.J., Hart, R.A., Hart, R.J., 1987. The Jamestown Ophiolite Complex, Barberton Mountain Belta Section through 3.5 Ga Oceanic-Crust. Journal of African Earth Sciences 6 (5), 681–730. Donaldson, C.H., 1974. Olivine Crystal Types in Harrisitic Rocks of the Rhum Pluton and Archean Spinifex Rocks. Geological Society of America Bulletin 85 (11), 1721–1726. Falloon, T.J., Crawford, A.J., 1991. The Petrogenesis of High-Calcium Boninite Lavas Dredged from the Northern Tonga Ridge. Earth and Planetary Science Letters 102 (3–4), 375–394.

References

563

Falloon, T.J., Green, D.H., McCulloch, M.T., 1989. Petrogenesis of high-Mg and associated lavas from the North Tonga Trench. In: Crawford, A.J. (Ed.), Boninites, pp. 357–395. Green, D.H., 1972. Archaean greenstone belts may include terrestrial equivalents of lunar maria? Earth and Planetary Science Letters 15 (3), 263–270. Grove, T.L., 1993. Corrections to expressions for calculating mineral components in Origin of calcalkaline series lavas at Medicine Lake Volcano by fractionation, assimilation and mixing and Experimental petrology of normal MORB near the Kane fracture zone; 22 degrees–25 degrees N, Mid-Atlantic Ridge; discussion. Contributions to Mineralogy and Petrology 114 (3), 422–424. Grove, T.L., Parman, S.W., Bowrig, S.W., Price, R.C., Baker, M.B., 2002. The role of an H2 O-rich fluid component in the generation of primitive basaltic andesites and andesites from the Mt. Shasta region, N California. Contributions to Mineralogy and Petrology 142 (4), 375–396. Hart, S.R., Dunn, T., 1993. Experimental cpx/melt partitioning of 24 trace elements. Contributions to Mineralogy and Petrology 113 (1), 1–8. Herzberg, C., 1992. Depth and Degree of Melting of Komatiites. Journal of Geophysical Research— Solid Earth 97 (B4), 4521–4540. Herzberg, C., 1995. Generation of Plume Magmas through Time—an Experimental Perspective. Chemical Geology 126 (1), 1–16. Hickey, R.L., Frey, F.A., 1982. Geochemical characteristics of boninite series volcanics implications for their source. Geochimica et Cosmochimica Acta 46 (11), 2099–2116. Hollings, P., Wyman, D., Kerrich, R., 1999. Komatiite-basalt-rhyolite volcanic associations in northern Superior Province greenstone belts; significance of plume-arc interaction in the generation of the proto continental Superior Province. In: Condie, K.C., Abbott, D.H. (Eds.), Oceanic Plateaus and Hotspot Islands; Identification and Role in Continental Growth. Submarine Plateaus and Hotspot Islands Young and Old; Identification and Role in Continental Growth 46 (1), 137–161. Hynes, A., Francis, D., 1982. Komatiitic basalts of the Cape Smith foldbelt, New Quebec, Canada. In: Arndt, N.T., Nisbet, E.G. (Eds.), Komatiites, pp. 159–170. Jahn, B.M., Sun, S.S., 1979. Trace element distribution and isotopic composition of Archean greenstones. In: Ahrens, L.H. (Ed.), Origin and Distribution of the Elements. Second Symposium on the Origin and Distribution of the Elements (11), pp. 597–618. Jochum, K.P., Arndt, N.T., Hofmann, A.W., 1991. Nb-Th-La in Komatiites and Basalts—Constraints on Komatiite Petrogenesis and Mantle Evolution. Earth and Planetary Science Letters 107 (2), 272–289. Kawamoto, T., Hervig, R.L., Holloway, J.R., 1996. Experimental evidence for a hydrous transition zone in the early Earth’s mantle. Earth and Planetary Science Letters 142 (3–4), 587–592. Kerr, A.C., Iturralde-Vinent, M.A., Saunders, A.D., Babbs, T.L., Tarney, J., 1999. A new plate tectonic model of the Caribbean; implications from a geochemical reconnaissance of Cuban Mesozoic volcanic rocks. Geological Society of America Bulletin 111 (11), 1581–1599. Kerrich, R., Wyman, D., Fan, J., Bleeker, W., 1998. Boninite series; low Ti-tholeiite associations from the 2.7 Ga Abitibi greenstone belt. Earth and Planetary Science Letters 164 (1–2), 303–316. Lahaye, Y., Arndt, N., Byerly, G., Chauvel, C., Fourcade, S., Gruau, G., 1995. The Influence of Alteration on the Trace-Element and Nd Isotopic Compositions of Komatiites. Chemical Geology 126 (1), 43–64. Lecuyer, C., Gruau, G., Anhaeusser, C.R., Fourcade, S., 1994. The Origin of Fluids and the Effects of Metamorphism on the Primary Chemical-Compositions of Barberton Komatiites-New Evidence from Geochemical (Ree) and Isotopic (Nd, O, H, Ar-39/Ar-40) Data. Geochimica et Cosmochimica Acta 58 (2), 969–984.

564

Chapter 16: Petrology and Geochemistry of Barberton Komatiites and Basaltic Komatiites

Lopezmartinez, M., York, D., Hanes, J.A., 1992. A Ar40 /Ar39 Geochronological Study of Komatiites and Komatiitic Basalts from the Lower Onverwacht Volcanics-Barberton Mountain Land, SouthAfrica. Precambrian Research 57 (1–2), 91–119. Lowe, D.R., 1999a. Geologic evolution of the Barberton Greenstone Belt and Vicinity. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic Evolution of the Barberton Greenstone Belt, South Africa. Geological Society of America Special Paper 329, 287–312. Lowe, D.R., 1999b. Petrology and sedimentology of cherts and related silicified sedimentary rocks in the Swaziland Supergroup. In: Lowe, D.R., Byerly, G.R. (Eds.), Geologic Evolution of the Barberton Greenstone Belt, South Africa. Geological Society of America Special Paper 329, 83– 114. Lowe, D.R., Byerly, G.R. (Eds.), 1999. Geologic Evolution of the Barberton Greenstone Belt, South Africa. Geological Society of America Special Paper 329, 319. McCulloch, M.T., Gamble, A.J., 1991. Geochemical and geodynamical constraints on subduction zone magmatism. Earth and Planetary Science Letters 102 (3–4), 358–374. McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. In: McDonough, W.F., Arndt, N.T., Shirey, S. (Eds.), IAVCEI General Assembly Symposium on Chemical Evolution of the Mantle. Chemical Evolution of the Mantle 120 (3–4), 223–253. Mysen, B.O., Boettcher, A.L., 1975. Melting of a hydrous mantle; I. Phase relations of natural peridotite at high pressures and temperatures with controlled activities of water, carbon dioxide, and hydrogen. Journal of Petrology 16 (3), 520–548. Nelson, D.R., Crawford, A.J., McCulloch, M.T., 1984. Nd-Sr isotopic and geochemical systematics in Cambrian boninites and tholeiites from Victoria, Australia. Contributions to Mineralogy and Petrology 88 (1–2), 164–172. Nisbet, E.G., Cheadle, M.J., Arndt, N.T., Bickle, M.J., 1993. Constraining the potential temperature of the Archaean mantle; a review of the evidence from komatiites. In: Campbell, I.H., Maruyama, S., McCulloch, M.T. (Eds.). The Evolving Earth 30 (3–4), 291–307. Nisbet, E.G., Walker, D., 1982. Komatiites and the structure of the Archaean mantle. Earth and Planetary Science Letters 60 (1), 105–113. Ohtani, E., Kawabe, I., Moriyama, J., Nagata, Y., 1989. Partitioning of elements between majorite garnet and melt and implications for petrogenesis of komatiite. Contributions to Mineralogy and Petrology 103 (3), 263–269. Parman, S.W., Dann, J.C., Grove, T.L., de Wit, M.J., 1997. Emplacement conditions of komatiite magmas from the 3.49 Ga Komati Formation, Barberton Greenstone Belt, South Africa. Earth and Planetary Science Letters 150 (3–4), 303–323. Parman, S.W., Grove, T.L., Dann, J.C., 2001. The production of Barberton komatiites in an Archean subduction zone. Geophysical Research Letters 28 (13), 2513–2516. Parman, S.W., Shimizu, N., Grove, T.L., Dann, J.C., 2003. Constraints on the pre-metamorphic trace element composition of Barberton komatiites from ion probe analyses of preserved clinopyroxene. Contributions to Mineralogy and Petrology 144, 383–396. Redman, B.A., Keays, R.R., 1985. Archean basic volcanism in the Eastern Goldfields Province, Yilgarn Block, Western Australia. Precambrian Research 30 (2), 113–152. Rudnick, R.L., Fountain, D.M., 1995. Nature and composition of the continental crust; a lower crustal perspective. Reviews of Geophysics 33 (3), 267–309. Schmitz, M., 2001. Ph.D. thesis. Massachusetts Institute of Technology, Cambridge. Smith, H.S., Erlank, A.J., 1982. Geochemistry and petrogenesis of komatiites from the Barberton greenstone belt, South Africa. In: Arndt, N.T., Nisbet, E.G. (Eds.), Komatiites, pp. 347–397.

References

565

Smith, H.S., Erlank, A.J., 1980. Geochemistry of some ultramafic komatiite lava flows from the Barberton Mountain Land, South Africa. In: Duncan, A.R., Gorton, M.P. (Eds.), Archean Geochemistry. Archean Geochemistry Field Conference 11 (3–4), 399–415. 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 (1–2), 1621– 1636. Stolper, E., Newman, S., 1994. The role of water in the petrogenesis of Mariana Trough magmas. Earth and Planetary Science Letters 121 (3–4), 293–325. Stone, W.E., Deloule, E., Larson, M.S., Lesher, C.M., 1997. Evidence for hydrous high-MgO melts in the Precambrian. Geology 25 (2), 143–146. Sun, S.S., McDonough, W.F., 1989a. Chemical and isotopic systematics of oceanic basalts; implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geological Society Special Publication 42, 313–345. Sun, S.S., Nesbitt, R.W., McCulloch, M.T., 1989b. Geochemistry and petrogenesis of Archaean and early Proterozoic siliceous high-magnesian basalts. In: Crawford, A.J. (Ed.), Boninites, pp. 148– 173. Upadhyay, H.D., 1982. Ordovician komatiites and associated boninite-type magnesian lavas from Betts Cove, Newfoundland. In: Arndt, N.T., Nisbet, E.G. (Eds.), Komatiites, pp. 187–198. Viljoen, M.J., Viljoen, R.P., 1969a. Evidence for the existence of a mobile extrusive peridotitic magma from the Komati Formation of the Onverwacht Group, Upper Mantle Project. Geological Society of South Africa Special Publication 2, 87–112. Viljoen, M.J., Viljoen, R.P., 1969b. Evidence for the existence of a mobile extrusive peridotitic magma from the Komati Formation of the Onverwacht Group. Geological Society of South Africa Special Publication 2, 55–86. Viljoen, M.J., Viljoen, R.P., 1969c. The geology and geochemistry of the Lower Ultramafic Unit of the Onverwacht Group and a proposed new class of igneous rocks. Upper Mantle Project. Geological Society of South Africa Special Publication 2, 55–85. Walter, M.J., 1998. Melting of garnet peridotite and the origin of komatiite and depleted lithosphere. Journal of Petrology 39 (1), 29–60. Williams, D.A.C., Furnell, R.G., 1979. Reassessment of part of the Barberton type area, South Africa. Precambrian Research 9 (3–4), 325–347. Wilson, A.H., Versfeld, J.A., 1994. The early Archaean Nondweni greenstone belt, southern Kaapvaal Craton, South Africa; Part II. Characteristics of the volcanic rocks and constraints on magma genesis. Precambrian Research 67 (3–4), 277–320. Xie, Q., Kerrich, R., Fan, J., 1993. HFSE/REE fractionations recorded in three komatiite-basalt sequences, Archean Abitibi greenstone belt; implications for multiple plume sources and depths. Geochimica et Cosmochimica Acta 57 (16), 4111–4118. Xie, Q., McCuaig, T.C., Kerrich, R., 1995. Secular trends in the melting depths of mantle plumes; evidence from HFSE/REE systematics of Archean high-Mg lavas and modern oceanic basalts. In: Ludden, J.N., Arndt, N.T. (Eds.), Mafic Magmatism through Time. Evolution of Mafic Magmatism through Time 126 (1), 29–42.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

567

Chapter 17

PRECAMBRIAN ARC ASSOCIATIONS: BONINITES, ADAKITES, MAGNESIAN ANDESITES, AND NB-ENRICHED BASALTS A. POLATa,1 AND R. KERRICHb a Max-Planck-Institut für b Department of

Chemie, Postfach 3060, D-55020 Mainz, Germany Geological Sciences, University of Saskatchewan, SK, Canada S7N 5E2

Boninitic lavas have recently been reported from several Precambrian terranes, including the ∼ 3.8 Ga Isua terrane of West Greenland; 2.8 Ga Opatica and the 2.7 Ga Abitibi terranes of the Superior Province; the 2.8 Ga North Karelian terrane of the Baltic Shield; and the 1.9 Ga Flin Flon terrane in the Trans-Hudson orogen. In the Isua belt, boninitic flows coexist with pillow basalts and picrites. Boninitic lavas, and low-Ti tholeiitic basalts, outcrop over a 300 km corridor in the Abitibi volcanic-plutonic subprovince. They are intercalated with a stratigraphically lower ocean plateau association of komatiites and basalts, and an upper volcanic arc association of tholeiitic to calc-alkaline arc basalts; accordingly there was contemporaneous eruption of neighbouring plume and arc magmas. The 2.8 Ga Opatica boninitic lavas are spatially and temporarily associated with arc-type volcanic rocks. The 2.8 Ga Baltic Shield boninitic rocks are related to a supra-subduction ophiolite complex. All of these Precambrian boninitic lavas share the low-TiO2, high Al2 O3 /TiO2 ratios, U-shaped REE patterns, and negative Nb but positive Zr anomalies of Phanerozoic counterparts; however, SiO2 contents are variable. Boninites of Phanerozoic age occur in ophiolites or intra-oceanic island arcs, such as the Izu-Bonin-Mariana arc system. These primary liquids are interpreted as second-stage high-temperature, low-pressure melting of a depleted refractory mantle wedge fertilized by fluids and/or melts, above a subduction zone. Precambrian boninitic lavas are likely products of the same conjunction of processes. Low-Ti tholeiites lack the LREE enrichment coupled with negative Nb anomalies of the boninites. They had a similar depleted wedge source, but without a subduction zone component. An association of adakites, magnesian andesites (MA), and Nb-enriched basalts (NEB) with “normal” tholeiitic to calc-alkaline basalts and andesites has recently been described from the 2.7 Ga Wawa and Confederation volcanic-plutonic terranes of the Superior Province. Cenozoic adakites are considered to form by slab melting; MA the product of hybridization of adakite liquids with the peridotitic mantle wedge; and NEB melting of the residue of the MA wedge source. This volcanic association is found in Cenozoic arcs characterized by shallow subduction of young, hot oceanic lithosphere. Archean equivalents likely formed under comparable tectonic settings. 1 Present address: Department of Earth Sciences, University of Windsor, Windsor, Ontario, Canada N9B 3P4.

DOI: 10.1016/S0166-2635(04)13017-X

568

Chapter 17: Precambrian Arc Associations

U-shaped REE patterns in conjunction with positive Zr anomalies of Archean and Phanerozoic boninites can be modeled by a depleted peridotitic wedge fertilized by adakite liquids and/or hydrous fluids in a convergent margin. Consequently, Phanerozoic type arcs were operating in Archean convergent margins. Imbrication of komatiite-basalt ocean plateau volcanic sequences with arcs solves the apparent Mg#, Ni deficit of some models for Archean upper continental crust. Higher geothermal gradients in Archean subduction zones may have played an important role for the growth of continental crust.

1. INTRODUCTION AND SCOPE Recent geochemical investigations of Precambrian orogenic belts have documented a great compositional diversity in subduction-related volcanic rocks. Together with “normal” bimodal tholeiitic to calc-alkaline suites, there are boninites and the association of adakites, magnesian andesites, and Nb-enriched basalts (Kerrich et al., 1998; Wyman, 1999a, 1999b; Polat and Kerrich, 2000; Polat et al., 2002; Boily and Dion, 2002; Shchipansky et al., 2004). The term “boninite” refers to a large variety of primary or near-primary magmas, with a wide variation of CaO/Al2 O3 (< 0.55 to > 0.75), defined by Crawford et al. (1989) as SiO2 > 53 wt% and Mg# > 60. Many Phanerozoic boninites are closely associated with ophiolites and intra-oceanic island arcs (Sun and Nesbitt, 1978; Taylor and Nesbitt, 1988; Smellie et al., 1995; Meffre et al., 1996; Bédard, 1999). Accordingly, recognition of Precambrian boninites has important implications for Precambrian geodynamic processes. The classification of Tertiary boninitic rocks is based mainly on major elements, such as CaO, K2 O, Na2 O, Al2 O3 , and SiO2 , and mineralogical and textural features (Cameron et al., 1979; Crawford et al., 1989). Given the fact that CaO, K2 O, Na2 O, and SiO2 are likely to be mobile during hydrothermal alteration and metamorphism, we use the term “boninitic” for older, Precambrian, metamorphosed lavas which share low TiO2 and Zr contents, high Al2 O3 /TiO2 ratios, and U-shaped REE patterns of fresh boninites (Coish, 1989; Brown and Jenner, 1989; Poidevin, 1994; Stern et al., 1995; Kerrich et al., 1998; Bédard, 1999; Boily and Dion, 2002), yet have lower SiO2 contents than in the boninite definition of Crawford et al. (1989). Studies of Tertiary boninites suggest that they are products of high-temperature (1000– 1300 ◦ C), low-pressure (< 10 kbar) second stage partial melting of a hydrous, depleted refractory mantle source above subducted oceanic lithosphere at convergent plate boundaries, such as the Izu-Bonin-Mariana subduction zones of the western Pacific (Crawford et al., 1989; Tatsumi and Maruyama, 1989; Stern et al., 1991; Pearce et al., 1992). Melting results from the combined effects of solidus depression by the presence of subduction-derived fluids and/or melts and steep geothermal gradients. Rocks with boninitic geochemical characteristics have also been reported from the Stillwater and Bushveld intra-continental rift intrusions (Crawford et al., 1989; Hatton and Sharpe, 1989). The occurrence of the adakite and magnesian andesite (MA) association sometimes with Nb-enriched basalts (NEB) has been described from certain Cenozoic arcs of “nor-

1. Introduction and Scope

569

mal” tholeiitic to calc-alkaline basalts, featuring subduction of young and/or hot oceanic lithosphere (Defant et al., 1992; Sajona et al., 1993; Defant and Kepezhinskas, 2001). Adakites are Al- and Na-enriched, intermediate to felsic calc-alkaline volcanic rocks having high Sr/Y and La/Ybcn ratios (Drummond et al., 1996; Martin, 1999). Magnesian andesites are primitive, intermediate calc-alkaline volcanic rocks, having high Mg#, and Cr and Ni contents (Kelemen, 1995; Sajona et al., 1996). Niobium-enriched basalts are mafic volcanic arc rocks possessing high Nb concentrations (> 7 ppm) and high Nb/Thpm and Nb/Lapm ratios (Defant et al., 1992; Sajona et al., 1993; Kepezhinskas et al., 1996). Both geochemical systematics and experimental petrology are consistent with adakite melts being generated by partial melting of subducted oceanic crust transformed into garnet-bearing amphibolite or eclogite (Martin, 1999; Defant and Drummond, 1990). Magnesian andesites are interpreted to be the product of hybridization of adakite liquids and/or subducted sediment with arc mantle wedge peridotite (Defant et al., 1992; Yogodzinski et al., 1995; Kelemen, 1995). The origin of NEB is a matter of debate; present theories suggest that they may be partial melts of phlogopite- and/or amphibolite-rich residual mantle previously metasomatized by adakitic melts in the generation of MA (Defant et al., 1992; Kepezhinskas et al., 1996), consistent with the association of these three distinctive volcanic types in space and time in Cenozoic arcs characterized by shallow, generally oblique subduction of young, hot ocean lithosphere. The association of adakite, MA, and NEB flows with tholeiitic to calc-alkaline basaltsandesites in intra-oceanic arcs has recently been recognized in two 2.7 Ga greenstone belts of the Superior Province (Hollings and Kerrich, 2000; Polat and Kerrich, 2001a). Boily and Dion (2002) documented the presence of temporarily and spatially associated boninitic and adakitic volcanic rocks in the 2.8 Ga Frotet-Evans greenstone belt of the Superior Province. Puchtel et al. (1999) and Shchipansky et al. (2004) reported adakite occurrences from the bimodal arc sequences of the 2.9 Ga Sumozero-Kenozero and 2.8 Ga North Karelian greenstone belts, respectively, in the Baltic Shield. Given the specific tectonic regime in which the Cenozoic association develops, the geochemical characteristics of Archean counterparts are of particular interest for understanding the early evolution of the crust-mantle system, and Archean subduction zone geodynamic processes (Polat and Kerrich, 2001a; Wyman et al., 2002). There are two unconstrained problems regarding the origin of these compositionally distinctive but spatially and temporarily associated Archean volcanic rocks. These are: (1) petrogenetic links between various rock types, and (2) the nature of Archean geodynamic processes that produced these rocks. In the Wawa subprovince, these volumetrically minor, syn-volcanic adakites, MA, NEB, and “normal” arc basalts to andesites predate the voluminous, syn- to late-tectonic tonalite-trondhjemitegranodiorite (TTG) batholiths that intrude all greenstone belt lithologies (Polat et al., 1998; Polat and Kerrich, 2001a). This paper reviews the field and geochemical characteristics of recently recognized Precambrian boninites, adakites, MA, NEB, and related arc volcanic rocks. On the basis of these new data, we attempt to evaluate the significance of these rocks in terms of Precambrian geodynamic, petrogenetic, and crustal growth processes.

570

Chapter 17: Precambrian Arc Associations

Fig. 1. Geological map of the eastern sector of the Isua greenstone belt. NWD: Northwest Tectonic Domain; CD: Central Tectonic Domain; SED: Southeast Tectonic Domain. Modified after Appel et al. (1998). Open circles show the locations of boninite samples.

2. EARLY ARCHEAN ISUA BONINITIC VOLCANIC ROCKS 2.1. Geological Setting The ∼ 3.8 Ga Isua greenstone belt is located in the Nuuk region of southwest Greenland (Fig. 1; Rosing et al., 1996; Appel et al., 1998). The region is dominated by the early

2. Early Archean Isua Boninitic Volcanic Rocks

571

Fig. 2. Boninitic pillow basalt outcrop in the Isua greenstone belt.

Archean Itsaq Gneiss Complex (Nutman et al., 1996). These orthogneisses were derived from tonalitic, granodioritic, and granitic precursors, and were intruded between 3.65 and 3.81 Ga (Nutman et al., 1996). The Isua greenstone belt contains the largest early Archean supracrustal units in the world. It is 35 km long and up to 4 km wide. The supracrustal units are characterized by lithotectonic successions of volcanic and sedimentary rocks, which are variably metamorphosed, metasomatised, and deformed (Gruau et al., 1996; Rosing et al., 1996; Appel et al., 1998). From recent detailed mapping the belt has been interpreted as a stack of fault-bounded, low- to high-strain lithotectonic domains (Komiya et al., 1999; Myers, 2001). These studies also revealed a complex history of poly-phase deformation and high-grade metamorphism. The low-strain domains contain well-preserved volcanic and sedimentary features, including basaltic pillow lavas, pillow breccia, heterogeneous volcanic breccia, and polymictic conglomerates (Fig. 2; Fedo, 2000; Myers, 2001). In contrast, primary volcanic and sedimentary structures in the high-strain domains are intensely deformed and transposed throughout the belt. In the low-strain domains, the variably deformed pillow basalts are intercalated with ultramafic units. The original stratigraphic relations between mafic (pillow basalts) and ultramafic units have been disrupted, and no primary volcanic textures such as spinifex have so far been recognized in ultramafic units. The sedimentary units consist mainly of banded iron formations, cherts, siliciclastic turbidites, and conglomerates (Fedo, 2000). Most supracrustal units were strongly deformed, then recrystallized under upper greenschist to amphibolite facies metamorphism. Mafic dykes cut the early schistosity and lineation (Appel et al., 1998). The northeastern part of the belt has been informally divided into three lithotectonic domains: Northwestern (NWD), Central (CD) and Southeastern (SED) (Fig. 1; Appel et al.,

572

Chapter 17: Precambrian Arc Associations

Fig. 3. Photomicrograph of the Isua boninites (a), Wawa adakites (b), magnesian andesites (c), and Nb-enriched basalts (d).

1998; Myers, 2001). Each domain is characterized by different lithologic associations and intensities of deformation. The Northwestern and Southeastern Tectonic Domains are composed of linear belts of intensely deformed garnet amphibolites of probable sedimentary origin, banded iron formations, cherts, ultramafic rocks, and quartzo-feldspathic schist. The Central Tectonic Domain consists of relatively less-deformed metavolcanic and metasedimentary rocks (Fig. 2). Quartzo-feldspathic schists of the Northwestern Tectonic Domain yield a U-Pb zircon age of about 3710 Ma (Nutman et al., 1997). Garnet-mica schists from the Northwestern Tectonic Domain yield an age of 3742 ± 49 Ma by Sm-Nd whole-rock isochron (Kamber et al., 1998). Volcanic rocks from the Central Tectonic Domain are composed predominantly of amphibolites, consisting mainly of tremolite + actinolite + hornblende + plagioclase + chlorite + talc schist (Fig. 3a). In addition, minor units of garnet + biotite + hornblende + carbonate are exposed in the domain. Volcanic rocks from the Northwestern and Southeastern Tectonic Domains are composed mainly of hornblende + garnet + biotite + carbonate schist, and chlorite and talc schists. Details of mineral assemblages were given in Myers (2001).

2. Early Archean Isua Boninitic Volcanic Rocks

573

Table 1. Summary of significant compositional and element ratios for Archean boninites, adakites, magnesian andesites (MA), basalts, andesites, and Nb-enriched basalts and andesites (NEBA)∗ Isua boninites SiO2 (wt%) 47–54 MgO 6.8–18.0 0.20–0.40 TiO2 8.2–11.9 Fe2 O3 Mg-number 0.60–0.80

Abitibi boninites 44–60 7.4–24.0 0.14–0.31 6.25–13.6 0.51–0.83

Wawa adakites 64–69 1.3–2.2 0.29–0.80 2.5–6.7 0.39–0.58

Wawa MA 56–64 3.4–6.4 0.47–1.83 5.7–10.5 0.50–0.64

Wawa basalts 46–54 3.6–6.4 0.6–2.1 8.6–18.3 0.35–0.55

Wawa andesites 57–63 1.5–2.9 0.5–1.7 5.6–14.9 0.29–0.52

Wawa NEBA 50–57 3.8–5.7 0.63–2.24 8.2–16.6 0.40–0.61

Cr (ppm) Ni Sc Zr Nb Th La Y Yb

60–1920 60–645 26–49 11.5–30.5 0.13–1.1 0.043–0.320 0.33–1.86 6.0–13.7 0.94–1.84

270–2300 110–930 32–71 9.7–42 0.27–1.70 0.03–0.26 0.34–1.59 7.4–24 1.13–3.62

11–166 8.0–49.0 4.3–18.7 109–153 2.17–6.53 2.40–7.90 15.3–51.5 6.4–15.3 0.44–1.17

106–531 21–229 12.3–26.6 81–221 2.6–12.9 1.08–6.90 9.6–61.6 8.8–21.6 0.72–1.81

14–362 27–72 17–50 62–138 2.7–5.2 0.56–3.51 6.7–24.6 13.3–35.0 1.0–3.8

9–165 4.0–81 10.7–49.2 99–165 3.9–6.9 1.00–4.60 8.3–44.6 9.5–31.4 0.81–3.35

14–221 8.5–90.1 24.2–44.6 110–278 7.3–16.2 1.42–3.83 11.9–37.2 18.4–59.4 1.20–5.82

(La/Sm)cn (La/Yb)cn (Gd/Yb)cn Zr/Y Ti/Zr (Zr/Sm)pm Al2 O3 /TiO2 (Nb/La)pm (Nb/Th)pm

0.55–1.39 0.16–0.79 0.26–0.64 1.28–2.33 72–116 1.1–1.7 45–95 0.23–0.76 0.23–1.200

0.72–1.2 0.20–0.60 0.29–0.62 0.94–2.2 71–117 0.88–1.49 47–104 0.53–1.12 0.82–1.39

3.4–4.8 20.5–42.9 2.3–4.6 9.3–21.7 13.9–36.9 0.71–1.79 21–61 0.12–0.27 0.10–0.22

2.34–3.83 5.26–27.9 1.50–4.87 5.86–11.68 26–135 0.37–1.81 9.2–30.1 0.06–0.53 0.09–0.69

1.37–2.93 1.88–17.31 1.28–3.07 3.2–7.9 37–92 0.56–1.63 6.1–25.1 0.14–0.51 0.13–0.86

1.54–3.92 2.70–22.88 1.14–3.32 3.43–12.18 20–96 0.77–1.94 8.5–30.1 0.14–0.61 0.15–0.67

1.42–2.79 2.65–18.19 1.25–4.05 3.76–8.91 22.4–100.2 0.64–1.46 6.6–24.1 0.23–0.67 0.39–1.03

∗ Data source as in Fig. 8.

2.2. Geochemistry Different lithotectonic domains of the Isua greenstone belt have distinct geochemical characteristics. This review deals only with the geochemistry of the Central Tectonic Domain where boninitic volcanic rocks have been identified (Fig. 1; Polat et al., 2002). The least altered metavolcanic amphibolites are characterized by high Mg# (0.60–0.80), MgO (7–18 wt%), Al2 O3 (14–20 wt%), Ni (60–645 ppm), and Cr (60–1920 ppm) contents, but low TiO2 (0.20–0.40 wt%), Zr (12–30 ppm), Y (6–14 ppm), and REE concentrations (Table 1). Collectively, these compositional features represent a coherent mafic to ultramafic suite. Chondrite-normalized REE patterns are concave upward (Fig. 4a). On primitive mantle-normalized trace element diagrams, they are characterized by relative depletion of Nb, but enrichment of Zr, relative to neighboring REE (Fig. 4c). Polat et al. (2002) showed that deformation, metamorphic alteration, and crustal contamination can be ruled

574

Chapter 17: Precambrian Arc Associations

Fig. 4. (a, b) Chondrite-normalized REE diagrams for the Isua metavolcanic rocks and Phanerozoic boninites. (c, d) Primitive mantle-normalized trace element diagrams for the Isua metavolcanic rocks and Phanerozoic boninites. Modified after Polat et al. (2002). Chondrite normalization values from Sun and McDonough (1989) and primitive mantle normalization values from Hofmann (1988).

out as the cause of the distinct and coherent chemical composition. These geochemical characteristics are comparable in most respects to those of Phanerozoic boninites. 2.3. Origin of the Isua Boninites It is generally believed that Phanerozoic boninitic magmas are produced by a two-stage process: extreme depletion of a mantle source by multiple melt extraction, followed by second-stage melting induced by enriched subduction components (Sun and Nesbitt, 1978; Cameron et al., 1983; Stern et al., 1991; Taylor et al., 1994). Accordingly, a two-component

2. Early Archean Isua Boninitic Volcanic Rocks

575

Fig. 4. (Continued.)

mixing model has been adopted to explain the origin of the ∼ 3.8 Ga Isua boninitic metavolcanic rocks (Polat et al., 2002). The depleted component was estimated by extrapolation of the HREE patterns of the Isua metavolcanic rocks, assuming monotonically decreasing chondrite- and primitive mantle-normalized LREE, Ti, Zr, Nb and Th abundances (see Taylor and Nesbitt, 1988). The enriched component was calculated from the average composition of ∼ 3.8 Ga tonalites of the southern Isua region (see Nutman et al., 1999). Trace element patterns produced by the mixing calculations are comparable to those of the Isua metavolcanic rocks and of Phanerozoic boninites (Fig. 5). We do not propose that the melts of the ∼ 3.8 Ga tonalites of the southern Isua region were involved directly in the petrogenesis of the Isua boninitic volcanic rocks. Rather, we assume that such tonalitic (adakite-like) melts originated in early Archean subduction zones, as exem-

576

Chapter 17: Precambrian Arc Associations

Fig. 5. Two component mixing model: (a) Chondrite-normalized REE and (b) primitive mantle-normalized immobile trace element mixing models. Adopted from Polat et al. (2002).

plified by the southern Isua tonalities, and the association of adakites with boninites in the Superior Province (Boily and Dion, 2002).

3. Late Archean Abitibi Boninite Series

577

3. LATE ARCHEAN ABITIBI BONINITE SERIES 3.1. Geological Setting The late Archean Abitibi greenstone belt is the largest greenstone belt in the Superior Province (Fig. 6; Jackson and Fyon, 1991; Calvert and Ludden, 1999, and references therein). The belt is tectonically bounded by the Opatica subprovince to the north and by the Pontiac subprovince to the south (Fig. 6). The Abitibi subprovince is separated from the Wawa subprovince to the west by the high-grade Kapuskasing structural zone, and its eastern contact is marked by the Proterozoic Grenville Province (Fig. 6). The belt has been divided into three major zones or domains on the basis of distinct volcano-sedimentary supracrustal sequences, related plutonism, and geochronology (Calvert and Ludden, 1999, and references therein). From north to south these zones are: (1) the Northern Volcanic Zone, (2) Central Granitic-Gneissic Zone, and (3) Southern Volcanic Zone. The older Northern Volcanic Zone (2730–2710 Ma) and the younger Southern Volcanic Zone (2705–2698 Ma) are separated by a major east-west trending, right lateral strike-slip fault zone, the Destor-Porcupine-Manneville fault zone, representing a terrane boundary (Mueller et al., 1996). According to Mueller et al. (1996), this fault zone records two distinct tectonic events: (1) an early arc-arc collision between 2697 and 2690 Ma, representing a compressional (thrusting) deformation phase; and (2) arc fragmentation between 2689–2680 Ma, recording a transcurrent deformation phase resulting from oblique plate convergence. Both phases of deformation were accompanied by igneous activity and greenschist to amphibolite facies metamorphism. Volcanic associations in the Abitibi subprovince include: (1) 2720–2707 Ma tholeiitic basalt and komatiite association; (2) 2750–2700 Ma arc tholeiite and calc alkaline association; (3) ca. 2716 Ma boninite series-depleted tholeiite association; (4) adakitehigh magnesian andesites (HMA); and (5) ca. 2674 Ma shoshonites (Wyman, 1999b; Wyman et al., 2002, and references therein). The first associations are oceanic plateaus derived from mantle plumes. The second and third associations are interpreted as intraoceanic arcs. The HMA are inferred to form by hybridization of adakite liquids with arc mantle. Shoshonites are late-tectonic and commonly related to regional translithospheric structures. They formed by second-stage melting of refractory sub arc mantle, fertilized by fluids and/or melts, during arc uplift and extension (Wyman and Kerrich, 1993; Dostal and Mueller, 1992). Arc-trench siliciclastic sedimentary rocks occurring between volcanic sequences were deposited between 2691 and 2673 Ma. The supracrustal units are intruded by syn- to post-tectonic granitoids decreasing in age from north to south (Jackson and Cruden, 1995). 3.2. Boninite Series-Depleted Tholeiite Association Detailed geochemical and field characteristics of the 2.7 Ga Abitibi boninitic lavas and related volcanic rocks have been discussed in Kerrich et al. (1998) and Wyman et al. (2002).

578

Chapter 17: Precambrian Arc Associations

Fig. 6. Subprovince map of the Superior Province, showing the locations of the Wawa and Abitibi subprovinces (a), and simplified geological map of the Wawa Subprovince, showing the locations of greenstone belts (b). Adopted from Williams et al. (1991).

4. Late Archean Opatica and Baltic Shield Boninites

579

Accordingly, only a summary of these compositional features and field relations is presented here (Table 1). The Abitibi boninite series are exposed at several locations over a strike length of 300 km between the northern and southern volcanic zones of the Abitibi subprovince. These mafic to ultramafic volcanic rocks often interfinger with flows of the plume-related basalt-komatiite association at lower stratigraphic levels, and primitive to evolved basalts of the younger Abitibi arc suite at higher stratigraphic levels. These stratigraphic relationships have been considered as evidence for interaction between a contemporaneous mantle plume and an intra-oceanic volcanic arc (Wyman, 1999b). The Abitibi boninites share the low concentrations of REE and HFSE, U-shaped REE patterns, and variably negative Nb-anomalies of Phanerozoic boninites. In addition, they have distinctive superchondritic Al2 O3 /TiO2 and Zr/Smcn ratios of Phanerozoic counterparts (Kerrich et al., 1998). Wyman (1999b) identified low-Ti tholeiites interfingered with the Abitibi boninites. These share the low Ti contents, high Al2 O3 /TiO2 ratios, but lack the U-shaped REE patterns and negative Nb-anomalies. 4. LATE ARCHEAN OPATICA AND BALTIC SHIELD BONINITES Recently, Boily and Dion (2002) and Shchipansky et al. (2004) reported occurrences of boninitic volcanic rocks in the 2.8 Ga Frotet-Evans greenstone belt of the Superior Province and the 2.8 Ga North Karelian greenstone belt of the Baltic Shield, respectively. The 250 km long Frotet-Evans greenstone belt occurs as a tectonic thrust of volcanic and sedimentary rocks in the predominantly plutonic Opatica subprovince. Volcanic rocks in the belt are composed predominantly of bimodal, tholeiitic to calc-alkaline lavas and adakitelike pyroclastic rocks. In contrast to the boninitic lavas of the Abitibi belt, the Frotet-Evans counterparts are not interstratified with plume-derived komatiite-tholeiitic basalt association. The origin of positive Zr-Hf anomalies on primitive mantle-normalized diagrams has been attributed to slab-derived adakitic melts. Boily and Dion (2002) proposed that the Frotet-Evans boninitic rocks formed in a supra-subduction zone during the extension of the Le Gardeur forearc. The 2.8 Ga North Karelian greenstone belt is exposed for 300 km along the boundary between the Karelian granite-greenstone and the Belomorian orogenic belt in the Baltic Shield (see Shchipansky et al., 2004). The North Karelian greenstone belt displays dismembered ophiolitic sequence including, gabbros and sheeted dykes. The geochemical characteristics of the North Karelian boninitic lavas are similar to those of the Troodos ophiolite, suggesting a supra-subduction zone origin. 5. LATE ARCHEAN WAWA ADAKITE-MANGNESIAN ANDESITE-NB-ENRICHED BASALT ASSOCIATION 5.1. Geological Setting The Wawa subprovince extends from the Vermilion district of Minnesota in the west to the Kapuskasing structural zone in the east (Williams et al., 1991; Fig. 6). Only key fea-

580

Chapter 17: Precambrian Arc Associations

tures that are directly relevant to the focus of this paper are described below. The Great Lakes Tectonic Zone separates the Wawa subprovince from the Marquette greenstone belt and the Minnesota Valley gneiss terrane to the south, whereas the Kapuskasing zone separates the Wawa from the formerly contiguous Abitibi belt to the east (Percival et al., 1994). The subprovince is composed of two linear trends of greenstone belts: the northern trend at the border with the Quetico subprovince comprising the Vermilion district, Shebandowan, Schreiber-Hemlo, Manitouwadge-Hornepayne, White River-Dayohessarah, and Kabinakagami greenstone belts; and the trend in the south-central portion of the subprovince, includes the Mishibishu, Michipicoten, and Gamitagama greenstone belts (Fig. 6b). These supracrustal zones are composed dominantly of the 2.75–2.70 Ga ocean floor tholeiitic basalt-komatiite association, and 2.72–2.65 Ga bimodal tholeiitic to calcalkaline arc association (Williams et al., 1991; Polat et al., 1998). The second association includes the recently recognized adakite-magnesian andesite (MA)-Nb-enriched basalts (NEB) association (Polat and Kerrich, 2001a). There are minor units of chert and banded iron formation, and gabbros within these volcanic sequences. Sedimentary rocks are predominantly siliciclastic turbiditic sandstones and shales, with minor conglomerate and carbonates. Generally, the former sequence structurally underlies the latter. Contacts between various volcanic-volcanic and volcanic-sedimentary units are for the most part tectonic, characterized by strike-slip or thrust faults (Polat et al., 1999). The linear trends of greenstone belts are separated and intruded by domains of syn- to post-kinematic tonalite-trondhjemite-granodioritic (TTG) plutons and sanukitoids (Williams et al., 1991; Corfu and Stott, 1998). Metamorphism is prevalently greenschist facies, with amphibolite facies proximal to the TTG batholiths. Primary igneous textures and pillow geometries of volcanic rocks are well preserved outside of major shear zones and intrusions. In greenschist facies rocks, chlorite and epidote are the dominant minerals in mafic rocks, whereas in their amphibolite facies equivalents, hornblende and actinolitic amphiboles, plagioclase, and epidote ± garnet are predominant. Mineralogically, dacites, rhyodacites, and rhyolites have quartz + alkali feldspar ± hornblende ± biotite ± muscovite (Fig. 3b). Andesitic flows are characterized predominantly by plagioclase + hornblende ± biotite ± quartz ± chlorite (Fig. 3c). Basalts are composed mainly of plagioclase + hornblende ± chlorite ± biotite ± quartz (Fig. 3d). Polat et al. (1999) interpreted the late Archean Wawa greenstone belts and formerly contiguous Abitibi belt as part of a 1000 km scale subduction-accretion complex that formed along an Archean intra-oceanic convergent plate margin during the trenchward migration of the magmatic arc axis. In this geodynamic framework, komatiite and associated tholeiitic basalt sequences were interpreted as dismembered fragments of oceanic plateau(s) captured by the arc; whereas LREE-enriched and variably HFSE-depleted bimodal volcanic sequences, with associated siliciclastic turbidites were interpreted as fragments of an intra-oceanic magmatic arc paired with trench turbidites. Subsequently, voluminous syn- to late-tectonic TTG batholiths intruded the imbricated ocean plateau-arc supracrustal units.

5. Late Archean Wawa Adakite-Mangnesian Andesite-Nb-Enriched Basalt Association

581

5.2. Geochemical Characteristics of Adakite-MA-NEB Association Detailed major and trace element, and Nd isotope characteristics of the Wawa adakites, MA, NEB, and “normal” tholeiitic to calc-alkaline basalts to andesites have been discussed in Polat and Kerrich (2001a, 2002). Accordingly, only a summary of these compositional features is presented here (Table 1). The adakite suite shares the high Al2 O3 and Na2 O contents, high La/Ybcn (21–43), Zr/Y (9–22), and Sr/Y (13–175) ratios, with large negative Nb and Ti anomalies of the Archean high-Al tonalite-trondhjemite-granodiorite (TTG) suite (Table 1). The magnesian andesites (MA) are distinctive in terms of relatively high Mg# (0.50–0.60), and Cr (110– 530 ppm) and Ni (21–229 ppm) contents relative to “normal” andesites (Table 1). They span ranges of SiO2 = 56–64 wt% and Mg# = 0.50–0.64. Like adakites, the MA have positively fractionated REE patterns (La/Ybcn = 5–28) and depletion of Nb and Ti relative to neighboring REE on primitive mantle-normalized diagrams. Niobium-enriched basalts and andesites (NEBA) range from 50–57 wt% SiO2 , 3.8– 5.7 wt% MgO, Mg# = 0.40–0.61, with high Nb (7–16 ppm) content relative to most arc basalts. As a group, NEBA plot to higher Nb, Zr, TiO2 , Fe2 O3 , and Y at a given value of La/Ybcn than the adakites and MA (Table 1). However, they have systematically lower Ni, Cr, Co, and SiO2 contents than the MA. Collectively, the Wawa NEBA are compositionally similar to Cenozoic Nb-enriched basalts (Polat and Kerrich, 2001a). “Normal” basalts plot mostly in the tholeiitic field on major element diagrams, and yet variably fractionated REE and high Zr/Y (3.2–7.9) ratios are more indicative of calcalkaline suites (Table 1). Andesites appear to form continuous trends with basalts on diagrams of major elements versus MgO, over the range of 57–63 wt% SiO2 and Mg# = 0.29–0.52 (Table 1). However, the two define separate trends for most trace elements (Polat and Kerrich, 2001a). The range of basalt to andesite compositions can be explained by progressive mixing of adakite liquids into a basalt end-member, yielding REE patterns that cross-over; this observation endorses coeval basalt and adakite magmatism (Polat and Kerrich, 2001a). In addition to their distinctive major and trace element compositions, the Wawa adakites, MA, and NEBA volcanic flows have also different Nd isotope characteristics (Fig. 7; Polat and Kerrich, 2002). The adakites are characterized by higher initial ε Nd (2670 Ma) values compared to the MA (+1.95 to +2.45 versus +0.14 to +1.167), consistent with distinct source reservoirs for the two volcanic suites (Fig. 7). The NEBA have more variable but intermediate initial ε Nd (2670 Ma) values (+1.11 to +2.05), falling between those of the adakites and magnesian andesites (Fig. 7). The “normal” tholeiitic to calc-alkaline basalts display variably high initial ε Nd (2670 Ma) values (+1.44 to +2.53). The initial ε Nd (2670 Ma) values (+1.53 to +2.44) of two “normal” andesite samples are comparable to those of basalts, suggesting that they may have derived from basalts by fractional crystallization. The results have been interpreted as a long term depleted mantle source for the basaltic precursors to adakites, hybridizing with a mantle wedge enriched up to 200 Ma in an earlier convergent margin.

582

Chapter 17: Precambrian Arc Associations

Fig. 7. Histograms for the initial ε Nd (2670 Ma) values of the Wawa adakites, MA, NEBA, and “normal” tholeiitic to calc-alkaline basalts and andesites. Adopted from Polat and Kerrich (2002).

6. PROTEROZOIC BONINITIC SUITES There are a few boninitic suites described from Proterozoic orogens, including the Paleoproterozoic Trans-Hudson orogen (Stern et al., 1995; Wyman, 1999b), Paleoproterozoic Bogoin-Boali greenstone belt, Central African Republic (Poidevin, 1994); and the Neoproterozoic Arabian-Nubian shield (Wolde et al., 1996; Teklay et al., 2002). All three boninitic suites are variably metamorphosed and deformed. On the basis of field relationships and geochemical characteristics, the Trans-Hudson and the Arabian-Nubian Shield boninitic suites were interpreted as parts of dismembered Proterozoic ophiolites formed in a subduction geodynamic setting. Similarly, the origin of the Bogoin-Boali boninitic suite

7. Conclusions

583

was attributed to subduction zone petrogenetic processes, including interaction of adakitic melts and depleted sub-arc mantle wedge, but an ophiolitic association has not been recognized in the belt (Poidevin, 1994).

7. CONCLUSIONS The least altered metavolcanic rocks from Paleoarchean to Neoproterozoic greenstone belts described above are geochemically comparable to many Phanerozoic boninites from the western Pacific ocean, such as the Izu-Bonin-Mariana fore-arc, and to boninite series volcanic rocks in Phanerozoic ophiolites, such as the Troodos and Betts Cove ophiolites. These similarities include: (1) high abundances of MgO, Ni, and Cr; (2) high Mg#; (3) low abundances of HFSE (Nb, Zr, Ti, Y); (4) high Al2 O3 /TiO2 but low Ti/Zr and Ti/V ratios; (5) negative fractionation of HREE; (6) U-shaped REE patterns; (7) enrichment of Zr relative to Sm and Nd; and (8) depletion of Nb relative to Th and La. The origin and tectonic setting(s) of boninitic rocks are among the least controversial of any volcanic rock type (Sun and Nesbitt, 1978; Cameron et al., 1979; Taylor et al., 1994), even though some of their specific geochemical characteristics are not well understood. Studies of Tertiary boninites suggest that they are products of high-temperature, lowpressure partial melting of a hydrous, depleted refractory mantle source above a subducted oceanic lithosphere at convergent plate boundaries, such as the Izu-Bonin-Mariana subduction zones of the western Pacific (Crawford et al., 1989; Tatsumi and Maruyama, 1989; Stern et al., 1991; Pearce et al., 1992). Given the geochemical similarities between Precambrian metavolcanic rocks discussed above and Phanerozoic boninites, the Precambrian boninitic volcanic rocks also appear to have had a two-stage petrogenetic history: (1) multistage partial melting of a mantle source with extreme depletion of highly incompatible to moderately incompatible elements; and (2) an enrichment of this source by slab-derived melts or fluids that are enriched in Th, LREE, Zr (Kerrich et al., 1998; Polat and Kerrich, 2001a). Given the presumably higher geothermal gradients in the early Archean mantle, the ∼ 3.8 Ga Isua boninitic volcanic rocks could have originated by the interaction of slab-derived, adakite-like melts with depleted sub-arc mantle (Fig. 5; see Drummond et al., 1996; Defant and Kepezhinskas, 2001). The 2.7 Ga Abitibi boninitic volcanic rocks are interlayered with plume-derived komatiites and tholeiitic rocks. This stratigraphic relationship has been considered as evidence for plume-arc interaction (Kerrich et al., 1998; Wyman et al., 2002). Therefore, arc-plume proximity may have played a key role for the generation of higher geothermal gradients in Abitibi subduction zone to produce boninitic melts. The 2.8 Ga Frotet-Evans boninitic volcanic rocks are interstratified with bimodal tholeiitic to calc-alkaline arc lavas, and were probably erupted in a fore-arc geodynamic setting. Both the Frotet-Evans and North Karelian boninitic suites are spatially and temporarily associated with adakitic volcanic rocks, suggesting that interaction between slabmelts and depleted sub-arc mantle peridotite played an important role for the generation of the both boninitic suites.

584

Chapter 17: Precambrian Arc Associations

The Isua and Abitibi boninitic suites differ from Phanerozoic counterparts in that they have lower SiO2 contents. However, it should be emphasized that SiO2 is susceptible to alteration. The Isua boninitic volcanic rocks have lower Nb/Th and Nb/La ratios than Abitibi counterparts, consistent with stronger second stage enrichment of the Isua boninitic rocks than Abitibi counterparts (Fig. 8). The geochemical characteristics of Archean Isua and Abitibi boninitic volcanic rocks are distinct from those of Archean plume-derived komatiite-basalt associations and subduction-derived adakite-MA-NEBA associations (Figs. 9, 10). The association of adakites, magnesian andesites, and Nb-enriched basalts has been reported from some late Archean greenstone belts (Hollings and Kerrich, 2000; Polat and Kerrich, 2001a). Many compositional characteristics of late Archean adakites, MA and NEBA are akin to Cenozoic counterparts. It appears that many Cenozoic adakites, MA, and NEB are closely associated with the subduction of young, hot oceanic slabs exemplified by the Philippine (Sajona et al., 1996), southern Andes (Stern and Kilian, 1996), Central America (Defant et al., 1992), and western Aleutian Islands convergent margins (Yogodzinski et al., 1995). Similarly, subduction of young, hot oceanic slabs may have been responsible for the generation and association of adakite, MA, and NEB in late Archean greenstone belts. The Wawa MA tend to have higher La/Smcn and Gd/Ybcn , but lower Nb/Th and Nb/La ratios than Abitibi counterparts (Fig. 9). These geochemical differences between the formerly contiguous subprovinces can be explained by the contemporaneous neighbouring plume and arc magmatism in the Abitibi, but contemporaneous yet spatially separated plume and arc magmatism in the Wawa subprovince (Dostal and Mueller, 1997; Wyman, 1999b; Wyman et al., 2002). In the Wawa subprovince, the spatial and temporal relationship between adakites, MA, and NEB is consistent with a genetic link between them (Polat and Kerrich, 2001a). This petrogenetic relation is endorsed by the spectrum of “normal” arc basalts and andesites resulting from mixing between basalt and adakite liquids, generating a spectrum of REE patterns that cross over (Polat and Kerrich, 2001a). “Normal” Wawa arc basalts plot with the low Ce-Yb trend defined by intra-oceanic arcs, whereas basalt-adakite mixtures plot on a negative slope to high Ce but low Yb (Fig. 10b). MA plot to higher Ce but variably lower Yb, consistent with hybridization of a mantle source by adakite liquids (Fig. 10b). It has been proposed that rapid mantle convection and subduction of young, hotter oceanic slabs provided optimized conditions for the production of adakite-MA-NEB association in the late Archean Wawa subduction zone. Accordingly, it was suggested that the late Archean Wawa adakites were derived mainly from slab melting, whereas the MA originated by hybridizing with sub-arc mantle wedge, and the NEB was attributed to melting of the residue of hybridized mantle wedge (Polat and Kerrich, 2001a). The 2.7 Ga Wawa adakites, MA, NEB, and “normal” intra-oceanic arc tholeiitic to calcalkaline basalts-andesites have distinctive Nd isotope characteristics (Polat and Kerrich, 2002). The variably large positive initial ε Nd values (+1.95 to +2.45) of the adakites signify that their basaltic precursors, with a short period of crustal residence, were derived from a long-term depleted, MORB-like mantle source(s). Variably lower initial ε Nd val-

7. Conclusions

585

Fig. 8. MgO (wt%) vs. Nb/Th (a) and TiO2 (wt%) vs. Zr (b) variation diagrams for Archean boninites, adakites, magnesian andesites (MA), komatiites (K) komatiitic basalts (KB), and oceanic plateau basalts (OPB). Data for Isua basalts from Polat et al. (2002), for adakites, magnesian andesites from Polat and Kerrich (2001a), and for komatiites and komatiitic basalts from Xie et al. (1993) and Polat et al. (1999), for Wawa OPB from Polat and Kerrich (2000), and for Abitibi MA from Derek Wyman (unpublished data).

586

Chapter 17: Precambrian Arc Associations

Fig. 9. Mg# vs. Gd/Ybcn (a) and La/Smcn (b) variation diagrams for Archean boninites, adakites, magnesian andesites (MA), komatiites (K) komatiitic basalts (KB), and oceanic plateau basalts (OPB). Data source as in Fig. 8. Chondrite normalization values from Sun and McDonough (1989).

7. Conclusions

587

Fig. 10. Yb (ppm) vs. Ce (ppm) variation diagram for modern arc lavas and Archean counterparts. Part (a) is adopted from Hawkesworth et al. (1993). Data source as in Fig. 8.

588

Chapter 17: Precambrian Arc Associations

ues in the magnesian andesites (+0.14 to +1.68), NEB (+1.11 to +2.05), and “normal” intra-oceanic arc tholeiitic to calc-alkaline basalts and andesites (+1.44 to +2.53) can be explained by melting of the isotopically depleted, heterogeneous sub-arc mantle wedge that had been variably hybridized by subduction-derived melts and/or fluids prior to initiation of arc volcanism. Collectively, the Nd-isotope systematics of the Wawa arc volcanic rocks can be explained by mixture of late Archean depleted upper mantle and previously recycled subduction-derived components.

8. IMPLICATIONS The geochemical characteristics of magmatism in modern subduction zones is controlled mainly by: (1) the age of the subducting plate, young versus old; (2) the nature of the overriding plate, continental versus oceanic; (3) the presence or absence of sediment subduction; (4) the subduction velocity, slow versus rapid; (5) angle of subduction, shallow versus steep; (6) convergence angle, orthogonal versus oblique; (7) previous history of the sub-arc mantle wedge; (8) depth of partial melting, shallow versus deep; and (9) the presence or absence of a paired back-arc basin (Defant and Drummond, 1990; Pearce and Peate, 1995). Of these factors, the thermal structure of subduction zones is likely to have the strongest effect on the geochemical characteristics of subduction-derived volcanic rocks (see Defant and Drummond, 1990; Defant and Kepezhinskas, 2001). 8.1. Genetic Link Between Adakites, Boninites, and Magnesian Andesites It is commonly believed that boninites and magnesian andesites originate in the sub-arc mantle wedge during the subduction of young, hot oceanic lithosphere (Drummond et al., 1996; Defant and Kepezhinskas, 2001). Under these conditions, the subducting oceanic slabs are likely to undergo melting rather than dehydration, to produce adakitic liquids (Defant and Drummond, 1990). Interaction between adakitic melts and sub-arc mantle peridotite results in generation of boninites or magnesian andesites. According to Pearce and co-workers (Pearce et al., 1992; Pearce and Peate, 1995), slab melting results in high Zr/Sm ratios, because Zr is not conservative in the melts. Archean boninites, adakites, and magnesian andesites have similar Zr/Sm ratios (Figs. 11, 12), suggesting a petrogenetic link between there three groups of rocks. The temporal and spatial relationship between the ∼ 2.8 Frotet-Evans and the Karelian boninitic flows and adakitic volcanic rocks also suggests a petrogenetic link between the two volcanic suites. In addition, slab melts tend to have high Al2 O3 /TiO2 ratios, this probably results from a combination of the consumption of plagioclase and conservative behaviour of Ti during slab melting (Defant et al., 1992; Pearce and Peate, 1995; Schiano et al., 1995). Thus, high Al2 O3 /TiO2 ratios and low Ti, Y and Yb contents in boninites and adakites suggest that these two suites are petrogenetically related (Figs. 11, 12). Similarly, moderate overlap between boninites and magnesian andesites on Ni vs Zr/Sm is consistent with a petrogenetic link between the two groups (Fig. 12).

8. Implications

589

Fig. 11. Selected element-element ratios vs. Gd/Ybcn variation diagrams. Data source as in Fig. 8. Chondrite normalization values from Sun and McDonough (1989).

590

Chapter 17: Precambrian Arc Associations

Fig. 12. Zr/Smcn vs. Ni (ppm), Zr/Smcn vs. Al2 O3 /TiO2 , and Y vs. Gd/Ybcn variation diagrams. Data source as in Fig. 8. Chondrite normalization values from Sun and McDonough (1989).

8. Implications

591

Boninites and magnesian andesites appear to be derived from the sub-arc mantle wedge through adakite-mantle interaction; however, they have significant differences in REE and HFSE systematics. The major geochemical differences between the two suites can be summarized as: (1) boninites have lower concentrations of REE and HFSE (Nb, Ti, Zr, Y) than magnesian andesites; and (2) magnesian andesites possess higher La/Smcn , Gd/Ybcn , and Zr/Y ratios than boninites (Fig. 11). These distinct compositional features of boninites and magnesian andesites can be explained by a combination of different source characteristics, and different degrees and depth of partial melting. Lower absolute abundances of REE and HFSE in boninites suggest that their mantle sources previously experienced larger degrees of melt extraction and/or lower degrees of slab-related enrichment than those of magnesian andesites. Negatively fractionated HREE patterns (Gd/Ybcn < 1) in boninites are consistent with low-pressure melting of highly to moderately incompatible element depleted mantle sources, whereas positively fractionated LREE (La/Smcn > 1) and HREE (Gd/Ybcn > 1) patterns in magnesian andesites are consistent with high-pressure melting of highly to moderately incompatible element enriched mantle sources. All reported Tertiary boninites (e.g., Izu-Bonin-Mariana boninites) have been interpreted as the products of intra-oceanic subduction zone geodynamic processes. Similarly, the origin of boninites associated with Phanerozoic ophiolites has been attributed to intraoceanic subduction zone processes (Sun and Nesbitt, 1978; Coish, 1989). If the geochemical characteristics of the Archean Isua, Abitibi, Frotet-Evans, and North Karelian boninitic volcanic rocks have the same geodynamic significance as Phanerozoic counterparts, then they likely originated in an intra-oceanic subduction zone-like tectonic setting, suggesting that Phanerozoic-like plate tectonic processes were operating as early as 3.8 Ga. Similarly, if the Neoarchean Superior Province and Baltic Shield adakites were generated by similar geodynamic processes as counterparts in Cenozoic arcs, it is likely that late Archean oceanic crust was also created and destroyed by modern plate tectonic-like geodynamic processes. 8.2. Archean Continental Growth In this section we discuss the significance of the higher geothermal gradients in Archean subduction zones for the generation of Archean continental crust. Ellam and Hawkesworth (1988) suggested that crust-forming processes in Archean subduction zones may have been different from those operating in modern counterparts. This difference was attributed to the higher geothermal gradients in the Archean mantle. Many continental growth models suggest that 60–70% of the present day continental crust formed by the end of the Archean (Taylor and McLennan, 1995, and references therein). It is generally accepted that Archean continental crust grew by accretionary and magmatic processes taking place at convergent plate boundaries (Moorbath, 1977; Kusky and Kidd, 1992; Taylor and McLennan, 1995; Kusky and Polat, 1999; Foley et al., 2002). Recent studies suggest that the accretion of plume-derived oceanic plateaus at con-

592

Chapter 17: Precambrian Arc Associations

vergent plate margins may have played an important role throughout Earth’s history in the growth of the continental crust (Stein and Hofmann, 1994; Rudnick, 1995; Condie, 1998; Puchtel et al., 1998; Kerrich et al., 1999, and references therein). Polat and Kerrich (2001b) showed from field relationships and geochemical considerations that late Archean continental crust in the southern Superior Province grew by mixing of oceanic plateau and subduction derived components (granitoids, tholeiitic to calc-alkaline bimodal volcanic rocks, trench turbidites) at convergent plate boundaries. Simple mixing calculations suggest that 6–12% oceanic plateau, mixing with a 88–94% arc magmas is required to produce late Archean continental crust in the southern Superior Province. TTG suites in the southern Superior Province are the major (∼ 75–90%) constituents of arc magmatism. However, relatively low MgO (0.5–3.0 wt%), Ni (5–70 ppm), and Cr (10–70 ppm) contents in these rocks (Feng and Kerrich, 1992) cannot account for the higher values of MgO (4.7 wt%), Ni (105 ppm) and Cr (180 ppm) in Archean upper continental crust (Taylor and McLennan, 1995). Given the fact that the Archean bulk continental crust has even higher MgO (5.9 wt%), Ni (130 ppm), and Cr (230 ppm), a more mafic, subduction-derived component is required to explain the composition. Polat and Kerrich (2001a, 2001b) suggested that magnesian andesites (MgO = 3.4–6.4 wt%; Ni = 20–230 ppm; Cr = 110–530 ppm) may have played an important role in the generation of the late Archean continental crust. However, MA are rare in the Archean rock record. They constitute volumetrically less than a few percent of the exposed crust in the Wawa and Abitibi subprovinces. The apparent MgO, Ni, and Cr deficit in the model Archean continental crust could be better explained by incorporation of a significant amount of MA or sanukitoids stored in the lower continental crust (cf. Kelemen, 1995; Smithies and Champion, 2000). In conclusion, studies in Archean greenstone-granitoid terranes suggest that high heat flow and subduction of young, hot oceanic slabs in the Archean provided optimized conditions for the generation of boninites, adakites, MA and NEB, and transformation of slabs and accreted oceanic plateaus into continental crust via partial melting under amphibolite to eclogite conditions.

ACKNOWLEDGEMENTS T.M. Kusky is thanked for the invitation to submit this contribution to the volume “Precambrian Ophiolites and Related Rocks”. The authors are grateful to D.A. Wyman for providing unpublished MA data from the Abitibi belt and for discussions. Comprehensive critiques by C. Herzberg and T. Kusky have resulted significant improvements to the paper. A. Polat thanks the Max-Planck-Institut (Mainz) and Alexander von Humbolt foundation for financial and logistical support. This is a contribution for the Isua Multi-disciplinary Research Project.

References

593

REFERENCES Appel, P.W.U., Fedo, C.M., Moorbath, S., Myers, J.S., 1998. Recognizable primary volcanic and sedimentary features in a low strain domain of the highly deformed, oldest known (3.7–3.8 Gry) Greenstone belt, Isua, West Greenland. Terra Nova 10, 57–62. Bédard, J.H., 1999. Petrogenesis of boninites from the Betts Cove Ophiolite, Newfoundland, Canada: Identification of subducted source components. Journal of Petrology 40, 1853–1889. Boily, M., Dion, C., 2002. Geochemistry of boninite-type volcanic rocks in the Frotet-Evans greenstone belt, Opatica subprovince, Quebec: Implications for the evolution of Archean greenstone belt. Precambrian Research 115, 349–371. Brown, A.V., Jenner, G.A., 1989. Geological setting, petrology and chemistry of Cambrian boninite and low-Ti tholeiite lavas in Western Tasmania. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Unwin Hyman, London, pp. 232–263. Calvert, A.J., Ludden, J.N., 1999. Archean continental assembly in the southeastern Superior province of Canada. Tectonics 18, 412–429. Cameron, W.E., Nisbet, E.G., Dietrich, V.J., 1979. Boninites, komatiites and ophiolitic basalts. Nature 280, 550–553. Cameron, W.E., McCollouch, M.T., Walker, D.A., 1983. Boninite petrogenesis: chemical and Nd-Sr constraints. Earth and Planetary Science Letters 65, 75–89. Coish, R.A., 1989. Boninitic lavas in Appalachian ophiolites: a review. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Unwin Hyman, London, pp. 264–286. Condie, K.C., 1998. Episodic continental growth and supercontinents: a mantle avalanche connection? Earth and Planetary Science Letters 163, 97–108. Corfu, F., Stott, G.M., 1998. Shebandowan greenstone belt, western Superior Province: U-Pb ages, tectonic implications, and correlations. Geological Society of America Bulletin 110, 1467–1484. Crawford, A.J., Fallon, T.J., Green, D.H., 1989. Classification, petrogenesis and tectonic setting of boninites. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Unwin Hyman, London, pp. 1– 49. Defant, M.J., Drummond, M.S., 1990. Derivation of some modern arc magmas by meting of young subducted oceanic lithosphere. Nature 347, 662–665. Defant, M.J., Jackson, T.E., Drummond, M.S., De Boher, J.Z., Bellon, H., Feigenson, M.D., Maury, R.C., Stewart, R.H., 1992. The geochemistry of young volcanism throughout western Panama and southeastern Costa Rica: an overview. Journal of the Geological Society of London 149, 569–579. Defant, M.J., Kepezhinskas, P., 2001. Evidence suggests slab melting in arc magmas. Eos, Transactions, American Geophysical Union 82, 65–69. Dostal, J., Mueller, W., 1992. Archean shoshonites from the Abitibi greenstone belt, Chibougamau (Québec, Canada): geochemistry and tectonic setting. Journal of Volcanology and Geothermal Research 53, 145–165. Dostal, J., Mueller, W.U., 1997. Komatiite flooding of a rifted Archean rhyolite arc complex: geochemical signature and tectonic significance of Stoughton-Roquemaure Group, Abitibi greenstone belt, Canada. Journal of Geology 105, 545–563. Drummond, M.S., Defant, M.J., Kepezhinskas, P.K., 1996. Petrogenesis of slab-derived trondhjemite-tonalite-dacite/adakite magmas. Geological Society of America Special Paper 315, 205– 215. Ellam, R.M., Hawkesworth, C.J., 1988. Is average continental crust generated at subduction zones? Geology 16, 314–317.

594

Chapter 17: Precambrian Arc Associations

Fedo, C.M., 2000. Setting and origin for problematic rocks from the > 3.7 Ga Isua greenstone belt, southern west Greenland: Earth’s oldest coarse clastic sediments. Precambrian Research 101, 69–78. Feng, R., Kerrich, R., 1992. Geochemical evolution of granitoids from the Archean Abitibi Southern Volcanic Zone and the Pontiac subprovince, Superior Province, Canada: Implications for tectonic history and source regions. Chemical Geology 98, 23–70. Foley, S., Tiepolo, M., Vannucci, R., 2002. Growth of early continental crust controlled by melting of amphibolite in subduction zones. Nature 417, 837–840. Gruau, G., Rosing, M., Bridgwater, D., Gill, R.C.O., 1996. Resetting of Sm-Nd systematics during metamorphism of > 3.7 Ga rocks: implications for isotopic models of early Earth differentiation. Chemical Geology 133, 225–240. Hatton, C.J., Sharpe, M.R., 1989. Significance and origin of boninite-like volcanic rocks associated with the Bushveld Complex. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Unwin Hyman, London, pp. 174–207. Hawkesworth, C.J., Gallagher, K., Hergt, J.M., McDermott, F., 1993. Mantle and slab contributions in arc magmas. Annual Review of Earth and Planetary Sciences 2, 175–204. Hofmann, A.W., 1988. Chemical differentiation of the Earth: the relationships between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297–314. Hollings, P., Kerrich, R., 2000. An Archean arc basalt-Nb-enriched basalt-adakite association: The 2.7 Ga Confederation assemblage of the Birch-Uchi greenstone belt, Superior Province. Contributions to Mineralogy and Petrology 139, 208–226. Jackson, S.L., Fyon, J.A., 1991. The western Abitibi subprovince in Ontario. In: Thurston, P.C., Williams, H.R., Sutcliffe, H.R., Stott, G.M. (Eds.), Geology of Ontario. Ontario Geological Survey Special Volume 4 (1), 405–484. Jackson, S.L., Cruden, A.R., 1995. Formation of the Abitibi greenstone belt by arc-trench migration. Geology 23, 471–474. Kamber, B.S., Moorbath, S., Whitehouse, M.J., 1998. Extreme Nd-isotope heterogeneity in the early Archean fact or fiction? Case histories from northern Canada and West Greenland—Reply. Chemical Geology 148, 219–224. Kelemen, P.B., 1995. Genesis of high Mg# andesites and the continental crust. Contributions to Mineralogy and Petrology 120, 1–19. Kepezhinskas, P., Defant, M.J., Drummond, M.S., 1996. Progressive enrichment of island arc mantle by melt-peridotite interaction inferred from Kamchatka xenoliths. Geochimica et Cosmochimica Acta 60, 1217–1229. Kerrich, R., Wyman, D.A., Fan, J., Bleeker, W., 1998. Boninite series: low Ti-tholeiite associations from the 2.7 Ga Abitibi greenstone belt. Earth and Planetary Science Letters 164, 303–316. Kerrich, R., Polat, A., Wyman, D.A., Hollings, P., 1999. Trace element systematics of Mg- to Fetholeiitic basalt suites of the Superior Province: implications for Archean mantle reservoirs and greenstone belt genesis. Lithos 46, 163–187. Komiya, T., Maruyama, S., Masuda, T., Nohda, S., Hayashi, M., Okamoto, K., 1999. Plate tectonics at 3.8–3.7 Ga: Field evidence from the Isua accretionary complex, Southern West Greenland. Journal of Geology 107, 515–554. Kusky, T.M., Kidd, W.S.F., 1992. Remnants of an Archean oceanic plateau, Belingwe greenstone belt, Zimbabwe. Geology 20, 43–46. Kusky, T.M., Polat, A., 1999. Growth of granite-greenstone terranes at convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73. Martin, H., 1999. Adakitic magmas: modern analogues of Archean granitoids. Lithos 46, 411–429.

References

595

Meffre, S., Aitchison, J., Crawford, A.J., 1996. Geochemical evolution and tectonic significance of boninites and tholeiites from Koh ophiolite, New Caledonia. Tectonics 15, 67–83. Moorbath, S., 1977. Ages, isotopes and evolution of Precambrian continental crust. Chemical Geology 20, 151–187. Mueller, W.U., Daigneault, R., Mortensen, J.K., Chown, E.H., 1996. Archean terrane docking: upper crustal collision tectonics, Abitibi greenstone belt, Quebec, Canada. Tectonophysics 265, 127– 150. Myers, J.S., 2001. Protoliths of the 3.7–3.8 Ga Isua greenstone belt, West Greenland. Precambrian Research 105, 129–141. Nutman, A.P., McGregor, V.R., Friend, C.R.L., Bennett, V.C., Kinny, P.D., 1996. The Itsaq Gneiss Complex of southern West Greenland; the world’s most extensive record of early crustal evolution (3900–3600 Ma). Precambrian Research 78, 1–39. Nutman, A.P., Bennett, V.C., Friend, C.R.L., Rosing, M.T., 1997. ∼ 3710 and  3790 Ga Ma volcanic sequence in the Isua (Greenland) supracrustal belt structural and Nd isotope implications. Chemical Geology 141, 271–287. Nutman, A.P., Bennett, V.C., Friend, C.R.L., Norman, M.D., 1999. Meta-igneous (non-gneissic) tonalites and quartz-diorites from an extensive ca. 3800 Ma terrain south of the Isua supracrustal belt, southern West Greenland: Constraints on early crust formation. Contributions to Mineralogy and Petrology 137, 364–388. Pearce, J.A., van der Laan, S., Arculus, R.J., Murton, B.J., Ishii, T., Peate, D.W., 1992. Boninite and Harzburgite from Leg 125 (Bonin-Mariana Fore-arc): A case study of magma genesis during the initial stage of subduction. In: Fryer, P., Pearce, J.A., Stokking, L.B., et al. (Eds.), Proceedings for the Ocean Drilling Program. Scientific Results 125, 623–659. Pearce, J.A., Peate, D.W., 1995. Tectonic implications of the composition of volcanic arc magmas. Annual Review of Earth and Planetary Sciences 23, 251–285. Percival, J.A., Stern, R.A., Skulski, T., Card, K.D., Mortensen, J.K., Begin, N.J., 1994. Minto block, Superior province: Missing link in deciphering assembly of the Craton at 2.7 Ga. Geology 22, 839–842. Poidevin, J.L., 1994. Boninite-like volcanic rocks from the Paleoproterozoic greenstone belt of Bogoin, Central African Republic: geochemistry and petrogenesis. Precambrian Research 68, 97– 113. Polat, A., Kerrich, R., Wyman, D.A., 1998. The late Archean Schreiber-Hemlo and White RiverDayohessarah greenstone belts, Superior Province: collages of Oceanic plateaus, oceanic arcs, and subduction-accretion complexes. Tectonophysics 294, 295–326. Polat, A., Kerrich, R., Wyman, D.A., 1999. Geochemical diversity in oceanic komatiites and basalts from the late Archean Wawa greenstone belts, Superior Province, Canada: trace element and Nd isotope evidence for a heterogeneous mantle. Precambrian Research 94, 139–173. Polat, A., Kerrich, R., 2000. Archean greenstone belt magmatism and continental growth-mantle evolution connection: constraints from Th-U-Nb-LREE systematics of the 2.7 Ga Wawa subprovince, Canada. Earth and Planetary Science Letters 175, 41–54. Polat, A., Kerrich, R., 2001a. Magnesian andesites, Nb-enriched basalt-andesites, and adakites from late Archean 2.7 Ga Wawa greenstone belts, Superior Province, Canada: implications for late Archean subduction zone petrogenetic processes. Contributions to Mineralogy and Petrology 141, 36–52. Polat, A., Kerrich, R., 2001b. Geodynamic processes, continental growth, and mantle evolution recorded in late Archean greenstone belts of the southern Superior Province, Canada. Precambrian Research 112, 5–25.

596

Chapter 17: Precambrian Arc Associations

Polat, A., Hofmann, A.W., Rosing, M., 2002. Boninite-like volcanic rocks in the 3.7–3.8 Ga Isua greenstone belt, West Greenland: Geochemical Evidence for intra-oceanic subduction zone processes in the Earth. Chemical Geology 184, 231–254. Polat, A., Kerrich, R., 2002. Nd-isotope systematics of ∼ 2.7 Ga adakites, magnesian andesites, and basalts-andesites, Superior Province: Evidence for shallow crustal recycling at Archean subduction zones. Earth and Planetary Science Letters 202, 345–360. Puchtel, I.S., Hofmann, A.W., Mezger, K., Jochum, K.P., Shchipansky, A.A., Samsonov, A.V., 1998. Oceanic plateau model for continental crustal growth in the Archean: a case study from the Kostomuksha greenstone belt, NW Baltic Shield. Earth and Planetary Science Letters 155, 57–74. Puchtel, I.S., Hofmann, A.W., Amelin, Y.W., Grabe-Schonberg, C.-D., Samsonov, A.V., Shchipansky, A.A., 1999. Sumozero-Kenozero greenstone belt, SE Baltic Shield: Isotope and trace element constraints. Geochimica et Cosmochimica Acta 63, 3579–3595. Rosing, M.T., Rose, N.M., Bridgwater, D., Thomsen, H.S., 1996. Earliest part of Earth’s stratigraphic record: A reappraisal of the > 3.7 Ga Isua (Greenland) supracrustal sequence. Geology 24, 43– 46. Rudnick, R.L., 1995. Making continental crust. Nature 378, 571–578. Sajona, F.G., Maury, R.C., Bellon, H., Cotton, J., Defant, M.J., Pubellier, M., 1993. Initiation of subduction and generation of slab melts in western Mindanao, Philippines. Geology 21, 1007– 1010. Sajona, F.G., Maury, R., Bellon, H., Cotton, J., Defant, M., 1996. High field strength element enrichment of Pliocene-Pleistocene island arc basalts, Zamboanga Peninsula, Western Mindanao (Philippines). Journal of Petrology 37, 693–726. Schiano, P., Clocchiatti, R., Shimizu, N., Maury, R.C., Jochum, K.P., Hofmann, A.W., 1995. Hydrous, silicate-rich melts in the sub-arc mantle and their relationship with erupted lavas. Nature 377, 595–600. Shchipansky, A.A., Samsonov, A.V., Bibikova, E.V., Babarina, I.I., Konilov, A.N., Krylov, K.A., Slabunov, A.I., Bogina, M.M., 2004. 2.8 Ga boninite-hosting partial suprasubduction zone ophiolite sequences from the North Karelian greenstone belt, NE Baltic Shield, Russia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 425–486. Smellie, J.L., Stone, P., Evans, J., 1995. Petrogenesis of boninites in the Ordovician Ballantrae Complex ophiolite, southwestern Scotland. Journal of Volcanology and Geothermal Research 69, 323– 342. Smithies, R.H., Champion, D.C., 2000. The Archean high-Mg diorite suite: Links to tonalitetrondhjemite-granodiorite magmatism and implications for early Archean crustal growth. Journal of Petrology 41, 1653–1671. Stern, R.J., Morris, J., Bloomer, S.H., Hawkins, J.W., 1991. The source of the subduction component in convergent margin magmas: trace element and radiogenic evidence from Eocene boninites, Mariana forearc. Geochimica et Cosmochimica Acta 55, 1467–1481. Stern, R.A., Syme, E.C., Bailes, A.H., Lucas, S.B., 1995. Paleoproterozoic (19–186 Ga) arc volcanism in the Flin Flon Belt, Trans-Hudson Orogen, Canada. Contributions to Mineralogy and Petrology 119, 117–141. Stern, C.R., Kilian, R., 1996. Role of the subducted slab, mantle wedge and continental crust in the generation of adakites from the Andean Austral Volcanic Zone. Contributions to Mineralogy and Petrology 123, 263–281. Stein, M., Hofmann, A.W., 1994. Mantle plumes and episodic crustal growth. Nature 372, 63–68.

References

597

Sun, S.-s., Nesbitt, R.W., 1978. Geochemical regularities and genetic significance of ophiolitic basalts. Geology 6, 689–693. Sun, S.-s., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geological Society of London Special Publication 42, 313–345. Tatsumi, Y., Maruyama, S., 1989. Boninites and high-Mg andesites: tectonic and petrogenesis. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Unwin Hyman, London, pp. 50–71. Taylor, R.N., Nesbitt, R.W., 1988. Light rare earth enrichment of supra subduction-zone mantle: evidence from the Troodos ophiolite, Cyprus. Geology 16, 448–451. Taylor, R.N., Nesbitt, R.W., Vidal, P., Harmon, R., Auvray, B., Croudace, I.W., 1994. Mineralogy, chemistry, and genesis of the boninite series volcanics, Chichijima, Bonin Islands, Japan. Journal of Petrology 35, 577–617. Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust. Reviews of Geophysics 33, 241–265. Teklay, M., Berhe, K., Reimold, W.U., Armstrong, R., Asmerom, Y., Watson, J., 2002. Geochemistry and geochronology of a Neoproterozoic low-K tholeiite-boninite association in Central Eritrea. Gondwana Research 5, 597–611. Williams, H.R., Stott, G.M., Heather, K.B., Muir, T.L., Sage, R.P., 1991. Wawa Subprovince. In: Thurston, P.C., Williams, H.R., Sutcliffe, H.R., Stott, G.M. (Eds.), Geology of Ontario. Ontario Geological Survey Special Volume 4 (1), 485–539. Wolde, B., Asres, Z., Desta, Z., Gonzalez, J., 1996. Neoproterozoic zirconium depleted boninite and tholeiite series rocks from Adola, southern Ethiopia. Precambrian Research 80, 261–279. Wyman, D.A., Kerrich, R., 1993. Archean shoshonitic lamprophyres of the Abitibi subprovince, Canada: Petrogenesis and tectonic setting. Journal of Petrology 34, 1067–1109. Wyman, D.A., 1999a. Paleoproterozoic boninites in an ophiolite-like setting, Trans-Hudson orogen, Canada. Geology 27, 455–458. Wyman, D.A., 1999b. A 2.7 Ga depleted tholeiite suite: evidence of plume-arc interaction in the Abitibi greenstone belt, Canada. Precambrian Research 97, 27–42. Wyman, D.A., Kerrich, R., Polat, A., 2002. Assembly of Archean cratonic mantle lithosphere and crust: plume-arc interaction in the Abitibi-Wawa subduction-accretion complex. Precambrian Research 115, 37–62. Xie, Q., Kerrich, R., Fan, J., 1993. HFSE/REE fractionations recorded in three komatiite-basalt sequences, Archean Abitibi greenstone belt: implications for plume sources and depths. Geochimica et Cosmochimica Acta 57, 4111–4118. Yogodzinski, G.M., Kay, R.W., Volynets, O.N., Koloskov, A.V., Kay, S.M., 1995. Magnesian andesite in the western Aleutian Komandorsky region: Implications for slab melting and processes in the mantle wedge. Geological Society of America Bulletin 107, 505–519.

This page intentionally left blank

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

599

Chapter 18

ARCHEAN GREENSTONE BELTS DO CONTAIN FRAGMENTS OF OPHIOLITES MAARTEN J. DE WIT CIGCES, Department of Geological Sciences, University of Cape Town, Rondebosch 7701, South Africa

1. INTRODUCTION Most Archean greenstone belts are severely tectonised so that reconstruction of their rock assemblages revealing original autochthonous relationships is a daunting task (de Wit and Ashwal, 1997; Kusky and Vearncombe, 1997). There are about 260 individual Archean greenstone belts worldwide. Of these about 40 (∼ 15%) have been studied in sufficient detail (and mapped at a scale of less than 1:100.000) to provide relatively reliable information about pre-2.5 Ga geological processes (de Wit and Ashwal, 1995, 1997). Greenstone belts represent some of the earliest records of Earth history, but they are not restricted to the Archean. For example, the large Neoproterozoic Arabian-Nubian shield has an Archean-like cratonic crust with at least 7 major greenstone belts, most of which comprise island arc-like successions and associated (but often dismembered) ophioliteassemblages (Berhe, 1997). Similarly, the Baltic shield contains greenstone belt sequences ranging in age from > 3.1 Ga (Mesoarchean) to 1.9 Ga (Mesoproterozoic). Some of the Mesoproterozoic greenstone belts share characteristics of many Archean greenstone belts (e.g., abundant komatiites), whilst others share characteristics of Phanerozoic ophiolites (Sorjonen-Ward et al., 1997; and this volume). There is no simple definition of a greenstone belt other than that they contain significant volumes of basaltic rocks metamorphosed at relatively low grades to yield “green” mineral assemblages (de Wit and Ashwal, 1995). A wide spectrum of tectonic environments is preserved within Archean greenstone belts, and many individual belts are mixtures of components from different tectonic environments and in particular from island arc terrains (de Wit and Ashwal, 1995, 1997; Kusky and Vearncombe, 1997). It is claimed nevertheless by some that oceanic crust-forming environments are not preserved amongst this mixture of tectonic regimes because in their views no rocks assemblages in Archean greenstone belt sequences exhibit sufficient features to warrant definitive classification as an ophiolite (Bickle et al., 1994; Hamilton, 1998). The difficulty in recognizing and even defining ophiolites has been acknowledged widely and is not addressed here (Anonymous, 1972; and this volume). In this short contribution I outline some probable and some possible ophiolite sequences that have been reported from a number of Archean greenstone belts DOI: 10.1016/S0166-2635(04)13018-1

600

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

Fig. 1. Global map of Archean greenstone belts from which ophiolite-like sequences have been reported and described. See Table 1 for references.

around the world (Fig. 1 and Table 1). I also comment on the likely tectonic implications of these examples to better resolve Archean processes. A vast number of geochemical analyses indicate that oceanic-like (MORB-type) basalts and alpine-type peridotites occur in a large number of the studied greenstone belts. To access this, the reader is referred to Sylvester et al. (1997), O’Hanley (1997), and Arndt et al. (1997) for overviews.

2. EVIDENCE FROM GEOLOGY 2.1. Ophiolite Complexes and Slivers in Mélanges of the 2.55–2.05 Ga Central Orogenic Belt of the North China Craton Most recently the first laterally extensive end-Archean ophiolite complex (the 50 × 5 km2 Dongwanzi ophiolite, 2505 Ma) has been documented in the North China Craton (Kusky et al., 2001; and this volume). The complex is surrounded by a mélange of ophiolitic fragments, ranging from deformed mantle peridotite with podiform chromitites to pillow lavas, set in sheared metasediments and amphibolites (Li et al., 2002, 2004). The Dongwanzi complex is one of a series of tectonic blocks with ophiolitic affinities that outcrop within the Zunhua tectonic zone of the 1600 km-long Central Orogenic Belt that is flanked to the east by the Qinglong foreland fold-and thrust-belt. The northern section of the Central Orogenic Belt also hosts the Quingjang greenstone belt (Yuehua and Wang, 1997).

Greenstone Belt /Craton Kalgoorlie Greenstone Belt, Yilgarn Craton, Kanowna Lake and Lake Cowan localities

Reference

Distinguishing features

Age

Alternative interpretations

Neoarchean 2675–2715 Ma

Proposed tectonic setting Extensional oceanic environment

Fripp and Jones, 1997

Basaltic-komatiitic sheeted dykes, sills with ophiolitic-type peridotite-gabbroic plutonic sequence. Pillow lavas, cherts. Contacts between lithologies are tectonic. Ductile magmatic fabrics in gabbros

Yellowknife Greenstone Belt, Slave Craton

Helmsteadt et al., 1986

Mafic sheeted dykes, microgabbros with contacts exposed. No ultra-mafics, no komatiites. MORB-like pillow sequences. Lower Kam Group (Chan Fm) locally resembles ophiolite sheeted duke complex. No komatiites Ophiolite allochthons, pillows lavas, dykes, gabbros, ultramafics

Neoarchean 2716–2686 Ma

Back-arc basin or proto-oceanic rift

Bickle et al., 1994

Point Lake Belt

Kusky, 1991

Neoarchean

Imbricated forearc thrust over continental crust

King and Helmsteadt, 1997

Cameron River Belt, Slave Craton

Kusky, 1990

Mafic pillow lavas, dyke-gabbro transition, sill complexes

Neoarchean

Possible oceanic crust of back-arc basin

2. Evidence from Geology

Table 1. Some Archean Greenstone Belts with ophiolite-like mafic-ultramafic assemblages

King and Helmsteadt, 1997; Corcoran et al., 2004 (continued on next page)

601

602

Table 1. (Continued) Reference

Distinguishing features

Age

Proposed tectonic setting Fore-arc supra-subduction zone or slivers of oceanic crust now in accretionary prism

Kusky et al., 2001; and this volume; Li et al., 2002; Yuehua and Wang, 1997

Complete “Penrose-type” ophiolite metamorphosed at amphibolite facies. Includes pillow lavas, sheeted dykes, gabbros and ultra-mafic complexes and harzburgitic tectonites. Part of the Central Orogenic Belt that includes the Shenyang greenstone belt and other ophiolite complexes. No komatiites

Neoarchean 2505 Ma

Cleaverville Greenstone Belt, NW Pilbara Craton

Otha et al., 1996; Barley, 1997

Series of duplexes of pillow basalts and diabases with MORB geochemistry overlain by cherts

3.1 Ga

Tectonic slides of upper oceanic crust

South Pass Greenstone Belt Wind River Range Wyoming Craton

Harper, 1985; Wilks and Harper, 1997

Metamorphosed pillow lavas, diabases, gabbros and ultra-mafics. Highly strained and disrupted by shear zones. Possible dismembered ophiolite

Mesoarchean 2800–2630 Ga

Possible oceanic crust of a back-arc basin

Alternative interpretations Zhai et al., 2002

(continued on next page)

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

Greenstone Belt /Craton Dongwanzi Complex Zunkuan mélange of the Central Orogenic Belt North China Craton

Greenstone Belt /Craton Barberton Greenstone Belt Kaapvaal Craton

Reference

Distinguishing features

Age

Proposed tectonic setting 1. Fore-arc supra-subduction complex 2. Oceanic basin— back-arc basin 3. Fore-arc supra-subduction zone with small mélanges

de Wit et al., 1987; Brandl and de Wit, 1997; Grove et al., 1997; Dann, 2001

3 ophiolitic allochthons of different ages of which 2 host komatiites. Pillow lavas and sheet flows, sheeted intrusions (mostly sills), peridotites, wehrlites, dunites, chromitites. Overlain by cherts and turbidites

Mesoarchean 1. 3480–3470 Ma 2. 3440–3340 Ma 3. 3300–3220 Ma

Pietersburg Greenstone Belt Kaapvaal Craton

de Wit et al., 1992; Brandl and de Wit, 1997

Pillow lavas, diabases, gabbros, ultra-mafics. Occasional komatiites. Sequence is tectonically disrupted into a number of Neoarchean thrust slices

Mesoarchean to Neoarchean 3450–2700 Ma

Oceanic back-arc basin, now in accretionary wedge and foreland basin

Isua Supracrustal belt, Greenland North Atlantic Craton

Maruyama et al., 1991; Myers, 2001

Pillow lavas, gabbros, ultramafics, BIF turbidites, felsic volcanics, calc-silicates. Highly deformed and metamorphosed

Paleoarchean

Oceanic crustal allochthons in subduction-related accretion complex

Alternative interpretations Lowe and Byerly, 1999; Dann, 2001

2. Evidence from Geology

Table 1. (Continued)

Nutman, 1997

(continued on next page)

603

604

Table 1. (Continued) Reference

Distinguishing features

Age

Proposed tectonic setting Immature oceanic arc or forearc supra-subduction zone ophiolite

Shchipansky et al., 2004

Boninites, pillow lavas, sheeted dikes, gabbro

2.8 Ga

Olondo greenstone belt, Aldan Shield

Puchtel, 2004

Komatiitic and tholeiitic pillow basalts, sheeted sill complex, gabbro, cumulate ultramafic rocks

3.0 Ga

Supra-subduction zone ophiolite

Wutai Shan, North China craton

Li et al., 2004

Metabasalts, chert, black smoker chimneys, gabbro

2.5 Ga

Forearc ophiolite fragment

Alternative interpretations

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

Greenstone Belt /Craton North Karelian terrane, Baltic Shield

2. Evidence from Geology

605

Thus, this orogenic belt contains mafic-ultramafic remnants that have traditionally been documented as an Archean greenstone belt, as well as tectonic slivers of ophiolites. From all accounts this is the first regional tectonic mélange zone comparable to modern accretionary prisms and trench complexes, until now assumed to be absent from the Archean (e.g., Hamilton, 1998). Although some of the field interpretations have been challenged (Zhai et al., 2002) there is little doubt that the components of a true ophiolite sequence (cf. Anonymous, 1972) are present over a great aerial extent of the Central Orogenic Belt and that now require mapping at the kind of scales (< 1:15,000) that will allow more detailed comparisons with modern ophiolites and possibly modern oceanic crust. 2.2. Sheeted Dykes of the Yellowknife Greenstone Belt, Slave Craton The Kam Group of the circa 2.7 Ga Yellowknife greenstone belt is probably best known for its extensive sheeted dyke complex that displays a gradual transition onto overlying tholeiitic pillow lavas and grades downward into isotropic fine-grained metagabbros (Helmsteadt et al., 1986; King and Helmsteadt, 1997; Bickle et al., 1994). The base of the sequence is not clearly observed, but likely consists of a shear zone at the top of the underlaying BIF/rhyolites and quartzites (e.g., the Dwyer Formation; personal observations; Kusky, 1987). This suggests that the upper part of an ophiolite sequence is exposed here. The sequence has many characteristics of the Mesozoic Rocas Verdes ophiolites (that also lack an exposed ultramafic base) in the southern Andes, which were formed in a back arc environment, and initially emplaced into the granitic roots of an active volcanic arc (Stern and de Wit, 2003). Other possible slices of dismembered ophiolitic sequences have been suggested to occur in at least two other greenstone belts of the Slave Craton (e.g., the Cameron River and the Point Lake Greenstone Belts; Kusky, 1990, 1991; King and Helmsteadt, 1997; Corcoran et al., 2004). 2.3. Sheeted Dykes and Gabbro-Peridotite Sequence of the Yilgarn Craton Fripp and Jones (1997) propose an ophiolitic origin for some of the greenstones of the Kalgoorlie terrane. Their detailed mapping (at scales of 1:1000 and 1:2500) in two wellexposed lake section of the southern part of the Kalgoorlie greenstone belt has revealed a complex of sheeted dykes and sills of both high-Mg and possibly dunitic (komatiite) compositions, as well as an extensive ophiolite-like peridotite to gabbroic plutonic sequence with igneous ductile-deformation fabrics. One cliff section on Island 2 of Lake Cowan displays a continuous 25 m-wide section of sheeted intrusive units with both one-way and two-way chilling. The ophiolitic units are believed to be part of regional greenstone assemblages throughout the Kalgoorlie terrane. Evidence for an earlier granitic basement on which these greenstones may have been deposited (and which may have been responsible for inherited continental chemical and mineralogical contamination of the mafic rocks) is thought to be unlikely. Seismic reflection data suggests that the greenstones are allochthonous. This in turn suggests that small contamination (of for example zircons) may have been derived from assimilation of associated turbidites. Fripp and Jones (1997, p. 435)

606

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

suggest that many of the basaltic and komatiitic rocks of the generally poorly exposed Kalgoorlie (and surrounding) terranes, may represent more sheeted intrusions than is recognized at present, and that many of the so-called ultramafic layered complexes may comprise components of Archean oceanic lower crust and the transition zone to the upper mantle. These suggestions certainly warrant a re-evaluation of the geology of many greenstone belts within the Yilgarn Craton. 2.4. Ophiolite Slivers of the Barberton Greenstone Belt, Kaapvaal Craton The Barberton greenstone belt comprises at least 3 major allochthonous thrust sheets with predominantly mafic-ultramafic rock assemblages (Fig. 2; Brandl and de Wit, 1997; de Wit et al., in preparation; but see Lowe, 1999, for a more autochthonous interpretation). Maficultramafic sections of the Barberton greenstone belt have been interpreted in the past as remnants of ophiolites (Annhaeusser et al., 1968) and in particular an ophiolite representative of oceanic or back-arc basin crust, named the Jamestown Ophiolite Complex (JOC; de Wit et al., 1987). New mapping and geochronology since then have cast significant doubts on this interpretation, in part because different components of the restored JOC sections are now known to comprise mafic-ultramafic sections of different ages. For example, some Alpine-type ultramafics (dunites and harzburgites, including dunitic tectonites) were restored as the lowest sequence of the JOC, despite the fact that they were separated from the upper sequence by a major shear zone (de Wit et al., 1987). Precise geochronology has since shown that these cumulates and tectonites belong to a maficultramafic sequence of a separate tectonic block (Kaapvalley allochthon, Fig. 2) flanking the NW margin of the greenstone belt that is ∼ 230 million years younger than the upper ophiolitic section with which they were previously linked (the Onverwacht allochthon, Fig. 2; de Wit et al., 1987; de Ronde and de Wit, 1994). In addition, the upper sequences of the JOC (that include the type Komati and Hooggenoeg Formations) do not contain a sheeted dyke section as was previously suggested by de Wit et al. (1987). Composite, and sometimes sheeted mafic-ultramafic intrusions, particularly massive tholeiites, wehrlites and pyroxenites are common in the lower parts of the section. de Wit et al. (1987) suggested that these were dykes, but they now appear to be mostly hypabyssal sills (Dann, 2000, 2001), although some extensive dykes are also present. The volume and relative age of the sills remains a matter of controversy (de Wit et al., 1987; Grove et al., 1997; Dann, 2000, 2001). Dann (2000, 2001) suggests that most of the sequence is extrusive and that the intrusives may be significantly younger than their host rocks. In contrast, de Wit et al. (1987) believe a significant proportion of the massive rocks are intrusives, near contemporaneous with the high-Mg basaltic pillow lavas. The chemistry of relict pyroxenes in some of the komatiites, when compared to those tested in experimental work, is compatible with such an intrusive interpretation (Parman et al., 1997, 2001, 2003; Parman and Grove, 2004). Recent geochemical and petrological work has shown that some of the associated komatiites have boninitic affinities and were probably generated during hydrous melting of depleted mantle in an arc-forearc environment (Grove et al., 1999; Parman et al., 2001, 2003; Grove and Parman, 2004; Parman and Grove, 2004). Thus the

2. Evidence from Geology

Fig. 2. Geological map of the Barberton Greenstone Belt, showing the 3 main allochthonous mafic-ultramafic sequences with ophiolite-like characteristics. Inset shows part of the Kromberg allochthon with the plutonic mafic-ultramafic sheeted sill intruding the mainly tholeiitic pillow lava-chert sequence. The sequence is about 2–3 km thick. 607

608

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

Fig. 3. Close up photo of part of the sheeted sills of the Kromberg Allochthon, showing one- and two-way chilling in tholeiitic intrusive units. Outcrop along the Komati River, located in Fig. 2.

lowermost mafic-ultramafic sequences of the Barberton greenstone belt probably comprise the upper ∼ 3 km of a 3.48–3.47 Ga supra-subduction complex, likely of ophiolitic nature. Overlying the supra-subduction complex (part of the Onverwacht allochthon) maficultramafic rocks comprise part of the Kromberg allochthon (Fig. 2). The type section of this sequence occurs along the Komatii River in the southeastern sector of the belt and the age of this sequence is bracketed between 3.45 and 3.36 Ga (de Ronde and de Wit, 1994; Lowe and Byerly, 1999). New mapping and geochemistry indicates that this sequence predominantly comprises a complex array of intrusive wehrlites, pyroxenites, dunites and massive tholeiites separating screens of tholeiitic pillow basalts and breccias, interbedded with minor cherts and BIF (Fig. 2 inset; de Ronde and de Wit, 1994; de Wit et al., in preparation). The pyroxenites and massive tholeiites are part of a vast intrusive sheeted-sill complex. Sheeted intrusions varying between several meters across to less than 10 cm thick, with one- and two-way chilled margins are well exposed midway along the section in the Komatii river (Figs. 2 and 3). These sheeted intrusions were initially also interpreted as sheeted dykes (de Wit et al., 1987) but the subsequent mapping suggests that they probably also represent sheeted sills. Neither the top nor the bottom of the mafic-ultramafic sequence is preserved in this section. Large dunite and peridotite massifs mostly with tectonic boundaries, have been interpreted as Alpine-type peridotite complexes and large thicknesses of tholeiitic pillow lavas are ubiq-

3. Evidence from Mantle Xenoliths

609

uitous throughout the sequence away from the type area (Barton, 1982; Paris, 1985; de Wit et al., 1987). Komatiites are relatively rare and the predominant chemistry of the extrusive/hypabyssal sequence is tholeiitic (de Wit et al., 1987; Lowe and Byerly, 1999). Thus, the Kromberg allochthon comprises a separate sequence that contains most of the components of an ophiolite (ss) including a tholeiitic sheeted complex albeit in the form of sills rather than dykes. In all, the Barberton greenstone belt contains at least three mafic-ultramafic allochthons each of which preserve the “Steinmann Trinity” and at least one of which conforms closer to the more rigorous definition of an ophiolite. Only continued mapping in this type Mesoarchean greenstone belt will reveal just how close. 3. EVIDENCE FROM MANTLE XENOLITHS A significant number of studies on Archean eclogite and peridotite xenoliths found in kimberlites and related rocks, and derived from mantle lithosphere underlying Archean cratons have been equated on geochemical grounds to subducted and metamorphosed oceanic crust. It is beyond the scope of this contribution to explore these findings further, but the interested reader should consult recent findings of Shirey et al. (2004) and references therein. Suffice it to state that the data from these xenoliths suggests that significant volumes of Archean accreted oceanic lithosphere and oceanic arcs are preserved in the mantle “keels” of most Archean cratons (Kusky, 1993; Saltzer et al., 2001; Foley et al., 2003). 4. DISCUSSION AND CONCLUSION Archean greenstone belts contain a large amount of tholeiitic pillow basalts (Hunter and Stowe, 1997; de Wit and Ashwal, 1995). Comparable volumes on the contemporary Earth are confined to its oceans and their arcs, plateaux and islands. “The world would have had to be astonishingly different in the Archean for such giant accumulations of pillow lavas to have formed on continents” (Burke, 1997). Studies of ophiolites have helped to resolve important questions about Phanerozoic oceanic crust. But the structure of Archean oceanic crust remains elusive. Archean ophiolites are rarely recognized for what they are, possibly because they do not necessarily fit a standard definition and/or expected model. Kusky et al. (2001) resolve that the restored crustal section of the Dongwanzi ophiolite (∼ 10 km) is substantially thicker than that of Phanerozoic ophiolite sequences. In contrast, de Wit et al. (1987) claimed that the Archean Jamestown ophiolite complex appeared to represent a relatively thin crustal section. Unresolved tectonic repetition does not allow a robust reconstruction of the various Barberton ophiolitic allochthons, but the most reliable restoration reveals a thickness of around 6–8 km. Both estimates appear to be in violation of simple models that predict much thicker oceanic crust on an Archean Earth with higher mantle temperatures, perhaps as thick as below Iceland (∼ 25 km) where a plume has elevated the mantle temperatures beneath the mid Atlantic ridge by some 200 ◦ C, approaching the

610

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

sort of temperatures expected of the upper Archean mantle (Bickle et al., 1994; Pollack, 1997; Grove and Parman, 2004). Yet spreading across Iceland is slow in contrast to the predicted fast spreading rates in Archean oceans (Abbott and Hofmann, 1984; Pollack, 1997; de Wit, 1998; Karson, 2001). It is not clear from modern oceanic environments what sort of crustal structure might evolve from voluminous magmatism at very fast spreading rates (Karson, 2001) and/or lower viscosities perhaps due to higher water content of Archean mantle (cf. Grove et al., 1997, 1999; Parman et al., 2001, 2003). Features specific to maficultramafic sequences of Archean greenstone belts should help resolve this. A most obvious difference between Paleoproterozoic-Recent ophiolites and that of the mafic-ultramafic crustal section of the Barberton greenstone belt sequence (and that of most Archean greenstone belt sequences) is the scarceness of gabbroic rocks in the latter (de Wit et al., 1987; de Wit and Ashwal, 1997). Gabbros reflect the transient storage of mantle melts before they transfer their evolved magma to form upper oceanic crust. The absence of significant volumes of gabbros indicates that Archean mafic mantle melts spend little time in magma chambers to construct a middle section of oceanic crust. This complements the generally high-Mg content of many greenstone belt basalts. High-Mg melts rarely bypass the magma storage chambers of Phanerozoic ophiolites and present oceanic crust (Elton, 1979). It is plausible that these characteristics of Archean basalts represent more rapid extensional processes of Archean tectonic environments. A similar suggestion has been put forward for the near-ubiquitous occurrence of the enigmatic ocelli (or variolites) in the mafic-ultramafic rocks of greenstone belts. These features are only rarely described from Phanerozoic ophiolites or oceanic crust, and must be providing an important message about Archean mantle melting and crystallization (de Wit and Ashwal, 1997). One suggestion is that they represent “frozen-in” differentiation processes (de Wit et al., 1987). This interpretation provides additional evidence that intermediate stages of magma storage in the form of gabbroic plutons may have been less frequent in the Archean than in the crustal framework of present oceans and island arcs. How these observations may be incorporated into models of Archean oceanic crust formation remains to be resolved. What is clear, however, is that Archean ophiolite-like sequences do occur in some Archean greenstone belts and their restored internal structures must be explored to further test and challenge theoretical models that imply oceanic crust is not preserved in Archean greenstone belts. Since at least 80% of known Archean greenstone belts have not been studied in sufficient details to partake in the discussions, its back to field-work we should go.

ACKNOWLEDGEMENTS Over the last 25 years a number of geologists familiar with Phanerozoic ophiolite sequences have visited the Barberton greenstone belt with me and provided constructive feedback on my interpretations of the rock sequences there as being ophiolite-like. I am grateful to them for sharing their critical views, in particular: C. Stern, R.A. Hart, W.S.F. Kidd, A.G. Smith, H. Helmsteadt, T.L. Grove, J.C. Dann, S. Bowring, A. Hynes, M. Searle,

References

611

H. Furnes, H. Staudigal, N. Banerjee, K. Meuhlenbachs, B. Robbins (in chronological order of the visits). Although some were more convinced than others, I believe they all agreed that the sequences could well be ophiolitic, although this interpretation of their opinions remains mine. Over these years, funds to study this greenstone belt were provided through the FRD (now NRF), as well as a number of exploration companies, which is gratefully acknowledged.

REFERENCES Abbott, D.H., Hofmann, S.E., 1984. Archean plate tectonics revisited: 1. Heat flow, spreading rate and the age of the subducting oceanic lithosphere and their effects on the origin and evolution of the continents. Tectonics 3, 429–448. Annhaeusser, C.R., Roering, C., Viljoen, M.J., Viljoen, R.P., 1968. The Barberton Mountain Land: a model of the elements and evolution of an Archean fold belt. Transactions of the Geological Society of South Africa 71, 255–270. Anonymous, 1972. Ophiolites. Geotimes 17, 2–25. Arndt, N.T., Albarede, F., Nisbet, E., 1997. Mafic and ultramafic magmatism. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 233–254. Barley, M.E., 1997. The Pilbara Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 657–664. Barton, C.M., 1982. Geology and mineral resources of northeast Swaziland (Barberton greenstone belts). Geological Survey of Swaziland Bulletin 10, 97. Berhe, S.M., 1997. The Arabian-Nubian Shield. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 761–771. Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts are not oceanic crust. Journal of Geology 102, 121–138. Brandl, G., de Wit, M.J., 1997. The Kaapvaal Craton, South Africa. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 581–607. Burke, K., 1997. Foreword. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. v–vii. Corcoran, P.L., Mueller, W.U., Kusky, T.M., 2004. Inferred Ophiolites in the Archean Slave Province. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 363–404. Dann, J.C., 2001. Vesicular komatiites, 3.5-Ga Komati Formation, Barberton greenstone belt, South Africa: inflation of submarine lavas and origin of spinifex zones. Bulletin of Volcanology 63, 426–481. Dann, J.C., 2000. The Komatii Formation, Barberton greenstone belt, South Africa. Part 1. New map and magmatic architecture. South African Journal of Geology 103, 47–68. de Ronde, C.E.J., de Wit, M.J., 1994. Tectonic history of the Barberton greenstone belt, South Africa: 490 million years of Archean crustal evolution. Tectonics 13, 983–1005. de Wit, M.J., 1998. On Archean granites, greenstones, cratons and tectonics: does the evidence demand a verdict? Precambrian Research 91, 181–226. de Wit, M.J., Ashwal, L.D., 1995. Greenstone Belts: what are they? South African Journal of Geology 98, 505–520.

612

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

de Wit, M.J., Ashwal, L.D., 1997. Convergence towards divergent models of greenstone belts. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. ix–xvii. de Wit, M.J., Hart, R.A., Hart, R.J., 1987. The Jamestown ophiolite complex, Barberton mountain belt: as section through 3.5 Ga oceanic crust. Journal of African Earth Sciences 6, 681–730. de Wit, M.J., Jones, M.G., Buchanan, D., 1992. The geology and evolution of the Pietersburg greenstone belt, South Africa. Precambrian Research 55, 123–153. Elton, D., 1979. High magnesia liquids as parental magmas for ocean floor basalts. Nature 279, 514–517. Fripp, R.E.P., Jones, M.G., 1997. Sheeted intrusions and peridotite-gabbro assemblages in the Yilgarn craton, western Australia: elements of Archaean ophiolites. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 422–437. Foley, S.F., Buhre, S., Jacob, D.E., 2003. Evolution of the Archean crust by delamination and shallow subduction. Nature 421, 249–252. Grove, T.L., Parman, S.W., 2004. Thermal evolution of the earth as recorded by komatiites. Earth and Planetary Science Letters 6975, 1–15. Grove, T.L., de Wit, M.J., Dann, J.C., 1997. Komatiites from the Komati type section, Barberton, South Africa. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 438–445. Grove, T.L., Parman, S.W., Dann, J.C., 1999. Conditions of magma generation for Archean komatiites from the Barberton Mountainland, South Africa. In: Fei, Y., Bertka, C., Mysen, B. (Eds.), Mantle Petrology: Field Observations and High Pressure Experimentation, a Tribute to Francis R. (Joe) Boyd. The Geochemical Society Special Publication 6, 155–167. Hamilton, W.B., 1998. Archean magmatism and deformation were not products of plate tectonics. Precambrian Research 91, 109–142. Harper, G.D., 1985. Dismembered Archean ophiolite, Wind River Mountains, Wyoming. Ophioliti 10, 297–306. Helmsteadt, H., Padgham, W.A., Brophy, J.A., 1986. Multiple dykes in the lower Kam Group, Yellowknife greenstone belt: evidence for Archean sea-floor spreading? Geology 14, 562–566. Hunter, D.R., Stowe, C.W., 1997. A historical review of the origin, composition, and setting of Archean greenstone belts. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 5–30. Karson, J.A., 2001. Oceanic crust when earth was young. Science 292, 1076–1079. King, J., Helmsteadt, H., 1997. The Slave Province, North-West Territories, Canada. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 459–479. Kusky, T.M., 1987. Comment on multiple dikes in the lower Kam Group, Yellowknife Greenstone Belt: Evidence for Archean sea floor spreading? Geology 15 (3), 280–282. Kusky, T.M., 1990. Evidence for Achaean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10, 820–841. Kusky, T.M., 1993. Collapse of Archean orogens and the origin of late- to post-kinematic granitoids. Geology 21, 925–929. Kusky, T.M., Vearncombe, J.R., 1997. Structural aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 90–124. Kusky, T.M., Li, J.H., Tucker, R.D., 2001. The Archean Dongwanzi ophiolite complex, North China craton: 2505 billion year old oceanic crust and mantle. Science 292, 1141–1142.

References

613

Li, J.H., Kusky, T.M., Niu, X.L., Feng, J., Polat, A., 2004. Neoarchean massive sulfide of Wutai Mountain, North China: A black smoker chimney and mound complex within 2.50 Ga-old oceanic crust. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 339–361. Li, J.H., Kusky, T.M., Huang, X., 2002. Archean podiform chromitites and mantle tectonites in ophiolite mélange, North China craton: a record of early oceanic mantle processes. GSA Today 12, 4–11. Lowe, D.R., 1999. Geological evolution of the Barberton Greenstone Belt and vicinity. Geological Society of America Special Paper 329, 287–312. Lowe, D.R., Byerly, G.D. (Eds.), 1999. Geological Evolution of the Barberton Greenstone Belt. Geological Society of America Special Paper 329, p. 319. Myers, J.S., 2001. Protoliths of the 3.8–3.7 Ga Isua Greenstone Belt, West Greenland. Precambrian Research 105, 129–141. Maruyama, S., Masuda, T., Nohda, S., Appel, P., 1991. The earliest records on oceanic and continental crust from the 3.8 accretionary complex, Isua, Greenland. In: 29th International Geological Congress, Abstracts, Kyoto, Japan, p. 1239. Nutman, A., 1997. The Greenland sector of the north Atlantic Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 665–674. O’Hanley, D.S., 1997. Serpentinites and rodingites as records of metasomatism and fluid history. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 164–175. Otha, H., Maruyama, S., Takahashi, E., Watanebe, Y., Kato, Y., 1996. Field occurrence, geochemistry and petrogenesis of the Archean mid-oceanic ridge basalts (AMORB) of the Cleaverville area, Pilbara Craton, Western Australia. Lithos 37, 199–221. Paris, I., 1985. The geology of the farms Josesfal, Dunbar and part of the Diepgezet in the Barberton Greenstone Belt. Unpublished Ph.D. thesis. University of the Witwatersrand, South Africa. Parman, S.W., Grove, T.L., 2004. Petrology and geochemistry of Barberton komatiites and basaltic komatiites: evidence of Archean fore-arc magmatism. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 539–565. Parman, S.W., Grove, T.L., Dann, J.C., 2001. The production of Barberton komatiites in an Archean subduction zone. Geophysical Research Letters 28, 2513–2516. Parman, S.W., Grove, T.L., Dann, J.C., de Wit, M.J., 1997. Emplacement conditions of komatiite magmas from the 3.49 Ga Komati Formation, Barberton greenstone belt, South Africa. Earth and Planetary Science Letters 150, 303–323. Parman, S.W., Shimizu, N., Grove, T.L., Dann, J.C., 2003. Constraints on the pre-metamorphic trace element composition of Barberton komatiites from ion probe analyses of preserved clinopyroxene. Contributions to Mineralogy and Petrology 144, 383–396. Pollack, H.N., 1997. Thermal characteristics of the Archean. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 223–232. Puchtel, I.S., 2004. 3.0 Ga Olondo greenstone belt in the Aldan Shield, E. Siberia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 405–424. Saltzer, R.L., Chatterjee, N., Grove, T.L., 2001. The special distribution of garnets and pyroxenes in mantle peridotites: Pressure-temperature history of peridotites forms the Kaapvaal craton. Journal of Petrology 42, 2215–2229. Shchipansky, A.A., Samsonov, A.V., Bibikova, E.B., Babarina, I.I., Konilov, A.N., Krylov, K.A., Slabunov, A.I., Bogina, M.M., 2004. 2.8 Ga boninite-hosting partial suprasubduction zone ophi-

614

Chapter 18: Archean Greenstone Belts Do Contain Fragments of Ophiolites

olite sequences from the North Karelian greenstone belt, NE Baltic Shield, Russia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 425–486. Shirey, S.B., Richardson, S.H., Harris, J.W., 2004. Ages, parageneses, and compositions of diamonds and evolution of the Precambrian mantle lithosphere of southern Africa. South African Journal of Geology 107, 119–130. Sorjonen-Ward, P., Nironen, M., Luukkonen, E.J., 1997. The Baltic Shield: Greenstone associations in Finland. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 677–698. Stern, C.R., de Wit, M.J., 2003. The Rocas Verdes Ophiolites, southernmost South America: remnants of progressive stages of development of ocean-type crust in a continental margin back-arc basin. In: Dilek, Y., Robinson, P.T. (Eds.), Ophiolites in Earth History. Geological Society of London Special Publication 218. Sylvester, P.J., Harper, G.D., Byerly, G.R., Thurston, P.C., 1997. Volcanic aspects. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 55–90. Wilks, M.E., Harper, G.D., 1997. Wind River Range, Wyoming Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 508–516. Yuehua, Y., Wang, W., 1997. North China Craton. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Oxford Univ. Press, UK, pp. 730–735. Zhai, M., Zhao, G., Zhang, Q., 2002. Is the Dongwanzi Complex an Archean ophiolite? http:// www.sciencemag.org/cgi/content/full/295/5557/923a.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Published by Elsevier B.V.

615

Chapter 19

NORTHERN PHILIPPINE OPHIOLITES: MODERN ANALOGUES TO PRECAMBRIAN OPHIOLITES? JOHN ENCARNACIÓN Department of Earth and Atmospheric Sciences, Saint Louis University, Saint Louis, MO 63103, USA

The northern Philippines is a possible modern analogue for some Precambrian greenstone belts. It has a ∼ 150 Myr history of multiple and overlapping periods of oceanic crust generation, arc volcanism, sedimentation, and deformation dominated by wrench tectonics. At least five ophiolite complexes of distinct ages make up most of the basement—all having a distinct suprasubduction zone signature. Arc plutons are predominantly of the diorite-tonalite series with minor alkali-feldspar bearing rocks. Sedimentary basins probably floored by oceanic crust are dominated by immature sediments and volcaniclastics and are locally up to ∼ 10 km thick. The whole arc and ophiolitic complex is in the process of being accreted to Eurasia, where it may be preserved in a broad “suture zone” between Eurasia and Australia and/or the Americas. 1. INTRODUCTION The Philippine islands constitute a mature island arc complex comprised mainly of ophiolites, island arc plutons, volcanics and volcaniclastics, and thick sedimentary basins filled with immature sediments. The term “island arc complex” is used because what is now the Philippines is a composite of more than one arc magmatic belt due to a long history of subduction whose polarity has probably changed more than once along the western and eastern side of the islands (in the current reference frame). However, the available evidence does not support the presence of any suture within the Philippines east of the Manila-Negros trench collision, at least in the northern part of the islands. Hence, the various arc volcanoplutonic belts on the islands were probably not juxtaposed by arc collisions, but rather were generated above the same suprasubduction zone setting by inward subduction from the western side and eastern sides. These magmatic belts are built on a basement of oceanic crust (ophiolites) that is itself composite. Based on good biostratigraphic ages on overlying sedimentary rocks and modern dating techniques (zircon U-Pb and 40 Ar-39 Ar dating), it has been shown that there are at least five generations of ophiolitic basement in the northern Philippines spanning an age range of ∼ 150 Myr. Based on areas that have been mapped in more detail, it appears that the ophiolites are not allochthonous slices that have been tectonically amalgamated. A model consistent with the field relations is one where DOI: 10.1016/S0166-2635(04)13019-3

616

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Table 1. Name and label in Fig. 1 a Lagonoy ophiolite b

Calaguas Island ophiolite

c

Dibut Bay ophiolite

d

Casiguran ophiolite

e

Montalban ophiolite

f

Zambales-Angat ophiolite

g

Itogon ophiolite

References David et al. (1997), Fernandez et al. (1994), Geary (1986), Karig (1983), Tamayo et al. (1998) Geary (1986), Geary et al. (1988), Geary and Kay (1989), Giese et al. (1986), Knittel (1989), Mitchell and Balce (1990) Billedo et al. (1996), Hashimoto et al. (1978), Tejada and Castillo (2002) Anonymous (1977), Billedo et al. (1995), Billedo et al. (1996), Tamayo et al. (2001) Arcilla (1991), Arcilla et al. (1989), Encarnación et al. (1993), Encarnación et al. (1999), Haeck (1987), Karig (1983) Abrajano et al. (1989), Arcilla (1991), Arcilla et al. (1989), Bachman et al. (1983), Encarnación et al. (1999), Encarnación et al. (1993), Evans (1985), Evans et al. (1991), Evans and Hawkins (1989), Florendo and Hawkins (1992), Fuller et al. (1991), Geary et al. (1989), Hawkins and Evans (1983), Karig (1983), Rossman et al. (1989), Schweller et al. (1984), Yumul (1996), Yumul et al. (1998) Anonymous (1977), Anonymous (1987), Balce et al. (1980), Bellon and Yumul (2000), Encarnación et al. (1993), Florendo (1994), Mitchell and Balce (1990)

List of ophiolites labeled on Fig. 1 and references pertaining to them.

each episode of oceanic crust generation occurs adjacent to or within older basement as forearc, backarc, or intraarc seafloor spreading type process. It has been pointed out that the Philippine island arc complex is “reminiscent of Precambrian greenstone belts” (e.g., Hall, 1996). In this paper, I outline some of the more salient features of the northern Philippines and its ophiolitic basement. I focus on the northern Philippines because it has been studied in more detail and is, therefore, more well-known. However, there are no known fundamental differences between the northern and southern Philippines, therefore the broad outline of the geology presented here is probably applicable to the south as well. For a more detailed discussion of each individual ophiolite terrane, the reader is referred to the original papers (see Table 1) and to a recent review paper focusing on the details of each individual ophiolite and their relationships to each other (Encarnación, in press). 2. REGIONAL TECTONIC SETTING The Philippines is located in Southeast Asia in the western Pacific at the juncture of the Philippine Sea plate and Eurasian plate (Rangin, 1991). It is a mature island arc that is in the process of being accreted to the Eurasian margin by subduction-related convergence along the east-dipping Manila and Negros trenches. Subduction along the western side of

2. Regional Tectonic Setting

617

Fig. 1. Present tectonic setting of the Philippines. Rectangle outlines the location of Fig. 2. PF— Philippine Fault; MN—Mindanao; M—Mindoro; P—Panay; LZ—Luzon; NPB—North Palawan Block.

the northern Philippines is generating the Luzon volcanic arc, which consists of the stratovolcanoes in the southern Zambales range (where Mt. Pintubo is located) and volcanic centers running northward along the east side of the Central Cordillera to small islands north of Luzon and into Taiwan. The Luzon arc has collided with the Eurasian margin in Taiwan and in the central Philippines along the islands of Mindoro and Panay (Fig. 1) (McCabe et al., 1982). Presumably as a result of this collision, subduction may be waning on the west side of the Philippines and convergence between the Philippine Sea plate and the Eurasian plate may be increasingly taken up along the west dipping Philippine trench and East Luzon trough (Cardwell et al., 1980; Hamilton, 1979; Lewis and Hayes, 1983;

618

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Fig. 2. General geology of the northern Philippines. See Fig. 1 for location. Main ophiolite exposure: a—Lagonoy ophiolite; b—Calaguas Islands ophiolite; c—Dibut Bay ophiolite; d— Casiguran ophiolite; e—Montalban ophiolite; f—Zambales and Angat ophiolites; g—Itogon ophiolite. CTVB—Central Valley Basin; CAVB—Cagayan Valley Basin; MB—Marinduque Basin; CSC—Cordon syenite complex. Adapted from (Anonymous, 1963, 1977, 1981, 1991; Letouzey and Sage, 1988).

3. Major Geologic Elements of the Northern Philippines and Their Characteristics

619

Rangin, 1991). An active arc in southeast Luzon has formed due to subduction along the Philippine trench; no volcanic arc has yet developed from underthrusting at the East Luzon trough. The latter appears to be exploiting an old subduction zone because there is a mature accretionary prism and Eocene plutons and arc volcanics in the northern Sierra Madre as shown in Fig. 2 (Lewis and Hayes, 1983). Oblique northwest convergence of the Philippine Sea plate is being accommodated partly by sinistral wrench faulting along the Philippine Fault system (Aurelio et al., 1991). The Marinduque basin (Fig. 2) is a pull apart basin associated with wrench faulting and is floored by young oceanic crust (Sarewitz and Lewis, 1991). It formed by seafloor spreading type processes reflected in symmetric magnetic anomalies. It is a good actualistic example of formation of younger ophiolitic basement within older ophiolitic and arc basement.

3. MAJOR GEOLOGIC ELEMENTS OF THE NORTHERN PHILIPPINES AND THEIR CHARACTERISTICS Ophiolites are scattered throughout the northern Philippines as shown in Fig. 2. Many of them have been disrupted mainly by wrench faulting. All of the ophiolites variably preserve residual mantle peridotite (in the form of partly serpentinized harzburgite), gabbroic rocks and ultramafic cumulates, sheeted dikes or dike swarms, pillow lavas and pillow breccias. Modern dating by the zircon U-Pb and 40 Ar-39 Ar techniques coupled with traditional biostratigraphic work has shown that there are at least five generations of ophiolite preserved in the northern Philippines. The oldest is the Lagonoy ophiolite in southeast Luzon (Fig. 2), which has a minimum age of Jurassic (Geary et al., 1988). It comprises the oldest basement known in the Philippines east of the Manila trench-Negros trench convergent margin. It has a distinct suprasubduction zone signature and may be a primitive island arc. The Calaguas Islands, Dibut Bay, Casiguran, and Montalban ophiolites are constrained as pre-Late Cretaceous or Early Cretaceous (see Table 1 for references). Together they may comprise a large section of Early Cretaceous oceanic crust that may have formed as backarc basin lithosphere associated with the Jurassic Lagonoy ophiolite/primitive arc. An interesting relationship established by careful mapping and biostratigraphy in the Northern Sierra Madre is the occurrence of a sequence of pillow lavas of Late Cretaceous age with a distinct arc signature overlying the Early Cretaceous back-arc basin-like pillow lavas of the Casiguran ophiolite (Billedo et al., 1996) (Fig. 2). The Eocene Zambales ophiolite, in western Luzon (Fig. 2) has been the most intensely studied and is the largest exposed ophiolite in the Philippines. It has a distinct suprasubduction zone signature with boninites, MORB-like crust and transitional arc like crust (Hawkins and Evans, 1983). The Zambales ophiolite may be a part of a much larger Eocene ophiolitic basement that most likely extends eastward underneath the Central Valley of Luzon and into the Southern Sierra Madre where it crops out again as the Eocene Angat ophiolite. Fig. 3 is a cross section across central Luzon that shows the probable geometry of the basement and the known geology in the Zambales range and western side of the Southern Sierra Madre. The Eocene ophiolitic basement may also extend northward into

620

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Fig. 3. East-West section across Luzon from the Zambales Mountains, across the Central Valley Basin and into the Southern Sierra Madre (compare with Fig. 1). It is inferred that most of the section is underlain by an ophiolitic basement that crops out as the Zambales ophiolite in the west and the Angat ophiolite in the east. Section is based on data from several sources (Arcilla et al., 1989; Bachman et al., 1983; Encarnación et al., 1993; Haeck, 1987; Hawkins and Evans, 1983).

northern Luzon. Fig. 4 shows the inferred extend of the various ophiolitic basement in Luzon. A detailed discussion for the basis of this model is discussed elsewhere (Encarnación, in press). The Itogon ophiolite in the Central Cordillera is Oligocene in age and is thought to have been generated during intraarc rifting (Florendo, 1994). Oligocene and Miocene alkaline plutons and volcanics in the Cordon Syenite Complex (CSC, Fig. 2), are thought to have been related to the same intraarc rifting event. Some of the lavas and dikes associated with the Itogon complex are similar to magmatic rocks in the Sumisu Rift (Florendo, 1994). Geochemical data for the various ophiolites is of variable quality and coverage. As mentioned earlier, the Zambales ophiolite has been studied the most and has a wide variety of data available. Although clearly a single slab of oceanic lithosphere, it contains MORBlike, transitional arc tholeiite, and boninitic rocks typical of a suprasubduction zone setting. A detailed discussion of all the available geochemical data from all the ophiolites in the northern Philippines can be found in Encarnación (in press). Not surprisingly, all of the ophiolites have evidence for a suprasubduction zone origin (e.g., Hawkins and Florendo, 1992) primarily in the form of variable enrichment of large ion lithophile elements and depletion in high field strength elements. Overlying and intruding many of the ophiolites are arc volcanics/volcaniclastics and arc plutons, respectively (Figs. 2 and 5). Abundant Cretaceous arc volcanics and volcaniclastics of andesitic and dacitic composition are the first major manifestation of arc volcanism in the northern Philippines. Eocene and Oligocene volcanics are also widespread and abundant. Many of these arc rocks are metamorphosed to lower greenschist facies assemblages. They are largely submarine and along with the volcanic sections of the ophiolites have been mapped as “Cretaceous-Paleogene metavolcanics” in the older maps and literature. The large batholiths exposed in the Central Cordillera and Northern Sierra Madre and to

4. Analog for Archean and Proterozoic Systems?

621

Fig. 4. Possible extent of ophiolitic basement in the northern Philippines. Dashed lines—outline of northern Philippines. Solid lines—inferred location of contacts between ophiolites of various ages. Compare with Fig. 2. The extrapolation of ophiolitic basement beyond the main exposures shown in Fig. 2 is based on isolated smaller exposures, structural, isotopic, and seismic data (Encarnación, in press). The contacts between the ophiolites are probably “intrusive” (i.e., younger ophiolites form by extension adjacent to, or within older ophiolite basement), although possibly modified by later wrench faulting (Encarnación, in press; Encarnación et al., 1993; Karig, 1983; Karig et al., 1986). The dominant ophiolitic basement is intruded by arc plutons, partly covered by volcanics and volcaniclastics, and downfaulted or downwarped into deep sedimentary basins that are filled with sediments and volcaniclastics.

a lesser extent, the Southern Sierra Madre, are largely hornblende diorites, quartz diorites and tonalites. These plutons are poorly studied and few of them have any published chemical analyses. Although dominated by diorite-tonalite series plutons, minor syenites, granodiorites, monzonites and lamprophyres have also been described. Most of these more alkali rich rocks are found in the southern Central Cordillera.

4. ANALOG FOR ARCHEAN AND PROTEROZOIC SYSTEMS? The key characteristics of the northern Philippines that may be relevant to Precambrian ophiolites and greenstone belts are summarized in Fig. 5. Petrologically, the crust is dominated by ophiolitic lithologies that would be classified as “greenstones”. Much of the arc volcanics and volcaniclastics that overly the ophiolitic basement may be difficult to differentiate from the ophiolitic volcanic section and indeed they may certainly be transitional in space and time. In some areas, a significant portion of the crust is made up of

622

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Fig. 5. Cartoon illustrating some of the key features of northern Philippine ophiolites and related rocks. Figure corresponds roughly to a section from the Manila trench off of the southern Zambales Range northeast to the Baguio region and east to the East Luzon trough. Ages of oceanic crust are based on zircon U-Pb ages and biostratigraphic ages (Billedo et al., 1996; Encarnación et al., 1993). Ages of intrusives are zircon U-Pb ages and K-Ar ages (Anonymous, 1977; Bellon and Yumul, 2000; Encarnación et al., 1993; Florendo, 1994; Wolfe, 1981). Note that the ∼ 70 Ma pillow lavas are unconformable on the ∼ 110 Ma ophiolite basement (Billedo et al., 1996).

diorite-tonalite intrusions as in many greenstone belts (e.g., de Wit and Ashwal, 1997). The thick ophiolite-floored sedimentary basins are also another key feature and because of their thickness, are likely to be preserved in the geologic record. The span of geologic activity preserved in the Philippines, covering 150 Myr is comparable to the span of activity recorded in some greenstone belts (de Wit and Ashwal, 1997; Kusky and Polat, 1999). Ophiolite generation, arc magmatism, and sedimentary deposition all spanned this ∼ 150 Myr period and occurred in various places in northern Luzon at various times. If this complex package were preserved in an ancient mountain belt, any attempt to understand the geology in terms of a “layer cake” stratigraphy would lead to erroneous conclusions. Interpretation of structural relationships of volcano-sedimentary belts in Archean greenstone terranes might benefit from a more realistic, albeit more complex, model based on modern analogues (e.g., Kusky and Vearncombe, 1997). In order to use the Philippine ophiolitic terranes as analogues for Precambrian greenstone belts, we need to project current plate motions and kinematics forward in time and predict what the geometry of the major units will be upon final accretion to the Eurasian margin. The Philippines has already partly collided with Eurasia along the Mindoro and Panay area. Convergence continues with eastward subduction of the South China Sea basin and Sulu Sea basin. Once these two small marginal basins are completely consumed by subduction, the Philippine arc complex will have been completely accreted to the Eurasian margin. The final stages of collision may likely be a bit more complex because westward subduction along the Philippine trench has started and the central part of the Philippines is already locked along the North Palawan Block margin. Hence, the final stages of collision

References

623

will probably require opening up of extensional basins, with possible formation of oceanic crust to accommodate heterogeneous strain. For example, Pubellier et al. (1996) suggested that the Philippines is undergoing ‘escape tectonics’ which allow the archipelago to deform. The formation of the oceanic Marinduque basin (Sarewitz and Lewis, 1991) is partly related to this intradocking deformation phase. As the westward subduction on the east side of the Philippines develops further, the northern Philippines will be intruded by yet another suite of plutons and will be blanketed by young arc volcanics. If closure of the Pacific ocean continues, the whole assemblage will be incorporated in a broad suture zone between Eurasia and Australia and/or the Americas. It will then be presumably overprinted by compressional structures and perhaps finally intruded by a post-orogenic intrusive suite. Its overall petrological, structural, and sedimentological characteristics would then be quite evocative of some Precambrian granite-greenstone belts.

REFERENCES Abrajano, T.A., Pasteris, J.D., Bacuta, G.C., 1989. Zambales ophiolite, Philippines: I. Geology and petrology of the critical zone of the Acoje massif. Tectonophysics 168, 65–100. Anonymous, 1963. Geological Map of the Philippines. Philippine Bureau of Mines, Manila. Anonymous, 1977. Report on geological survey of northeastern Luzon, consolidated report. Japan International Cooperation Agency-Metal Mining Agency of Japan, Tokyo, p. 106. Anonymous, 1981. Geology and Mineral Resources of the Philippines. Bureau of Mines and Geosciences, Ministry of Natural Resources, Manila, p. 406. Anonymous, 1987. Geology and mineralization in the Baguio area, northern Luzon. 5. United Nations Department of Technical Cooperation for Development, Manila. Anonymous, 1991. Report on the Mineral Exploration, Mineral Deposits and Tectonics of Two Contrasting Geologic Environments in the Republic of the Philippines, Terminal Report. Japan International Cooperation Agency, Metal Mining Agency of Japan, Mines & Geosciences Bureau, R.P., Tokyo. Arcilla, C.A., 1991. Lithologic, age, and structural study of the Angat Ophiolite, Luzon, Philippines. M.Sc. thesis. University of Illinois, Chicago, p. 107. Arcilla, C.A., Ruelo, H.B., Umbal, J., 1989. The Angat ophiolite, Luzon, Philippines: lithology, structure, and problems in age interpretation. Tectonophysics 168, 127–135. Aurelio, M.A., Barrier, E., Rangin, C., Müller, C., 1991. The Philippine Fault in the Late Cenozoic tectonic evolution of the Bondoc-Masbate-N. Leyte area, Central Philippines. Journal of Southeast Asian Earth Sciences 6, 221–238. Bachman, S.B., Lewis, S.D., Schweller, W.J., 1983. Evolution of a forearc basin, Luzon Central Valley, Philippines. The American Association of Petroleum Geologists Bulletin 67, 1143–1162. Balce, G.R., Encina, R.Y., Momongan, A., Lara, E., 1980. Geology of the Baguio region and its implications on the tectonic development of the Luzon Central Cordillera. Geology and Palaeontology of Southeast Asia 21, 265–287. Bellon, H.a.Y., Yumul Jr., G.P., 2000. Mio-Pliocene magmatism in the Baguio mining district (Luzon, Philippines): age clues to its geodynamic setting. Comptes Rendus de l’Academie des Sciences Serie II Fascicule A—Sciences de la Terre et des Planetes 331, 295–302.

624

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Billedo, E., et al., 1995. The geology and structures of San Ildefonso Peninsula, Casiguran, Aurora Province: Their relationship to the Northern Sierra Madre orogeny. Journal of the Geological Society of the Philippines 50, 37–60. Billedo, E., Stephan, J.F., Delteil, J., Bellon, H., Sajona, F., Feraud, G., 1996. The pre-Tertiary ophiolitic complex of northeastern Luzon and the Polillo group of islands. Philippines. Journal of the Geological Society of the Philippines 51, 95–114. Cardwell, R.K., Isacks, B.L., Karig, D.E., 1980. The spatial distribution of Earthquakes, focal mechanism solutions, and subducted lithosphere in the Philippine and northeastern Indonesian Islands. In: Hayes, D.E. (Ed.), The Tectonic and Geologic Evolution of Southeast Asian Seas and Islands. In: Geophysical Monographs, vol. 23. American Geophysical Union, Washington, DC, pp. 1–35. David, S.J., et al., 1997. Geology and tectonic history of Southeastern Luzon, Philippines. Journal of Asian Earth Sciences 15, 435–452. de Wit, M., Ashwal, L.D., 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, p. 809. Encarnación, J., in press. Multiple ophiolite generation preserved in the northern Philippines and the growth of an island arc complex. Tectonophysics (special issue on “Ophiolites and Continental Margins of the Pacific Rim and Caribbean Region”). Encarnación, J., Mukasa, S.B., Evans, C., 1999. Subduction components and the generation of arclike melts in the Zambales ophiolite, Philippines: Pb, Sr, Nd isotopic constraints. Chemical Geology 156, 343–357. Encarnación, J.P., Mukasa, S.B., Obille, E.J., 1993. Zircon U-Pb geochronology of the Zambales and Angat ophiolites, Luzon, Philippines: Evidence for Eocene arc-back arc pair. Journal of Geophysical Research 98, 19,991–20,004. Evans, C.A., 1985. Magmatic ‘metasomatism’ in peridotites from the Zambales ophiolite. Geology 13, 166–169. Evans, C.A., Castaneda, G., Franco, H., 1991. Geochemical complexities preserved in the volcanic rocks of the Zambales ophiolite, Philippines. Journal of Geophysical Research 96, 16251–16262. Evans, C.A., Hawkins, J.W., 1989. Compositional heterogeneities in upper mantle peridotites from the Zambales range ophiolite, Luzon, Philippines. Tectonophysics 168, 23–41. Fernandez, M.V., Revilla, A.P., David, S.J., 1994. Notes on the Cretaceous carbonates in Catanduanes Island and Caramoan Peninsula. Journal of the Geological Society of the Philippines 49, 241–261. Florendo, F.F., 1994. Tertiary intra-arc rifting in the northern Luzon terrane, Philippines. Tectonics 13, 623–640. Florendo, F., Hawkins, J.W., 1992. Comparisons of the geochemistry of volcanic rocks of the Zambales ophiolite, northern Luzon, Philippines, implications for tectonic setting. Acta Geologica Taiwanica 30, 172–176. Fuller, M., Haston, R., Lin, J., Richter, B., Schmidtke, E., Almasco, J., 1991. Tertiary paleomagnetism of regions around the South China Sea. Journal of Southeast Asian Earth Sciences 6, 161–184. Geary, E.E., 1986. Tectonic significance of basement complexes and ophiolites in the northern Philippines: results of geological, geochronological and geochemical investigations. Ph.D. thesis. Cornell University, Ithaca, p. 221. Geary, E.E., Harrison, T.M., Heizler, M., 1988. Diverse ages and origins of basement complexes, Luzon, Philippines. Geology 16, 341–344. Geary, E.E., Kay, R.W., 1989. Identification of an Early Cretaceous ophiolite in the Camarines NorteCalaguas Islands basement complex, eastern Luzon, Philippines. Tectonophysics 168, 109–126.

References

625

Geary, E.E., Kay, R.W., Reyonolds, J.C., Kay, S.M., 1989. Geochemistry of the mafic rocks from the Coto Block, Zambales ophiolite, Philippines, trace element evidence for two stages of crustal growth. Tectonophysics 168, 43–63. Giese, U., Knittel, U., Kramm, U., 1986. The Paracale Intrusion: geologic setting and petrogenesis of a trondhjemite intrusion in the Philippines island arc. Journal of Southeast Asian Earth Sciences 1, 235–245. Haeck, G.H., 1987. The geologic and tectonic history of the central portion of the Southern Sierra Madre, Luzon, Philippines. Ph.D. thesis. Cornell University, Ithaca. Hall, R., 1996. Reconstructing Cenozoic SE Asia. In: Hall, R., Blundell, D. (Eds.), Tectonic Evolution of Southeast Asia. Geological Society Special Publications 106, 153–184. Hamilton, W., 1979. Tectonics of the Indonesian Region. U.S. Geological Survey Professional Paper 1078. Hashimoto, W., Aoki, N., David, P.P., Balce, G.R., Alcantara, M., 1978. Discovery of Nummulites from the Lubingan crystalline schist exposed east of Bongabon, Nueva Ecija, Philippines and its significance on the geologic development of the Philippines. Geology and Palaeontology of Southeast Asia 19, 57–63. Hawkins, J.W., Evans, C.A., 1983. Geology of the Zamabales range, Luzon, Philippines Islands: ophiolite derived from an island arc-back arc basin pair. In: Hayes, D.E. (Ed.), Tectonics and Geologic Evolution of Southeast Asian Seas and Islands. American Geophysical Union Geophysical Monographs 27 (2), 95–123. Hawkins, J.W., Florendo, F., 1992. Supra-subduction zone magmatism: implications for the origin of Philippine ophiolites. Acta Geologica Taiwanica 30, 163–171. Karig, D.E., 1983. Accreted terranes in the northern part of the Philippine archipelago. Tectonics 2, 852–855. Karig, D.E., Sarewitz, D.R., Haeck, G.D., 1986. Role of strike-slip faulting in the evolution of allochthonous terranes in the Philippines. Geology 14, 852–855. Knittel, U., 1989. Comment on “Diverse ages and origins of basement complexes, Luzon, Philippines”. Geology 17, 669. Kusky, T., Polat, A., 1999. Growth of granite-greenstone terranes at convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73. Kusky, T., Vearncombe, J.R., 1997. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 95–127. Letouzey, J., Sage, L., 1988. Geological and Structural Map of Eastern Asia. American Association of Petroleum Geologists, Tulsa. Lewis, S.D., Hayes, D.E., 1983. The tectonics of northward propagating subduction along eastern Luzon, Philippine islands. In: Hayes, D.E. (Ed.), The Tectonic and Geologic Evolution of Southeast Asian Seas and Islands. American Geophysical Union Geophysical Monograph 27 (2), 57–78. McCabe, R., Almasco, J., Diegor, W., 1982. Geologic and paleomagnetic evidence for a possible Miocene collision in western Panay, central Philippines. Geology 10, 325–329. Mitchell, A.H.G., Balce, G.R., 1990. Geological features of some epithermal gold systems, Philippines. Journal of Geochemical Exploration 35, 241–296. Pubellier, M., Quebral, R., Aurelio, M., Rangin, C., 1996. Docking and post-docking escape tectonics in the southern Philippines: Tectonic Evolution of Southeast Asia. In: Hall, R., Blundell, D. (Eds.), Tectonic Evolution of Southeast Asia. Geological Society of London Special Publication 6, 511– 523. Rangin, C., 1991. The Philippine mobile belt: a complex plate boundary. Journal of Southeast Asian Earth Sciences 6, 209–220.

626

Chapter 19: Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites?

Rossman, D.L., Castaneda, G.C., Bacuta, G.C., 1989. Geology of the Zambales ophiolite, Luzon, Philippines. Tectonophysics 168, 1–22. Sarewitz, D.R., Lewis, S.D., 1991. The Marinduque intra-arc basin, Philippines: Basin genesis and in situ ophiolite development in a strike-slip setting. Geological Society of America Bulletin 103, 187–203. Schweller, W.J., Roth, P.H., Karig, D.E., Bachman, S.B., 1984. Sedimentation history and biostratigraphy of ophiolite-related Tertiary sediments, Luzon, Philippines. Geological Society of America Bulletin 95, 1333–1342. Tamayo, A.R.J., et al., 2001. Preliminary geochemical and mineral data from the Isabela-Aurora ophiolite, northeastern Luzon, Philippines. InterRidge News 10 (2), 50–53. Tamayo, A.R.J., Yumul, G.P.J., Santos, R.A., Jumawan, F., Rodolfo, K.S., 1998. Petrology and mineral chemistry of a back-arc upper mantle suite: example from the Camarines Norte ophiolite complex, south Luzon. Journal of the Geological Society of the Philppines 53 (1–2), 1–23. Tejada, M.L.G., Castillo, P.R., 2002. In search of common ground: Geochemical study of ancient oceanic crust in eastern Philippines. In: Goldschmidt 2002. Cambridge Publications, Davos, Switzerland, p. A767. Wolfe, J.A., 1981. Philippine geochronology. Journal of the Geological Society of the Philippines 20, 1–30. Yumul, G.P.J., 1996. Varying mantle sources of supra-subduction zone ophiolites: REE evidence from the Zambales ophiolite complex, Luzon, Philippines. Tectonophysics 262, 243–262. Yumul, G.P.J., Dimalanta, C.B., Faustino, D.V., de Jesus, J.V., 1998. Upper mantle-lower crust dikes of the Zambales ophiolite complex (Philippines): distinct short-lived, subduction-related magmatism. Journal of Volcanology and Geothermal Research 84, 287–309.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

627

Chapter 20

THE RESURRECTION PENINSULA OPHIOLITE, MÉLANGE AND ACCRETED FLYSCH BELTS OF SOUTHERN ALASKA AS AN ANALOG FOR TRENCH-FOREARC SYSTEMS IN PRECAMBRIAN OROGENS TIMOTHY M. KUSKYa , ROSE GANLEYa, JENNIFER LYTWYNb AND ALI POLATc a Department

of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO 63103, USA b Department of Geosciences, University of Houston, Houston, TX 77058, USA c Department of Geology, University of Windsor, Ontario, Canada

Southern Alaska’s Mesozoic-Cenozoic Chugach-Prince William terrane is an unusual forearc in that it contains belts of graywacke-dominated flysch, mélange, and ophiolitic fragments all intruded by a suite of tonalite-trondhjemite-granodiorite plutons, and large parts of the accretionary prism are metamorphosed to the greenschist, amphibolite, or granulite facies. The overall structural geometry, abundance and types of rocks and rock suites present, the petrogenetic relationships between rock suites, and the metamorphic style are all strongly reminiscent of Archean granite-greenstone terranes. As such, the southern Alaska forearc represents one of the world’s best modern analogs to early stages in the evolution of Archean granite-greenstone terranes. In this contribution, we examine the regional geology of the flysch, mélange, and accreted ophiolites, as well as details of the geology of the 57 ± 1 Ma Resurrection Peninsula ophiolite of southern Alaska’s Chugach terrane as a remarkable analog to some Archean greenstone belts. The Resurrection ophiolite formed in a near-trench environment as the Kula-Farallon ridge was being subducted beneath North America. The magmatic sequence includes pillow lavas, sheeted dikes, gabbros, trondhjemites, and a poorly-exposed ultramafic section. The lavas show mid-ocean ridge basalt and arc-like geochemical signatures, interpreted to reflect compositionally diverse melts derived from near-fractional melting of a variably depleted mantle source, mixed with variable amounts of assimilated continentally-derived flysch. A sedimentary sequence overlying the ophiolite preserves a continuous record of turbidite sedimentation deposited on the ophiolite as it was transported to North America and emplaced in the Chugach accretionary prism. The top of the sedimentary section is truncated by the Fox Island shear zone, a 1-km thick, greenschist-facies, west-over-east thrust related to the emplacement of the ophiolite into the accretionary wedge. The Fox Island shear zone is intruded by a 53.4 ± 0.9 Ma granite, showing that the ophiolite formed, was transported to DOI: 10.1016/S0166-2635(04)13020-X

628

Chapter 20: The Resurrection Peninsula Ophiolite

the North American continent, overthrust by a major accretionary prism-related thrust, and intruded by granite all within 3.6 ± 1.4 Ma. Geological relationships in the southern Alaska forearc are instructive, in that if similar relationships were found in an Archean granite-greenstone terrane, they would probably currently be interpreted to reflect calc-alkaline mafic-felsic volcanic-plutonic complexes intruded and erupted through a complex metasedimentary sequence. As such, the belt would probably be interpreted as an arc sequence. Many Precambrian forearc ophiolites and accretionary prisms may have gone unrecognized because the processes of forearc ophiolite emplacement and intrusion by near-trench magmas at triple junctions has been poorly documented. The near absence of subduction mélanges in Archean terranes has been used to argue that accretionary wedges are not present in Archean granite-greenstone terranes. Models for Archean plate tectonics predict that the planet may have had a greater plate boundary length, with more, smaller, faster moving plates than at present. One consequence of such a model is that subducting plates would have tended to be thickly sedimented, with their proximity to active collisions along numerous plate boundaries. Subduction of these thickly sedimented plates would tend to produce Archean accretionary wedges dominated by relatively coherent terranes and few mélanges, consistent with the geological record preserved in Archean granite greenstone terranes. 1. INTRODUCTION AND REGIONAL GEOLOGY Southern Alaska is composed of a series of belts of Paleozoic, Mesozoic, and Cenozoic age (Fig. 1a) (Plafker et al., 1994). The Wrangellia composite terrane north of the Border Ranges fault system (Fig. 1a) includes intraoceanic Paleozoic and Mesozoic magmaticarc assemblages of the Peninsular, Wrangellia, and Alexander terranes, locally overlain by thick carbonate successions (Plafker et al., 1989, 1994). The Southern Margin composite terrane to the south is an accretionary complex formed as a result of late Mesozoic and Cenozoic subduction and offscraping of sediments derived largely from the Wrangellia terranes to the north (Plafker et al., 1994). Included in this composite terrane are the Chugach, Prince William, and Yakutat terranes of the Prince William Sound area, the Ghost Rocks Formation on Kodiak Island, and Chugach metamorphic complex in the eastern Chugach Mountains (Fig. 1). The Aleutian thrust fault system in the south separates the Southern Margin composite terrane from the Pacific plate (Plafker et al., 1994). The Chugach-Prince William terrane (Fig. 1a) is a complexly deformed Mesozoic and Cenozoic accretionary prism (Plafker et al., 1977; Tysdal et al., 1977; Tysdal and Case, 1979; Nelson et al., 1985; Bradley et al., 1994, 1999a (with marginal notes, http://wrgis.wr.usgs.gov/open-file/of99-18/); Pavlis and Sisson, 1995; Kusky et al., 1997a, 1997b; Kusky and Bradley, 1999) that is up to 100 km wide along the Gulf of Alaska and extends along strike for over 2000 km along the Gulf of Alaska continental margin from Baranof Island in southeastern Alaska to the Sanak and Shumagin Islands in the southwest. The inboard part of the Chugach accretionary prism, the McHugh Complex (Fig. 1b), is a Permian-Cretaceous mélange, whereas the further outboard

1. Introduction and Regional Geology

629

Fig. 1. (a) Map of southern Alaska showing locations of the Chugach, Prince William, and Wrangellian terranes and other tectonic elements discussed in text. (b) Map of the Kenai Peninsula showing locations of the Resurrection Peninsula and Knight Island ophiolites.

630

Chapter 20: The Resurrection Peninsula Ophiolite

part is a belt of Upper Cretaceous deformed flysch of the Valdez Group representing deep-sea fan turbidites (Fig. 1b) (Nelson and Nelson, 1993; Bradley and Kusky, 1992; Plafker et al., 1994; Kusky et al., 1997a; Kusky and Bradley, 1999). Outboard of the Valdez Group is a belt of somewhat less deformed Paleocene to Eocene flysch of the Orca Group of the Prince William terrane (Fig. 1b) (Plafker, 1969; Tysdal and Case, 1979; Jones et al., 1981; Plafker et al., 1985a, 1985b; Nelson et al., 1987). The geology of the Gulf of Alaska region includes the two most outboard units of the accretionary wedge deposits exposed above sea level, the Valdez and Orca Groups. These units are similar in lithology as both are composed of graywacke, siltstone, and shale, and local conglomerates and volcanic rocks. The dominant depositional mode was by gravity flow as turbidity currents moving westward along the trench axis (Winkler, 1976; Nilsen and Zuffa, 1982), although some east and northeastward directed paleocurrents have been documented as well (Plafker et al., 1994; Kusky et al., 1997b; D. Bradley, personal communication, 1999a, 1999b). In general, both the Valdez and Orca Groups become richer in quartz, potassium feldspar, and plagioclase, and poorer in volcanic lithic fragments to the east along the margin (Dumoulin, 1987). These rocks are interpreted to represent Late Cretaceous to Paleocene slope, fan, basin plain, and trench-fill turbidites (Nilsen and Zuffa, 1982) or deep-sea sedimentary fans later transported during accretion (Clendenin, 1991; Plafker et al., 1994). Both the Valdez and Orca Groups are variably folded with slatey cleavage developed in the finer grained rocks (Tysdal et al., 1977). Metamorphism from zeolite to low-greenschist facies is widespread but upper greenschist to amphibolite facies occurs locally near shear zones and as contact metamorphism around intrusions (Tysdal and Case, 1979; Moore et al., 1983; Plafker et al., 1989; Sisson et al., 1989; Bol and Gibbons, 1992). The Contact fault (Fig. 1b) is mapped as a major terrane boundary separating the Late Cretaceous rocks of the Valdez Group from Paleocene to Eocene rocks of the Orca Group (Tysdal and Case, 1979; Plafker et al., 1985a, 1985b). In eastern Prince William Sound the Contact fault separates poly-deformed, strongly foliated, middle greenschist facies rocks of the Valdez Group from less-deformed and metamorphosed rocks of the Orca Group. However, in the western Gulf of Alaska, the location of the boundary is obscured by the absence of an abrupt change in composition and metamorphic grade (Dumoulin, 1987, 1988; Gilbert et al., 1992). Sparse fossil control on ages along a belt approximately 60 km wide through the region add further ambiguity in locating a position for the Contact fault (Jones and Clark, 1973; Tysdal et al., 1977; Plafker et al., 1985a, 1985b; Nelson et al., 1985; Bol and Roeske, 1993; Kusky and Young, 1999). Several partial and dismembered ophiolite sequences outcrop in the western Gulf of Alaska and Prince William Sound, the most notable of which are the Knight Island (Fig. 2) and Resurrection Peninsula ophiolites (Fig. 3). The Paleogene Resurrection and Knight Island ophiolites are relatively undeformed and are the most complete ophiolite sequences in Alaska (Nelson et al., 1987; Plafker et al., 1994). The Resurrection Peninsula ophiolite is located on the Resurrection Peninsula south of Seward, and the Knight Island ophiolite is located 85 km east of Seward on Knight Island (Fig. 1b). Gravity data (Case et al., 1966; Saltus et al., 1999 (ftp://greenwood.cr.usgs.gov/pubs/open-file-reports/ofr-97-0520/

1. Introduction and Regional Geology

631

Fig. 2. Geologic map of the Resurrection Peninsula ophiolite and overlying sedimentary sequence. Modified after Nelson et al. (1987), Tysdal and Case (1979), and the authors’ mapping.

632

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 3. Geologic map of the Knight Island ophiolite (after Nelson et al., 1985) showing the main ophiolitic units and sample locations.

data) suggest that the exposures of ophiolites in the region may be continuous below sea level from south of Knight Island to Glacier Island northward (Fig. 1b), and then several tens of kilometers east, and extend to approximately 10 km depth (Crowe et al., 1992; Nelson and Nelson, 1993) (Fig. 1b). Minor ultramafic rocks occur as pods and lenses of harzburgite and dunite in the mafic sections of the ophiolites (Tysdal et al., 1977; Tysdal and Case, 1979; Nelson et al., 1987; Bol et al., 1992; Crowe et al., 1992; Barker et al., 1992; Bradley and Kusky, 1992; Nelson and Nelson, 1993; Plafker et al., 1994). Both ophiolites are similar chemically and petrographically (Tysdal et al., 1977; Nelson and Nelson, 1993; Lytwyn et al., 1997). Other outcrop belts of mafic rocks in

2. The Sanak-Baranof Magmatic Belt

633

the Valdez and Orca Groups occur as tectonic slices of pillow lavas, sheeted dikes, gabbros, and interbedded tuffaceous units within the flysch, but lack exposures of welldeveloped layered gabbro and basal peridotite sections characteristic of deeper levels of full ophiolite sections (Nelson et al., 1987). Pillow basalt, minor diabase, and gabbro form 15–20 percent of the exposed rocks of the Orca group, which has been metamorphosed to zeolite facies (with some local greenschist facies) during emplacement, compaction, and dewatering of the accretionary prism (Goldfarb et al., 1986; Barker et al., 1992).

2. THE SANAK-BARANOF MAGMATIC BELT Plutons of the Sanak-Baranof belt intrude a complexly deformed, Mesozoic and Cenozoic accretionary prism—the Chugach-Prince William composite terrane—along the seaward margin of the Peninsular-Wrangellia-Alexander composite terrane (Fig. 1). The inboard part of the prism is a mélange of variably metamorphosed basalt, chert, argillite, graywacke, plus minor limestone and ultramafic rocks (Triassic to mid-Cretaceous McHugh Complex and equivalents). Farther outboard are belts of strongly deformed Upper Cretaceous and Lower Tertiary flysch, assigned to the Valdez and Orca Groups, respectively. The Orca Group includes several belts of mafic and ultramafic rocks that have been interpreted as having formed at a spreading ridge just before ridge subduction (Bol et al., 1992). Penetrative deformation in the accretionary prism (thrust imbrication, folding, mélange formation) and regional metamorphism (typically prehnite-pumpelleyite to greenschist facies) occurred during and shortly after offscraping and underplating during the Cretaceous and Early Tertiary. Near-trench plutons were emplaced into the already deformed accretionary prism. Paleocene to Eocene plutons of the Sanak-Baranof belt outcrop discontinuously along the entire 2200 km length of the Chugach-Prince William terrane (Bradley et al., 2003; Kusky et al., 2003; Sisson et al., 2003). The plutons are mainly granodiorite, granite, and tonalite (Hudson, 1983). Some of the plutons are elongate parallel to structural grain of the accretionary prism, although they are discordant at their ends. Some are enormous—the Kodiak batholith, for example, is approximately 1650 square kilometers in area. Smaller intermediate to silicic dikes are plentiful in some regions, such as the Anchorage and Seldovia quadrangles (Winkler et al., 1984; Winkler, 1992; Bradley and Kusky, 1992). They crosscut everything but a few late, brittle faults in the accretionary prism.

3. FOREARC OPHIOLITES OF THE SOUTHERN ALASKA CONVERGENT MARGIN The Resurrection Peninsula and Knight Island ophiolites, structurally emplaced into the forearc accretionary prism of the Chugach and Prince William terranes of southern Alaska,

634

Chapter 20: The Resurrection Peninsula Ophiolite

are interpreted as remnants of the Kula-Farallon ridge that was subducted beneath the North American continental margin in the Early Tertiary (Bol et al., 1992; Lytwyn et al., 1997; Kusky and Young, 1999). Subduction of the Kula-Farallon ridge and large poleward migration of the Chugach-Prince William composite terrane are fundamental events in the interpretation of the history of plate motions across the northeast Pacific Basin for the Tertiary (Hillhouse and Gromme, 1977; Byrne, 1979; Stone et al., 1982; Plumley et al., 1983; Hillhouse et al., 1985; Engebretson et al., 1985; Stock and Molnar, 1988; Lonsdale, 1988; Atwater, 1989). The formation and emplacement of the Resurrection Peninsula ophiolite may serve as a model for other ophiolites emplaced in forearc trench-ridge-trench settings of all ages, including the Precambrian, when ridge-trench encounters were probably more common than at present. Additionally, ophiolites emplaced in this and similar triple junction settings are some of the few examples of ophiolitic crust that was clearly formed at an oceanic spreading center, and not in arc- or forearc-related environments. Ophiolite emplacement in the forearc is not the only consequence of Tertairy ridge subduction in southern Alaska; other effects of ridge subduction include near-trench magmatism (Marshak and Karig, 1977; Moore et al., 1983; Bradley et al., 1993, 1999b, 2003; Harris et al., 1996; Lytwyn et al., 1997; Kusky et al., 2003), anomalous forearc deformation (Kusky et al., 1997a), high-temperature, low-pressure metamorphism (Sisson and Pavlis, 1993; Pavlis and Sisson, 1995), gold mineralization (Bradley et al., 1994; Haeussler et al., 2003) and a cessation of arc magmatism (e.g., DeLong and Fox, 1977; Bradley et al., 2003). Many of the above events are diachronous along strike. This is highlighted especially well by dating of the near-trench magmatic rocks, which show a west to east age progression from 61 to 50 Ma, consistent with emplacement during passage of the Kula-Farallon ridge as it migrated beneath the southern Alaska margin (Bradley et al., 1993, 1999b; Kusky et al., 1997a). The 57 ± 1 Ma (Nelson et al., 1989) age of the Resurrection Peninsula ophiolite is broadly coeval with this suite of Eocene near-trench granitoids in the Gulf of Alaska region, suggesting that it too is related to passage of the Kula-Farallon-North America triple junction. The close association of the Resurrection Peninsula ophiolite to these events indicates the importance of understanding the process of its emplacement as it relates to the development of the Chugach/Prince William accretionary wedge, and consequently, to our understanding of ridge-trench encounters in general. This understanding may find general applicability for many Precambrian forearc ophiolites. Paleomagnetic data from mafic rocks of the Resurrection Peninsula ophiolite suggest that the ophiolite and the composite Chugach-Prince William terrane, into which it was emplaced soon after formation, originated along the Wrangellian margin of North America, far to the south of their present position, perhaps as far as the present latitude of northern Washington (Bol et al., 1992; Bradley et al., 1993). Assimilation of flysch-like sediments by volcanics and dikes, in addition to turbidites interbedded with pillow basalts, indicate that the Resurrection Peninsula ophiolite formed in close proximity to the trench axis where it was soon subducted (Lytwyn et al., 1997). During its transport history, the Resurrection Peninsula ophiolite thus passed through a series of changing geologic environments related

3. Forearc Ophiolites of the Southern Alaska Convergent Margin

635

to both migration down the ridge flank and away from the ridge as well as its increasing proximity to the North American continent. This study addresses general questions related to ridge-trench encounters and ophiolite emplacement with time, using the Resurrection ophiolite as an analog for some Precambrian ophiolites emplaced at convergent margins. Other issues including increased volcanism in sediments related to injection of mid-ocean ridge basalts (MORB) at the base of the prism, and the emplacement of MORB/melted sediment hybrid magmas in structures associated with ophiolite emplacement are also described, and their implications for studies of Precambrian ophiolites are discussed. 3.1. Geology of the Resurrection Peninsula Ophiolite The Resurrection Peninsula ophiolite (Fig. 2) consists of a west-dipping sequence of sedimentary rocks, pillow basalts, sheeted dikes, and massive and layered gabbro (Fig. 4). The ophiolite is metamorphosed to greenschist facies (Nelson et al., 1989; Lytwyn et al., 1997; Kusky and Young, 1999). Tysdal et al. (1977) mapped the Resurrection Peninsula ophiolite as intruding the Valdez Group flysch and inferred a Late Cretaceous age for the ophiolite based on a few widely scattered fossils from the Valdez Group. Nelson et al. (1989) reported a U/Pb (zircon) age of 57 ± 1 Ma for a plagiogranite that intrudes the dike complex and is also cut by dikes (Fig. 4) and thus serves to date active spreading of the ophiolite during Early Eocene. The Resurrection Peninsula ophiolite is therefore contemporaneous with the Orca Group. Ultramafic rocks occur as pods of serpentinized peridotite and pyroxenite in the gabbroic section, and as tectonic blocks in flysch lying in fault contact with the ophiolite, the flysch having been thrust in a west-southwest direction over the ophiolite (Nelson et al., 1987; Bol et al., 1992). The mostly serpentinized ultramafic rocks of Resurrection Peninsula were originally clinopyroxenite, dunite, and peridotite (Nelson et al., 1987). The gabbroic rocks of the Resurrection Peninsula occur as a mafic pluton of at least 25 km2 in area that makes up the eastern side of the peninsula (Tysdal et al., 1977; Nelson et al., 1987, 1989). The structurally lowest part of the gabbro is a layered unit in the east containing locally well-developed, west-dipping magmatic mineral layering of alternating light and dark layers of pyroxene and feldspar. The layered unit is separated from a massive unit to the west by a block of Valdez flysch and interbedded volcanic and sedimentary rocks (Fig. 2). To the west, the massive gabbro contains an increasing percentage of dikes which further grade into the sheeted dike unit occupying the central section of the peninsula (Tysdal et al., 1977; Nelson et al., 1985). The dikes of the sheeted dike unit (Fig. 2) generally strike north along the central part of the peninsula, although the strike is more westward in the southern part of the peninsula (Tysdal et al., 1977; Nelson and Nelson, 1993; Bol et al., 1992). Most dikes are vertical, with some local variation and crosscutting of preexisting dikes at low angles (Fig. 4b). Dikes range from 0.1–5 m thick but average 0.3–1 m, increase in percentage upward, with both one-way and two-way chill margins observed on individual dikes in approximately equal proportions (Nelson et al., 1987; Bol et al., 1992). The dikes are variably aphanitic,

636

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 4. Photographs of: (a) deformed argillites and sandstone beds of the Humpy Cove Formation overlying the Resurrection Peninsula ophiolite, (b) basaltic pillow lavas of the Resurrection Peninsula ophiolite, (c) Sheeted dike complex from Killer Bay, (d) Mafic dike of sheeted complex cutting trondhjemite dated at 57 ± 1 Ma (Nelson et al., 1989).

porphyritic, and diabasic (Nelson et al., 1987). Both the gabbro and the sheeted dikes are intruded by 57 ± 1 Ma old plagiogranite (Nelson et al., 1987). To the west, a gradational zone 10 m thick separates the sheeted dikes and pillow basalts (Bol et al., 1992). Diabasic dikes exhibit intersertal, intergranular, and subophitic textures. The mineralogy is dominated by anhedral to euhedral plagioclase occurring as randomly oriented laths, some extensively embayed. Subophitic augite is common whereas pristine olivine is rare. Alteration minerals include (in order of decreasing abundance) chlorite (after olivine), actinolite, epidote (after plagioclase), and carbonate. Pillow basalts, minor amounts of massive basalt flows, and broken pillow breccias (Figs. 2, 3, and 4) form a unit 1000 m thick on the western flank of the peninsula (Tysdal et al., 1977; Nelson et al., 1987). Pillows average 0.5 m in diameter, strike north, and dip about 30◦ to 45◦ to the west, with interpillow spaces filled locally with red

3. Forearc Ophiolites of the Southern Alaska Convergent Margin

637

and green chert (Nelson et al., 1987). Pillow basalts of the Resurrection and Knight Island ophiolites contain interbedded terrigeneous flysch deposits (Fig. 4), suggesting continuous sedimentation during formation of the ophiolites at a source of high sediment input such as a continental margin (Tysdal et al., 1977; Nelson et al., 1987; Bol et al., 1992). Sheeted dike complexes containing chilled margins within the two ophiolites suites, however, suggest that the ophiolites formed in an extensional setting such as a spreading ridge, which indicates a close proximity of a spreading center with a continental margin (Tysdal et al., 1977; Coleman, 1977; Nelson et al., 1987; Nicholas, 1989; Nelson and Nelson, 1993; Bol et al., 1992). Pillow basalts include aphanitic, porphyritic, and hypocrystalline varieties. Subhedral to euhedral plagioclase phenocrysts are enclosed within a fine-grained quenched groundmass or devitrified glass containing abundant needle-like, randomly oriented plagioclase microlites. Some plagioclase phenocrysts are embayed and many crystals are normally zoned. Some show corroded, altered cores, but most are remarkably unaltered. Clinopyroxene occurs rarely as phenocrysts and more commonly as microphenocrysts in hypocrystalline basalts and large seriate grains in holocrystalline samples. Olivine is rare. Although several samples of pillow basalt appear virtually unaltered, others are slightly metamorphosed as evidenced by chlorite (after clinopyroxene) and/or amygdules filled with secondary chlorite, zeolites (natrolite?), and carbonate. Rare veins of prehnite also are found. Investigations of paleomagnetic data by Bol et al. (1992) on rocks of the Resurrection Peninsula ophiolite suggest that it (and presumably the Chugach terrane) was 13 ± 9◦ south of its present position with respect to North America, at 57 Ma when the ophiolite formed along the Kula-Farallon ridge. The 57 Ma age of the ophiolite, in conjunction with geologic relationships led Bol et al. (1992) to conclude that the Resurrection Peninsula ophiolite represents an accreted fragment of the Kula-Farallon ridge that was emplaced shortly after its formation. 3.2. The Knight Island Ophiolite The Knight Island Ophiolite (Figs. 1 and 3) is a fault-bounded complex within isoclinally folded flysch of the Orca Group. Pillow basalt and sheeted dikes, variably affected by greenschist-facies metamorphism, dominate the complex, while ultramafic rocks are found only as boulders along the beach and as xenoliths within the sheeted dike complex (Richter, 1965; Tysdal et al., 1977; Nelson et al., 1985, 1989; Bol et al., 1992; Crowe et al., 1992; Nelson and Nelson, 1993). The pillow and massive basalts form a unit over 5000 m thick with interpillow spaces filled with breccia, agglomerate, and minor red and green chert (Tysdal et al., 1977; Nelson and Nelson, 1993). Individual sheeted dikes up to several meters thick locally intrude Orca flysch, while small gabbroic plutons intrude both the dikes and adjacent sedimentary rocks (Tysdal et al., 1977). The alteration minerals found in pillow lavas and dikes are consistent with zeolite to lower greenschist-facies hydrothermal metamorphism recognized within the upper sections of both ophiolites (Wiltse, 1973; Tysdal et al., 1977; Nelson et al., 1989; Lytwyn et al., 1997).

638

Chapter 20: The Resurrection Peninsula Ophiolite

3.3. Valdez/Orca Metabasalts East of Resurrection Peninsula and Knight Island, a belt of metabasaltic and metadiabasic rocks (Fig. 1) is discontinuously exposed from Valdez to east of Yakutat, Alaska (Lull and Plafker, 1990). These metabasalts and metadiabases consist of pillow lavas, breccias, and dikes intruding and interbedded with mostly Valdez Group accretionary metasediments (Lull and Plafker, 1990) and were therefore generally regarded as Late Cretaceous (84– 66 Ma) in age. Recent discovery of geochemically similar rocks in the Orca Group, however, demonstrate that at least some of these metabasalts may be Eocene in age. We collectively refer to these rocks as Valdez/Orca metabasalts and later present geochemical analyses and modeling, suggesting a link with the Resurrection/Knight Island ophiolites and the time of ridge subduction.

4. GEOCHEMISTRY OF THE RESURRECTION PENINSULA AND KNIGHT ISLAND PILLOW BASALTS AND SHEETED DIKES 4.1. Effects of Greenschist-Facies Metamorphism Sample preparation for analysis of major, minor, and trace elements by inductively coupled plasma (ICP) spectrometry was as described by Lytwyn et al. (1997). Static hydrothermal alteration of the pillow lavas and sheeted dikes, as evidenced by zeolite to lower greenschist-facies metamorphism in both ophiolites, could result in mobility of certain elements such as Na, K, Ca, Sr, Ba, and Rb. Although groundmass alteration and devitrification is evident in some analyzed samples, phenocryst and microphenocryst phases are generally unaltered. The extent of clinopyroxene and plagioclase alteration in diabases exhibiting somewhat greater metamorphism is still less than 15 percent. Elements with low (less than three) ionic potentials (ratios of atomic number/ionic radius) such as Sr, K, Ba, and Rb are considered mobile in aqueous fluids relative to those having larger ionic potentials (greater than three) such as REE and high field strength elements (e.g., Pearce and Peate, 1995). Sr is typically enriched (displays positive anomalies on extended spiderdiagrams) in extensively altered metabasites (Pearce and Cann, 1973), whereas most of the ophiolitic samples, although enriched in Ba, exhibit negative Sr anomalies (Fig. 5) which is more consistent with plagioclase fractionation. Hydrothermal alteration (Fig. 5) causes Mg uptake in basalt (Mottl, 1983). Resurrection/Knight Island samples (e.g., K-13 and RP-390B) with relatively high MgO (8.25 and 9.02 wt%), however, also have the lowest abundances of incompatible trace elements (Fig. 5) which further supports the idea that their geochemistry is largely primary and little affected by hydrothermal processes. 4.2. Compositional Range and Tectonic Setting Pillow lavas and sheeted dikes from the Resurrection Peninsula and Knight Island ophiolites range from basalt to basaltic andesite (Table 1). Plots of major, minor, and trace

4. Geochemistry of the Resurrection Peninsula and Knight Island Pillow Basalts and Sheeted Dikes

639

Fig. 5. Spiderdiagrams of (A) Resurrection Peninsula, (B) Knight Island, and (C) Valdez/Orca samples from this study normalized to the chondritic values of Anders and Grevesse (1989). Fig. 2c also includes metabasalt sample APR164C from the Valdez Group (Lull and Plafker, 1990). Elements are arranged from left to right in order of increasing compatibility in solid mantle phases.

640

Table 1. Inductively coupled plasma whole rock analysis of ophiolitic pillow lavas and sheeted dikes RP86B233B (Dike) 51.93 1.64 14.41 11.58 0.23 6.75 8.10 3.04 1.17 0.16 99.01 2.34 122 115 40 39 329 303 NA NA 7.0 18.5 13.6 4.04 1.33 5.4 6.4 4.2 3.82 0.60 10.41

RP86B246B (Pillow) 53.22 1.99 14.02 10.60 0.21 5.92 8.07 5.18 0.24 0.22 99.67 1.59 161 122 53 39 61 343 NA NA 8.3 22.2 17.2 5.15 1.72 6.9 8.2 5.4 4.97 0.77 9.53

RP86B322B (Pillow) 51.36 1.37 14.43 10.39 0.19 7.21 10.47 3.23 0.44 0.15 99.24 1.67 103 226 34 42 112 303 NA NA 5.1 13.7 10.7 3.21 1.22 4.5 5.4 3.6 3.24 0.50 9.34

RP86B361B (Pillow) 55.71 1.86 15.89 9.68 0.13 5.50 6.30 4.23 0.20 0.20 99.70 2.50 167 163 46 37 133 298 NA NA 9.8 24.4 16.8 4.92 1.48 6.3 7.4 4.8 4.40 0.69 8.70

RP86B365B (Pillow) 51.50 1.58 15.36 9.82 0.16 8.30 6.40 4.22 1.06 0.19 98.59 2.86 146 95 46 36 434 275 NA NA 9.4 23.1 16.2 4.58 1.46 5.9 6.9 4.5 4.17 0.63 8.83

RP86B375B (Pillow) 48.90 1.58 18.77 10.34 0.22 5.54 9.71 4.46 0.58 0.22 100.32 5.79 154 164 37 34 324 300 193 59 7.5 19.8 14.3 3.81 1.13 4.8 5.9 3.8 3.57 0.51 9.29

RP86B390B (Pillow) 50.51 1.31 16.13 9.51 0.15 8.25 11.64 2.00 0.10 0.15 99.75 2.09 103 142 30 35 53 251 NA NA 6.7 16.8 11.3 3.21 1.09 4.1 4.8 3.1 2.81 0.44 8.55

RP86B408B (Pillow) 54.17 1.23 15.30 9.63 0.15 8.24 8.96 2.38 0.33 0.20 100.59 2.88 111 177 30 33 156 236 NA NA 7.4 18.0 12.0 3.36 0.99 4.2 4.8 3.1 2.81 0.42 8.66

RP86B420B (Pillow) 51.30 1.32 16.22 9.60 0.16 7.16 8.45 3.68 1.35 0.16 99.40 2.59 124 181 34 37 521 258 NA NA 7.0 17.7 12.4 3.61 1.24 4.6 5.5 3.5 3.19 0.48 8.63

RP86B451B (Pillow) 53.11 1.48 15.74 10.33 0.17 6.09 5.61 5.36 1.00 0.18 99.07 1.94 146 119 42 33 270 288 NA NA 9.1 22.8 15.1 4.14 1.32 5.5 6.4 4.2 3.94 0.62 9.29

RP86BK-2 471B (Pillow) (Dike) 52.66 53.94 2.10 0.99 13.49 15.44 13.78 8.37 0.23 0.13 6.00 7.12 7.50 8.35 3.32 5.10 0.80 0.15 0.25 0.15 100.13 99.74 2.45 2.25 172 87 118 78 58 27 41 33 263 24 359 233 NA NA NA NA 8.2 4.1 23.2 11.2 18.8 7.9 5.58 2.51 1.88 0.83 7.5 3.5 9.0 4.2 5.9 2.8 5.40 2.60 0.83 0.42 12.39 7.52 (continued on next page)

Chapter 20: The Resurrection Peninsula Ophiolite

SiO2 TiO2 Al2 O3 Fe2 O3 * MnO MgO CaO Na2 O K2 O P2 O5 Sum LOI (%) Zr Sr Y Sc Ba V Cr Ni La Ce Nd Sm Eu Gd Dy Er Yb Lu FeO*

RP86B207B (Pillow) 50.23 1.44 16.69 8.55 0.17 7.55 9.85 2.99 1.61 0.14 99.22 2.91 104 123 34 46 237 264 NA NA 5.4 15.3 11.5 3.40 1.29 4.6 5.6 3.6 3.33 0.51 7.69

SiO2 TiO2 Al2 O3 Fe2 O3 * MnO MgO CaO Na2 O K2 O P2 O5 Sum LOI (%) Zr Sr Y Sc Ba V Cr Ni La Ce Nd Sm Eu Gd Dy Er Yb Lu FeO*

K-4 (Pillow) 51.17 1.21 15.47 9.85 0.18 7.99 10.52 3.60 0.17 0.13 100.29 1.74 100 175 31 39 81 285 NA NA 4.0 11.5 9.2 2.83 1.06 3.9 4.9 3.2 2.98 0.46 8.86

K-8 (Pillow) 50.40 1.19 16.18 10.38 0.16 8.42 10.53 3.44 0.14 0.09 100.93 2.22 72 139 30 43 75 262 NA NA 3.4 9.9 8.0 2.55 0.99 3.8 4.6 3.1 2.88 0.46 9.33

K-11 (Pillow) 50.15 1.20 16.14 9.87 0.15 7.87 12.06 2.15 0.05 0.12 99.76 1.65 85 118 30 38 24 244 NA NA 3.8 11.1 8.8 2.75 1.03 3.9 4.7 3.1 2.85 0.44 8.87

K-13 (Pillow) 52.02 0.79 15.88 7.18 0.14 9.02 11.78 2.33 0.06 0.09 99.29 2.44 71 152 20 28 50 159 NA NA 3.4 9.6 6.8 2.08 0.70 2.7 3.2 2.0 1.90 0.31 6.45

K-19 (Pillow) 53.84 1.13 15.62 9.37 0.13 7.92 7.25 5.05 0.06 0.11 100.48 2.73 88 95 29 32 36 245 264 195 4.0 12.3 9.5 2.61 1.04 3.7 4.5 3.0 2.84 0.41 8.42

K-21 (Pillow) 51.23 1.13 15.70 9.34 0.14 7.07 11.21 4.00 0.08 0.12 100.02 3.02 84 144 25 32 35 236 324 121 3.3 10.8 8.4 2.15 0.85 3.2 3.9 2.6 2.40 0.34 8.40

K-24 (Pillow) 48.25 1.29 17.06 9.59 0.17 8.47 11.70 2.55 0.07 0.13 99.28 2.46 91 153 29 35 46 259 421 164 2.7 10.3 9.5 2.65 0.96 3.8 4.7 3.0 2.84 0.40 8.62

K-26 (Pillow) 50.81 1.21 16.30 9.45 0.18 8.43 12.01 2.79 0.10 0.13 101.41 1.83 77 126 29 35 28 232 NA NA 3.8 10.9 8.7 2.69 0.97 3.8 4.6 3.1 2.74 0.43 8.86

K-35 (Pillow) 53.83 1.87 15.45 11.51 0.17 4.93 5.26 5.34 0.30 0.26 98.92 2.42 198 118 53 30 143 329 NA NA 7.6 22.7 17.8 5.20 1.37 6.9 8.2 5.5 5.12 0.80 10.35

K-37 (Pillow) 40.44 1.96 18.37 13.59 0.20 4.99 14.74 3.78 0.07 0.32 98.46 8.39 123 158 46 45 67 381 NA NA 7.3 17.3 13.8 4.17 1.31 5.8 7.0 4.7 4.19 0.65 12.22

K-42 (Pillow) 50.09 1.21 16.29 10.17 0.15 9.68 10.88 2.04 0.20 0.12 100.83 3.79 92 135 26 33 90 254 356 178 3.6 11.5 9.2 2.54 1.01 3.5 4.2 2.7 2.56 0.37 9.14

641

Samples from Resurrection Peninsula indicated by RP and Knight Island by K. NA, not analyzed.

K-30 (Pillow) 50.23 1.75 15.69 11.31 0.15 6.31 10.79 3.35 0.05 0.19 99.82 3.17 133 115 43 37 12 301 NA NA 5.6 16.3 13.3 4.12 1.47 5.6 6.8 4.5 4.12 0.65 10.17

4. Geochemistry of the Resurrection Peninsula and Knight Island Pillow Basalts and Sheeted Dikes

Table 1. (Continued)

642

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 6. Samples plotted on binary diagrams of (A) FeO* /MgO versus SiO2 showing the calc-alkaline (CA) and tholeiitic (Th) fields of Miyashiro (1974) and (B) K2 O versus SiO2 showing the low-K (tholeiitic), medium-K (calc-alkaline), and high-K (calc-alkaline) subdivisions of Le Maitre et al. (1989) and Rickwood (1989). Symbols in these and subsequent diagrams are as follows: solid circles, Resurrection Peninsula pillow lavas and sheeted dikes from this study; open circles, Resurrection Peninsula pillow lavas and sheeted dikes from Crowe et al. (1992); solid triangles, Knight Island pillow lavas from this study; open triangles, Knight Island pillow lavas and sheeted dikes analyzed by Crowe et al. (1992) and Nelson and Nelson (1993); pluses, Valdez/Orca metabasalts from this study; crosses, Valdez metabasalts from Lull and Plafker (1990) and Crowe et al. (1992). Field of basalts and andesites from the Taitao Ophiolite, South America, from Kaeding et al. (1990).

4. Geochemistry of the Resurrection Peninsula and Knight Island Pillow Basalts and Sheeted Dikes

643

Fig. 7. Samples plotted on various discriminant diagrams using same symbols as Fig. 5. Open squares represent greenstones (Hill, 1979) from the Ghost Rocks Formation. The abbreviations are as follows: OFB, ocean floor basalts; LKT, low-K tholeiites; CAB, calc-alkaline basalts; WPB, within-plate basalts; MORB, mid-ocean ridge basalts; IAT, island-arc tholeiites; OIT, ocean-island tholeiites; OIA, ocean-island alkalic basalts. Discriminant diagrams are from the following sources: Ti/Zr/Y and Ti versus Zr from Pearce and Cann (1973), TiO2 /MnO/P2 O5 from Mullen (1983), and Zr/Y versus Zr from Pearce and Norry (1979).

elements show only modest compositional differences between the two ophiolites (e.g., Figs. 5–9) with the Resurrection samples somewhat more enriched in alkalis and light rare earth elements (LREE—Fig. 10). Binary plots of major and minor elements and ratios reveal significant compositional scatter in the ophiolitic data when utilizing SiO2 (Fig. 6) and MgO (Fig. 8) as measures of differentiation (discussed below). Chondrite-normalized spiderdiagrams (Fig. 5) show that Resurrection and Knight Island samples generally have flat to slightly LREE-enriched patterns and overall REE abundances ranging between 10X and 40X chondritic. Samples also display overall negative Sr, Eu, and Ti anomalies (relative to adjacent REE), positive Zr anomalies, and negative

644

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 8. Geochemical binary plots of Resurrection Peninsula and Knight Island pillow lavas and sheeted dikes on coordinates of (a) SiO2 , (b) TiO2 and (c) CaO/Al2 O3 versus MgO. Data symbols same as Fig. 5. The field of Valdez/Orca metabasalts (lighter shaded) defined from data by Lull and Plafker (1990) and Crowe et al. (1992), while the field of Ghost Rocks greenstones (darker shaded), Kodiak Island, is based on the data of Hill (1979). Liquid lines of descent for perfect fractional crystallization (PFX) of selected parental liquids RP-365B, RP-390B, and K-13 at pressures of 1 bar and 10 kbar were calculated using an algorithm after Weaver and Langmuir (1990). The compositional range of Resurrection and Knight Island samples cannot be related through fractionation of a common parental magma.

to positive Ba anomalies (Fig. 5). When normalized to normal mid-ocean ridge basalts (N-MORB) as in Fig. 13, Resurrection and Knight Island lavas and dikes show relative enrichments in highly incompatible elements (Th, K, Rb, and Ba) (Crowe et al., 1992; Nelson and Nelson, 1993) and display slight LREE enrichments similar to transitional (Ttype) mid-ocean ridge basalts (normalization values from Sun and McDonough, 1989). Major and minor element plots of FeO* /MgO and K2 O versus SiO2 indicate that Resurrection and Knight Island lavas and dikes include both tholeiitic and calc-alkaline varieties

5. Valdez/Orca Metabasalts

645

Fig. 9. Comparisons of samples discussed in the text with calculated MORB parental magmas (after Niu and Batiza, 1991) on coordinates of (a) SiO2 /FeO* versus MgO and (b) TiO2 versus SiO2 /FeO* . Heavy lines represent the range of calculated parental magma compositions produced through polybaric melting of rising peridotitic material beginning at different pressures of melting (8–20 kbar) and continuing to around 4 kbar pressures. Lines that are near orthogonal to the melting trends connect equal values of F (degrees of melting) at intervals of 0.05. Liquid lines of descent (thin lines) for selected parental liquids were determined with an algorithm after Weaver and Langmuir (1990). The selected parental magmas represent F = 0.24 following initial melting at 18 kbar, F = 0.18 following initial melting at 14 kbar, and F = 0.20 following initial melting at 10 kbar. Data sources for the fields of Ghost Rocks greenstones and Valdez/Orca metabasalts same as Figs. 3 and 4.

(Fig. 5) as is also indicated on the AFM (A = Na2 O + K2 O, F = FeO + Fe2 O3 , M = MgO) diagram (Crowe et al., 1992; Nelson and Nelson, 1993). Some discriminant diagrams, utilizing trace elements (e.g., Ti, P, Zr, Y, REE, etc.) considered relatively immobile during low-temperature alteration and metamorphism up to greenschist facies (Pearce and Cann, 1973; Pearce and Norry, 1979; Pearce et al, 1981; Pearce, 1982; Mullen, 1983), define Resurrection and Knight volcanics and dikes as mid-ocean ridge basalts (MORB) and ocean floor tholeiites (Fig. 4). The Th-Hf-Ta discriminant diagrams, however, identify Resurrection and Knight Island lavas and dikes as calc-alkaline basalts and island-arc tholeiites (Crowe et al., 1992; Nelson and Nelson, 1993).

5. VALDEZ/ORCA METABASALTS Valdez/Orca metabasalts analyzed in this study (Table 2) and Lull and Plafker (1990) are generally more depleted in incompatible trace elements (REE, Ti, Zr, Y, etc.) than lavas and dikes from the Resurrection and Knight Island ophiolites (Fig. 5). Valdez/Orca basalts include both tholeiitic and calc-alkaline varieties (Figs. 6A, 6B, and 7).

646

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 10. N-MORB normalized spiderdiagrams comparing Knight Island lavas and dikes (shaded field) with a basaltic glass (D42-4) from the Chile Ridge (Klein and Karsten, 1995) and a calc-alkaline basalt from the southern volcanic zone (SVZ) of the Andes (Hickey et al., 1986). Knight Island data from Nelson and Nelson (1993) were used in order to compare Ta, Th, Nb, Cs, Rb, and U not analyzed in this study. All the spiderdiagrams patterns are broadly similar in terms of enrichments in highly incompatible elements (e.g., Cs and Ba) and negative anomalies in certain HFS elements (e.g., Th and Ta). Normalization values for N-MORB from Sun and McDonough (1989).

6. SIGNIFICANCE OF MAGMATIC ROCKS Geochemical analysis of the igneous rocks of the Resurrection and Knight Island ophiolites as well as related igneous rocks of the Valdez and Orca Groups show that they are chemically similar to mid-ocean ridges or primitive island arcs (Crowe et al., 1992;

6. Significance of Magmatic Rocks

647

Fig. 10. (Continued.)

Nelson and Nelson, 1993; Lytwyn et al., 1997). All the authors, however, have realized the arc-like chemistry of these rocks to be incompatible with the field evidence in the Gulf of Alaska, as no arc remnants are exposed outboard of the Gulf of Alaska. This observation is critical for the interpretation of ancient sequences, where the accretionary prism has experienced a collision and original relationships are not as clear. The arc-like chemistry of the most-outboard accreted ophiolites serves as a lesson to those who would use geochemistry alone to attempt to discriminate tectonic environments. Crowe et al. (1992) suggested sediment contamination to explain the variable volcanic rock compositions in an accretionary setting, having identified subduction of transforms and ridges as instances where MORB type magmas might be contaminated by sediments. Nelson and Nelson (1993) believing a contractional forearc setting is incompatible with a strong MORB-like signature and the presence of sheeted dikes, suggested that the migration of magma along a partially subducted ridge to an unsubducted part of the ridge can explain the mixing of sediments and MORB magma. Lytwyn et al. (1997) determined that mafic rocks of the Resurrection Peninsula and Knight Island have similar petrographic and geochemical compositions, which indicates a similar origin and tectonic history. The mafic rocks formed at a distance from the paleotrench have a strong MORB geochemical signature, whereas contemporaneous magmas derived from the same mantle melting column but intruded through the accretionary prism have more calc-alkaline signatures, reflecting near-fractional melting of parental magmas and assimilation of sedimentary material from the accretionary prism. These refractory arctype magmas came from shallow melts that are unable to mix with deeper less refractory magmas once ridge subduction occurred, suggesting ophiolite genesis at a spreading ridge near a subduction zone (Lytwyn et al., 1997, 2000; Kusky and Young, 1999; Bradley et al., 2003). With a greater number of plates and an even greater number of triple junctions in the Archean, calc-alkaline ophiolites generated at ridges near subduction zones were probably much more common in the earlier record.

648

Chapter 20: The Resurrection Peninsula Ophiolite

Table 2. Inductively coupled plasma whole rock analysis of Valdez and Orca metabasalts Latitude Longitude SiO2 TiO2 Al2 O3 Fe2 O3 * MnO MgO CaO Na2 O K2 O P 2 O5 Sum Zr Sr Y Sc Ba V Cr Ni La Ce Nd Sm Eu Gd Dy Er Yb Lu FeO*

82AMH-52A 60◦ 58 40 145◦ 53 37 50.20 0.47 15.08 9.33 0.22 10.98 12.00 1.54 0.12 0.05 99.99 29 51 15 42 23 248 594 179 0.68 2.2 1.8 0.62 0.26 1.5 2.2 1.7 1.64 0.26 8.40

82AMH-40C 60◦ 50 57 146◦ 34 32 51.20 0.79 15.52 9.14 0.14 9.20 12.61 2.02 0.11 0.08 100.81 49 86 24 43 26 292 307 99 2.10 5.9 4.7 1.48 0.57 2.6 3.4 2.5 2.50 0.41 8.22

82AMH-43A 60◦ 50 39 146◦ 31 50 50.88 0.88 15.03 10.82 0.17 8.58 11.82 1.66 0.10 0.08 100.02 52 75 25 44 27 289 283 96 1.6 5.6 5.1 1.70 0.65 2.9 3.8 2.7 2.67 0.43 9.73

7. SEDIMENTARY ROCKS OF THE RESURRECTION PENINSULA Sedimentary rocks associated with the Resurrection ophiolite include typical Valdeztype flysch rocks west of the Fox Island shear zone, and an unusual group of thinly bedded flysch (Humpy Cove Formation of the Orca Group, Kusky and Young, 1999) east of the Fox Island shear zone (Fig. 2).

7. Sedimentary Rocks of the Resurrection Peninsula

649

7.1. Valdez Group, West Shore of Resurrection Bay Generally north-striking beds of the Valdez Group flysch along the western shore of Resurrection Bay comprise massive coarse-grained sandstone, containing carbonate nodules, with interbedded 1–3 cm thick siltstone and shale couplets. In the central parts of the bay several massive units of black mud-matrix debris flows or olistostromes with blocks of graywacke up to 0.5 m in length out crop intermittently. Deformation within the thick mudstone units decreases to the north where numerous quartz veins dominate the outcrop. Folded quartz-veins and boudins show west-over-east asymmetry. 7.2. Humpy Cove Formation of the Orca Group Sedimentary rocks of the Resurrection Peninsula deposited above the ophiolite include variably metamorphosed graywacke, siltstone, and shale. Because these rocks rest conformably over a 57 Ma ophiolite, and are cut by 53.4 Ma plutonic rocks, they are temporally correlated with the Orca Group. However, because of their unusual lithologic nature and unique tectonic setting the name “Humpy Cove Formation” of the Orca Group was proposed for these rocks (Kusky and Young, 1999), including description of a formal type section extending along the north shore of Humpy Cove and extending through the north shore of Fox Island (Fig. 2). Detailed descriptions and chemical analyses of these rocks are provided by Kusky and Young (1999). The turbidites resting conformably over the ophiolite typically comprise monotonously bedded 1–5 cm thick graded argillite/shale couplets with 10–20 cm thick beds of siltstone and thicker (0.2–1 m) sandstone/graywacke beds. Most of the shales are metamorphosed slates up to greenschist facies, but for the purposes of the sedimentological description the term “shale” is retained. Beds strike to the north-northeast and dip west. Small-scale sedimentary structures such as flute marks, cross-laminations, ripples, and parallel-laminations are only rarely present, precluding designation of Bouma facies. Diagenetic carbonate concretions, approximately 5–20 cm in diameter are abundant in the graywacke of Thumb Cove, forming bedding-parallel horizons within the units (Young, 1997). These nodules form elongate, darkly weathering pits, and weather to a chalky brown color and typically appear slightly flattened where the exposure is three-dimensional. Local depositional sequences show variations in the proportion of coarse- to finegrained horizons. These variations include an increasing thickness of sandstone beds to 10 m or more, without a corresponding increase in finer-grained layers, and rhythmically interbedded, 1–5 cm thick siltstone and shale couplets in sequences ranging to approximately 10 m thick. A slatey cleavage is typically developed in the fine-grained rocks. Where black shales are present, the layers tend to be thin interbedded shale and siltstone couplets with only minor beds of sandstone, if present at all. The black shales typically have a fissile fabric. Volcanic tuffs interbedded with turbidites form volumetrically minor lenses that pinch and swell, making them difficult to use as marker beds across widely separated outcrops. The proportion of volcanic beds increases up-section where they form a notable portion of the rocks of Fox Island (Fig. 2).

650

Chapter 20: The Resurrection Peninsula Ophiolite

Except in the Fox Island shear zone, bedding is fairly coherent, with localized deformation of bedding typically expressed as open to isoclinal folding, slumping confined to individual strata, and minor shortening or extension of more competent beds. Pervasive quartz veins filling bedding-parallel cleavage planes also show isoclinal folding in places. 7.3. Age and Duration of Sedimentation of the Humpy Cove Formation The 57 ± 1 Ma age of the Resurrection Peninsula ophiolite represents a maximum age for turbidites deposited conformably over the ophiolite. A 40 Ar/39Ar cooling age on biotite of 53.4±0.9 Ma for the Hive Island stock (Fig. 2) that intrudes along a continuation of the Fox Island shear zone south of Fox Island (Bradley et al., 1999b) provides a constraint on the minimum age for sediments structurally below the shear zone. This age is a biotite cooling age for the pluton; consequently, the pluton may have been emplaced well before 53.4 ± 0.9 Ma. Thus we estimate that the sediments above the ophiolite, up to the Fox Island shear zone, took approximately 3.6 Ma to accumulate, from 57 to 53.4 Ma (PaleoceneEocene). This estimate for the time duration of sedimentation involves some additional assumptions. First, the 57 ± 1 Ma age for the ophiolite is from a trondhjemite that may have intruded late in the seafloor spreading history of the ophiolite, and could conceivably even be coeval with some of the tuffaceous rocks in the Humpy Cove Formation. However, the trondhjemite intrudes the sheeted dike complex and is also cut by dikes of the sheeted complex, so it presumably represents an accurate age for the ophiolite. There also may be an unknown time delay between the age of the ophiolite and initiation of sedimentation of the Humpy Cove Formation, although this too is unlikely since flysch of the lowermost Humpy Cove Formation is interbedded with pillow lavas of the ophiolite. The upper age bracket on the age of the Humpy Cove Formation is provided by the 53.4 ± 0.4 Ma 40 Ar/39 Ar age of an intruding pluton; since this is a cooling age, the age of intrusion is older. In the area, there is typically about a 2 Ma time difference between U/Pb and 40 Ar/39 Ar ages for plutonic rocks of this age (Bradley et al., 1999b). Taking these assumptions and possible sources of error into account, the interval bracketed for deposition of the Humpy Cove Formation could be as long as 58 to 53 Ma (5 m.y.), or as short as 56 to 53.8 Ma (2.2 m.y.). Accordingly, we adopt 3.6 ± 1.4 Ma as the best estimate for the duration of sedimentation of the Humpy Cove Formation.

8. FOX ISLAND SHEAR ZONE The Fox Island Shear Zone is a greater-than 1-km wide north-striking ductile shear zone that juxtaposes deformed and metamorphosed sedimentary and volcanic rocks of the Valdez Group to the west over lower-grade metasedimentary Orca Group rocks lying above the ophiolite to the east (Fig. 2). The eastern unit on Fox Island comprises interbedded sedimentary and volcanic rocks of similar metamorphic grade to rocks of the Resurrection

8. Fox Island Shear Zone

651

Peninsula (Fig. 2). Sedimentary rocks are primarily blue-gray to black shales interbedded with thin siltstone and silty sandstone layers with locally massive sandstone units of 1–2 m thickness. Tuffs form individual units ranging from approximately 0.5 to several meters thick. The tuffs are lenticular with finely laminated light green to light brown patches. The fine-grained groundmass contains plagioclase phenocrysts, quartz augen, and elongate lenses a few centimeters long that appear to be flattened volcanic breccia or pumice. Where these lenses are present, they are either light green or brown and the groundmass is brown or light green. Some of the quartz augen (less than one centimeter) have rotated asymmetric tails indicating a west over east sense of shear. Very fine-grained black shales dominate outcrops on the northeastern side of the island. These rocks are interbedded with thin beds of sandy siltstone and typically have a fissile fabric that shows flow along disrupted and sheared bedding planes. Thin zones of folding and boudinage occur throughout the section. Stratigraphically above the black shales, volcanic tuffs increase in both the number of beds and in thickness of individual layers. A few volcanic flows or sills are present near the southern end of the section. In cliffside exposures, alternating tuff, mudstone, and graywacke beds are coherent and planar. Shear zones that are typically several tens of centimeter wide cut the northeast side of the island. Siltstone beds are pulled apart and isoclinally folded. Bedding on either side of these zones is relatively undeformed and planar. Isoclinally folded layers include doubly plunging folds indicating rotation of fold axes into the extensional direction. Motion along these shear zones is west over east with rotation of the fold axial surface down to the northwest. Similar orientations of beds and foliations on Fox Island as well as folds for all of Resurrection Bay indicate are compatible with west over east thrusting (e.g., Kusky et al., 1997b; Kusky and Young, 1999). A different style of deformation is evident approximately 500 m south, near the sand spit that juts out from the eastern side of Fox Island (Fig. 2). Alternating 1-cm thick siltstone and 2-cm thick mudstone beds are chaotically folded but do not exhibit the fissile fabric or flow along bedding surfaces that are observed in the zone of folding to the north. Quartz veins filling cleavage planes are axial planar to small folded quartz veins that cut the outcrop. Those perpendicular to cleavage are flattened, whereas those oriented along cleavage planes are extended, indicating strong flattening. Minor sandy beds are boudinaged, whereas the tops and bottoms of folded horizons appear welded or annealed, suggesting that soft-sediment deformation may have preceded tectonic flattening in this zone. Central and Western Fox Island consists predominantly of metavolcanic and metasedimentary rocks within the Fox Island shear zone (Fig. 11) that records ductile, west-overeast thrusting. Rocks of the central and western units are more deformed and have a higher metamorphic grade than rocks to the east. Mafic and pelitic rocks exhibit a schistosity close to the Fox Island shear zone (Fig. 11). Secondary minerals in sedimentary rocks include biotite (giving a purple hue to the siltstones and mudstones), muscovite, and lesser amounts of chlorite. However, due to the wide range of temperatures and pressures over which these minerals are stable, we cannot specify a unique temperature and pressure under which these

652

Chapter 20: The Resurrection Peninsula Ophiolite

Fig. 11. Photographs of high strain fabrics and porphyroclasts indicating west-over-east thrusting in the Fox Island shear zone. (a) Sigma-shaped asymmetric porphyroclasts of quartz indicating a west-over-east sense of shear (top of the photo moved to the upper left). (b) Doubly-plunging fold in graywacke/argillite in a shear zone on the eastern side of Fox Island. (c) Staircase trajectory in sigmoidal-shaped quartz layers in high-grade psammite of the western belt in the Fox Island shear zone, also indicating west-over-east shearing (top of the photo moved to the left).

8. Fox Island Shear Zone

653

rocks formed from these data. In comparison, eastern unit sedimentary rocks contain minor biotite, muscovite, and calcite. Locally abundant chlorite in the volcanic rocks provides a greenish tint and greasy feel to the rocks. One-to-three centimeter chlorite and amphibole porphyroclasts locally produce a schistose fabric. Recrystallised quartz-augen are present in all rock types. The central and western units are distinguished from one another on the basis of lithology and degree of deformation. The central unit is characterized by thick sandstone sequences, whereas the western unit is dominated by volcanic tuffs and sills. Deformation increases across the central and western units to the west. The units are separated by a mylonitic shear zone that roughly follows the changing lithology from the sandstone-rich to the volcanic-rich unit and is mapped as the thrust surface in figures showing the shear zone. Deformation within the central unit is characterized by abundant quartz veins filling bedding-parallel cleavage planes. Secondary foliations typically are isoclinally folded, whereas porphyroclasts, such as sigmoidal augen, provide excellent kinematic indicators (Fig. 11). Disruption of bedding occurs as boudinage of sandy and silty layers with finer grained rocks exhibiting flow fabrics around the boudins. Minor shortening of beds is evident as small-scale ramps duplexing competent layers. All fabrics exhibiting ductile flow features indicate west-over-east kinematics (Fig. 11). As is observed on Resurrection Peninsula, minor but ubiquitous brittle late down-to-thewest normal fabrics are locally superimposed on rocks of the central unit. This deformation is manifested as quartz-filled tension gashes in siltstone and sandstone, and as kink banding in finer grained rocks. Metamorphic rocks of the western unit were highly deformed along the ductile thrust fault that separates the central from the western portions of the island. Disrupted bedding is nearly ubiquitous as multiple duplexed packages are ramped one upon the other making identification of individual beds nearly impossible. Thin quartz veins filling planar foliations are typically isoclinally folded whereas thicker quartz veins are elongated in sigmoidal augen with strongly elongated shear-sense indicators (Fig. 11). Mylonitic C-S fabrics also indicate west-over-east shearing. Thrusting produced alternating layers of volcanic and pelitic rocks at all scales. Rusty weathering, broken beds of graywacke form lozenges in thrust-imbricated packages 2– 3 m in length and 0.5–1 m thick. These duplexed thrust packages are separated by 2 to 3-cm thick shear zones expressed by finer grain size and a closely spaced foliation. Similar deformation is observed in duplexed thrust packages at the scale of tens of meters made up of many smaller ramped packages. Motion along the Fox Island shear zone produced some large-scale structures. In Sunny Cove the Fox Island shear zone is associated with upright, steeply west-plunging isoclinal folds. Very large (∼ 10 m) laterally extensive boudins are visible in the cliff faces along the south side of the island. Some of these boudins can be traced laterally through a portion of the island and are observed where the beds reappear on the opposite side.

654

Chapter 20: The Resurrection Peninsula Ophiolite

9. INTERPRETATION OF SEDIMENTARY FACIES OF THE HUMPY COVE FORMATION Flysch of the Resurrection Peninsula may be expected to show features within the graded beds that illustrate the type of flow regime present when they were deposited. However, primary depositional features such as cross-bedding, parallel-laminations, ripple marks, flute casts, and load casts are only rarely observed, precluding assignment of typical Bouma Ta-Te classification schemes, or submarine fan facies models (Mutti and Ricci Lucchi, 1978; Walker, 1979, 1992; Howell and Normark, 1982; Miall, 1984). Examination of units where deformation appears relatively minor indicates that these features remain absent. Depositional features that may be expected even in a low flow regime apparently were not preserved during initial deposition. Sandstone units in Thumb Cove increase in thickness above the contact with the ophiolite, forming a coarsening-upward sequence. This is expected in settings such as that interpreted here for the Humpy Cove Formation, in which the depositional site is conveyed from a ridge to an abyssal or outer trench slope environment, into a zone of active terrigeneous sedimentation in a trench. Channel features are present only at the top of the Humpy Cove Formation, with erosional features at the base of the sandstone bodies absent, indicating that the lateral extent of the sandstones is not controlled by channels, but by a general waning of the transport process (e.g., Mutti and Ricci Lucchi, 1978; Howell and Normark, 1982; Shanmugam and Moiola, 1985). Thin-bedded sandy siltstones and shales interbedded with pillow basalts of the ophiolite suggest deposition away from the main sink of coarse-grained terrigeneous sediments. These sandy siltstones were deposited during active magmatism at the ridge when the ophiolite was at a distance from the trench, and therefore represent a distal depositional environment. Deposition along the active ridge axis suggests that the turbidity current must have greatly reduced its velocity during its encounter with the topographically high ridge, permitting only the finer-grained suspended particles to travel up the ridge crest. Therefore sedimentation is expected to occur as fine-grained, thinly-bedded turbidites.

10. SEDIMENTATION AT A TRIPLE JUNCTION DURING A RIDGE-TRENCH INTERACTION Subduction of the Kula-Farallon ridge beneath the North American continental margin is well documented (Marshak and Karig, 1977; Sisson and Hollister, 1988; Lonsdale, 1988; Bol et al., 1992; Bradley et al., 1993, 2003; Sisson et al., 1994; Lytwyn et al., 1997; Kusky et al., 1997a, 1997b; Sisson et al., 2003). Subduction of the Kula-Farallon ridge accompanied by eastward migration of the trench-ridge trench triple junction along the trench axis (Bradley et al., 1993, 1999b) formed a topographic high that moved east along the trench axis and which may have acted as a barrier that inhibited sediment transport (Fig. 12). As sediment transport was dominantly (though not universally) westward along

10. Sedimentation at a Triple Junction during a Ridge-Trench Interaction 655

Fig. 12. Schematic diagram showing sedimentation patterns around postulated tectonic setting for the formation and history of the Resurrection Peninsula ophiolite. The ophiolite formed along the Kula-Farallon spreading center at 57 ± 1 Ma and was transported toward and emplaced into the Chugach accretionary prism. Continentally-derived flysch filtered along the ridge axis became interbedded with pillow lavas, and some may have escaped out along transform offsets to add a terrigeneous component to those sediments deposited on top of the ophiolite as it was transported toward North America (Humpy Cove Formation). As the ophiolite entered the trench, turbidite sedimentation became dominated by thicker, more massive turbidites with a stronger biogenic component, suggesting higher marine productivity (upwelling or shallow water?) in the source area. CCD is carbonate compensation depth.

656

Chapter 20: The Resurrection Peninsula Ophiolite

the trench axis (Nilsen and Zuffa, 1982; Plafker et al., 1994; Kusky et al., 1997b), several things may have occurred during the eastward migration of the ridge (Fig. 12). Uplift and erosion of the accretionary prism over the migrating ridge would redeposit previously deposited sediments on either side of the migrating ridge (e.g., DeLong and Fox, 1977; Forsythe and Nelson, 1985; Nelson et al., 1993). Sediment being transported westward along the trench axis would encounter the ridge axis and begin to slow, potentially depositing coarser material than would otherwise be deposited if there were no barrier to sediment transport. Where sediments may have jumped the ridge, the effect may be analogous to channel overbank deposits allowing only the finer grained sediments to spill over (Fig. 12). Sediment sources perpendicular to the trench axis would supply additional input to the main sedimentary feeder channels (Nilsen and Zuffa, 1982), although sediment input on the western side of the ridge axis may still be expected to travel west along the trench to be deposited away from the ridge. A transition from black shales and siltstones to turbidites with fewer black shales and then massive sandstones is recorded in the stratigraphic profiles of the Humpy Cove Formation. This change in lithology occurs by approximately 1400 m above basalt in all three stratigraphic sequences. Most chemical profiles also exhibit large concentration changes at 1400 m with decreases in terrigeneous elements and increases in biogenic elements that is best illustrated in the Sc and Sr chemical profiles (Kusky and Young, 1999). The close association between variations in chemical stratigraphy and changing lithologies between the three sequences at 1400 m suggests that the same event is recorded in each of the sequences. The transition from a regime dominated by thin-bedded turbidites to one dominated by thicker sandstone beds is here interpreted as a transition from distal to more proximal turbidite facies associated with transport of the ophiolitic basement to the Humpy Cove Formation toward the continent (Fig. 12). Turbidites interbedded with pillow lavas, and those just above the pillow basalts in Humpy Cove, are all thinly bedded, suggesting that they are the result of low flow velocity or distal turbidity currents. The transition to thicker-bedded sandstone turbidites stratigraphically up section is interpreted as increasing proximity to the continental margin where an increased volume of sediment was deposited. Early in the history of the ophiolite, distal turbidity currents produced a few pulses of thin-bedded siltstones, each approximately 10 m thick along the still active ridge. Since these also show a strong terrigeneous component, it is likely that they represent distal turbidites transported along the axial rift valley of the Kula-Farallon spreading center (Fig. 12). Sources could include both sediments transported down submarine canyons cutting through the accretionary prism, and turbidites that initially flowed along the trench, with sufficient velocity to jump and flow along the ridge (Fig. 12). After the ophiolite was conveyed away from the ridge, a 1000-m thick sequence of thin-bedded turbidites containing black shales, siltstones, and thin graywacke beds accumulated above the pillow basalts in Humpy Cove. Continued transport, both away from the ridge and toward the continent, gradually resulted in a depositional regime of increasingly proximal turbidite sedimentation characterized by lighter-colored shale beds with a greater proportion of sandstones

11. A Neogene Analog: Chile Rise and Taito Ophiolite

657

making up individual turbidites. This period of sedimentation may mark the transition of the location of the ophiolite from the outer trench slope to the floor of the trench where axially transported turbidites rapidly accumulated (Fig. 12). 11. A NEOGENE ANALOG: CHILE RISE AND TAITO OPHIOLITE The emplacement of the Taito ophiolite during subduction of the Chile Rise beneath the South American margin represents a Neogene example of ophiolite emplacement during ridge subduction (Herron et al., 1981; Schweller et al., 1981; Forsythe and Nelson, 1985; Cande and Leslie, 1986; Forsythe et al., 1986; Forsythe and Prior, 1992; Nelson and Forsythe, 1989; Bangs et al., 1992; Nelson et al., 1993; Lagabrielle et al., 1994; Sisson et al., 1994). Comparison with the Resurrection Peninsula ophiolite illustrates differences in sedimentation resulting from emplacement under different conditions: the Chile Rise strikes nearly parallel to the trench, whereas the Kula-Farallon ridge had a more oblique strike to the trench. The Taito ophiolite was emplaced at 3 Ma during passage of the Chile Rise along the south American margin near the Gulfo de Penas basin (Nelson et al., 1993). The Taito sedimentary cover comprises thin marine sedimentary rocks overlying conglomeratecontaining clasts of the pre-Jurassic metamorphic basement structurally beneath the ophiolite. The absence of a thick sedimentary cover sequence is interpreted to be the result of ophiolite emplacement at the location where the Chile Rise came into contact with the continent. Passage of the ridge beneath the accretionary prism produced near trench plutonic rocks and localized uplift that exposed the prism above sea level shedding sediments to either side of the trench prior to obduction. The buoyant, topographically high ridge, maintained a very shallow-dipping angle of subduction. The close proximity of the Taito ophiolite to the point of contact between the Chile Rise and the continent suggests that the ophiolite was emplaced at very shallow levels in the accretionary prism. In contrast, the thick sequence of sedimentary rocks over the Resurrection Peninsula ophiolite and the ductile shearing along the Fox Island shear zone suggests that this ophiolite was emplaced into the accretionary prism at depth and exposed at the surface only after uplift and erosion. This agrees with the idea that the ophiolite was transported a maximum of approximately 90 km from the ridge crest during its 3.6 ± 1.4 Ma transport history.

12. IMPLICATIONS FOR UNDERSTANDING RIDGE-TRENCH ENCOUNTERS IN THE PRECAMBRIAN The southern Alaska convergent margin consists of belts of accreted PaleozoicCenozoic flysch, mélange, and ophiolites. Paleogene near-trench plutons intruded the margin diachronously from 61 Ma in the west to 50 Ma in the east during migration of a trench-ridge-trench triple junction. Recognizing these plutons as products of ridge subduction has several implications for forearc evolution and interpretation of linear belts of

658

Chapter 20: The Resurrection Peninsula Ophiolite

plutons in ancient orogens. Forearcs are not necessarily places characterized exclusively by high-P, low-T metamorphic series and a lack of plutonism, but may contain high-T, low-P metamorphism in association with belts of plutonic rocks if the forearc was affected by ridge subduction (Fig. 13). Similarly, belts of magmatic rocks in ancient orogens may not necessarily represent individual arc terranes, but could be a paired arc/forearc system that experienced ridge subduction (Fig. 13) (Kusky et al., 2003). The record of ridge subduction events varies considerably depending on plate geometry and rates of triple junction and slab window migration (e.g., Sisson et al., 2003). However, some of the hallmark signatures of ridge subduction in forearcs include along strike diachronous intrusion of TTG (tonalitic, trondhjemitic, granodiorite), to granitic plutons, high-T metamorphism, diachronous gold mineralization, and belts of anomalous complex faulting (Fig. 13). Structural, thermal and magmatic aspects of the Chugach terrane are similar to the geology of Archean granite-greenstone terranes. In both, deformation is locally melt-dominated, and plutons follow a low-K series from diorite to trondhjemite. Metamorphism is of a high-T, low-P series. Most Archean granite-greenstone terranes acquired their first-order structural and metamorphic characteristics at convergent plate margins, where large accretionary wedges similar in aspect to the Chugach, Makran, and Altaids grew through offscraping and accretion of oceanic plateaux, oceanic crustal fragments, juvenile island arcs, rifted continental margins, and pelagic and terrigeneous sediments. Some suites of TTG in these terrains appear to have been generated during ridge subduction events, suggesting that ridge subduction is an important process in continental growth. Ridge subduction was likely an important process in the Archean, when the total number of plates was higher, and the number of ridge-trench encounters was greater. The southern Alaska margin serves as a relatively modern example of processes important in Archean forearc evolution and continental growth.

13. FLYSCH AND MÉLANGE DOMINATED ACCRETIONARY PRISMS: PHANEROZOIC AND PRECAMBRIAN EXAMPLES Many contemporary and ancient accretionary prisms are dominated by two fundamentally different types of structural belts-mélange and relatively coherent flysch terrains (e.g., Fig. 1b). However, mélange terranes are very rare in Archean convergent margin sequences, and only slightly more common in Early Proterozoic orogens. Controls on the accretion of flysch and mélange terranes at convergent margins are poorly understood. The Chugach terrane of southern Alaska represents the Mesozoic outboard accretionary margin of the Wrangellian composite terrane. It consists of two major lithotectonic units, including Triassic-Cretaceous mélange of the McHugh Complex, and Late Cretaceous flysch of the Valdez Group. Mapping on the Kenai Peninsula, and in the eastern Chugach Mountains has clarified the nature of the boundary between the mélange and flysch terrains. In most places, the contact is a gradational or interleaved boundary between chaotically deformed mélange of argillite, chert, greenstone, and graywacke of the McHugh Complex, and a less chaotically deformed mélange of argillite and graywacke of the Valdez Group. The latter

13. Flysch and Mélange Dominated Accretionary Prisms: Phanerozoic and Precambrian Examples

659

Fig. 13. Three-dimensional sketch indicating the relative position and timing of ridge subduction, formation of the slab window, generation of the Nuka-Aialik, and Harris Bay plutons and their intrusion into the Resurrection ophiolite, and activation of the late out of sequence thrust faults including the Fox Island shear zone.

is known as the Iceworm mélange (Kusky et al., 1997a, 1997b), and is interpreted it as a contractional fault zone (Chugach Bay thrust) along which the Valdez Group was thrust beneath the McHugh Complex. The McHugh Complex had already been deformed and metamorphosed to prehnite-pumpellyite facies prior to formation of the Iceworm mélange.

660

Chapter 20: The Resurrection Peninsula Ophiolite

(a)

(b) Fig. 14. Schematic diagram showing possible different accretion styles for (a) a thinly and (b) a thickly sedimented oceanic plate. These differences may account for the paucity of mélanges in Archean terranes.

We interpret accretion of the mélanges of the McHugh Complex as a normal response to subduction of oceanic crust bearing a thin pelagic sedimentary veneer (Fig. 14), whereas accretion of the coherent thrust packages of the Valdez Group is interpreted as a normal wedge response to subduction of oceanic crust bearing a thick pile of loosely consolidated submarine fan and trench sediments (Fig. 14). Thrusting of the McHugh Complex over the Valdez Group and the formation of the Chugach Bay thrust and Iceworm mélange may represent a dramatic critical taper adjustment to these changing subduction regimes.

13. Flysch and Mélange Dominated Accretionary Prisms: Phanerozoic and Precambrian Examples

661

Davis et al. (1983) defined a popular Coloumb wedge model for the mechanics of accretionary prisms in which the stress is everywhere as large as possible, with the wedge thickening toward the rear because of the increased strength of rock in this direction. For a wedge of given strength (determined by rock rheology, basal resistance, pore fluid pressure, etc.), the forces that resist motion of the wedge are balanced by the forces that control the shape of the wedge (e.g., the thickness of the wedge determines the cross-sectional area across which the principal horizontal stress can act). In this way, the combination of surface slope plus basal decollement dip (critical taper) is maintained by deformation of the wedge, as material is added to the toe or base of the wedge. Platt (1986) analyzed the dynamics of accretionary wedges with negligible yield strength and demonstrated that underplating can oversteepen the surface slope, causing extension in the upper part of the rear of the wedge. This regains the equilibrium critical taper and causes uplift of metamorphic rocks in the rear of the accretionary wedge. These models are directly applicable to the Chugach accretionary wedge in that the McHugh complex is interpreted as material off-scraped intermittently during Triassic-Early Cretaceous subduction, with the establishment of a critically tapered wedge (Fig. 14). Attempted subduction of the thick sedimentary veneer of the Valdez Group during Maastrichtian-Paleocene time dramatically upset the critical taper by both rapid frontal accretion and increased underplating (Fig. 14b). The wedge’s response to this episode of rapid accretion was to deform internally to regain the critical taper. Rapid frontal accretion would cause the wedge to thicken to maintain the critical taper, and in the case of the Chugach terrane much of this thickening was accommodated on the Chugach Bay thrust, emplacing the older part of the wedge (McHugh Complex) over the younger part (Valdez Group). Formation of the Iceworm mélange may have been aided by fluids released from dewatering of the recently accreted Valdez Group escaping upward along the lower boundary of the relatively impermeable McHugh Complex. The Chugach Bay thrust may also have formed where it did because the protoliths of the Iceworm mélange represent a muddy facies of the Valdez Group with considerably more pore fluids that in other parts of the Valdez Group. Elevated pore fluid pressure within part of the accretionary wedge decreases the effective strength of the rock, providing a horizon for enhanced localized deformation. Increased underplating of the wedge during accretion of the Valdez Group caused enhanced uplift of the rear of the accretionary wedge, probably with associated extensional faulting, (e.g., Platt, 1986), providing a mechanism for the uplift and exposure of the metamorphosed McHugh Complex, and its emplacement over the Valdez Group on the Chugach Bay thrust (Fig. 14). Differences in accretion style analogous to the McHugh/Valdez dichotomy are suggested to be characteristic of subduction of thinly versus thickly sedimented oceanic plates, and we cite here a few additional examples to emphasize the generality of this model. Late Cretaceous uplift and erosion of the Mongolia-Okhotsk intracontinental collision zone in Asia shed a large amount of sediments to the south and east (Klimetz, 1983; Nanayama, 1992), which triggered rapid frontal accretion and growth of accretionary complexes along the northwest Pacific rim (Kimura, 1994). Kimura (1994) also documented that rapid underplating by attempted subduction of thick sedimentary caps on the downgoing plates caused uplift of metamorphic rocks of the Kamuikotan complex of Hokkaido

662

Chapter 20: The Resurrection Peninsula Ophiolite

and the Susunai complex of Sakhalin. Taira et al. (1988) have documented rapid growth of the Nankai accretionary wedge in response to the Izu collision. Tertiary uplift and erosion of the Himalayan mountain range from the India-Asia collision caused rapid growth of the Makran and Sunda accretionary prisms from sediments supplied rapidly to the trench by the Indus and Ganges Rivers (Graham et al., 1975; Burbank et al., 1996; Sengör and Natal’in, 1996). Similarities are also seen between the model presented here and the ongoing collision of the Yakutat terrane with the southern Alaska margin. This collision has caused uplift and erosion of the Chugach and St. Elias Mountains, deposition of a thick sequence of Miocene and younger clastic sediments in the Aleutian trench, and rapid accretion at the toe of the Aleutian accretionary prism (Plafker et al., 1994). Climate, as well as collisions, may play a significant role in episodic growth and structural style of accretionary prisms (e.g., Hay, 1996; von Huene and Scholl, 1991). Bangs and Cande (1997) have shown how a thickly-sedimented segment of the Chile trench south of the Chile triple junction in the area strongly influenced by glacial climate corresponds to rapid growth and accretion in the accretionary prism. The subducting Chile ridge acts as a topographic barrier to sediments shed into the trench by glacial erosion in the south, and growth of the prism north of the triple junction is less rapid than that to the south of the triple junction. Periods of rapid accretion were also noted during intervals when glaciation extended farther north and erosion accelerated trench sedimentation north of the triple junction. Subduction of the Chile ridge causes significant erosion of the accretionary complex, making the pre-ridge subduction record south of the northward migrating triple junction difficult to interpret. We suggest that, in general, subduction of slabs with a thin veneer of sedimentary cover may lead to subduction erosion (e.g., von Huene and Scholl, 1991) or to the formation of mélanges in accretionary wedges, whereas subduction of thickly-sedimented slabs may generally result in large-scale accretion of relatively coherent packages separated by zones of shearing and type I mélange formation. Using a constant slip-rate model, subduction of an oceanic slab with a thin sedimentary veneer will result in significantly higher shear strains in the boundary layer between the upper and lower plates than would subduction of an oceanic slab with a thick sedimentary cover. For example, if we assume one kilometer of thrusting per million years and we accommodate this strain in a 100-m thick sedimentary section, then shear strains of ten will be typical for the sedimentary cover and will likely involve significant mixing between layers. If the subducting oceanic plate has a 1-km thick sedimentary cover, then shear strains will be about one, and we may expect strains that are lower than in the mélange and perhaps more typical of a foreland fold-thrust belt. This model may also explain a fundamental temporal change in the style of subduction zone tectonism. Many workers have noted the near absence of subduction mélanges from Archean terranes, even those interpreted as ancient accretionary wedges (e.g., Kusky, 1989, 1990, 1991). Models for Archean plate tectonics predict that the planet may have had a greater plate boundary length, with more, smaller, faster moving plates than at present (e.g., Pollack, 1997). One consequence of such a model is that subducting plates would have tended to be thickly sedimented, with their proximity to active collisions along numerous plate boundaries. Subduction of these thickly sedimented plates would tend to

14. Ridge Subduction Model for Evolution Archean Greenstone Belts, TTG Magmas, and VMS Deposits

663

produce Archean accretionary wedges dominated by relatively coherent terranes and few mélanges, consistent with the geological record preserved in Archean granite greenstone terranes (e.g., de Wit and Ashwal, 1997).

14. RIDGE SUBDUCTION MODEL FOR EVOLUTION ARCHEAN GREENSTONE BELTS, TTG MAGMAS, AND VMS DEPOSITS Archean greenstone belts are 1- to 100-km scale terranes of linear to curvilinear volcanic and sedimentary supracrustal sequences mostly with tectonic boundaries. These greenstone belts typically record poly-phase deformation, greenschist to amphibolite facies metamorphism, multiple phases of granitoid intrusion, deposition of giant volcanogenic massive sulphide (VMS), banded iron formation, and gold deposits, consistent with complex geodynamic interactions between various geological processes operation at Archean subduction zones (Kusky and Polat, 1999). The formerly contiguous late Archean Abitibi-Wawa subprovince, Superior Province, is the largest Archean greenstone-granitoid terrane globally, exposed over a strike length of greater than 1000 km, and was built by the amalgamation of volcanic and sedimentary rocks and intrusion of granitoids with diverse geochemical compositions (Stott, 1997; Polat and Kerrich, 2001). The formation of the Wawa-Abitibi greenstone-granitoid terrane occurred over a relatively short period, from 2.75 to 2.65 Ga, diachronously from north to south by lateral spread of subduction-accretion complexes (Percival et al., 1994; Calvert and Ludden, 1999; Polat and Kerrich, 2001). Many geological features of the Wawa-Abitibi greenstone-granitoid terrane can be better explain by complex interactions between intra-oceanic arcs, and mantle plumes and subducted ridges. For example, the subduction of ridge systems followed by opening asthenospheric windows in the late Archean Wawa-Abitibi subduction zone may have provided optimized thermal conditions for the formation of temporally and spatially associated ultramafic to felsic volcanic rocks (e.g., boninites, picrites, adakites, magnesian andesites, rhyolites, etc.), intrusion of TTG plutons, amphibolite metamorphism, and the generation of VMS, BIF and lode gold deposits. 14.1. Ridge Subduction, T TG Petrogenesis and Continental Growth It is generally accepted that Archean continental crust grew by accretionary and magmatic processes taking place at convergent plate boundaries (Taylor and McLennan, 1995; Kusky and Polat, 1999; Polat and Kerrich, 2001). Many continental growth models suggest that 60–70 percent of the present-day continental crust was formed by the end of the Archan (see Taylor and McLennan, 1995; Rudnick, 1995). Polat and Kerrich (2000) showed from field relationships and geochemical considerations that late Archean continental crust in the southern Superior Province grew by mixing of oceanic plateau and subduction derived components (granitoids, tholeiitic to calc-alkaline bimodal volcanic rocks, trench turbidites) at convergent plate boundaries. Simple mixing calcula-

664

Chapter 20: The Resurrection Peninsula Ophiolite

tions suggest that 6–12 percent oceanic plateau, mixing with a 88–94 percent arc magmas, is required to produce late Archean continental crust in the southern Superior Province. Several geochemical studies have suggested that the growth of the Archean continental crust took place in convergent margin settings mainly by intrusion of TTG, which constitute the larger portions of the Archean continental crust (greater than 70 percent). These TTG magmas are thought to result from reworking of basaltic crust which was either subducted or accreted (Moorbath, 1977; Taylor and McLennan, 1985, 1988; Tarney and Jones, 1994; Polat and Kerrich, 2000); TTG are genetically linked to eclogites through reworking of basaltic crust (Rollinson, 1997; Rapp et al., 2003). However, the reworking mechanism(s) by which basaltic crust is transferred to TTG and an eclogitic residue in Archean convergent margins has not been fully understood. The origin of Archean TTG is mainly attributed to melting of subducted oceanic crust (Martin, 1999; Drummond et al., 1996; Rapp et al., 1999). Melting of subducted slabs alone, however, cannot explain the presence of the eclogitic residue in the continental lithosphere because the denser refractory eclogitic slabs would have been recycled into the deep mantle (MacDonough, 1991). To explain the presence of eclogites and TTG suits in the continental lithosphere, here we propose that partial melting of accreted and/or underplated oceanic plateaus and normal oceanic crust in the accretionary wedges and/or on the base forearc by upwelling of an asthenospheric window, following a ridge subduction episode, played a major role in the generation and preservation of the Archean continental crust. The partial melting of accreted basaltic crust, which was metasomatized by previous slab-derived melts and/or fluids, occurred under amphibolite to eclogite metamorphic conditions. Diaprically upwelling tonalite, trondhjemite, granodiorite, diorite, mozonite, and sanukitoid melts derived from partial melting of basaltic crust on the base of thickened accretionary complexes formed the continental crust. The complementary eclogitic residues contributed to the sub-continental lithospheric mantle and further growth of the lithospheric mantle resulting from underplating mantle plumes (see Wyman et al., 2002). TTG suites that intruded Archean oceanic island arc accretionary complexes formed the nuclei of intra-oceanic island arcs, subsequent accretion oceanic island arcs and oceanic plateaus, and oceanward migration of the arc-trench system resulted in the lateral growth of Archean continental crust in subduction zones. 14.2. Ridge Subduction and VMS Deposits There are several present-day examples of VMS deposits resulting from ridge subduction. The Bransfield Strait of the Antarctic Peninsula, for example, is currently undergoing extension due to subduction of the Antarctic-Phoenix spreading centre (Lawver et al., 1995). There is an extensive hydrothermal activity in Bransfield Strait, resulting in deposition of VMS (Lawver et al., 1995). Development of a mantle window following the ridge subduction may have triggered the extension of the forearc oceanic lithosphere by thermal erosion. An upwelling of mantle window beneath the extending forearc oceanic crust could have triggered intensive hydrothermal activity and deposition of the VMS in

15. Conclusions

665

Phanerozoic orogenic belts. Archean VMS deposits likely to have been formed by similar geodynamic processes. On the basis of striking similarities between the geological characteristics of Archean greenstone belt and those of the Cenozoic Alaskan subductionaccretion complex (see Crowe et al., 1992; Nelson and Nelson, 1993; Kusky et al., 2003; Sisson et al., 2003), we propose that many Archean VMS deposits formed during ridgesubduction events.

15. CONCLUSIONS Recognizing the Sanak-Baranof belt as a product of ridge subduction has several implications for forearc evolution and the interpretation of linear belts of plutons in ancient mountain belts. Forearcs are not necessarily exclusively places characterized exclusively by high-P, low-T metamorphic series and a lack of plutonism, but may contain high-T, lowP metamorphism in association with belts of plutonic rocks if the forearc was affected by ridge subduction. Similarly, belts of magmatic rocks in ancient mountain belts may not necessarily represent individual accreted island arc terranes, but could be a paired arc/forearc system that experienced a ridge subduction event. The record of ridge subduction events may vary considerably in individual examples depending on the plate geometry and rates of triple junction and slab window migration. However, some of the hallmark signatures of ridge subduction in forearcs include ophiolite emplacement, time-transgressive intrusion of tonalitic, trondhjemitic, granodiorite, to granitic plutons, high temperature metamorphism, and diachronous gold mineralization, and belts of anomalous complex faulting. Faults can control pluton emplacement, both as inactive zones of structural weakness, and as active dilational bends (Kusky et al., 2003). Structural, thermal and magmatic aspects of the Chugach terrane are similar to the geology of Archean greenstone-granodiorite terranes (Pavlis et al., 1988; Kusky, 1989; Barker et al., 1992; Kusky and Polat, 1999). In both, deformation is locally meltdominated, and plutons follow a low-K series from diorite to trondhjemite (Pavlis et al., 1988). Metamorphism is of a high-temperature low-pressure series. Ridge subduction was most likely an important process in the Archean, when the total number of plates was higher, and the number of ridge-trench encounters was greater. The southern Alaska margin then may serve as a relatively “modern” example of processes that were likely to have been important in Archean forearc evolution and continental growth. The Resurrection Peninsula and Knight Island ophiolitic lavas/dikes exhibit both calcalkaline and tholeiitic geochemistry possibly reflecting fractionation from multiple, compositionally diverse parental magmas coupled with variable assimilation of flysch metasedimentary rock. Parental magmas may be modeled as near-complete mixtures of geochemically diverse, near-instantaneous melts derived from a polybaric mantle column through melting of variably depleted mantle source (Lytwyn et al., 1997). Compositional variation among the parental magmas may primarily reflect different degrees of overall mantle melting. Slight alkali and LREE enrichments in Resurrection Peninsula and Knight Island lavas/dikes may indicate assimilation of flysch by invading N-MORB, suggesting

666

Chapter 20: The Resurrection Peninsula Ophiolite

formation along an oceanic (possibly Kula-Farallon) ridge system in close proximity to a trench. In contrast, refractory metabasalts from the Valdez/Orca accretionary complex possibly formed through inefficient mixing of near-instantaneous melts extracted mostly from shallower, more depleted regions of the mantle melting column (Lytwyn et al., 1997). Basalts from the Resurrection and Knight Island ophiolites, Ghost Rocks Formation, southern Kenai Peninsula and possibly Valdez/Orca accretionary complex may be related to high-angle subduction of the Kula-Farallon ridge system. Mantle convection beneath the mid-ocean ridge portion permits near-instantaneous melts from shallower, more refractory mantle sources to efficiently mix with melts from deeper, more fertile sources to produce MORB-like parental magmas. Ridge subduction, however, limits the top of the melting column to greater depths beneath the overiding plate and leads to inefficient pooling (mixing) of near-instantaneous melts as reflected in refractory basalts along the forearc. Basalts generated along a subducting spreading center, and erupted within or seaward of the forearc accretionary prism, may acquire a calc-alkaline overprint through assimilation of sediments and crustal rocks by upward-migrating melts along the asthenospheric window/accretionary prism interface. Resurrection Peninsula sediments from Thumb Cove, Humpy Cove, and Fox Island, located directly above the Resurrection Peninsula ophiolite, show a progression from thin turbidite deposition characterized by black shale, siltstone, and predominantly thin (10– 20 cm) sandstone beds, to thick massive turbidites characterized by shale and siltstone couplets with predominantly thick (greater than 50 cm) sandstone beds. This transition is interpreted as a shift from distal, deeper water turbidites deposited nearest the ophiolite as it migrated away from the ridge, to more proximal turbidite deposition as the ophiolite was transported into the trench and was emplaced in the accretionary prism forming along the North American continental margin. The Fox Island shear zone is a west-over-east ductile thrust that places older turbidites of the Valdez Group above younger turbidites deposited on the Resurrection Peninsula ophiolite. The shear zone is the upper bounding fault along which the ophiolite was emplaced, cutting sediments deposited above the ophiolite. It developed after passage of the Kula-Farallon ridge when MORB and sediments mixed at the base of the prism and created the granitic melts which intrude the shear zone a few kilometers south of Fox Island. Metamorphism and ductile flow fabrics are interpreted as the result of shearing at depths of less than 20 km (Kusky et al., 2003) aided by the heat of nearby plutons.

ACKNOWLEDGEMENTS This work was supported by NSF Grants EAR-9304647, and EAR- 9706699 awarded to T. Kusky, and by the U.S. Geological Survey. We thank D. Bradley, S. Nelson, S. Bloomer, S. Dolan, and K. Dietrich for stimulating discussions and assistance in the field.

References

667

REFERENCES Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoritic and solar. Geochim. Cosmochim. Acta 53, 197–214. Atwater, T., 1989. Plate tectonic history of the northeast Pacific and western North America. In: Winterer, E.L., Hussong, D.M., Decker, R.W. (Eds.), The Eastern Pacific Ocean and Hawaii, The Geology of North America, vol. N. Geological Society of America, pp. 1–39. Bangs, N.L., Cande, S.C., 1997. Episodic development of a convergent margin inferred from structures and processes along the southern Chile margin. Tectonics 16, 489–503. Bangs, N., Cande, S., Lewis, S.D., Miller, M., 1992. Structural framework of the Chile margin at the Chile ridge collision zone. Proceedings of the Ocean Drilling Program Initial Report 141, 11–21. Barker, F., Farmer, G.L., Ayuso, R.A., Plafker, G., Lull, J.S., 1992. The 50 Ma granodiorite of the eastern Gulf of Alaska: Melting in an accretionary prism in the forearc. Journal of Geophysical Research 97, 6757–6778. Bol, A.J., Gibbons, H., 1992. Tectonic implications of out-of-sequence faults in an accretionary prism, Prince William Sound, Alaska. Tectonics 11, 1288–1300. Bol, A.J., Roeske, S.M., 1993. Strike-slip faulting and block rotation along the Contact Fault System, Eastern Prince William Sound, Alaska. Tectonics 12, 49–62. Bol, A.J., Coe, R.S., Grommé, C.S., Hillhouse, J.W., 1992. Paleomagnetism of the Resurrection Peninsula, Alaska: Implications for the tectonics of southern Alaska and the Kula-Farallon Ridge. Journal of Geophysical Research 97, 17213–17232. Bradley, D.C., Kusky, T.M., 1992. Deformation history of the McHugh Complex, Seldovia Quadrangle, south-central Alaska. In: Bradley, D.C., Ford, A.B. (Eds.), Geologic Studies in Alaska by the U.S. Geological Survey during 1990. United States Geological Survey Bulletin 1999, 17–32. Bradley, D.C., Haeussler, P.J., Kusky, T.M., 1993. Timing of early Tertiary ridge subduction in southern Alaska. In: Till, A. (Ed.), Geologic Studies in Alaska by the U.S. Geological Survey during 1992. United States Geological Survey Bulletin 2068, 163–177. Bradley, D.C., Haeussler, P.J., Kusky, T.M., Goldfarb, R., 1994. Near trench magmatism, deformation, and gold mineralization during Paleogene ridge subduction in Alaska. Paper presented at SUBCON Conference, United States Geological Survey, Catalina Island, California. Bradley, D.C., Kusky, T.M., Haeussler, P.J., Karl, S.M., Donley, D.T., 1999a. Geologic map of the Seldovia quadrangle, scale 1:250 000. United States Geological Survey Open File Report 99, 18. Bradley, D.C., Parrish, R., Clendenin, B., Lux, D., Layer, P., Heizler, M., Donley, D.T., 1999b. New geochronological evidence for the timing of Early Tertiary ridge subduction in southern Alaska. United States Geological Survey Professional Paper. Bradley et al., 2003. Burbank, D.W., Beck, R.A., Mulder, T., 1996. The Himalayan foreland basin. In: Yin, A., Harrison, T.M. (Eds.), The Tectonic Evolution of Asia. Cambridge Univ. Press, pp. 149–188. Byrne, T., 1979. Late Paleocene demise of the Kula-Pacific spreading center. Geology 7, 341–344. Calvert, A.J., Ludden, J.N., 1999. Archean continental assembly in the southeastern Superior province of Canada. Tectonics 18, 412–429. Cande, S., Leslie, R., 1986. Late Cenozoic tectonics of the southern Chile trench. Journal of Geophysical Research 91, 471–496. Case, J.E., Barnes, D.F., Plafker, G., Robbins, S.L., 1966. Gravity survey and regional geology of the Prince William Sound epicenter region, Alaska. United States Geological Survey Professional Paper 543-C.

668

Chapter 20: The Resurrection Peninsula Ophiolite

Clendenin, W.S., 1991. Thermal history and vertical tectonics of the southern Alaska convergent margin. Ph.D. dissertation. Brown University, Providence, RI. Coleman, R.G., 1977. Ophiolites. Springer-Verlag, New York. Crowe, D.E., Nelson, S.W., Brown, P.E., Shanks III, W.E., Valley, J.W., 1992. Geology and geochemistry of volcanogenic massive sulfide deposits and related igneous rocks, Prince William Sound, south-central Alaska. Economic Geology 87, 1722–1746. Davis, D., Suppe, J., Dahlen, F.A., 1983. Mechanics of fold-thrust belts and accretionary wedges. Geophys. Res. 88, 1153–1172. DeLong, S.E., Fox, P.J., 1977. Geological consequences of ridge subduction. In: Talwani, M., Pitman III, W.C. (Eds.), Island Arcs, Deep Sea Trenches, and Back Arc Basins, vol. 1. AGU, Washington, DC, pp. 221–228. de Wit, M.J., Ashwal, L.D., 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35. Oxford Science Publications, Oxford. Drummond, M.S., Defant, M.J., Kepezhinskas, P.K., 1996. Petrogenesis of slab-derived trondhjemite-tonalite-dacite/adakite magmas. Transactions of the Royal Society of Edinburgh: Earth Science 87, 205–215. Dumoulin, J.A., 1987. Sandstone composition of the Valdez and Orca Groups, Prince William Sound, Alaska. United States Geological Survey Bulletin 1774. Dumoulin, J.A., 1988. Sandstone petrographic evidence and the Chugach-Prince William Terrane boundary in Southern Alaska. Geology 16, 456–460. Engebretson, J.M., Cox, A., Gordon, R.G., 1985. Relative plate motions between oceanic and continental plates in the Pacific basin. Geological Society of America Special Paper 206. Forsythe, R., Nelson, E., 1985. Geological manifestations of ridge collision: Evidence from the Golfo De Penas-Taito Basin, southern Chile. Tectonics 4, 477–495. Forsythe, R., Prior, D., 1992. Cenozoic continental geology of South America and its relations to the evolution of the Chile triple junction. Proceedings of the Ocean Drilling Program Initial Report 141, 23–31. Forsythe, R.D., Nelson, E.P., Carr, M.J., Kaeding, M.E., Herve, M., Mpodozis, C., Soffia, J.M., Harambour, S., 1986. Pliocene near-trench magmatism in southern Chile: A possible manifestation of ridge collision. Geology 14, 23–27. Gilbert, S.A., Casey, J.F., Bradley, D.C., Kusky, T.M., 1992. Geochemistry of silicilastic rocks in the Peninsular, Chugach, and Prince William Terranes: Implications for the tectonic evolution of south central Alaska (abstract). Geological Society of America Abstracts with Programs 24, A305. Goldfarb, R.J., Leach, D.L., Miller, M.L., Pickthorn, W.J., 1986. Geology, metamorphic setting, and genetic constraints of epigenetic lode-gold mineralization within the Cretaceous Valdez Group, south-central Alaska. In: Keppie, J.D., Boyle, R.W., Haynes, S.J. (Eds.), Turbidite Hosted Gold Deposits. Geological Society of Canada Special Paper 32, 87–105. Graham, S.A., Dickinson, W.R., Ingersoll, R.V., 1975. Himalayan-Bengal model for flysch dispersal in Appalachian-Ouachita system. Geological Society of America Bulletin 86, 273–286. Haeussler, P., Bradley, D.C., Goldfarb, R., 2003. Brittle deformation along the Gulf of Alaska margin in response to Paleocene-Eocene triple junction migration. In: Sisson, V., Roeske, S., Pavlis, T. (Eds.), Geology of a Transpressional Orogen Developed During Ridge-Trench Interaction Along the North Pacific Margin. Geological Society of America Special Paper 371, 119–140. Harris, N.R., Sisson, V.B., Wright, J.E., Pavlis, T.L., 1996. Evidence for Eocene mafic underplating during fore-arc intrusive activity, eastern Chugach Mountains, Alaska. Geology 24, 263–266. Hay, W.W., 1996. Tectonics and climate. Geologische Rundschau 85, 409–437.

References

669

Herron, E., Cande, S., Hall, B., 1981. An active spreading center collides with a subduction zone: A geophysical survey of the Chile margin triple junction. In: Klum, L.D., Dymond, J., Dausch, E.J., Hussong, D.M. (Eds.), Nazca Plate: Crustal Formation and Andean Convergence. Memoirs of the Geological Society of America 154, 683–701. Hickey, R.L., Frey, F.A., Gerlach, D.C., Lopez-Escobar, L., 1986. Multiple sources for basaltic arc rocks from the southern volcanic zone of the Andes (34◦ –41◦ S): Trace element and isotopic evidence for contributions from subducted oceanic crust, mantle, and continental crust. J. Geophys. Res. 91, 5963–5983. Hill, M.D., 1979. Volcanic and plutonic rocks of the Kodiak-Shumagin Shelf, Alaska: Subduction deposits and near-trench magmatism. Ph.D. thesis. Univ. of California, Santa Cruz, p. 274. Hillhouse, J.W., Gromme, C.S., 1977. Paleomagnetism of the Triassic Nikolai greenstone, McCarthy Quadrangle, Alaska. Canadian Journal of Earth Science 14, 2578–2592. Hillhouse, J.W., Gromme, C.S., Csejtey, B., 1985. Tectonic implications of paleomagnetic poles from lower Tertiary volcanic rocks, south central Alaska. Journal of Geophysical Research 90, 12 523–12 536. Howell, D.G., Normark, W.R., 1982. Sedimentology of submarine facies. In: Scholle, P.A., Spearing, D. (Eds.), Sandstone Depositional Environments. American Association of Petroleum Geologists Memoirs 31, 361–404. Hudson, T., 1983. Calc-alkaline plutonism along the Pacific rim of southern Alaska. In: Roddick, J.A. (Ed.), Circum-Pacific Plutonic Terranes. Geological Society of America Memoir 10, 159–169. Jones, D.L., Clark, S.H.B., 1973. Upper Cretaceous (Maestrichtian) fossils from the Kenai-Chugach Mountains, Kodiak and Shumagin Islands, southern Alaska. United States Geological Survey Journal of Research 1, 125–136. Jones, D.L., Siberling, N.J., Berg, H.C., Plafker, G., 1981. Tectonostratigraphic terrane map of Alaska. United States Geological Survey Open File Report 81, scale 1:250 000. Kaeding, M., Forsythe, R.D., Nelson, E.P., 1990. Geochemistry of the Taitao ophiolite and neartrench intrusions from the Chile margin triple junction. J. S. Am. Earth Sci. 3, 161–177. Kimura, G., 1994. The latest Cretaceous-early Paleogene rapid growth of accretionary complex and exhumation of high pressure series metamorphic rocks in northwestern Pacific margin. Journal of Geophysical Research 99, 22147–22164. Klein, E.M., Karsten, J.L., 1995. Ocean-ridge basalts with convergent-margin geochemical affinities from the Chile Ridge. Nature 374, 52–57. Klimetz, M.P., 1983. Speculations on the Mesozoic plate tectonic evolution of eastern China. Tectonics 2, 139–166. Kusky, T.M., 1989. Accretion of the Archean Slave Province. Geology 17, 63–67. Kusky, T.M., 1990. Evidence for Archean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10, 820–841. Kusky, T.M., Bradley, D.C., 1999. Kinematics of mélange fabrics: Examples and Applications from the McHugh Complex, Kenai Peninsula, Alaska. Journal of Structural Geology 21, in press. Kusky, T.M., Polat, A., 1999. Growth of granite-greenstone terranes at convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73. Kusky, T.M., Young, C., 1999. Emplacement of the Resurrection Penninsula ophiolite in the southern Alaska forearc during a ridge-trench encounter. Journal of Geophysical Research 104, 29,025– 29,054.

670

Chapter 20: The Resurrection Peninsula Ophiolite

Kusky, T.M., Bradley, D.C., Haeussler, P.J., 1997a. Progressive deformation of the Chugach accretionary complex, Alaska, during a Paleogene ridge-trench encounter. Journal of Structural Geology 19, 139–157. Kusky, T.M., Bradley, D.C., Haeussler, P.J., Karl, S., 1997b. Controls on accretion of flysch and mélange belts at convergent margins: Evidence from the Chugach Bay thrust and Iceworm mélange, Chugach Terrane, Alaska. Tectonics 16, 855–878. Kusky, T.M., Bradley, D.C., Donley, D.T., Rowley, D., Haeussler, P., 2003. Controls on intrusion of near-trench magmas of the Sanak-Baranof belt, Alaska, during Paleogene ridge subduction, and consequences for forearc evolution. In: Sisson, V.B., Roeske, S., Pavlis, T.L. (Eds.), Geology of a Transpressional Orogen Developed during a Ridge-Trench Interaction along the North Pacific Margin. Geological Society of America Special Paper 371, 269–292. Lagabrielle, Y., Le Moigne, J., Maury, R.C., Cotten, J., Bourgois, J., 1994. Volcanic record of the subduction of an active spreading ridge, Taito Peninsula (southern Chile). Geology 22, 515–518. Lawver, L.A., Keller, R.A., Fisk, M.R., 1995. Bransfield Strait, Antarctic Peninsula active extension behind a dead arc. In: Taylor, B. (Ed.), Backarc Basins and Magmatism. Plenum Press, New York, pp. 315–344. Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyer Le Bas, M.J., Sabine, P.A., Schmid, R., Sorensen, H., Streckeisen, A., Woolley, A.R., Zanettin, B., 1989. A Classification of Igneous Rocks and Glossary of Terms. Blackwell, Oxford. Lonsdale, P., 1988. Paleogene history of the Kula plate: Offshore evidence and onshore implications. Geological Society of America Bulletin 100, 733–754. Lull, J.S., Plafker, G., 1990. Geochemistry and paleotectonic implications of metabasaltic rocks in the Valdez Group, southern Alaska. United States Geological Survey Bulletin 1946, 29–38. Lytwyn, J.N., Casey, J.F., Gilbert, S., Kusky, T.M., 1997. Arc-like mid-ocean ridge basalt formed seaward of a trench-forearc system just prior to ridge subduction: An example from subaccreted ophiolites in southern Alaska. Journal of Geophysical Research 102, 10225–10243. Lytwyn, J., Gilbert, S., Casey, J., Kusky, T.M., 2000. Geochemistry of near-trench intrusives associated with ridge subduction, Seldovia Quadrangle, southern Alaska. Journal of Geophysical Research 105 (B12), 27,957–27,978. Marshak, R.S., Karig, D.E., 1977. Triple junctions as a cause for anomalously near-trench igneous activity between the trench and volcanic arc. Geology 5, 233–236. Martin, H., 1999. Adakitic magmas: modern analogues of Archean granitoids. Lithos 46, 411–429. McDonough, W.F., 1991. Partial melting of subducted oceanic crust and isolation of its residual eclogitic lithology. Philosophical Transactions of the Royal Society of London A 335, 407–418. Miall, A.D., 1984. Principles of Sedimentary Basin Analysis. Springer-Verlag, New York. Miyashiro, A., 1974. Volcanic rock series in island arcs and active continental margins. Am. J. Sci. 274, 321–355. Moorbath, S., 1977. Ages, isotopes and evolution of Precambrian continental crust. Chemical Geology 20, 151–187. Moore, J.C., Byrne, T., Plumley, P.W., Reid, M., Gibbons, H., Coe, R.S., 1983. Paleogene evolution of the Kodiak Islands, Alaska: Consequences of ridge-trench interaction in a more southerly latitude. Tectonics 2, 265–293. Mottl, M.J., 1983. Metabasalts, axial hot springs, and the structure of hydrothermal systems at midocean ridges. Geol. Soc. Am. Bull. 94, 161–180. Mullen, E.D., 1983. MnO/TiO2 /P2 O5 : A minor element discriminant for basaltic rocks of oceanic environments and its implications for petrogenesis. Earth Planet. Sci. Lett. 62, 53–62.

References

671

Mutti, E., Ricci Lucchi, F., 1978. Turbidites of the northern Apennines: Introduction to facies analysis. International Geological Review 20, 125–166. Translated by T.H. Nilsen. Nanayama, F., 1992. Stratigraphy and facies of the Paleocene Nakanogawa Group in the southern part of central Hokkaido (in Japanese). Journal of Geological Society of Japan 98, 1041–1059. Nelson, E.P., Forsythe, R.D., 1989. Ridge collision at convergent margins: Implications for Archean and post-Archean crustal growth. Tectonophysics 161, 307–315. Nelson, E., Forsythe, R., Diemer, J., Allen, M., Urbino, O., 1993. Taito ophiolite: A ridge collision ophiolite in the forearc of southern Chile (46◦ S). Revista Geológica de Chile 20, 137–164. Nelson, S.W., Nelson, M.S., 1993. Geochemistry of ophiolitic rocks from Knight Island, Prince William Sound, Alaska. In: Bacon, C.D., Till, A.B. (Eds.), Geologic Studies in Alaska by the U.S. Geological Survey. United States Geological Survey Bulletin 2068, 130–142. Nelson, S.W., Dumoulin, J.A., Miler, M.L., 1985. Geologic map of the Chugach National Forest, Alaska. United States Geological Survey Misc. Field Study Map MF-1645-B, scale 1:250 000. Nelson, S.W., Miller, M.L., Dumoulin, J.A., 1987. Resurrection and Knight Island ophiolites and recent faulting on Montague Island, southern Alaska. In: Centennial Field Guide, vol. 1, Cordilleran Section. Geological Society of America, pp. 433–438. Nelson, S.W., Miller, M.L., Dumoulin, J.A., 1989. Resurrection Peninsula Ophiolite. In: Nelson, S.W., Hamilton, T.W. (Eds.), Guide to the Geology of Resurrection Bay, Eastern Kenai Fjords Area, Alaska Guidebook. Geological Society of Alaska, Anchorage, pp. 10–20. Nicholas, A., 1989. Structure of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer Academic, Norwell, MA. Nilsen, T.H., Zuffa, G.G., 1982. The Chugach Terrane, a Cretaceous trench-fill deposit, southern Alaska. In: Leggett, K.J. (Ed.), Trench Forearc Geology. Blackwell, Cambridge, MA, pp. 213– 227. Niu, Y., Batiza, R., 1991. An empirical method for calculating melt compositions produced beneath mid-ocean ridges: Application for axis and off-axis (seamounts) melting. J. Geophys. Res. 96, 21,753–21,777. Pavlis, T., Sisson, V.B., 1995. Structural history of the Chugach metamorphic complex in the Tana River region, eastern Alaska, a record of Eocene ridge subduction. Geological Society of America Bulletin 107, 1333–1355. Pavlis, T.L., Monteverde, D.H., Bowman, J.R., Rubenstone, J.L., Reason, M.D., 1988. Early Cretaceous near-trench plutonism in southern Alaska: A tonalite-trondhjemite intrusive complex injected during ductile thrusting along the Border Ranges Fault system. Tectonics 7, 1179–1199. Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plate boundaries. In: Thorpe, R.S. (Ed.), Andesites. Wiley, New York, pp. 525–548. Pearce, J.A., Cann, J.R., 1973. Tectonic setting of basic volcanic rocks determined using trace element analysis. Earth Planet. Sci. Lett. 19, 290–300. Pearce, J.A., Norry, M.J., 1979. Petrogenetic implications of Ti, Zr, Y and Nb variations in volcanic rocks. Contrib. Mineral. Petrol. 69, 33–47. Pearce, J.A., Peate, D.W., 1995. Tectonic implications of the composition of volcanic arc magmas. Annu. Rev. Earth Planet. Sci. 23, 251–285. Pearce, J.A., Alabaster, T., Shelton, A.W., Searle, M.P., 1981. The Oman ophiolite as a Cretaceous arc-basin complex: Evidence and implications. Philos. Trans. R. Soc. London Ser. A 300, 299– 317. Percival, J.A., Stern, R.A., Skulski, T., Card, K.D., Mortensen, J.K., Begin, N.J., 1994. Minto block, Superior province: Missing link in deciphering assembly of the Craton at 2.7 Ga. Geology 22, 839–842.

672

Chapter 20: The Resurrection Peninsula Ophiolite

Plafker, G., 1969. Tectonics of the March 27, 1964, Alaska earthquake. United States Geological Survey Professional Paper 543I, 1–74. Plafker, G., Jones, D.L., Pesagno, E.A., 1977. A Cretaceous accretionary flysch and mélange terrane along the Gulf of Alaska margin. Accomplishment during 1976. United States Geological Survey Circular 751-B, B41–B43. Plafker, G., Keller, G., Barron, J.A., Blueford, J.R., 1985a. Paleontologic data on the age of the Orca Group, Alaska. United States Geological Survey Open File Report 85-429, 24. Plafker, G., Keller, G., Nelson, S.W., Dumoulin, J.A., Miller, M.L., 1985b. Summary of data on the age of the Orca Group. In: Bartsch-Winkler, S.W. (Ed.), The United States Geological Survey in Alaska: Accomplishments during 1984. United States Geological Survey Circular 967, 74–76. Plafker, G., Nokleberg, W.J., Lull, J.S., 1989. Bedrock geology and tectonic evolution of the Wrangellia, Peninsular, and Chugach terranes along the Trans-Alaska Crustal Transect in the Chugach Mountains and southern Copper River Basin, Alaska. Journal of Geophysical Research 94, 4255–4295. Plafker, G., Moore, J.C., Winkler, G.R., 1994. Geology of the southern Alaska margin. In: Plafker, G., Berg, H.C. (Eds.), The Geology of North America, vol. G 1, The Geology of Alaska. Geological Society of America, pp. 389–467. Platt, J.P., 1986. Dynamics of orogenic wedges and the uplift of high pressure metamorphic rocks. Geological Society of America Bulletin 97, 1106–1121. Plumley, P.W., Coe, R.S., Byrne, T., 1983. Paleomagnetism of the Paleocene Ghost Rocks Formation, Prince William Terrane, Alaska. Tectonics 2, 295–314. Pollack, H.N., 1997. Thermal characteristics of the Archean. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35. Oxford Science Publications, Oxford, pp. 223–232. Polat, A., Kerrich, R., 2000. Archean greenstone belt magmatism and the continental growth-mantle evolution connection: Constraints from Th-U-Nb-LREE systematics of the 2.7 Ga Wawa subprovince, Superior Province, Canada. Earth and Planetary Science Letters 175, 41–54. Polat, A., Kerrich, R., 2001. Geodynamic processes, continental growth, and mantle evolution recorded in late Archean greenstone belts of the southern Superior Province, Canada. Precambrian Research 112, 5–25. Rapp, R.P., Shimizu, N., Norman, M.D., Applegate, G.S., 1999. Reaction between slab-derived melts and peridotite in the mantle wedge: Experimental constraints at 3.8 GPa. Chemical Geology 160, 335–356. Rapp, R.P., Shimizu, N., Norman, M.D., 2003. Growth of early continental crust by partial melting of eclogite. Nature 425, 605–609. Richter, D.H., 1965. Geology and mineral deposits of central Knight Island, Prince William Sound, Alaska. Alaska Division of Mines and Minerals Geologic Report 16. Rickwood, P.C., 1989. Boundary lines within petrologic diagrams with use oxides of major and minor elements. Lithos 22, 247–263. Rollinson, H., 1997. Eclogite xenoliths in West African kimberlites as residues from Archean granitoids crust formation. Nature 389, 173–176. Rudnick, R., 1995. Making continental crust. Nature 378, 571–577. Saltus, R.W., Hudson, T.L., Connard, G.G., 1999. A new magnetic view of Alaska. GSA Today 93, 1–6. Schweller, W.J., Klum, L.D., Prince, R.A., 1981. Tectonics, structure, and sedimentary framework of the Peru-Chile trench. In: Klum, L.D., Dymond, J., Dausch, E.J., Hussong, D.M. (Eds.), Nazca

References

673

Plate: Crustal Formation and Andean Convergence. Memoirs of the Geological Society of America 154, 323–350. Sengör, A.M.C., Natal’in, B.A., 1996. Paleotectonics of Asia: fragments of a synthesis. In: Yin, A., Harrison, T.M. (Eds.), The Tectonic Evolution of Asia. Cambridge Univ. Press, pp. 486–640. Shanmugam, G., Moiola, R.J., 1985. Submarine fans and related turbidite sequences. In: Bouma, A.H. (Ed.), Frontiers in Sedimentary Geology. Springer-Verlag, New York, pp. 27–35. Sisson, V.B., Hollister, L.S., 1988. Low-pressure facies metamorphism in an accretionary sedimentary prism, southern Alaska. Geology 16, 358–361. Sisson, V.B., Pavlis, T.L., 1993. Geologic consequences of plate reorganization: An example from the Eocene southern Alaska fore arc. Geology 21, 913–916. Sisson, V.B., Hollister, L.S., Onstott, T.C., 1989. Petrologic and age constraints on the origin of a low-pressure/high temperature metamorphic complex, southern Alaska. Journal of Geophysical Research 94, 4392–4410. Sisson, V.B., Pavlis, T.L., Prior, D.J., 1994. Penrose Conference Report: Effects of triple junction interactions at convergent margins. GSA Today 4 (10), 248–249. Sisson, V.B., Pavlis, T.L., Roeske, S.M., Thorkelson, D.J., 2003. Introduction: An overview of ridgetrench interactions in modern and ancient settings. In: Sisson, V.B., Roeske, S.M., Pavlis, T.L. (Eds.), Geology of a Transpressional Orogen Development during Ridge-Trench Interaction along the North Pacific Margin. GSA Special Paper 371, 1–18. Stock, J., Molnar, P., 1988. Uncertainties and implications of the late Cretaceous and Tertiary positions of North America relative to the Farallon, Kula, and Pacific plates. Tectonics 7, 1339–1384. Stone, D.B., Panuska, B., Packer, D.R., 1982. Paleolatitude versus time for southern Alaska. Journal of Geophysical Research 87, 3697–3707. Stott, G.M., 1997. The Superior Province, Canada. In: de Wit, M., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 480–507. Sun, S.-s., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins. Geol. Soc. Spec. Publ. 42, 313–345. Taira, A., Katto, J., Tashiro, M., Okamura, M., Kodama, K., 1988. Cretaceous to Miocene accretionary prism. Mod. Geol. 12, 5–46. Tarney, J., Jones, C.E., 1994. Trace element geochemistry of orogenic igneous rocks and crustal growth models. Journal of the Geological Society of London 151, 855–868. Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition and Evolution. Blackwell, Cambridge, p. 312. Taylor, S.R., McLennan, S.M., 1988. The significance of the rare earths in geochemistry and cosmochemistry. In: Gschneidner Jr., K.A., Eyring, L. (Eds.), Handbook on the Physics and Chemistry of Rare Earths, vol. 2. Elsevier, New York, pp. 485–578. Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust. Review of Geophysics 33, 241–265. Tysdal, R.G., Case, J.E., 1979. Geologic map of the Seward and Blying Sound Quadrangles, Alaska. United Stated Geological Survey Misc. Inv. Ser. Map I-1150, scale 1:250 000. Tysdal, R.G., Case, J.E., Winkler, G.R., Clark, S.H.B., 1977. Sheeted dikes, gabbro, and pillow basalt in flysch of coastal southern Alaska. Geology 5, 377–383. von Huene, R., Scholl, D.W., 1991. Observations at convergent margins concerning sediment subduction, subduction erosion, and the growth of continental crust. Reviews of Geophysics 29, 279–316.

674

Chapter 20: The Resurrection Peninsula Ophiolite

Walker, R.G., 1979. Turbidites and associated coarse clastic deposits. In: Walker, R.G. (Ed.), Facies Models. Ainsworth, Mitchner, ON, pp. 91–104. Walker, R.G., 1992. Turbidites and submarine fans. In: Walker, R.G., James, N.P. (Eds.), Facies Models: Response to Sea Level Change. Geol. Assoc. Canada, Ontario, pp. 239–263. Weaver, J.S., Langmuir, C.H., 1990. Calculation of phase equilibrium in mineral-melt systems. Comput. Geosci. 16, 1–19. Wiltse, M.A., 1973. Fe-Cu-Zn massive sulfide deposits in an ancient outer arc ridge-trench slope environment. Geol. Soc. Am. Abstr. Programs 5, 122–123. Winkler, G.R., 1976. Deep Sea fan deposition of the Lower Tertiary Orca Group, Eastern Prince William Sound, Alaska. In: Miller, T.P. (Ed.), Recent and Ancient Sedimentary Environments in Alaska, Proceedings of the Alaska Geological Society, Anchorage, 1975. Alaska Geological Society, pp. 189–204. Winkler, G.R., 1992. Geologic map and summary geochronology of the Anchorage 1◦ × 3◦ Quadrangle, southern Alaska. U.S. Geol. Surv., Misc. Investigation Series, Map I-2283, scale 1:250 000. Winkler, G.R., Miller, M.L., Hoekzema, R.B., Dumoulin, J.A., 1984. Guide to the Bedrock Geology of a Traverse of the Chugach Mountains from Anchorage to Cape Resurrection. Alaska Geological Society, p. 40. Wyman, D., Kerrich, R., Polat, A., 2002. Assembly of Archean cratonic mantle Lithosphere and crust: Plume-arc interaction in the Abitibi-Wawa subduction-accretion complex. Precambrian Research 115, 37–62. Young, C.P., 1997. The Resurrection Peninsula ophiolite: Emplacement in the southern Alaska forearc and controls on sedimentation during ophiolite transport. M.A. thesis. Boston University, Boston, MA.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

675

Chapter 21

PHANEROZOIC ANALOGUES OF ARCHAEAN OCEANIC BASEMENT FRAGMENTS: ALTAID OPHIOLITES AND OPHIRAGS 1 AND B.A. NATAL’IN A.M.C. SENGÖR ¸

˙ Maden Fakültesi, Jeoloji Bölümü, Ayazaˇga 34469, Istanbul, ˙ ITÜ Turkey

To the memory of Academician Alexandr Valdemarovich Peive for his advocacy of ophiolites as records of past oceans and his recognition that the Altaids was a Phanerozoic factory for the continental crust

Ophiolites have long been regarded as rock associations typical of orogenic belts. In the framework of the theory of plate tectonics, they have been interpreted as remnants of the igneous basement of former oceans, the closure of which generated the orogenic belts. Such remnants occur in orogens either in the form of a complete suite displaying a sequence from pillow lavas and pillow lavas, sills and dykes, through a sheeted dyke complex to gabbros, peridotites and dunites to tectonised harzburgites, or they occur as dismembered pieces of such a suite. Following the 1972 Penrose definition, we restrict the term ophiolite to the more or less complete suite and call its various fragments formed by processes incorporating them into the continental crust ophirags. In the minds of most geologists, orogenic belts are linear/arcuate, long and narrow zones of intense deformation. That is why, irregularly shaped areas of widespread ‘orogenic deformation’ interspersed with abundant fragments of the members of the ophiolite association in the Precambrian, but especially in the Archaean, have been thought of as products of processes no longer operative. However, the geology of the Altaid orogenic system in Asia greatly resembles in its overall map aspects, lithological content, structural characteristics, and in the distribution and types of fragments of floors of former oceans to the Archaean granite-greenstone terrains. In the Altaids, ophiolites are now encountered in three main settings: (1) Ophiolites that occur as basement of ensimatic arcs, (2) Ophiolites and ophirags that occur in former forearcs now entrapped within transform sutures: (a) Ophiolites as backstop to accretionary wedges, (b) Ophirags within accretionary wedges, and (3) Ophiolites and ophirags in collisional suture zones that have usually evolved from members of the second category. Ophiolites and ophirags have a widespread distribution within the orogenic edifice. This distribution was brought about by processes that shaped the Altaid edifice, namely, generation of supra1 Also at ˙ITÜ Avrasya Yerbilimleri Enstitüsü, Ayazaˇga 34469, ˙Istanbul, Turkey.

DOI: 10.1016/S0166-2635(04)13021-1

676

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

subduction zone forearc basements created by pre-arc spreading, back-arc basin opening, subduction-accretion, trench-linked strike-slip faulting including arc slicing and arc shaving faults and associated ocean floor spreading processes, and collision of buoyant pieces along suture zones. Without appreciating the nature and sequence of these processes and their superimposition, it is impossible to understand the rules that govern the distribution of oceanic basement fragments in the Altaids. These processes have led to a tremendous degree of structural shuffling of previously distant environments and a large degree of dismembering of formerly more complete ocean floor fragments. The preservation is highly selective and favours upstanding and buoyant segments of ocean floors. Such pieces are embedded most commonly in metapelitic/metapsammitic or, more rarely, in serpentinitic matrices in mélange/wildflysch complexes. We contend that the same rules apply to the greenstone belts of the Precambrian and greatly hinder their deciphering in the absence of biostratigraphic control.

1. INTRODUCTION The purpose of this paper is to present an overview of the distribution, current tectonic position and mode of origin of the remnants of oceanic basement rocks in the Altaids as likely analogues of similar rocks in many Precambrian, but mainly Archaean, orogenic belts. We contend, following Burke et al. (1976), Sengör ¸ et al. (1993), Sengör ¸ and Natal’in (1996a, 1996b), and Kusky and Polat (1999), that the Archaean greenstone belts represent mostly subduction-accretion complexes that became arc basements by migration of magmatic arc fronts as a consequence of subduction back-stepping. The processes that created the Archaean greenstone belts were no different, therefore, from those now creating magmatic arcs and associated subduction-accretion complexes, with the exception of those processes related to the higher heat output of the planet in the Archaean. Before we describe the geology of the Altaids and of the oceanic basement remnants in them, we review, from a historical perspective, the reasons why the recognition of the ArchaeanAltaid similarity has long been hindered. These reasons are in part associated with the role ophiolites were believed to play in orogenic processes.

2. OPHIOLITES AND OROGENY: A HISTORICAL REVIEW AND CRITIQUE 2.1. Ophiolites and Plate Tectonics The concept of ophiolite gained great popularity after the advent of plate tectonics, because ophiolites were thought to represent pieces of crust and upper mantle of the floors of now vanished oceans (see, for example, Coleman, 1977, parts I and II; also see the contributions in Dilek and Newcomb, 2003). Ophiolites were hailed as guides to the solution of two main problems: (1) identification of places where former oceans had vanished (i.e., suture zones)

2. Ophiolites and Orogeny: A Historical Review and Critique

677

and (2) understanding of the nature of the oceanic crust and upper mantle and all the ensimatic constructs such as island arcs, guyots, oceanic plateaux and fracture zone crust and mantle that may get preserved as pieces of ‘oceanic crust and mantle’. An associated problem concerned the mechanism by which ophiolites were brought into juxtaposition with continental crust. All three of these problems were initially approached within the context of the large-scale geometric properties of plate tectonics and its immediate predecessor, the two-dimensional spreading-subduction hypothesis of Hess (1962) and Dietz (1961). 2.2. Ophiolites as Ocean Floor Remnants It quickly became accepted that the basic ophiolitic pseudostratigraphy correlated well with the geophysical layering of the oceanic crust. Initially, such correlations (e.g., Dietz, 1963) were based on a crude (and largely incorrect) oceanic model of mainly serpentinitic layer 3 underlying a basaltic + sedimentary layer 2, which then underlay a largely unconsolidated sediment cover forming the layer 1 (Hill, 1957; Hess, 1962). However, Bishopp (1952) already had suggested that the Troodos Massif in Cyprus, displaying a clear transition downwards from pillow lavas and pillow lavas and dykes, through a sheeted dyke complex to gabbros, peridotites and dunites to tectonised harzburgites, was most likely a piece of oceanic crust. Accumulating observations on other ophiolites in the world revealed a similar sequence (e.g., Davies, 1968, in Papua New Guinea; Reinhardt, 1969, in Oman; Bailey et al., 1970, in California; for a general assessment, see Peive, 1969), which rapidly led to a corrected and more detailed correlation (see the summaries in Coleman, 1971, Fig. 1, and Dewey and Bird, 1971, Figs. 1 and 2, and references therein). In the seventies a number of models were proposed for the detailed geometry and kinematics of spreading centres on the basis of ophiolite data (e.g., Greenbaum, 1972, 1977; Gass and Smewing, 1973; Sleep, 1975; Coleman, 1977; Dewey and Kidd, 1977). Dewey (1974, 1976) suggested, on the basis of the short time interval between generation and obduction, and of the reconstructed tectonic setting, of the Caledonian/Appalachian ophiolites, that most major ophiolite nappes may have originated in back-arc basins rather than in major oceans. While much ink was being spilled in correlating ophiolites and the oceanic crust (in major oceans and in marginal basins), Miyashiro (1973) made the seminal suggestion, on the basis of geochemistry, that the Troodos ophiolite may have been born in an island arc setting. It was later pointed out that the Troodos could not have been a ‘normal’ island arc, because the sheeted dykes clearly indicate an environment of very considerable extension, similar to mid-oceanic spreading centres, not habitually encountered in active island arcs (e.g., Saunders et al., 1980). This dilemma was resolved, when Taylor and Karner (1983) showed that ‘back-arc basin spreading’ had occurred in the Marianas and the South Scotia arcs before an arc was established. Pearce et al. (1984) called this process ‘pre-arc spreading’ and showed that most Neo-Tethyan ophiolite nappes south and east of the Carpathians had indeed formed in such environments of ‘pre-arc spreading’. Sengör ¸ (1990) pointed out that most large ophiolite nappes in the world probably formed in such a setting and that this may be a unique mechanism of forming giant ophiolite nappes.

678

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

As giant ophiolite nappes were further autopsied, however, it turned out that not only supra-subduction zone pre-arc environments, but also fracture zone environments were represented in them. Saleeby (1977) first drew attention to this fact by interpreting the Kings-Kaweah ophiolitic mélange belt in California as a fossil fracture zone (also see Saleeby et al., 1978; Saleeby, 1981). Shortly afterwards, Karson and Dewey (1978), Simonian and Gass (1978), and Lemoine (1980) documented segments of fracture zones preserved in ophiolite nappes. This strengthened a speculation that Dewey had first emphasised in 1975, namely that the vicissitudes of finite plate evolution inevitably lead to former transform faults turning into subduction zones. This had been implicit in the classical Dewey and Bird (1971) paper (see their Fig. 4), but the realisation that ophiolites preserve sections of fracture zones greatly encouraged the interpretation that such subduction zones as the Marianas and the Aleutians, which are nearly orthogonal to former magnetic sea-floor spreading anomalies in the upper plate, most probably nucleated on former fracture zones (Casey and Dewey, 1984). One such case may now be just in the process of origination in the Gorringe Bank on the Azores Fracture Zone (Auzende et al., 1984). Thus obduction began to be thought also to be related to fracture zone locations in the oceans. It was also thought clear that along the trend of mountain ranges long and narrow belts of ophiolitic material with dominantly steep fabric, popularised by Gansser’s (1964, 1966) ‘Indus suture’, marked where two continents had been apposed after the intervening oceans had entirely disappeared by subduction. Both in the Himalaya and the Middle East, large, coherent ophiolite nappes were shown to root into such suture zones (Gansser, 1964; Ricou, 1971). Hamilton (1969) (also Hamilton, 1970) and later Ernst (1970) and Hsü (1971) followed an earlier speculation by Dietz (1963) by arguing that ophiolitic mélange was a chaotic mixture of ophiolite pieces, sedimentary rocks of diverse origin and metamorphic rocks of HP/LT type, mixed and further churned along a subduction zone (see Sengör, ¸ 2003, for the evolution of the mélange concept). They argued that present-day subduction-accretion complexes at the snout of subduction zones were most likely sites of mélange origin and accumulation. Although Dietz (1963) had approached the problem of the mixing of oceanic basement rocks with what he called ‘eugeosynclinal prisms’ from the viewpoint of a world-wide review, both Hsü and Hamilton considered the mélange problem from a Pacific Ocean perspective, where both active and fossil subduction-accretion complexes still largely face the ocean. However, earlier, Gansser (1955) had shown that ophiolitic mélange (called ‘coloured mélange’ by him: A. Gansser, personal communication, 7th May 2003) also occurred along narrow belts of intense deformation in the Alpide belts of the Middle East. When plate tectonics was applied to the interpretation of these orogenic belts, it was thought that these narrow mélange belts were simply sutures (Dewey and Bird, 1970; Dewey et al., 1973). This line of thought was hardly unexpected in view of a widespread conviction that ‘Alpine-type serpentinites occur in linear swarms of subparallel or en échelon masses along the axes of old and new belts of folded mountains, . . .’ (Dietz, 1963, p. 947). In fact, the main Zagros crush belt was shown to have along it both mélange stringers and coherent ophiolite nappes (Stöcklin, 1968). Mélange belts also turned up un-

2. Ophiolites and Orogeny: A Historical Review and Critique

679

der many major ophiolite nappes (e.g., Gansser, 1964; Reinhardt, 1969) corroborating the view that mélanges marked sites of sutures. 2.3. Ophiolites as Suture Markers: A Historical Perspective It is important to underline the fact that all authors, including Dewey (2003), have emphasised the role of the ophiolites in the classical Alpine- and Himalayan-type (sensu Sengör, ¸ 1990, 1992) of long and narrow orogenic belts that result from continental collision. Even those authors writing from a non-collisional, Pacific, viewpoint have depicted the ophiolitic mélange belts as long and narrow units attached to the prows of arcs (for a recent example, see Moores, 1998). The consideration of ophiolites along narrow, well-defined zones is hardly surprising in a retrospective interpretation of the history of the ophiolite concept in the early 20th century. In 1904, Eduard Suess published a two-and-a-half page note, with no figures, in the proceedings of the French Academy of Sciences (Suess, 1904). In it, he noted the discovery of overthrusts in the Alps by French and Swiss geologists. He then compared the frontal regions of overthrusts with the internal ogives at one of the termini of the ice cap in Greenland between Kangersuk and the Nasasuak nunatak and suggested that such ogives were homologous to the arcs of overthrusts (arc de charriage). Suess emphasised that deep moraines are brought to the surface along such listric thrust faults. Suess argued that to understand the nature of this ‘grand phenomenon’ of overthrusting one must look at the ‘roots’ of the nappes. He noted that the roots of the grand Alpine nappes exhibit much mafic and ultramafic rock and he (incorrectly) correlated the ultramafics of the Ivrea Zone with some of the Penninic ophiolites (because he had remained ignorant of the associated submarine lavas in the ophiolites as opposed to the entirely plutonic/metamorphic subcontinental Ivrea body: see Bailey’s point in Bailey (1944, p. 755)). Suess’ next point concerned the exotic blocks of the Himalaya and the associated mafic and ultramafic rocks. He supported the view that they belong to an allochthonous unit expelled from the north and northeast. The implication that a phenomenon similar to the one seen in the Alps is clear. In a short paragraph, he compared the island festoons of the western Pacific with thrust fronts. At the end of his note, Suess mentioned Auguste Daubrée’s studies (Daubrée, 1879, 1888; see Sengör, ¸ 2003) showing that ultramafic rocks represent the deeper parts of the earth and emphasised their frequent association with fronts of nappes, and the observation that they never appear in forelands of mountain belts. Suess left it there, but his implication is clear: orogeny somehow brings rocks that usually form deep in the earth to the surface by means of very large overthrusts. He noted the association of jumbled rock masses, which we now call mélange, with this phenomenon (Sengör, ¸ 2003). This is the point that Gustav Steinmann followed up and showed that in the extremely deformed and jumbled rock associations of the so-called Aufbruchszone (= zone of piercing) in Graubünden in eastern Switzerland, the bottom of the immense Austroalpine nappe system was exposed and exhibited a thin zone of extremely deformed and jumbled rock

680

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

association (composite thickness about 700 m, but in many places reduced to mere slivers: Trümpy, 1975). Before the advent of the nappe theory, the Aufbruchszone was assumed to represent a submarine ridge, separating the Western Alps from the Eastern Alps and characterised by the ‘eruption’ of gabbros and peridotites. Steinmann had selected this area for a detailed study by himself and his students. Schardt’s classic 1893 publication reinterpreting the entire Préalpes as a nappe prompted him to test the theory of nappes on Bündner examples. He found that although Schardt’s and later Lugeon’s parallelisation of the nappe structures in western Switzerland with those of eastern Switzerland was incorrect in detail, the broad outlines of the nappe theory not only stood the tests, but made many of Steinmann’s own observations intelligible. Among Steinmann’s observations was the intimate association of radiolarites, deep-sea muds and radiolarite-bearing pelagic limestones with gabbros and peridotites. Steinmann noted that this association, previously vaguely ascribed to eruptive processes along the Aufbruchszone, in reality represented an extremely deformed association that stemmed from the deepest part of the Alpine ‘geosyncline’ (he estimated the maximum water depth of the ‘geosyncline’, on the basis of the present-day analogues of the radiolarites and the associated deep-sea sedimentary rocks, to be more than 4000 m: Steinmann (1905, p. 55); see Trümpy (2003) for an excellent historical assessment). At this depth, gabbros, peridotites and the deep sea sediments somehow had come together and were then uplifted by the medial Cretaceous by some 5 km! Steinmann ascribed this uplift to an early phase of Alpine orogeny that had folded up the middle part of the Alpine geosyncline (he regarded the Helvetic and the Austroalpine shelf areas as the northern and the southern parts, respectively, of the same geosyncline). Continued shortening finally generated the two largest nappes of the Alps from the deepest parts of the geosyncline: the ophiolite (by which Steinmann meant only the ultramafic rocks and the gabbros in 1905) and deep-sea sediment-filled northern part formed the Rhaetic Nappe, whereas the relatively shallower southern part formed the Austroalpine Nappe. The Austroalpine Nappe overrode the Rhaetic Nappe and literally squashed it (‘ausgequetscht’, Steinmann, 1905, p. 58). For that reason, Termier (1903, p. 762) had called the Austroalpine Nappe the traîneau écraseur (crushing sledge). This crushing sledge had such an effect on its underlying nappe, according to Steinmann, that ‘the structure of the middle or the “piercing zone” [‘Aufbruchszone’] became dominated by numerous smaller overthrusts and squashing, so much so, that a complete agreement with the confusing structure of the northern Swiss klippes originated and that, even while mapping at a scale of 1:25,000, only “squashing zones” [Quetschzonen] can be discerned, which almost earn the designation of a friction breccia [Reibungsbrekzie] at large’ (Steinmann, 1905, p. 10). This confusing, mega-breccia structure (which the Swiss geologist Joos Cadisch called the Aroser Schuppenzone [= the Zone of Imbrication of Arosa] in Cadisch et al. (1919, p. 362)) and the stratigraphic similarities led Steinmann to correlate the Rhaetian Nappe with the northern Swiss klippes and the Préalpes. As yet, the father of the Steinmann trinity was not aware of the importance of the third member of the trinity, namely the pillow lavas. In his classic 1927 paper he completed the trinity (but without naming it as such and without admitting the extrusive nature of the ‘diabases’! see Bailey (1936, 1944) and Trümpy (2003)) and pointed out the settings in which it was

2. Ophiolites and Orogeny: A Historical Review and Critique

681

encountered (see Bailey, 1936, pp. 1718–1722). They were what we today recognise as the suture zones of the Mediterranean and the mélange belts associated with them. Thus, Suess’ prophetic point that ultramafic rocks seemed to be associated with major thrusts culminated, in the light of the continuing Alpine discoveries, in Steinmann’s point that the trinity he noticed pointed to the presence of oceanic depths, now represented by the extremely deformed zones of confused structure along thrust belts. Already in his 1905 paper Steinmann pointed out that the ophiolites (then thought of as only the ultramafics and the gabbros) and the associated deep-sea deposits (i.e., the ophiolitic mélange, a term Steinmann did not use) not always formed ‘middle nappes’ as in the Alps. In some mountain belts (as in the Himalaya) they constituted the highest nappes. Steinmann (1905, p. 64) pointed out that ophiolites issued most commonly from the ‘internal’ parts of geosynclines. Steinmann’s point that ophiolites were associated with deep-sea sedimentary rocks and thus must have issued from the deepest part of the Alpine geosyncline found a sympathetic reader in Suess, although Suess was no believer in geosynclines (Suess, 1888, pp. 263–264, 1909, p. 722 and note 52; cf. Sengör, ¸ 1982a, 1982b, 1998). He had satisfied himself that the Alps formed from the destruction of the Tethys, which he had regarded as an ocean, no different in structure from the present-day oceans. But he agreed with Steinmann that ophiolites, representing former deep-sea environments, were confined to a narrow zone in a mountain belt and repeated that they never occurred on the foreland. However, Suess also made the important observation, which was largely ignored after him, that ophiolites never occupy an axial position in mountain belts (Suess, 1909, p. 644). He also cautioned against an overenthusiastic acceptance of the idea that all ophiolites had issued from the deep sea by pointing out that the ophiolites in the western Pyrenees were not associated with deep sea sedimentary rocks, although, there also, significant dislocation accompanied them (Suess, 1909, p. 646). Suess regarded ophiolitic igneous rocks as sills, largely confined to surfaces of tectonic movement, but also underlined that in many cases the suspicion that they had been transported in a solid state seemed unavoidable: ‘This suspicion disappears [i.e. turns into a certainty] concerning the large surface of movement on which the Tibetan slabs were transported onto parts of the Himalaya’ (Suess, 1909, p. 646). In 1911, in a short note, Émile Argand demonstrated that in the Pennine zone of the Swiss and Italian Alps, the higher one looked in the nappe edifice, the more frequently one encountered ophiolites. This interpretation, when viewed in terms of Lugeon’s rule of undoing the Alpine deformation, namely, the higher a nappe in the edifice, the more southerly must be its original palaeogeographic provenance, led Argand (1916) to combine the ideas of Suess (1904, 1909) and Steinmann (1905) in the framework of his embryonic tectonics to interpret Alpine ophiolites as submarine effusions issuing along thrust faults into the active flank of an asymmetrically shortening geosyncline (Fig. 1). Like Argand, Leopold Kober, one of the most influential tectonicians of the first half of the 20th century, also followed the views of Suess and Steinmann in his very popular textbook Der Bau der Erde (1921, pp. 34–36). By 1928, when the second, enlarged edition of his book appeared, his enthusiasm for the close association of thrusts and the ophiolites seems to have somewhat waned, but, following Steinmann (1905, 1927), he still held fast onto their deep-sea origin.

682

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

By 1922, however, Argand had changed his mind. He no longer viewed geosynclines as products of shortening, because he had become a follower of Wegener’s continental drift theory (Sengör, ¸ 1982a, 1982b, 1982c, 1998). In this new view, geosynclines marked sites of crustal stretching. When stretching went far enough the continental crust ruptured and the area of extension became underlain by the sima (Fig. 2). Argand saw the provenance of ophiolites in this simatic basement. He pointed out that the frequent association of the ‘green rocks’ with abyssal and bathyal sedimentary rocks could thus be explained. He also noted that sedimentary rocks of shallower provenance also appeared commonly mixed with these. Argand ascribed such mixing to submarine slumps (Argand, 1924, p. 299). Staub’s (1924, 1928) scheme (Fig. 3) was a direct derivative of Argand’s earlier interpretation (1916), but Staub, coming from an eastern Switzerland experience, was not a partisan of the eccentric position of ophiolite effusion channels with respect to the geosynclinal axis. Abandoning Suess and Argand, he liberally distributed them throughout the axial part of the Tethys. It had thus become common knowledge by the teens of the 20th century that ophiolites had something to do with orogeny. Kober (1921, p. 35) quoted Suess with emphasis that they always occurred within mountain belts and never on forelands. But again, not every orogenic belt had them: Steinmann (1905, p. 63) pointed out that in the Central Andes, for example, he had been able to find no ophiolites despite intensive search. He thus distinguished mountains of Alpine-type from mountains of Cordilleran-type. Only the former, which Steinmann thought encompassed an entire geosyncline, had ophiolites. The Cordilleran-type mountain belts, Steinmann speculated, may have been only the margin of a geosyncline or they my have grown out of shallow geosynclines. In a widely cited paper, Hess (1939) pointed out that the dominant member of the Steinmann trinity was serpentinites. He believed that they were intrusive, mainly disposed of in two parallel belts 60 miles distant from the axis of any orogen in which they are located, and that their time of intrusion always coincided with the first major deformation of the orogenic belt in which they occurred. He challenged the view that they originated in geosynclines by showing that they also occurred in island arcs, which Hess believed had no geosynclinal precursors. According to Hess, geosynclines formed only after the first major deformation in a given mountain belt and were essentially foredeeps. He was thus in agreement with Franz Eduard Suess (1937), Eduard Suess’ son. Another extremely infulencial tectonician of the first half of the twentieth century, Hans Stille, offered, in 1939, the first systematic account of ophiolite genesis and significance within the framework of his geotectonic cycle. According to him, ophiolites occurred during the preparatory phase of mountain-building, when the maternal geosyncline began subsiding. He called this ‘initial magmatism’ (he had thought of calling it ‘ophiolitic magmatism’, but had decided against it, because not all initial magmatism in geosynclines was exclusively ophiolitic: see Stille, 1939). He believed that they issued along the axes of subsiding geosynclines and there mixed with deep-sea sedimentary rocks (Fig. 4). The essence of Stille’s scheme was that ophiolites were confined to geosynclines. He thus simply followed his countryman Steinmann.

2. Ophiolites and Orogeny: A Historical Review and Critique

Fig. 1. Argand’s (1916, Pate II, Fig. 1) famous block diagram showing the conditions in the Alpine ‘geosyncline’ at the time of its pre-paroxysmal stage. Note the ophiolite ‘effusions’ à la Suess and Steinmann (v and v ). We have copied this translated version from Collet (1935, Plate I).

Fig. 2. Origin of geosynclines and oceans illustrated on the example of the Ionian Sea by Argand (1924, Figs. 19 and 19 bis). Note the surfacing of the simatic rocks at the toe of the continental apron and their subsequent incorporation into an orogenic belt during shortening. 683

684

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 3. Staub’s (1924, Fig. 61, 1928, Fig. 1) view of the origin of ophiolites in the Alpine Tethys.

In 1936, Stille had divided the North American Cordilleran geosyncline into a western and an eastern trough. The western trough was thought to contain thicker piles of dominantly clastic/silicic sedimentary rocks closely and abundantly associated with mafic and intermediate ‘initial’ volcanics (Fig. 5). The eastern trough was believed to hold a thinner pile of more carbonate rich sedimentaries and had no initial volcanics. Stille called the western trough ‘pliomagmatic’ and the eastern trough ‘miomagmatic’. In 1940, he called the pliomagmatic trough a eugeosyncline and the miomagmatic trough a miogeosyncline. Ophiolites occurred only in eugeosynclines. As the eu- and miogeosynclinal couples formed the great mother troughs of mountain belts, the orthogeosynclines, Stille thus ultimately conceded the eccentric position of ophiolite feeding channels within a large orthogeosynclinal system. By the end of the fifties, it had thus become clear that ophiolites (1) represented a mafic/ultramafic association commonly closely coupled with deep-sea sedimentary rocks, (2) they were confined to orogenic belts, and (3) they generally occurred as tabular bodies that came into being early during the life cycle of a mountain belt in a volcanically active part of the maternal geosyncline, but they were also present as smaller bodies in enormous piles of eugeosynclinal sedimentary rocks. Only Hess made the, in retrospect, critical observation that on continents although ‘alpine-type peridotites occur only in alpine-type mountain structures’. ‘They appear to occur ubiquitously throughout the earliest Precambrian rocks’ (Hess, 1955, p. 394). All of these observations, except the last one, were easily converted to plate tectonic interpretations, once it was recognised that tabular ophiolite bodies were in fact nappes and the vast thicknesses of eugeosynclinal sedimentary rocks with associated ophiolitic magmatics were tectonically stacked sequences in accretionary complexes. Papers by Gass and Masson-Smith (1963) and Dietz (1963), respectively, already had the essences of these ideas. 2.4. Ophiolites as Distributed Fragments in Continental Structure: Historical Perspective Argand (1924) already surmised that sialic rafts drifting in sima would accumulate at their prows oceanic and continental sedimentary and igneous material in part mixed by sedimentary gliding processes and in part by tectonism. He pointed out that the closure of former oceans north of Tibet must have given rise to complicated accretionary ‘cushions’ between colliding continents (‘bourrelet complexe’: see Fig. 6 herein). He thought that one such complexly folded cushion, combining the continental slope accumulation of the colliding

2. Ophiolites and Orogeny: A Historical Review and Critique

685

Fig. 4. Sequential cross-sections to illustrate schematically Stille’s view of the ‘geotectonic cycle’ and accompanying phenomena, such as the ‘magmatological cycle’ (see Stille, 1940, esp. p. 21) drawn after his descriptions and sketches in a number of his publications to show the place of ophiolitic magmatism in his scheme (from Sengör, ¸ 1982a, Fig. 1.9). (a) Consolidated crust before a regeneration that will form new orthogeosynclines between cratons. (b) Preparatory phase of orogeny: during this ‘anorogenic’ time epeirogenic movements (caused by epeirogenic stresses, Se , which are allegedly much weaker than the orogenic stresses So ) give rise to geosynclines and geanticlines. In geosynclines, the base of the crust is pushed into progressively deeper regions in the earth. As a result of the accompanying increase in temperature, its base weakens and its effective thickness to carry stresses decreases. At this stage initial magmatism (simatic, mostly ophiolitic) loads the geosyncline with heavy magma. As a result of subsidence, the area of the geosyncline becomes more ‘mobile’ than the bordering geanticlines (cratons). Stille never explained where the simatic magmas come from and why. It is conceivable that he might have thought of the ophiolitic ‘feeders’ as dyke systems parallel with the shortening direction. In 1939, he pointed out that not all initial magmatism was mafic. He gave examples of intermediate even felsic ‘initial magmatic rocks’. His scheme would indeed allow such magma mixing in the geosynclinal subsidence stage. (c) During an orogenic phase stresses throughout the globe allegedly increase. The crust fails at its weakest point which is the geosyncline. The geosyncline becomes compressed between the two stable blocks (cratons). This, as a rule, gives rise to a symmetric orogen. As a consequence of shortening, the crust in an orogen thickens, its base melts and this gives rise to synorogenic plutonism, which is sialic, in contrast to the simatic initial magmatism.

686

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 5. Hans Stille’s view of the origin of ophiolites in the North American Cordilleran geosyncline (from Stille, 1936, Fig. 2, 1940, Fig. 60).

continents, now forms the material of the Kuen-Lun range (Fig. 6). Although he did not explicitly say so, from his description of the nature of continental slopes being regions of the mixture of simatic rocks, abyssal, bathyal and shallower sedimentary rocks, it is not unreasonable to assume that in the structure of the Kuen-Lun he expected to see shreds of the ‘sima’ embedded in his ‘complexly folded cushion’. How wide the ophiolite-bearing cushions may be depended on how big the ocean was, from which the cushion had been swept off. Argand did not further elaborate on this idea. In 1955, Hess made the seminal observation that alpine-type peridotites ‘appear to occur ubiquitously throughout the earliest Precambrian rocks’ (Hess, 1955, p. 399), as we have mentioned above. He thought that the processes which created that picture were fundamentally different from those which created the Phanerozoic orogen-bound peridotite belts. He thought that the Precambrian situation arose because entire areas of ocean had somehow been converted into continents. Dietz (1963) reinterpreted Hess’ observation by assuming that successive orogenies had incorporated ocean-floor igneous rocks into what he called the eugeosynclinal prisms along continental margins. Like Hess, he related this to J. Tuzo Wilson’s (1949) idea of peripheral growth of the continents. While Hess (1955) thought that Wilson’s mechanism probably contributed to the vertical growth of the continents, Dietz (1963) pointed out that tectonic patterns of Precambrian terrains clearly indicated lateral growth also. Both Hess and Dietz, thought, however, that the processes that generated Precambrian terrains were no longer operative either in kind (Hess) or in extent (Dietz). They thought so, because they were

2. Ophiolites and Orogeny: A Historical Review and Critique

687

Fig. 6. Argand’s cross-section across the Tibetan high plateau showing part of the Tien-Shan, Serindia (the Tarim Basin), Kuen-Lun and the Himalaya, in addition to the high plateau itself. The bourrelet complexe (shown by dense stippling added to Argand’s figure by us) represents, according to Argand, an earlier accretionary complex that had formed long the southern margin of the ‘Serindian Block’ than the mainly Mesozoic one shown by Argand himself with sparse stippling (‘The tectonic products arisen from the axial zone of the Tethys’: Argand, 1924, p. 348). Modified from Argand (1924, Fig. 13).

not familiar with regions that resembled the Precambrian terrains they were considering. The world geological opinion has largely followed either Hess or Dietz until the beginning of the 1990s. The main purpose of this paper is to describe one major region in which ophiolites occur much as they do in Precambrian, specifically Archaean, terrains and to argue that the processes responsible for creating the tectonic patterns of these regions were substantially the same. 2.5. The Term Ophiolite: Can Its Original Usage Be Taken as a Guide to Its Present Usage? The term ophiolite was introduced by the great French geologist Alexandre Brongniart in 1813 (for a detailed history, see Amstutz, 1980). When first introduced, Brongniart used it essentially for rocks made up dominantly of serpentinite enveloping different species of minerals and he gave both compositional and textural criteria for their recognition. In 1827, Brongniart gave a much more extended discussion of the term ophiolite and there too he applied the term only to a variety of serpentinites associated with diverse types of minerals. In an 1821 paper on the northern Apennines, Brongniart described the ophiolites of what is now known as the Liguride units. There he showed the close association of serpentinites of diverse types, gabbros, mafic volcanics and cherts. Brongniart’s usage of the term ophiolite was strictly mineralogical/petrographical. It can be extended to cover the present usage only in retrospect, but not by considering Brongniart’s publications alone. The present usage (Penrose field conference participants, 1972) covers an association that goes from ultramafics to mafic volcanics. Brongniart’s term applies strictly to the ultramafic part. It is clear that the original usage as intended by Brongniart (1813, 1821, 1827) cannot be taken as a guide to the present usage. But confining the term ophiolite only to the Penrose definition (Penrose field conference participants, 1972) would restrict it to only a few cases where the Penrose sequence is pre-

688

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

served intact. What would we then call the other remnants of ocean floors now embedded in the continents? We could naturally describe each according to its petrography, but descriptive terms are not efficient communicators of interpretations. Under what term would we unite separate slivers of ultramafics, gabbros, basalts, etc., if we believe they are were plucked off now vanished oceanic crust and upper mantle? Thus, by the end of the eighties, two major classes of oceanic basement fragments had been identified: one was the giant ophiolite nappes, which were clearly generated atop subduction zones in environments of pre-arc spreading and many had been obducted across former fracture zones. The other class was far more heterogeneous: it consisted mainly of a diverse array of oceanic basement scraps in mélanges representing a full spectrum of the ‘Penrose ophiolite definition’ (Penrose field conference participants, 1972), but, in many mélange belts, being dominated by the mafic end of the spectrum. The numerous ‘diabase-phyllitoid’ and ‘slate-diabase’ associations of the Tethysides are examples of such ‘mafic dominance’ (e.g., Yilmaz and Sengör, ¸ 1985). Dewey (2003) recently called the first class ophiolites sensu stricto and the second, ophiolites sensu lato. He further argued that ‘Ophiolites sensu lato can be thought of as a simple random record of the Wilson Cycle of opening, widening, narrowing, and closing oceans but ophiolite complexes sensu stricto are events related to particular and special tectonic configurations’ (Dewey, 2003). We think that Dewey’s (2003) suggestion of separating ophiolites sensu lato, including all occurrences of oceanic basement rocks, from ophiolites sensu stricto, i.e., those ophiolites showing the complete Penrose sequence is very apposite. However, it would be handier to find single-word designations for these two classes. We here suggest to keep ophiolite for those bodies preserving the complete Penrose sequence as has become customary since 1972. For separated fragments of this complete sequence, also including sea-mounts, oceanic plateaux, and fracture zone scarps, and dominated by basalts and gabbros, and rarer ultramafic rocks (Dewey, 2003) including both lherzolitic and harzburgitic types, we suggest the term ophirag, from the Greek oφιs (= serpent, snake) and ραγos (= tatter, shred, sliver). Thus we shall henceforth call ophiolites sensu stricto in Dewey’s sense ophiolites and his ophiolites sensu lato ophirags. Ophiolites and ophirags together make up remnants of oceanic basement rocks incorporated into continents. As we shall see below, by far the largest number of the Altaid oceanic basement remnants are ophirags; the same is true for most Archaean occurrences of oceanic basement remnants.

3. OUTLINE GEOLOGY OF THE ALTAIDS 3.1. Historical Perspective: The Problem of the Altaids The Altaids (Fig. 7) constitute one of the world’s largest and most complicated orogenic systems. Their name derives from the Altay Mountains in Russia, China and Mongolia. It was coined a century ago, in the first part of the third volume of Das Antlitz der Erde (Face of the Earth) by Eduard Suess (1901, pp. 246–250) to designate the mountain chains that formed around what he called the ‘ancient vertex of Asia’ (Fig. 8). He noted that the

3. Outline Geology of the Altaids

689

Altaids constituted a series of ever widening, north-concave arcs around the ancient vertex and were delimited to the south by the chains that arose from the mid- to late Mesozoic and early Cainozoic elimination of a broad Mesozoic ocean, the Tethys (cf. Sengör ¸ and Natal’in, 1996a; Sengör, ¸ 1998). Suess characterised both the structural coherence of the Altaids (disposition in more or less concentric arcs, lack of a foreland) and their compositional peculiarity (paucity of large gneiss terrains, dominance of slates, schists, cherts with some basalts and serpentinites) and likened them to the ‘waves in the open sea’ as opposed to ‘waves breaking on a beach’ that resembled more the mountain ranges with pronounced forelands such as the Alps and the Himalaya (Sengör, ¸ 1998, p. 248). In the 20th century, the greatest problem the Altaids posed has been their unusual shape and internal composition. They resembled none of the more familiar great mountain ranges of the globe such as those along the Alpine-Himalayan ranges or the Cordilleras in the Americas. That parts of them were like parts of the North American Cordillera or Japan was not considered significant. That they greatly resembled the Archaean and some Proterozoic regions was not at all recognised. On the one hand, such Archaean and Proterozoic terrains were thought not to have Phanerozoic counterparts and, on the other, the Altaids, which in reality were similar to those Precambrian areas, were thought unique. Complex models of geosynclines were proposed for the interpretation of the Altaids, but few could go beyond being ad hoc solutions (in this, the difficulties of interpretation were remarkably similar to those in the Precambrian regions; but again, this similarity was not recognised). The early plate tectonic models for the Altaids were also ad hoc (with the very remarkable exception of the classic Uralides paper by Hamilton, 1970), in that they invoked head-on collisions for any length of suture that could be identified with no obvious genetic connexion among the various colliding entities (e.g., Kropotkin, 1972; Zonenshain, 1973; Burrett, 1974). The neutral names used for Suess’ Altaids, such as the Ural-Amurian or the Ural-Mongolian foldbelt or the Central Asian foldbelt in the literature betray this interpretative sterility. Suess’ Altaids disappeared from the literature, along with the recognition of their structural and evolutionary coherence. 3.2. Present Interpretation of the Altaids The initial step in the rehabilitation of Suess’ Altaid model was the recognition that much of Central Asia, particularly the mountain ranges around the Tarim Basin and those extending into Kazakhstan consisted dominantly of subduction-accretion complexes (Sengör ¸ and Okuro˘gulları, 1991; Sengör, ¸ 1992) and that those north of the Tarim basin constituted a single, unified orogenic system (Hamilton, 1970; Sengör ¸ et al., 1993). There are very few and spatially restricted neat, long and narrow ophiolitic sutures and no forelands (except at the outermost periphery: see Fig. 7) against which parts of the orogen abut across a clear structural front except in the extreme south (Fig. 7). The shape of the orogen is irregular and does not have the familiar linear/ arcuate, long and narrow shapes of most Phanerozoic orogenic belts. The trend of the Altaid orogen within this confusing region could most easily be followed by tracing out the magmatic arc fronts (for methodology, see Sengör ¸ and Okuro˘gulları, 1991;

690

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 7. Generalised tectonic map of the Altaids and related surrounding tectonic units (from Sengör ¸ and Natal’in, 1996a, Fig. 21.18). The Uralide and Baykalide orogenic systems have not been further subdivided. Dotted line shows the limit of the post-Altaid cover. Key to numbers: 1. Valerianov-Chatkal (pre-Altaid continental basement, Altaid accretionary complex and magmatic arc), 2. Turgay (pre-Altaid continental basement, Altaid accretionary complex and magmatic arc, buried under 2-km-thick Mesozoic-Cainozoic sedimentary cover), 3. Baykonur-Talas (pre-Altaid continental basement, early Palaeozoic Altaid accretionary complex and magmatic arc), 4.1. Djezkazgan-Kirgiz (pre-Altaid continental basement, Altaid Palaeozoic magmatic arc and accretionary complex), 4.2. Jalair-Nayman (pre-Altaid continental crust, early Paleozoic marginal-sea complex, early Palaeozoic magmatic arc and accretionary complex), 4.3 (or 16) Borotala (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex), 5. Sarysu (Altaid accretionary complex and magmatic arc), 6. Atasu-Mointy (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex), 7. Tengiz (pre-Altaid continental basement, Vendian-early Palaeozoic magmatic arc and accretionary complex), 8. Kalmyk Köl-Kökchetav (Vendian-early Palaeozoic magmatic arc and accretionary complex), 9. Ishim-Stepnyak (pre-Altaid continental basement, Vendian-early Palaeozoic magmatic arc and accretionary complex), 10. Ishkeolmes (early Palaeozoic magmatic arc and accretionary complex), 11. Selety (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex), 12. Akdym (Vendian-early Palaeozoic magmatic arc and accretionary complex 13. Boshchekul-Tarbagatay (early Palaeozoic magmatic arc and accretionary complex), 14. Tekturmas (Ordovician-medial Palaeozoic accretionary complex, medial Devonian-early Carboniferous magmatic arc), 15. Junggar-Balkhash (early-late Palaeozoic magmatic arc, medial and late Palaeozoic accretionary complex), 16. see unit 4.3, 17. Tar-Muromtsev (early Palaeozoic magmatic arc and accretionary complex), 18. Zharma-Saur (Palaeozoic magmatic arc, early Palaeozoic accretionary complex), 19. Ob-Zaisan-Surgut (late Devonian-early Carboniferous accretionary complex, strike-slip fault-bounded fragments of the lte Devonian-early Carboniferous magmatic arc, late Palaeozoic volcanic arc), 20. Kolyvan-Rudny Altay (early Palaeozoic accretionary wedge, early and medial-late Palaeozoic magmatic arc), 21. Gorny Altay (early Palaeozoic accretionary complex and magmatic arc superimposed by medial Palaeozoic magmatic arc; in the South Altay sector, medial palaeozoic accretionary complex with fore-arc basin), 22. Charysh-Chuya-Barnaul (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex, medial Palaeozoic fore-arc basin and magmatic arc), 23. Salair-Kuzbas (pre-Altaid continental basement, Vendian-early Palaeozoic magmatic arc and accretionary complex, Ordovician-Silurian fore-arc basin, Devonian pull-apart basin, late Palaeozoic foredeep basin), 24. Anuy-Chuya (early Palaeozoic magmatic arc and accretionary complex), 25. Eastern Altay (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex with large seamount fragments), 26. Kozhykov (early Palaeozoic magmatic arc and accretionary complex), 27. Kuznetskii Alatau (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex), 28. Belyk (Vendian-medial Cambrian magmatic arc and accretionary complex), 29. Kizir-Kazyr (Vendian-medial Cambrian magmatic arc and accretionary complex), 30. North Sayan (Vendian-early Palaeozoic magmatic arc and accretionary complex), 31. Utkhum-Oka (pre-Altaid continental basement, early Palaeozoic magmatic arc and accretionary complex), 32. Ulugoi (Vendian-early Cambrian magmatic arc and accretionary complex), 33. Gargan (pre-Altaid continental basement, early Palaeozoic magmatic arc and Vendian-early Palaeozoic accretionary complex), 34. Kitoy (early Palaeozoic magmatic arc), 35. Dzhida (early Palaeozoic magmatic arc and accretionary complex), 36. Darkhat (pre-Baykalide continental basement, Riphean magmatic arc and accretionary complex), 37. Sangilen (Baykalide microcontinent that collided with Darkhat unit in the Riphean and experienced strike-slip displacement during the early Palaeozoic Altaid evolution), 38. Eastern Tannuola (early Palaeozoic magmatic arc and accretionary complex), 39. Western Sayan (early Palaeozoic magmatic arc and accretionary complex), 40. Kobdin (pre-early and medial Palaeozoic magmatic arc and accretionary complex), 41. Ozernaya (Vendian-early Cambrian magmatic arc and accretionary complex), 42. Han-Taishir (pre-Altaid continental basement, Vendian-early Cambrian magmatic arc and accretionary complex), 43. Tuva-Mongol (equivalent to the central and eastern parts of Suess’ ‘ancient vertex of Asia’; see Fig. 8 herein): 43.1. Tuva-Mongol arc massif (pre-Altaid continental crust and Vendian to Permian magmatic arc), 43.2. Khangay-Khantey (Vendian-Triassic accretionary complex and magmatic arc), 43.3. South Mongolia (Ordovician-early Carboniferous accretionary complex), 44. South Gobi (pre-Altaid continental basement, Palaeozoic magmatic arc and accretionary complex).

3. Outline Geology of the Altaids

691

692

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Sengör, ¸ 1993; for detailed application to the Altaids, especially Sengör ¸ and Natal’in, 1996a, Figs. 21.19–21.25). It seemed that magmatic fronts jumped forward within growing subduction-accretion complexes. This growth also enlarged the continent. When this model was applied to the Altaids as a whole, it appeared that they consisted of the fragments of at most two large magmatic arcs (Fig. 7) stacked together dominantly by arc-subparallel strike-slip faults (both arc-slicing and arc-shaving varieties (Natal’in and Sengör, ¸ 1996): Fig. 9) and not by head-on collisions as previously assumed (Sengör ¸ et al., 1993; Sengör ¸ and Natal’in, 1996a). These strike-slip faults were seen to have enormous offsets along them, amounting to hundreds, in many cases more than 1000 km. One pair of them, in particular, the Irtysh and the Gornostaev, appear as broad zones of intense shear with offsets around 2000 km or more (Fig. 7). These two are veritable keirogens dividing the Altaid orogenic system into two main domains: The Kazakhstan-Tien Shan domain in the west and the Altay-Mongol domain in the east. The Altay-Mongol domain itself is naturally divided into two by the continental Tuva-Mongol axis (unit 43.1 in Fig. 7) that encloses the Khangai and Khantey mountains (forming unit 43.2 in Fig. 7). The accretionary complexes lying west and northwest of the Tuva-Mongol continental axis form the Altay-Sayan sector and the Tuva-Mongol fragment itself, together with the Palaeozoicearly Mesozoic accretionary complexes forming the Khangai-Khantey mountains, southern Mongolia (unit 43.3. in Fig. 7) and the Gobi (unit 44 in Fig. 7), constitute the MongolOkhotsk sector (Fig. 7). Although the Tuva-Mongol continental fragment is the only major Precambrian core in the Altay-Mongol domain of the Altaids, there are at least 9 substantial pre-Altaid continental slivers in the Kazakhstan-Tien Shan domain (Fig. 7). These follow the disposition of the magmatic arc fronts here and they are separated from one another by subduction-accretion complexes with age ranges from Vendian to Carboniferous. These

Fig. 8. Tectonic map of a part of Central Asia showing Suess’ ‘ancient vertex of Asia’ and associated units after Obrutschew (1926, Plate 11). Suess never presented a map of the ancient vertex. Obrutschew was one of his most loyal followers in the definition of it. For a description of the ancient vertex of Asia, see Suess (1901, ch. 3) and Obrutschew (1926, pp. 15–17). Also see Stille (1958, pp. 61–66). For Obrutschew’s loyalty to Suess’ interpretation of the ancient vertex as a Precambrian unit, see Stille (1958 p. 65). Stille wrote (1958, p. 64) with emphasis, and ‘against the divergent opinions cited in the literature’, that Suess himself had viewed the ancient vertex as a younger structure than the Angaran Shield. It is unclear why Stille thought so (he indicates no specific writing by Suess to support his statement), because Suess expressly says that the ancient vertex is a high-lying remnant of the Angara Shield: ‘The larger part of the western half of the vertex was faulted down at an early period, and thus gave rise to the amphitheatre of Irkutsk’ (Suess, 1901, p. 98). Suess correctly recognized that what is now called the Tuva-Mongol continent (Sengör ¸ et al., 1993; Sengör ¸ and Natal’in, 1996a; see Fig. 7 herein, unit 43) was no younger than the basement of Angara Shield, even including Archaean rocks. What he did not know was that the Khangay-Khantey accretionary complex reached in age from the Vendian into the Mesozoic (cf. Sengör ¸ and Natal’in, 1996a).

3. Outline Geology of the Altaids

693

694

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 9. Arc subparallel ‘trench-linked’ (Woodcock, 1986) strike-slip faults and their effects on the geometry of arc and subduction-accretion systems. (A) Arc slicing faults cut through the entire arc system and are proper plate boundaries. They may repeat or elide arcs. Their best active examples are seen in the Philippine Fault System (Rutland, 1968; Cardwell et al., 1980; Hall, 2002), in the Palu Fault that slices through the northern arm of Sulawesi (Celebes) (Cardwell et al., 1980), and in the northern Caribbean (Mann et al., 1991). In the Sangilen unit (unit No. 37 in Fig. 7; ophiolite No. 10 in Fig. 14), arc repetition created an ophiolitic zone that resembles a collisional suture, but is in reality a transform suture, exhibiting ophiolites entrapped by an arc repeating strike-slip fault. (B) Arc shaving faults affect only the fore-arc and shave bits of the fore-arc and transport them coastwise along the trench. The great Atacama fault in the Andean fore-arc is the best-known example (Sengör, ¸ 1990). Note how the forearc area is repeated and thus enlarged in front of arc-shaving faults and are elided and narrowed behind them. This is how the Altay region has been vastly enlarged, while South Mongolia was constantly being shaven along the immense Irtysh strike-slip system. In some cases hybrid faults originate. The Great Sumatra fault (Sieh and Natawidjaja, 2000), for example, mostly shaves the arc, but does so along the volcanic axis and in places wanders behind the axis to incorporate arc magmatics into the fore-arc ‘sliver plate’ (Jarrard, 1986). Some of the Altaid trench-linked strike-slip systems in the Altay-Mongol domain were of this hybrid type. If pull-apart geometries originate along any of such trench-linked, arc-subparallel strike-slip faults, new oceanic crust may be formed along short spreading centres trending at high angles to the arc, as, for example, in the Cayman Trough or in the Andaman Sea. The Han-Taishir ophiolite in the unit 42 of the Altaids, for example, may have formed in such an environment.

arc segments originally formed a single arc along the Ural/Yenisey margin of the then combined Russian and Angara cratons (Fig. 10). This arc became detached in the Vendian/Cambrian and formed what we call the Kipchak Arc (Fig. 11). In the internal parts of the Kazakhstan-Tien Shan sector, the ages of the accretionary complexes do not reach beyond the Ordovician-Silurian. It looks as if by Devonian time this part was wholly assembed by strike-slip stacking of the Kipchak Arc (Fig. 12; Sengör ¸ et al., 1993; Sengör ¸ and Natal’in, 1996a, 1996b). Similarly the Altay-Sayan sector was also internally assembed by the end of the Silurian (Sengör ¸ et al., 1993; Sengör ¸ and Natal’in, 1996a, 1996b). The internal Kazakhstan-Tien Shan sector formed a small continent by itself (with a narrow

3. Outline Geology of the Altaids

695

Fig. 10. Vendian palaeotectonic reconstruction of the Altaids (630–530 Ma). The legend shown here applies to all reconstructions in Figs. 10–16.

696

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 11. Early Cambrian palaeotectonic reconstruction of the Altaids (530–520 Ma). For legend, see Fig. 10.

attachment to the Russian craton via the Valerianov-Chatkal arc) whereas the Altay-Sayan sector was assembled as a tight attachment of the Angara craton and its Baykalide frame. These two sectors were initially united by an ensimatic segment of the Kipchak Arc now making up the Boshchekul-Tarbagatay and Zharma-Saur arc fragments (Sengör ¸ and Natal’in, 1996a). This ‘bridge’ eventually formed the backstop in front of which the enormous flysch wedge of the Junggar-Balkhash accretionary complex accumulated (Fig. 13). The Boshchekul-Tarbagatay/Zharma-Saur ‘bridge’ had a sliding attachment to the periAngara accretionary systems. As the bridge tightened in the form of a south-concave pincer, its Angara end moved along the previously-assembled units in a right-lateral sense along the Irtysh shear zone. At the same time the Valerianov-Chatkal unit was moving in a left-lateral sense along the eastern margin of the Urals forming the Denisov-Oktabyarsk suture. The driving motive of these movements were the rotation of the Angara and the

3. Outline Geology of the Altaids

697

Fig. 12. Early Devonian palaeotectonic reconstruction of the Altaids (420–390 Ma). For legend, see Fig. 10.

698

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 13. Early Carboniferous palaeotectonics of the Altaids (350–325 Ma). For legend, see Fig. 10.

Russian cratons (that had separated from each other in the Vendian) towards one another that progressively narrowed the oceanic area between them. This motion culminated in the late Carboniferous by the elimination of all marine areas from the Kazakhstan-Tien-Shan domain and the Altay-Sayan sector (with minor exceptions in the western Tien-Shan). The Tarim fragment had collided with the Valerianov-Chatkal, but mainly with the DjezkazganKirgiz unit (Sengör ¸ and Natal’in, 1996a) earlier to terminate the orogenic evolution of the Kazakhstan-Tien Shan segment (Fig. 14). In the late Carboniferous and the early Permian the Kazakhstan-Tien Shan domain (and, with it, the Russian craton) moved right-laterally for some 2000 km with respect to the Altay-Mongol domain and the Angara craton along the Irtysh keirogen. This motion created the Nurol extensional basin north of the Kolyvan Mountains (Fig. 15). In the late Permian a major reversal of this motion occurred along the Gornostaev keirogen. This left-lateral motion generated the major extensional basins of Nadym in the north and the Junggar-Alakol-Turfan system in the south (Allen et al., 1995;

3. Outline Geology of the Altaids

699

Fig. 16). The main tenets of the evolution here outlined and graphically depicted in Figs. 10–16 have been corroborated to a remarkable degree by the recent palaeomagnetic work by Rob Van der Voo and his colleagues in the Tien Shan and in the Kazakhstan orocline (Collins et al., 2001; Bazhenov et al., 2002, 2003; Van der Voo et al., 2002). The Mongol-Okhotsk sector continued its evolution until the Jurassic. The Solonker Ocean separating it from the Manchurides in northeastern China and the Russian Far East remained open until the early Permian (Fig. 15). The final closure and closure-related easterly extrusion of part of the Khangai-Khantey, including segments of the Tuva-Mongol arc massif (Bindeman et al., 2002), was mainly a result of the Cimmeride collisions in China and lasted into the Jurassic (Sengör ¸ and Natal’in, 1996a). The Altaid evolution displays an immense spatial and temporal continuity rarely seen in this clarity by the geologist in the evolution of a continent by orogeny. The Altaid orogeny was locally and intermittently arrested by massive keirogenic movements and local taphrogeny associated with it. The present state of the orogen is one of considerable disruption into innumerable little mosaic pieces, but it is the former continuity of its tectonic units that enables us to understand its structure and to reconstruct its evolution. This evolution betrays 0.4 million km3 /a continental growth during the evolution of the Altaids, which may account for 40% of the total Palaeozoic growth (Sengör ¸ and Natal’in, 1996b). This estimate has been largely corroborated by new geochemical work (Chen et al., 2000; Jahn, 2002) and it had been inferred in a different tectonic framework by Peive and his collaborators (Peive et al., 1972). Another important observation is the lack of any episodicity in the Altaid evolution as a whole. Changes in structural regimes such as switch from subduction to strike-slip or local collisions as seen in the western Sayan mountains did introduce local episodicity, but there never was any cessation of motion within the orogen as a whole accommodating the continuous approach of the Russian and the Angaran cratons. Although most of the high-pressure metamorphics and backstop ophiolites seem to have been generated in the early history of the orogenic system, nothing comparable with the Tethyan ‘episodes’ of high-pressure metamorphism or ophiolite obduction is seen in the Altaids. This is probably a consequence of the lack of continental collisions and the great size of the ocean consumed in the Altaids, as opposed to the Tethysides: the small size of the ocean in the Tethysides dictated a narrow time interval for the subduction-related metamorphism and the arrival at subduction zones of long continental margins triggered nearsynchronous obductions of ophiolites along considerable strike-length. The abundance of ophiolite preservation in the earlier history of the orogenic system is probably a consequence of flysch-choking of the subduction zones that hindered plucking off ocean-floor pieces as seen in the still-active Makran accretionary complex in Iran and Pakistan, where the preserved ophiolites and ophirags are almost entirely of Cretaceous age (McCall and Kidd, 1982). The Altaid evolution has an actualistic analogue in the present-day western Pacific and it also greatly resembles the Pan-African evolution in North Africa and Arabia (Sengör ¸ and Natal’in, 1996b). It is also similar in style to most of the Archaean granite-greenstone terrains (Sengör ¸ and Natal’in, 1996b; Burke, 1997). The only difference we see between the Archaean terrains and the Altaid evolution seems a consequence of the higher oceanic

700 Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 14. Late Carboniferous palaeotectonics of the Altaids (325–300 Ma). For legend, see Fig. 10.

3. Outline Geology of the Altaids

Fig. 15. Early Permian palaeotectonics of the Altaids (300–250 Ma). For legend, see Fig. 10. 701

702 Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Fig. 16. Late Permian palaeotectonics of the Altaids (250–245 Ma). For legend, see Fig. 10.

4. Oceanic Basement Fragments in the Altaids: Some Representative Examples

703

geothermal gradients in the Archaean. The immense flattening of supracrustal units such as pillow-lavas observed in some Archaean terrains (0
k
1) accompanied by amphibolitegrade metamorphism (Dr. John Myers, personal communication, 1999) indicates considerable extension in a hot environment, the equivalents of which have not yet been reported from the Altaids.

4. OCEANIC BASEMENT FRAGMENTS IN THE ALTAIDS: SOME REPRESENTATIVE EXAMPLES In the Altaids all known types of oceanic basement fragments, except large pre-terminal ophiolite nappes similar to those in Oman, Papua New Guinea or Newfoundland (contrary to certain claims such as those of Buchan et al., 2001: see below), occur in a wide variety of tectonic positions. Altaid ophiolites and ophirags may be classified according to the tectonic environments in which they now occur in the following manner: (1) Ophiolites that occur as arc basement of ensimatic arcs; (2) Ophiolites and ophirags that occur in former forearcs now entrapped within transform sutures: (a) Ophiolites as backstop to accretionary wedges; (b) Ophirags within accretionary wedges; (3) Ophiolites and ophirags in collisional suture zones. In the Altaid edifice, classes (1) and (2a) may in places include the same body. In very rare cases a single body may fit all three classes except (2b) because of the peculiarities of the Altaid evolution. In the descriptions below there are examples of such bodies that fit more than one category. It is impossible to give an exhaustive description or even a simple list of all occurrences. There must be vast numbers of individual bodies located in the three tectonic environments indicated above. It is impossible to walk a long profile across the Altaids without encountering some ophirag or ophiolite. Instead we have chosen a few examples from among these myriads to highlight their characteristics. In our choice we have been guided not only by the mere availability of the data, but also according to how well each locality is known by the international community. We chose those that have been visited by the largest number of geologists both from local countries and from afar. Among those we have preferred those that have been recently (re)described in internationally widely circulated journals, preferably in English, so that our preferred interpretations may be compared with existing ones in the literature by the widest possible readership. 4.1. Ophiolites that Occur as Arc Basement of Ensimatic Arcs Some ophiolites in the Altaids form basements of ensimatic arcs representing upper plate oceanic basement of former Altaid intraoceanic subduction zones. We mention below those in units 10, 13, 25, 26 and 42 (Figs. 7 and 17). This naturally does not mean that they do not occur in other ensimatic arc remnants. We do not mention them here, because some

704

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

are now buried under younger cover and their presence is only surmised by geophysical means (especially under the West Siberian Basin: see Fig. 7) and about some others we simply have not yet come across an adequate publication. Yet others, as in the ZharmaSaur accretionary complex (unit 18 in Fig. 7), are so disrupted by later tectonism that it is difficult to tell whether they represent a once-coherent ophiolite disrupted in situ or ophirags assembled together from disparate areas in an accretionary complex mainly by arc-shaving strike-slip tectonism. In the following accounts all unreferenced geological description is from Sengör ¸ and Natal’in (1996a). Unit 10 (Ishkeolmes). This is a remarkable ensimatic arc, sitting on the Middle Cambrian Kuyanbai ophiolite complex (1 in Fig. 17). The ophiolite is topped by komatiitic basalts and boninites described along the western side of the unit (Spiridonov, 1991), i.e., near the fore-arc/arc contact. West of it is an accretionary complex including Ordovician cherts, shales and Middle to Upper Ordovician flysch. Unit 13 (Boshchekul-Tarbagatay). This is a middle Cambrian to Silurian arc system constructed atop an ophiolite unit yielding isotopic ages between 568 and 525 Ma. Clearly a 43 Ma spreading history may be implied or the unit may have brought together complete ophiolite fragments of different ages. Parts of the ophiolite are metamorphosed giving rise to garnet amphibolites. Atop this ophiolite are volcanic rocks consisting of rhyolites and dacites and, together with these arc magmatics, the ophiolite now builds the arc massif of the Boshchekul-Tarbagatay arc system (Peive and Mossakovsky, 1982; Khromykh, 1986; Borisenok et al., 1989; Yakubchuk and Degtyarev, 1991), very similar to the present-day ensimatic arc of the Marianas. The arc produced in the late Ordovician magmatic rocks as evolved as granodiorites, similar to those known from the ensimatic arc of the KingsKaweah ophiolite in California (Jason B. Saleeby, personal communication, 2002). Collins et al. (2002) corroborated the ensimatic arc interpretation by pointing out that in this unit the sediments had not been fed from any pre-existing continental source. Unit 25 (Eastern Altay). This unit has boninites as the lowest unit (Simonov et al., 1994; Berzin and Kungurtsev, 1996) and Dobretsov and Buslov (in press) believe that they sit on an ophiolite basement (2 in Fig. 14). The boninites are of Vendian-earliest Cambrian age. The arc evolved to a calc-alkalic composition in the early to medial Cambrian. Unit 26 (Kuznetskii Alatau). The Kuznetskii Alatau range has an ophiolite basement (3 in Fig. 14) on which supposedly the best-known ensimatic arc of the entire Altaid edifice was constructed (Dr. Alexander V. Vladimirov, personal communication, 2002). The basement is of late Precambrian age and the arc constructed on top of it has an age range from

Fig. 17. Some of the major ophiolite and ophirag occurrences in the Altaids. The numbers are referred to in the text. A vast number of smaller occurrences could not be shown at the chosen scale. Also not shown are those under the cover of the West Siberian Basin, which are known through drilling and geophysical means (for the extent of the post-Altaid cover, see Fig. 7). Compare this map with the maps of Precambrian greenstone belts in the end-papers in de Wit and Ashwal (1997).

4. Oceanic Basement Fragments in the Altaids: Some Representative Examples

705

706

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

the Vendian to the Middle Cambrian, separated into two sequences by a Lower Cambrian reefal limestone unit (Volkov, 1988; Kungurtsev, 1993). Unit 42 (Han-Taishir). This unit has in its northern half a pre-Altaid piece of continental crust dated by single zircon evaporation ages on a xenocryst as 1715 Ma. The granite carrying the xenocryst was emplaced at 1127 Ma (Kröner et al., 2001). To this basement is attached the complete Han-Taishir (transliterated as Khantaishir in some publications) ophiolite (4 in Fig. 17). In fact, the ophiolite has been pieced together from various separate fault-bounded blocks and inclusions in a serpentinitic mélange (Khain et al., 2003). Several generations of dykes, sills and gabbro intrusions indicate a protracted history of the ophiolite formation that in turn indicates formation of the ‘ophiolite’ in different tectonic settings. Boninites have been reported from the sheeted dyke complex, although it is impossible to guess from the descriptions to which generation of dykes the boninites actually belong (Simonov et al., 1994). The plagiogranites from the ophiolite gave a zircon age of 568 Ma (Khain et al., 2003). Khain et al. (2003) interpret the Han-Taishir ophiolite to have resulted from forearc spreading, whereas Zonenshain and Kuzmin (1978) had originally interpreted the geochemical data from these ophiolites to indicate a back-arc basin origin. Taking into account the complex overprinting relationships in the dyke complex, its protracted history and its proximity to the continental margin, we consider the Han-Taishir ophiolites to have formed in an Andaman Sea-type situation with spreading next to the continent which shortly later nucleated on top an ensimatic arc that provided the boninites. The presence of the Neoproterozoic (?) Shargyngol dyke complex in the pre-Altaid continental piece (Badarch et al., 2002), if indeed of appropriate age, may be a part of this margin-parallel shear regime. 4.2. Ophiolites and Ophirags that Occur in Former Forearcs Now Entrapped within Transform Sutures Ophiolites as Backstop to Accretionary Wedges Unit 37 (Sangilen). This unit consists of a Baykalian continental nucleus and an Altaid accretionary complex that grew in front of it (present north). The Agardag ophiolites to the north (5 in Fig. 17), recently have been dated as 579 Ma by Sm/Nd (Pfänder et al., 2002). They occur in blocks embedded in a mélange, where previous studies have resulted in the recognition of several types of rock associations differing in mineralogical composition and major-element geochemistry among the gabbros (Izokh et al., 1988). Sedimentary rocks in the mélange are represented by schists, sandstones, cherts and carbonate rocks. Pillow basalts, dykes and microgabbro from a region (7.5 × 20 km) stretching parallel with the mélange zone reveal geochemical and isotope features similar to OIB, island arc and back-arc tectonic settings. Pfänder et al. (2002) have defined the nature of the Agardag ophiolite on the basis of the island arc and back-arc signatures. As a result of their study, the whole of the forearc region of the Sangilen unit (at most 40 km wide) has been interpreted as an independent back-arc/fore-arc Agardag terrane (Badarch et al., 2002), the tectonic connexions of which, as in most ‘terrane interpretations’, are entirely obscure. The close association of the island arc and back arc lavas now separating the accretionary com-

4. Oceanic Basement Fragments in the Altaids: Some Representative Examples

707

plex from the continental nucleus of the Sangilen unit leads us to interpret the ophiolites, in the light of our large-scale Altaid tectonic model (Sengör ¸ and Natal’in, 1996a), as the product of a pre-arc spreading event in front of the continental nucleus of the Sangilen unit and now forming a backstop to the accretionary complex, whose age ranges from Vendian to the early Cambrian. It is thus similar to the Coast Range ophiolite in California and it is, like the Coast Range ophiolite south of San Francisco (W.R. Dickinson, personal communication, 2001, and in preparation), in places highly disrupted. Unit 30 (North Sayan). The North Sayan unit was indicated to be an ensimatic arc in Sengör ¸ and Natal’in (1996a). Indeed, the Vendian (?) through Lower Cambrian andesites, basalts, dacites, hyaloclastites, tuffs, cherts and reef limestones constitute the arc massif. In the accretionary wedge to the (present) south of this arc, in the Boruss mélange belt (6 in Fig. 17), Simonov et al. (1994) and Dobretsov et al. (1995) report ophiolites with geochemical signatures indicating island arc boninites and back-arc basin basalts. These ophiolites are bereft of a sheeted dyke complex. We interpret these sequences as parts of an ophiolitic forearc in front of which an accretionary complex had developed that include high-pressure schists, metapeiltes, metacherts, marbles and greenschists with ages around 450 to 400 Ma. The structure of this forearc complex is now much disrupted owing to considerable coastwise transport and oroclinal bending of the entire Western Sayan Mountains (including the Northern Sayan (30 in Fig. 7) and the Western Sayan (39 in Fig. 7) units. Unit 39 (Western Sayan). This unit also consists of an ensimatic arc massif and a large accretionary wedge. The arc basement is exposed to the west of the unit and consists of island arc-boninite-bearing Kurtushiba ophiolite (7 in Fig. 17). The ophiolite consists of a package of nappes thrust onto glaucophane-bearing accretionary complex rocks. The lower part of the package consists of three nappes each of which being represented by volcanic and sedimentary rocks. Their thickness individually ranges from 1.5 to 3 km. It is clear that with such thicknesses, if they are indeed stratigraphic, we cannot have a normal oceanic crustal succession. It is probably some sort of a pre-arc spreading product. Above is an ultramafic/gabbro/diabase thrust sheet that is 7 km thick. Its internal structure is said to be ‘weakly disturbed’. The nappe includes metamorphic peridotites, cumulate ultramafics and gabbros followed by isotropic gabbros and diabases and volcanic rocks. The diabase section contains sheeted dykes in the central part of the ophiolite outcrop and sills in its northeastern sector. The entire succession is terminated by Lower Cambrian volcanic rocks of island arc type forming a small area in the central and southern part of the ophiolite belt (Berzin and Kungurtsev, 1996). Farther east are coeval arc magmatic rocks including basalts, andesites, felsic volcanics, tuffs and conglomerates and reef limestones. Westwards, shallow-marine rocks of the forearc basin appear. Within the forearc area are MORB-type ophirags that probably represent offscraping from the downgoing slab. The Kurtushiba ophiolite exposed along the western margin of the arc massif thus resembles, in its tectonic position, to the ophiolite believed to underlie the Great Valley Sequence in California.

708

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Ophirags Within Accretionary Wedges Ophirags occurring as accretionary wedge scraps are the most common types of oceanic basement fragments preserved in the Altaid edifice. Our Fig. 17 shows those that can be shown at the chosen scale. Naturally, there are many more in all of the Altaid accretionary complexes. In the following, we discuss only those that have been recently studied in terms of their regional geology, petrology and geochemistry. Unit 4.1 (Djezkazgan-Kirgiz). In this unit, the ophirags of Kara-Archa (8 in Fig. 17), Kenkol (9 in Fig. 17), Toluk (10 in Fig. 17), Karachi-Karakty (11 in Fig. 17), Karadzhorgo (12 in Fig. 17), and Archaly (13 in Fig. 17), have been studied geochemically with a view to determining their tectonic setting of origin. Fossil finds in the associated pelagic sedimentary rocks indicate that they are of Cambrian-Middle Ordovician age. Basaltic rocks associated with these ophirags are mostly of normal MORB type, but some, such as Toluk, reveal E- and T-MORB features indicating ocean island remnants (Lomize et al., 1997). In Toluk and Kaarchi-Karakty there are basalts exhibiting arc-related features; therefore they have been compared with SSZ ophiolites. The study of spinel from the peridotite in the KarachiKarakty body also indicates mid-oceanic and supra-subduction zone tectonic settings (Demina et al., 1995), possibly indicating a pre-arc spreading event. In the Kara-Archa body, arc-related Ordovician volcanic rocks overlie the ophiolitic pseudostratigraphy. This region has been interpreted as an ensimatic arc. Lomize et al. (1997) thought that the arctrench system of the Djezkazgan-Kirgiz unit stopped operating in the Caradocian. However, there is clear evidence to the south of these ophirags that the subduction-accretion complex continued its growth until the early Carboniferous (Sengör ¸ and Natal’in, 1996a). The Southern Tien Shan Accretionary Complex in China. Farther east, in China, a piece of accretionary complex formed south of several units of the Altaids in the KazakhstanTien Shan domain after these units had been assembled by strike-slip stacking by the early Devonian (Figs. 7 and 12). This late-forming unit was left nameless by Sengör ¸ and Natal’in (1996a). The Youshugou ophirag (14 in Fig. 17) in the southern Tien Shan occurs in this nameless accretionary complex and is of pre-Middle Devonian age. Its mafic dykes display an REE pattern interpreted to indicate mixed magma sources (Allen et al., 1992). A first group of magma is characterised by depleted HFSE, characteristic of supra-subduction ophiolites. A second group has enriched HFSE content that is typical for off-spreading sites such as seamounts. However, the sampling is from a mélange and the two separate magma sources may in fact indicate two separate bodies of ophirags stemming from two contrasting tectonic setting now mixed in the mélange (Allen et al., 1992). The rest of the known southern Tien Shan ophirags in China (15 in Fig. 17) are reported in the Changawuzi, Kule, Heiyinshan and Kumishi regions. Their basalts are of MORB type and Chen et al. (1999) consider them as having originated in mid-ocean ridge environments. An Ar/Ar plateau age from a pyroxene from the gabbro section in one of these bodies gives an age of 439.4 ± 26.7 Ma (Llandoverian). Blueschists from the mélange yield 351 ± 2 Ma (Famennian). The nameless accretionary complex of the southern Tien-Shan clearly has accumulated pieces of oceanic basement from diverse environments. Normal oceanic crustal fragments plus ocean islands were clearly swept into the subduction zone in front of it. Some of the

4. Oceanic Basement Fragments in the Altaids: Some Representative Examples

709

supra-subduction zone signatures may indicate Mariana-type outbowed migratory arcs or they may have been like the eastern Aleutians barring embayments formed by indentations in the pre-Altaid continental piece that forms the backstop of the unit. The present state of knowledge about its structural geology unfortunately does not allow us to choose among these possibilities. Unit 14 (Tekturmas). Ophirags to the southeast of the Tekturmas unit have been recently studied from the viewpoint of their regional geology, petrology and geochemistry by Wang et al. (in press). These are the Dabut (16 in Fig. 17) and the Honguleleng (17 in Fig. 17) ophirags. Darbut has a Sm-Nd age of 395 Ma (Emsian) and is closely associated with Ordovician radiolarian cherts in the subduction mélange of the Tekturmas unit. The geochemical discrimination diagrams indicate that the Dabut ophirag represents a MORB-type ophiolite later enriched in an ocean island setting probably by plume contribution. It may be that the original ocean floor was of Ordovician age onto which the later hot-spot edifice was built. It is clear that the Dabut ocean island was clipped off the downgoing plate and added as an ophirag into the subduction mélange. The Honggueleng ophirag also has MORB signatures, but also island arc signatures. If one considers the reconstruction shown in Fig. 13, the reason for this becomes readily apparent. Pieces of unit 18 (Zharma-Saur) representing an ensimatic Ordovician island arc were fed laterally into the accretionary complex of unit 14 by arc-shaving strike-slip faults to become ophirags within it. Thus, ocean floor ophirags and island arc ophirags were brought into close proximity within an accretionary complex by means of arc-shaving strike-slip faults. Unit 15 (Junggar-Balkhash). In the interior position of the Junggar-Balkhas unit, Wang et al. (in press) studied the Karameili (18 in Fig. 17) and the Mayila (19 in Fig. 17) ophirags. Of these, the Karameili ophirags give a pure MORB signature and they are clearly just plucked off pieces of the downgoing oceanic crust. The Mayila ophirags are also of MORB-type, but they are enriched by a plume source and probably are pieces of former ocean islands. Their tectonic history was probably not dissimilar to those of the Dabut and the Honggueleng complexes (compare Figs. 13 and 17). Unit 19 (Ob-Zaysan-Surgut). The Ob-Zaysan-Surgut unit is one of the largest tectonic units of the Altaids. It was transported from the present-day southeastern Mongolia to its present position along the giant Irtysh keirogen. The Aermentai ophiolites are located within this unit (20 in Fig. 17) and give a Sm-Nd isochron age of 561 ± 41 Ma (latest Precambrian or earliest Cambrian) according to Hu et al. (2000). By contrast, Wang et al. (in press) report a Sm-Nd age of 479 ± 27 Ma (Arenigian), measured by Liu and Zhang (1993) on the same ophiolite. A whole rock isochron age of the cumulate gabbro, diabase and andesite is said to give 561 ± 41 Ma (Huang et al., 1997). It seems that the ophiolite is probably Cambro-Ordovican in age and gives geochemical signatures indicating an ensimatic island arc setting. This inferred setting is consistent with the reconstruction given in Fig. 11, except that the arc was probably formed offshore of the Tuva-Mongol fragment and was later moved by coastwise transport to its present position as suggested by Sengör ¸ and Natal’in (1996a) along arc-slicing strike-slip faults. It was during this process that they

710

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

become incorporated into the large accretionary complex of the Ob-Zaysan-Surgut unit. The presence of early Devonian adakites not too far away from this ophiolite complex support their location in a setting not too dissimilar to the Aleutian arc system exhibiting considerable strike-slip. Unit 25 (Eastern Altay). Dobretsov and Buslov (in press) have claimed that more than 50% of the Eastern Altay unit is made up of stranded seamounts of considerable size, in fact, fragments of a veritable plateau. However, during the IGCP 283 international excursion in 1993 we noted that the mafic rocks and their shallow-water limestone cover appeared in extremely deformed ophirags incorporated into a vast subduction mélange terrain. Much of the mélange Dobretsov and Buslov (in press) interpret as debris flows accumulated along the aprons of seamounts. This naturally gave the impression that seamounts (or an oceanic plateau) had been preserved entire and had, in fact, choked and stopped the subduction. However, we thought that their ‘olistostromes’ were, in fact, tectonic mélanges. Associated pillow lavas were highly deformed and dismembered, indicating large amounts of internal strain. Therefore, the seamounts themselves were deformed and dismembered and incorporated into a mélange wedge together with normal MORB-type basalts, boninites, high pressure rocks and pelagic sedimentary rocks such as cherts during active subduction. What the eastern Altay unit exhibits is nothing more than diverse types of ophirags stemming from different provenances now embedded in a subduction-accretion mélange. Unit 35 (Dzhida). Small ophirag lenses (22 in Fig. 17) are scattered in the accretionary wedge of the Dzhida unit. Stratigraphic estimates limit their age to Vendian-Early Cambrian time. Volcanic rocks of these ophirags reveal an oceanic signature, and they have been interpreted as relicts of oceanic crust (Gordienko, 1987; Kuzmin et al., 1995; Parfenov et al., 1995). Some subalkalic basalt may represent fragments of seamounts. To the south of the accretionary wedge there is a body of an ensimatic magmatic arc of Vendian-early Cambrian age in which felsic magmatism was active till the end of the Cambrian. A Neoproterozoic ophiolitic basement is inferred for this arc (Badarch et al., 2002) though direct evidence for this suggestion is lacking. The accretionary wedge/magmatic arc disposition as well as the nature of ophirags clearly indicate a north facing for the arc and several recent reconstructions have so portrayed it (e.g., Parfenov et al., 1995; Sengör ¸ and Natal’in, 1996a). However in a more recent paper the Dzhida arc has been shown as a south-facing arc without any compelling reason (Badarch et al., 2002). Unit 43.2 (Khangay-Khantey). Khangay-Khantey is a vast collage of accretionary complexes of various ages including various ophiolites and ophirags. In fact, there is no intact ophiolite in this entire zone. However, in the Bayankhongor area a large ophiolitic outcrop extends for 300 km in a NW-SE orientation exhibiting a serpentinitic mélange including a complete ophiolite suite but in form of collections of ophirags (21 in Fig. 17: Buchan et al., 2001, 2002). A 569 ± 21 Ma Sm-Nd age on pyroxene and whole rock from ophiolitic gabbro (Kepezhinskas et al., 1991) indicates a late Precambrian (Vendian) age for this ‘ophiolite’. Khain et al. (2003) reported a similar age (571 ± 4 Ma). The ‘ophiolite’ is interpreted to have formed as a result of sea-floor spreading on the basis of enriched MORB signatures (Kepezhinkas et al., 1991; Badarch et al., 2002; Buchan et al., 2001, 2002). The

4. Oceanic Basement Fragments in the Altaids: Some Representative Examples

711

shallow water sedimentary rocks covering it may indeed indicate a plume environment for its original production and may render a clue to its preservation. The presence of extensive serpentinitic mélanges in it is very similar to the Kings-Kaweah ophiolite in California and may indeed suggest a fracture zone origin that originally may have localised an ocean island similar to the Coastal Complex in Newfoundland (Karson and Dewey, 1978). Buchan et al. (2001, 2002) maintain that the Bayankhongor ophiolitic mélange marks a suture zone, because ‘It seems unlikely that an ophiolite fragment 300 km long would remain intact and unmixed with the rest of the rocks in an accretionary wedge’ (Buchan et al., 2001, p. 459). This is a surprising statement in view of (1) the extremely disrupted and mélanged aspect of the Bayankhongor occurrence (Buchan et al., 2001), (2) the presence of similar areas in zones of accretion, such as the 500 km-long Kings-Kaweah ophiolite belt in the western foothills of the Sierra Nevada (Saleeby, 1977, 1981; Saleeby et al., 1978), the entire Coastal complex in the Coast Ranges in California that is coherent to the north of San Francisco for 200 km and incoherent but continuous (like the Bayankhongor suite) for 300 km south of it (e.g., Evarts, 1977; Contenius et al., 2000), and coherent slabs of considerable size within accretionary complexes in Guatemala (Aubouin et al., 1984) and Alaska (Plafker et al., 1994), (3) the location of the Bayankhongor ophiolite within a steep zone of metapelites and metapsammites, which, by the admission of Buchan et al. (2001) mostly show an incoherent mélange or broken formation style of deformation. There is no evidence northeast of the Bayakhongor ophiolitic belt of a continental piece onto which it could have been ‘obducted’. The alleged ‘passive continental margin’ of Badarch et al. (2002, p. 94) is described by them to include ‘lenses of limestone, sandstone, chert, tuff, minor felsic volcanic material and vesicular basalt’ which have been ‘intensively deformed and telescoped into a tectonic mélange’. The entire style of deformation is one of close to isoclinal folding with much strike-slip faulting with greenschist metamorphism typical of accretion complexes. To the northeast of the Bayankhongor ophiolite, the strongly deformed DevonianCarboniferous turbidites, erroneously interpreted as passive continental margin deposits by Buchan et al. (2001) and Badarch et al. (2002), continue for more than 1500 km into Russian territory. The turbidites contain slivers/beds of mafic and intermediate volcanic rocks and red pelagic cherts as recognised by Buchan et al. (2001) in Mongolia (Marinov et al., 1973; Zonenshain, 1972). Fault-bounded blocks and sheets of basalts, cherts and sedimentary rocks metamorphosed in greenschist facies appear among them but mainly along the margins of the Devonian-Carboniferous turbidite units (see Figs. 7 and 17). The age of these rocks has been interpreted as Vendian-Cambrian or as early Paleozoic, although evidence for these ascriptions is sparse. Majority of researchers agree that the DevonianCarboniferous and Vendian-Cambrian or Lower Paleozoic rocks of the Khangai-Khantey belong to subduction-accretion complexes (Sengör ¸ et al., 1993; Gusev and Khain, 1995; Sengör ¸ and Natal’in, 1996a, 1996b; Zorin, 1999; Parfenov et al., 1999). Finds of small lenses of ophirags support their conclusions (e.g., 23 in Fig. 17). In Mongolia, the age of the ophirgs/ophiolites is poorly constrained as pre-Silurian (Parfenov et al., 1999; Tomurtogoo, 1997). Pillow lavas and sheeted dykes reveal MORB-type features.

712

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

The negative ε Nd signatures from the Palaeozoic and Mesozoic granites intruding Badarch et al.’s (2002) ‘passive continental margin’ most likely tap the highly deformed and thickened turbidite wedge of the Khangai-Khantey subduction-accretion complex that had been fed by the Precambrian Tuva-Mongol continental fragment (Zonenshain, 1973; Sengör ¸ and Natal’in, 1996a) much like the negative ε Nd signatures known from the Lachlan subduction-accretion complexes in the Tasman orogenic belt in Australia. The structural sections in the paper by Buchan et al. (2001) clearly indicate that the Bayakhongor ophiolite is an intra-accretionary complex ophiolite slab, much disrupted during incorporation. We are not even sure that it was indeed originally a single slab. In the Russian part of the Khangai-Khantey unit, the subduction-accretion complex includes Triassic turbidites. Fault-bounded blocks and sheets of strongly deformed metabasalts, cherts, and turbidites metamorphosed in greenschist facies as well as gabbro and ultramafic rocks are mapped as the Riphean Kulinda Suite (24 in Fig. 17; Suite as used in the Russian Stratigraphic nomenclature is exactly equivalent to the formation as defined in the International Stratigraphic Guide: Salvador, 1994, p. 42; de Wever and Popova, 1997, p. 394). Its age is not constrained by fossil findings or isotopic dating but inferred on the basis of long-distance correlation. Basalts of the Kulinda Suite reveal NMORB and T-MORB signatures indicating presence of fragments of normal oceanic crust and off-spreading magmatic constructions (Gusev and Peskov, 1993, 1996). Similar types of ophiolitic rocks have been established in the Molodovsk (25 in Fig. 17) and Gorbits regions (26 in Fig. 17). Younger, Devonian?-Triassic? ophiolites of the Ust-Tura region (27 in Fig. 17) represent a back-arc basement (Gusev and Peskov, 1996). This short overview shows that along the whole length of the Khangai-Khantey unit there is a unity of rock types and type of ophiolites/ophirags that mainly occur as slivers in mélanges and belong to N-MORB and OIB types. Only along the margins of the unit arc-related settings are known. Unit 43.3 (South Mongolian). Small tectonic lenses of ophirags are abundant in the South-Mongolian unit. They are commonly described as blocks in mélanges or imbricate thrust sheets (Marinov et al., 1973; Ruzhentsev et al., 1985, 1987; Ruzhentsev and Pospelov, 1992). Ophirags in the South Mongolian unit are associated both with tholeiitic and island arc volcanic rocks, shallow-marine and pelagic sediments. We are not aware of any recent detailed studies of these ophirags. They may be of both oceanic and island arc origin as it is indicated by recent detailed studies of sedimentary rocks (e.g., Lamb and Badarch, 1997). Unit 44 (South Gobi). New petrologic and isotopic studies are available for ophiolites of this unit. Petrographic, geochemical (major elements) and REE studies of the Hegeshan ophirags (28 in Fig. 17) of the Nei Mongol region, China, have shown a similarity with rocks of mid-ocean ridges (Nozaka and Liu, 2002). These ophirags represent basement fragments of the floor of the former Solonker ocean that closed in the Permian (Fig. 15; Zhang et al., 1984, Wang and Liu, 1986, 1991; Sengör ¸ and Natal’in, 1996a; Chen et al., 2000). However, metamorphic minerals of amphibolites yield unexpectedly young K-Ar ages of 110 and 130 Ma (early Cretaceous: Nozaka and Liu, 2002). The ophirags formed much earlier as it is shown by the Sm-Nd whole rock age of 403 ± 27 Ma

5. Conclusions

713

(early Devonian: Chen et al., 2000). At the same time Robinson et al. (1999) report inhomogeneity of mineral composition and the geochemistry of the Hegeshan ophirags. Major element and trace element studies allowed them to distinguish rocks that originated in arc and backarc settings and in zones of within plate basaltic magmatism. The latter can be fragments of seamounts. We must note, however, that the Hegeshan ophirags occur in a tectonic mélange and tectonic settings, inferred solely on geochemical grounds from small fragments, need the structural relationships to be established to be entirely credible (e.g., Hsü et al., 1991). 4.3. Ophiolites and Ophirags in Collisional Suture Zones Unit 13 (Boshchekul-Tarbagatay). We have already pointed out that the BoshchekulTarbagatay arc system was built on an oceanic foundation. This arc system, that had originated as a single arc in the Cambrian (Sengör ¸ and Natal’in, 1996a), represents in fact a double arc, namely the Boshchekul-Tarbagatay sensu stricto (unit 13.1) and the BayanaulAkbastau (unit 13.2). These two arcs are now separated by the Maikain-Balkybek ophiolitic suture zone that forms a 700 km-long (i.e., almost exactly as long as the present-day Okinawa back-arc or the Mariana inter-arc/back-arc basin) and very narrow, linear/arcuate steep belt of ophiolites (Yakubchuk and Degtyarev, 1991). Sengör ¸ and Natal’in (1996a) had interpreted this suture to represent a now vanished early Ordovician intra-arc/backarc basin, the ophiolites of which had been clearly generated in a supra-subduction zone environment. The marginal basin closed by the early to medial Llandoverian. Unit 4.2 (Jalair-Nayman). A similar long and narrow, linear/arcuate pre-Ordovician ophiolite belt separates the Jalair-Nayman unit (unit No. 4.2) into two moieties. The preOrdovician ophiolites contain hornblende in the cumulate gabbro section and that is why Sengör ¸ and Natal’in (1996a) had interpreted them as of supra-subduction zone origin. Cambro-Ordovician turbidites cover the ophiolite stratigraphically. The basin closed during the Ordovician-(?)Silurian interval.

5. CONCLUSIONS The bold tone of many of the interpretations we present of the original tectonic settings of the ophiolites and the ophirags in the Altaid edifice in this paper must have struck the reader. In reality, our mood in formulating our interpretations was one of much diffidence. The reason for this diffidence was lack of precise and accurate observations in an orogenic system that is immense and occupies a terrain not entirely conducive to easy work. However, even where observations are plenty and multifarious, our diffidence was not entirely alleviated. The reason for this is the extremely disrupted state of the entire orogenic edifice. The scale of disruption ranges from literally millimetric scale in shear zones to tens or even hundreds of kilometres where large tectonic units are disrupted between the large cratons of Russia and Angara. Our bold presentation aims at making our choices clear with a view to encouraging field checks.

714

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

What makes the tectonic interpretation of the oceanic basement remnants difficult is that most of them occur as ophirags and torn from their oiriginal places of incorporation into the continental crust. Processes of incorporation of oceanic basement rocks into the continental crust are complicated enough. Therefore, when intending to reconstruct the nature of ophiolitic rock assemblages in the Altaids (and in Turkic-type orogens: Sengör ¸ and Natal’in, 1996b), their original structure and original disposition with respect to neighbouring units, one must account for not only the complicated processes of incorporation into continents, but also for further deformation, tearing of the assemblages in pieces and their following distribution in areas with length scales in excess of 1000 km. All we see in the Altaids is that there are no privileged linear/arcuate zones along which ophiolites crowd as in familiar Alpine- and Himalayan-type collisional orogenic belts. Neither do they commonly define narrow, but long belts of accretionary complexes as in the Cordilleran belts. They appear to have formed in or adjacent to such accretionary complexes only after the pre-assembly form of the Altaid collage is reconstructed (see Figs. 10–16 herein). A glance at Fig. 17 shows that they are now literally everywhere within the orogenic collage! They are literally everwhere in form of small bodies compared with the intact, large ophiolite nappes of Oman, Turkey, Newfoundland or Papua New Guinea. In that respect they greatly resemble the Archaean ultramafics and mafic rocks. Moreover, many of the Altaid ophiolites have been interpreted as ensimatic island arc remnants or products of pre-arc spreading events. That is why they have thicker and more varied mafic, in places even intermediate and felsic, volcanic components. Such components range from mafic komatiites and boninites to even rhyolites, with island arc tholeiite and andesite dominance. This variety is also one we encounter in the Archaean greenstone belts (for a few recent cases relevant for rock-types mentioned in this paper, see Polat and Kerrich, 2001; Polat et al., 2002). When we add the reported slivers of oceanic plateaux and guyots in the Altaid ophirags, the resemblance to the Archaean terrains with their thick komatiitic lava piles becomes essentially complete. Perhaps the most valuable lesson a Precambrian geologist would learn from the Altaids is how careful she or he must be in judging the present setting and tectonic history of any piece of oceanic basement. In the Altaids, oceanic basement fragments became incorporated into continental structure by a truly bewildering diversity of mechanisms ranging from the familiar, cartoon-style, head on subduction-related trapping and offscraping effects through multifarious oblique-subduction-related strike-slip events and their associated complexities including arc-slicing and arc-shaving faults, releasing and restraining bends along them, numerous kinds of secondary structures that form along the paths of emerging shear zones such as Riedel and anti-Riedel shears, P- and X-shears, rotations around vertical axes of both blocks and their bounding structures, to a variety of collisional and post-collisional processes that both help incorporate ocean floor basement remnants into continents and, once incorporated, further deform, dismember and distribute them. Alternating head-on and oblique subduction events would repeatedly superpose their effects onto the incorporated and preserved basement fragments until they become almost totally bereft of any indications as to their original provenance.

5. Conclusions

715

Hamilton (1992 and 1998) has complained that mélange belts were lacking in the Archaean. We think that this complaint is not justified. Sengör ¸ has been shown such jumbled rock associations resembling Phanerozoic mélange belts both in the Superior Province by Ali Polat (see Polat and Kerrich, 1999) and in the Yilgarn craton by Nick Archibald (see Sengör ¸ and Natal’in, 1996b). However, the extreme disruption seen in the greenstone belts that interleaves and intercalates ultramafics, mafic volcanic and plutonic rocks and sedimentary rocks commonly creates geometries that greatly resemble the mélange belts of the Phanerozoic. The problem in recognising Archaean mélanges is the associated rock types (ultramafic-mafic igneous rocks and dominantly pelites and cherts with little to no neritic limestone knockers), extreme deformation usually leading to complete transposition of original lithological contacts and the commonly accompanying, multi-phase amphibolite grade metamorphism that make the recognition of original rock types and geometries extremely difficult. In Isua, for instance, John Myers showed Sengör, ¸ in a deformed pillow lava, individual pillows with thicknesses of only a few tens of cm and lengths exceeding a few tens of metres! Under such conditions only a most meticulous, large-scale mapping by a geologist experienced in mélange tectonics can possibly decipher a mélange. But the resemblance does not stop at the rock association level. The structures seen in the Altaids and in the Archaean greenstone belts are very similar pointing to the operation of similar processes and in similar sequence (see Sengör ¸ and Natal’in, 1996b and the many contributions in de Wit and Ashwal, 1997). We are unable to agree with Hamilton’s (1992, 1998) defense of interpretations of simple deformation of greenstone belts. Sengör ¸ has first-hand experience in some of the Canadian and Yilgarn greenstone belts and in the Isua region of Greenland, which he was able to visit under the guidance of local experts of structural geology who had mapped them. In all of them the tectonic deformation is intense and multiphase and not confined to the surroundings of rising igneous diapirs as maintained by Hamilton. All of them very greatly resemble the tectonic style we are familiar with from the Altaids. In the Yilgarn, even the occurrence and sequence of strike-slip faulting as established by Nick Archibald (see Sengör ¸ and Natal’in, 1996b) are very reminiscent of the Altaids. Most Altaid oceanic basement fragments are ophirags. There is not in them a single giant ophiolite nappe of the kind we know from Oman or from Papua New Guinea or Turkey or Newfoundland. This is also the case in the Archaean. The entire intact Archaean terrain preserved today on the face of the earth occupies some 6% of the entire land surface, in other words an area about 8 million square kilometres (Condie and Sloan, 1998, Fig. 8.3; we think this is an overestimate!). The Altaid orogenic system, by contrast occupies some 8.5 million square kilometres. Therefore, the entire Altaid System covers almost exactly as much area as the entire area occupied by the intact Archaean crust! Since the whole of the Altaids do not contain a single giant ophiolite nappe, is it so surprising that the Archaean does not either? If only the Altaids were to be preserved 2500 million years hence, the geologists of that remote era (assuming that they would have the same means as we do today) would not have the slightest notion of the types of mountain-building, namely Alpine, Himalayan and the Andean, on which most of our theories of orogeny have so far been built.

716

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

When one considers that the style of the Altaids is not unique to them in the Phanerozoic, but is represented by the Mesozoic-Cainozoic tectonics of the Nipponides (Sengör ¸ and Natal’in, 1996a, 1996b), of the North American Cordillera in the Mesozoic (Burchfiel et al., 1992), of the eastern part of a part of the Cimmerides in the Palaeozoic and the Mesozoic (Sengör, ¸ 1984), and of the South American Cordillera in the Palaeozoic-early Mesozoic and also in part of the eastern part of the Gondwanides in the late Palaeozoicearly Mesozoic (Sengör ¸ et al., 2001), we realise that the Archaean tectonic style is very much well and alive today. During the Phanerozoic, only in few places strike-slip faulting has gathered such an immense chunk of subduction-accretion complexes into a continental form as in the Altaids and that is why the great similarity of the Altaid style Turkic-type orogeny to Archaean terrains is not readily obvious. But a simple glance at Proterozoic tectonics in such places is in the Pan African system in northeast Africa and Arabia or in the Mazatzal System in North America convinces us that much of the continental crust has been built by Turkic-style orogens and that they have remained operative from the earliest Archaean to the present-day (Kusky and Polat, 1999).

ACKNOWLEDGEMENTS We thank Tim Kusky for inviting this contribution and waiting for it with endless and good-humoured patience. We are grateful to Brian F. Windley for helpful discussions on the geology of the Altaids. We also thank Bor-ming Jahn for his leadership in generating and administering international projects to test various tectonic models for the Altaids, from which we have derived great benefits. Rob van der Voo kindly informed us before publication of the results of his and his colleagues’ palaeomagnetic work in Central Asia. Evgeny V. Khain has discussed with us the tectonics of the Altay-Mongolian sector of the Altaids and generously shared his unpublished data and ideas on the Eastern Sayan and Central Mongolian ophiolites. V.E. Khain and Warren B. Hamilton have been, as always, helpful with discussions and excellent advice. Sengör ¸ is indebted to Ali Polat for leading him in the Schreiber-Hemlo belt in the Superior Province in Canada, to Ali Polat and John Myers for showing him part of the Isua region in western Greenland and to Nick Archibald for showing him a cross-section across the Yilgarn craton in Australia. He is also grateful to William R. Dickinson, for, among numerous other things, showing him the Californian Coast Ranges under his own incomparable guidance.

REFERENCES Allen, M.B., Windley, B.F., Chi, Z., 1992. Paleozoic collisional tectonics and magmatism of the Chinese Tien Shan, central Asia. Tectonophysics 220, 89–115. Allen, M.B., Sengör, ¸ A.M.C., Natal’in, B.A., 1995. Junggar, Turfan, and Alakol basins as Late Permian to? Early Triassic sinistral shear structures in the Altaid orogenic collage, Central Asia. Journal of the Geological Society of London 152, 327–338.

References

717

Amstutz, G.C., 1980. The early history of term Ophiolites and its evolution until 1945. In: Panayiotou, A. (Ed.), Ophiolites—Proceedings of the International Ophiolite Symposium, Cyprus, 1979. Ministry of Agriculture and Natural Resources Geological Survey Department, Republic of Cyprus, pp. 149–152. Argand, E., 1911. Sur la répartition des roches vertes mesozoiques dans les Alpes Pennines avant la formation des plis couchés. Rech. Sci. Phys. Natur. Genève 31, 575. Argand, E., 1916. Sur l’arc des Alpes occidentales. Eclogae Geologicae Helvetiae 14, 145–191. Argand, E., 1924. La tectonique de l’Asie. In: Congrés Géologiques International, Comptes Rendus de la XIIme session, Premier Fascicule. H. Vaillant-Carmanne, Liége, pp. 171–372. Aubouin, J., Bourgois, J., Azema, J., 1984. A new type of active margin: the convergent-extensional margin, as exemplified by the Middle America Trench off Guatemala. Earth and Planetary Science Letters 67, 211–218. Auzende, J.-M., Ceuleneer, G., Cornen, G., Juteau, T., Lagabrielle, Y., Lensch, G., Mevel, C., Nicolas, A., Prichard, H., Ribeiro, A., Ruellan, E., Vanney, J.R., 1984. Intraoceanic tectonism on the Gorringe Bank: observations by submersible. Geological Society of London Special Publication 13, 113–120. Badarch, G., Cunningham, W.D., Windley, B.F., 2002. A new terrane subdivision for Mongolia: implications for the Phanerozoic crustal growth of Central Asia. Journal of Asian Earth Sciences 21, 87–110. Bailey, E.B. (Sir), 1936. Sedimentation in relation to tectonics. Geological Society of America Bulletin 47, 1713–1726. Bailey, E.B. (Sir), 1944. Mountains that have travelled over volcanoes. Nature 154, 752–756. Bailey, E.H., Blake Jr., M.C., Jones, D.l., 1970. On-land Mesozoic oceanic crust in California Coast Ranges. U.S. Geological Survey Professional Paper 700-C, C70–C81. Bazhenov, M.L., Levashova, N., Collins, A., Van der Voo, R., Degtyarev, K.E., 2002. The Kazakhstan orocline (C. Asia), Part I: paleomagnetism and paleolatitudinal motion. Geological Society of America Abstracts with Programs 34 (6), 491. Bazhenov, M.L., Collins, A.Q., Degtyarev, K.E., Levashova, N.M., Mikolaichuk, A.V., Pavlov, V.E., Van der Voo, R., 2003. Paleozoic northward drift of the Northern Tien Shan (Central Asia) as revealed by Ordovician and Carboniferous paleomagnetism. Tectonophysics 366, 113–141. Berzin, N.A., Kungurtsev, L.V., 1996. Geodinamicheskaya interpretatsiya geologicheskikh kompleksov Altae-Sayanskoi oblasti. Geologiya i Geofizika 37 (1), 63–81. Bindeman, I.A., Vinogradov, V.I., Valley, J.W., Wooden, G.L., Natal’in, B.A., 2002. Archean protolith and accretion of crust in Kamchatka: shrimp dating of zircons from Sredinny and Ganal massifs. Journal of Geology 110, 271–289. Bishopp, D.W., 1952. Some new features of the Geology of Cyprus. 19th International Geological Congress. Algiers 17, 13–18. Borisenok, V.I., Ryazantsev, A.V., Degtyarev, K.E., Yakubchuk, A.S., 1989. Paleozoiskaya geodinamika tsentralnogo Kazakhstana. In: Pushcharovskiy, Y.M. (Ed.), Tektonicheskie issledovaniya pri sredne- i krupnomasshtabnom geokartirovanii. Nauka, Moscow, pp. 81–95. Brongniart, A., 1813. Essai d’une classification minéralogique des roches mélangées. Journal of Mines 199, 1–48 (offprint pages). Brongniart, A., 1821. Sur le gisement ou position relative des ophiolites, euphotides, jaspes, etc. dans quelques parties des Apennines. Ann. Mines Rec. Mém. Exploit. Mines 6, 177–238. Brongniart, A., 1827. Classification des Roches Homogènes et Héterogènes. Levrault, Paris, p. 144.

718

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Buchan, C., Cunningham, D., Windley, B.F., Tomurhuu, D., 2001. Structural and lithological characteristics of the Bayankhongor ophiolite zone, Central Mongolia. Geological Society of London 158, 445–460. Buchan, C., Pfänder, J., Kröner, A., Brewer, T.S., Tomurtogoo, O., Tomurhuu, D., Cunningham, D., Windley, B.F., 2002. Timing of accretion and collisional deformation in the Central Asian Orogenic Belt, implications of granite geochronology in the Bayankhangor ophiolite zone. Chemical Geology 192, 23–45. Burchfiel, B.C., Cowan, D.S., Davis, G.A., 1992. Tectonic overview of the Cordilleran orogen in the western United States. In: Burchfiel, B.C., Lipman, P.W., Zoback, M.L. (Eds.), The Cordilleran Orogen: Conterminous U.S. The Geology of North America G-3. Geological Society of America, Boulder, pp. 407–479. Burke, K., 1997. Foreword. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. Clarendon, Oxford, pp. v–vii. Burke, K., Dewey, J.F., Kidd, W.S.F., 1976. Dominance of horizontal movements, arc and microcontinental collisions during the later permobile regime. In: Windley, B.F. (Ed.), The Early History of the Earth—Based on the Proceedings of A Nato Advanced Study Institute held at the University of Leicester, 5–11 April, 1975. Wiley, London, pp. 113–129. Burrett, C.F., 1974. Plate tectonics and the fusion of Asia. Earth and Planetary Science Letters 21, 181–189. Cadisch, J., Leupold, W., Eugster, H., Brauchli, R., 1919. Geologische Untersuchungen in Mittelbünden. Vierteljahrsschrift der Naturforschenden Gesellschaft in Zürich 66 (Heim Festschrift), 359–417 + one foldout map. Cardwell, R.K., Isacks, B.L., Karig, D.E., 1980. The spatial distribution of earthquakes, focal mechanism solutions, and subducted lithosphere in the Phillipine and northeastern Indonesian islands. In: Hayes, D.E. (Ed.), The Tectonic and Geologic Evolution of Southeast Asian Seas and Islands. In: Geophysical Monographs, vol. 23. American Geophysical Union, pp. 1–35. Casey, J.F., Dewey, J.F., 1984. Initiation of subduction zones along transform and accreting plate boundaries, triple junction evolution, and forearc spreading centres—implications for ophiolitic geology and obduction. Geological Society Special Publication 13, 269–290. Chen, B., Jahn, B.-M., Wilde, S., Xu, B., 2000. Two contrasting Paleozoic magmatic belts in northern Inner Mongolia, China: petrogenesis and tectonic implications. Tectonophysics 328, 157–182. Chen, C., Lu, H., Jia, D., Cai, D., Wu, S., 1999. Closing history of the southern Tianshan oceanic basin, western China: an oblique collisional orogeny. Tectonophysics 302, 23–40. Coleman, R.G., 1971. Plate tectonic emplacement of upper mantle peridotites along continental edges. Journal of Geophysical Research 76, 1212–1222. Coleman, R.G., 1977. Ophiolites—Ancient Oceanic Lithosphere? Springer-Verlag, Berlin, IX + 229 pp. Collet, L., 1935. The Structure of the Alps, second ed. Edward Arnold, London, xvi + 304 pp. + 12 plates. Collins, A.Q., Levashova, N.M., Degtyarev, K.E., Bazhenov, M.L., Van der Voo, R., 2001. Permian paleomagnetism of eastern Kazakhstan: Implications concerning the amalgamation of Eurasia. EOS 82 (47), F315. Collins, A., Shagatin, K.L., Degtyarev, K.E., Levashova, N., Bazhenov, M.L., Van der Voo, R., 2002. Early Paleozoic island arc volcanism of the Boshekul-Chingiz Arc (Kazakhstan) and sedimentary provenance: Absence of a continental source. Geological Society of America Abstracts with Programs 34 (6), 341–342.

References

719

Condie, K.C., Sloan, R.E., 1998. Origin and Evolution of Earth—Principles of Historical Geology. Prentice Hall, Upper Saddle River, viii + 498 pp. Constenius, K.N., Johnson, R.A., Dickinson, W.R., Williams, T.A., 2000. Tectonic evolution of the Jurassic-Cretaceous Great Valley forearc, California: Implications for the Franciscan thrustwedge hypothesis. Geological Society of America Bulletin 112, 1703–1723. Daubrée, A., 1879. Études Synthetiques de Géologie Expérimentale. Dunod, Paris, III + 828 pp. Daubrée, A., 1888. Les Régions Invisibles du Globe et des Espaces Célestes—Eaux Souterraines, Tremblements de Terre, Météorites. Bibliothèque Scientifique Internationale LXVII. Félix Alcan, Paris, [I] + 202 pp. Davies, H.L., 1968. Papuan ultramafic belt. In: 23rd International Geological Congress 1, pp. 209– 220. de Wever, P., Popova, I., 1997. Cenozoic radiolaria from European Platform: a review. Geodiversitas 19, 383–469. de Wit, M.J., Ashwal, L.D. (Eds.), 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, xxvii + 809 pp. Demina, L.I., Lomize, M.G., Avdonin, A.B., 1995. Geodinamicheskie kharakteristiki peridotitov Prisongkelya (Severnyi Tyan Shan) Vestnik Moskovskogo Universiteta. Geologiya 1 (4), 91–100. Dewey, J.F., 1974. Continental margins and ophiolite obduction: Appalachian Caledonian System. In: Burk, C.A., Drake, C.L. (Eds.), The Geology of Continental Margins. Springer-Verlag, Berlin, pp. 933–950. Dewey, J.F., 1975. Finite plate evolution: some implications for the evolution of rock masses on plate margins. American Journal of Science 275-A, 260–284. Dewey, J.F., 1976. Ophiolite obduction. Tectonophysics 31, 93–120. Dewey, J.F., 2003. Ophiolites and lost oceans: rifts, ridges, arcs, and/or scrapings?. In: Dilek, Y., Newcomb, S. (Eds.), Ophiolite Concept and the Evolution of Geological Thought. Geological Society of America Special Paper 373, 153–158. Dewey, J.F., Bird, J.M., 1970. Mountain belts and the new global tectonics. Journal of Geophysical Research 75, 2625–2647. Dewey, J.F., Bird, J.M., 1971. Origin and emplacement of the ophiolite suite: Appalachian ophiolites in Newfoundland. Journal of Geophysical Research 76, 3179–3206. Dewey, J.F., Kidd, W.S.F., 1977. Geometry of plate accretion. Geological Society of America Bulletin 88, 960–968. Dewey, J.F., Pitman, W.C. III, Ryan, W.B.F., Bonnin, J., 1973. Plate tectonics and the evolution of the Alpine System. Geological Society of America Bulletin 84, 3137–3180. Dietz, R.S., 1961. Continent and ocean basin evolution by spreading of the sea-floor. Nature 190, 854–857. Dietz, R.S., 1963. Alpine serpentinites as oceanic rind fragments. Geological Society of America Bulletin 74, 942–957. Dilek, Y., Newcomb, S. (Eds.), 2003. Ophiolite Concept and the Evolution of Geological Thought. Geological Society of America Special Paper 373, xii + 504 pp. + 1 foldout plate. Dobretsov, N.L., Berzin, N.A., Buslov, M.M., 1995. Opening and tectonic evolution of the PaleoAsian Ocean. International Geology Review 37, 335–360. Dobretsov, N.L., Buslov, M.M., in press. Fragments of oceanic islands in accretion-collision areas of Gorny Altai and Salair, southern Siberia, Russia: early stages of continental crust growth of the Siberian continent in Vendian-Early Cambrian time. Journal of Asian Earth Sciences.

720

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Ernst, W.G., 1970. Tectonic contact between the Franciscan mélange and the Great Valley sequence—crustal expression of a late Mesozoic benioff zone. Journal of Geophysical Research 75, 886–902. Evarts, R.C., 1977. The geology and petrology of the Del Puerto ophiolite, Diablo Range, Central California Coast Ranges. In: Coleman, R.G., Irwin, W.P. (Eds.), North American Ophiolites. State of Oregon, Department of Geology and Mineral Industries. Bulletin 95, 121–139. Gass, I.G., Masson-Smith, D., 1963. The geology and gravity anomalies of the Troodos massif, Cyprus. Philosophical Transactions of the Royal Society of London A 255, 417–467. Gass, I.G., Smewing, J.D., 1973. Intrusion, extrusion and metamorphism at constructive margins: evidence from the Troodos massif, Cyprus. Nature 242, 26–29. Gansser, A., 1955. New aspects of the geology in central Iran. In: Proceedings of the Fourth World Petroleum Congress, Section I/A/5, Paper 2, pp. 279–300. Gansser, A., 1964. Geology of the Himalayas. Wiley–Interscience, New York, p. 289. Gansser, A., 1966. The Indian Ocean and the Himalayas: a geological interpretation. Eclogae Geologicae Helvetiae 59, 831–848. Gordienko, I.V., 1987. Paleozoysky magmatizm i geodinamika Tzentralno-Aziatskogo skladchatogo poyasa. Nauka, Moscow, p. 238. Greenbaum, D., 1972. Magmatic processes at ocean ridges: evidence from the Troodos massif, Cyprus. Nature; Physical Science 238, 18–21. Greenbaum, D., 1977. The chromitiferous rocks of the Troodos ophiolite complex, Cyprus. Economic Geology 72, 1175–1194. Gusev, G.S., Khain, V.E., 1995. O sootnosheniyakh Baikalo-Vitimskogo, Aldano-Stanovogo i Mongolo-Okhotskogo terrreinov (yug srednei Sibiri). Geotektonika (5), 68–82. Gusev, G.S., Peskov, A.I., 1993. Novye dannye o rifeiskikh ofiolitakh Prishilkinskogo segmenta Mongolo-Okhotskogo shva. Doklady Akademii Nauk 333 (2), 220–223. Gusev, G.S., Peskov, A.I., 1996. Geokhimiya i usloviyaobrazovaniya ofiolitov vostochnogo Zabaikalya. Geokhimiya (8), 723–737. Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions model and animations. Journal of Asian Earth Sciences 20, 353– 431. Hamilton, W., 1969. Mesozoic California and the underflow of Pacific mantle. Geological Society of America Bulletin 80, 2409–2430. Hamilton, W., 1970. The Uralides and the motion of the Russian and Siberian platforms. Geological Society of America Bulletin 81, 2553–2576. Hamilton, W., 1992. Evolution of the Archaean mantle and crust. In: The Geology of North America, v. C-2: Precambrian: Conterminous U.S. Geological Society of America, Boulder, pp. 597–636. Hamilton, W., 1998. Archean tectonics and magmatism. In: Ernst, W.G., Nelson, C.A. (Eds.), Integrated Earth and Environmental Evolution of the Southwestern United States, pp. 5–43 (The Clarence A. Hall Jr. volume). Bellweather Publishing, for the Geological Society of America, Columbia. Hess, H.H., 1939. Island arcs, gravity anomalies and serpentinite intrusions. In: 17th International Geological Congress, Moscow, vol. 2, pp. 263–283. Hess, H.H., 1955. Serpentines, orogeny, and epeirogeny. In: Poldervart, A. (Ed.), Crust of the Earth (A Symposium). Geological Society of America Special Paper 62, 391–408. Hess, H.H., 1962. History of ocean basins. In: Engel, A.E.G., et al. (Eds.), Petrological Studies: A Volume in Honor of A.F. Buddington. Geological Society of America, pp. 599–620.

References

721

Hill, M.N., 1957. Recent geophysical exploration of the ocean floor. Physics and Chemistry of the Earth 2, 129–163. Hsü, K.J., 1971. Franciscan mélanges as a model for eugeosynclinal sedimentation and understhrusting tectonics. Journal of Geophysical Research 76, 1162–1170. Hsü, K.J., Wang, Q., Hao, J., 1991. Geologic evolution of the Naimonides: A working hypothesis. Eclogae Geologicae Helvetiae 84, 1–31. Hu, A., Jahn, B.-M., Zhang, G., Chen, Y., Zhang, Q., 2000. Crustal evolution and Phanerozoic crustal growth in northern Xinjiang: Nd isotopic evidence, Part I. Isotopic characterization of basement rocks. Tectonophysics 328 (1–2), 15–51. Huang, X., Jin, C.W., Sun, B.S., Jun, Y.B., Zhang, R.H., 1997. Study on the age of Aermantai ophiolite, Xinjiang by Nd-Sr isotope geology (in Chinese). Acta Petrologica Sinica 13, 85–91. Izokh, A.E., Vladimirov, A.G., Stupakov, C.I., 1988. Magmatizm Agardagskoi shovnoi zony. In: Solovyev, V.A. (Ed.), Geologo-Petrologicheskie Issledovaniya Yugo-Vostochnoj Tuvy. IGIG SO AN SSSR, Novosibirsk, pp. 19–74. Jahn, B.-M., 2002. Generation of the juvenile crust in the Central Asian Orogenic Belt. In: Wu, F.Y., Wilde, S.A., Jahn, B.-M. (Eds.), IGCP-420 Continental Growth in the Phanerozoic: Evidence from Central Asia. Fourth Workshop, Abstracts and Guidebook, Changchun, pp. 57–68. Jarrard, R.D., 1986. Relations among subduction parameters. Reviews of Geophysics 24, 217–284. Karson, J., Dewey, J.F., 1978. Coastal Complex, western Newfoundland: an early Ordovician oceanic fracture zone. Geological Society of America Bulletin 89, 1037–1049. Kepezhinskas, P.K., Kepezhinskas, K.B., Pukhtel, I.S., 1991. Lower Paleozoic oceanic crust in Mongolian Caledonides: Sm-Nd isotope and trace element data. Geophysical Research Letters 18, 1301–1304. Khain, E.V., Bibikova, E.V., Salnikova, E.B., Kröner, A., Gibsher, A.S., Didenko, A.N., Degtyarev, K.E., Fedotova, A.A., 2003. The Palaeo-Asian ocean in the Neoproterozoic and early Palaeozoic. New geochronologic data and palaeotectonic reconstructions. Precambrian Research 122, 329– 358. Khromykh, B.F., 1986. Novye dannye o vend-paleozoiskoi evolyutsii i metallogenii Boshchekulskogo Rudnogo Rayona. Izvestiya Akademii Nauk Kazakhskoy SSR (6), 20–34. Kober, L., 1921. Der Bau der Erde. Gebrüder Borntraeger, Berlin, II + 324 pp. Kröner, A., Tomurtogoo, O., Badarch, G., Windley, B.F., Kozakov, I.K., 2001. New zircon ages and significanse for crustal evolution in Mongolia. In: Sklyarov, E.K. (Ed.), Assembly and Breakup of Rodinia Supercontinent. Evidence from South Siberia, Irkutsk, pp. 142–145. Kropotkin, P.N., 1972. Eurasia as a composite continent. Transactions of the American Geophysics Institute 53, 80. Kungurtsev, L.V., 1993. Geodynamic complexes and formation of the northern structure of AltaiSayan area, Forth International symposium on geodynamic evolution of Paleoasian ocean. IGCP Project 283, Report No. 4, Novosibirsk, pp. 93–96. Kusky, T., Polat, A., 1999. Growth of granite-greenstone terranes at convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73. Kuzmin, M.I., Gordienko, I.V., Almukhamedov, A.I., Antipin, V.S., Bayanov, V.D., Filimonov, A.D., 1995. Paleookeanicheskie kompleksy Dzhidinskoi zony kaledonid. Geologiya i Geofizika 36 (1), 3–18. Lamb, M.A., Badarch, G., 1997. Paleozoic sedimentary basins and volcanic-arc systems of Southern Mongolia: New stratigraphic and sedimentologic constraints. International Geology Review 39, 542–576.

722

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Lemoine, M., 1980. Serpentinites, gabbros, and ophicalcites in the Piemont-Ligurian domain of the Western Alps: possible indicators of oceanic fracture zones and of associated serpentinites protrusions in the Jurassic-Cretaceous Tethys. Archives des Sciences 33, 103–115. Liu, W., Zhang, X.B., 1993. The characteristics and geological significance of the UlungurZhaisangpo tectonic melange belt. In: Tu, G.Z. (Ed.), New Improvement of Solid Geosciences in Northern Xinjuang. Science Press, Beijing, pp. 217–228. Lomize, M.G., Demina, L.I., Zarschikov, A.A., 1997. Kirgizsko-Terskeiskii paleookeanicheskii bassein (Tyan Shan) [Kirgiz-Tersk paleooceanic basin (Tien Shan)]. Geotektonika 6, 35–55. Mann, P., Draper, G., Lewis, J.F., 1991. An overview of the geologic and tectonic development of Hispaniola. In: Mann, P., Draper, G., Lewis, J.F. (Eds.), Geologic and Tectonic Development of the North America-Caribbean Plate Boundary in Hispaniola. Geological Society of America Special Paper 262, 1–28. Marinov, N.A., Zonenshain, L.P., Blagonravov, V.A., 1973. Geologiya Mongolskoi Narodnoi Respubliki. Tom 1. Stratigrafiya. Nedra, Moscow, 584 pp. McCall, G.J.H., Kidd, G.G.W., 1982. The Makran, southeastern Iran: the anatomy of a convergent plate margin active from Cretaceous to present. In: Leggett, J.K. (Ed.), Trench-Forearc Geology. Geological Society of London Special Publication 10, 387–397. Miyashiro, A., 1973. The Troodos ophiolite complex was probably formed in an island arc. Earth and Planetary Science Letters 19, 218–224. Moores, E.M., 1998. Opholites, the Sierra Nevada, “Cordilleria,” and orogeny along the Pacific and Caibbean margins of North and South America. In: Ernst, W.G., Nelson, C.A. (Eds.), Integrated Earth and Environmental Evolution of the Southwestern United States, pp. 44–58 (The Clarence A. Hall Jr., volume). Bellweather Publishing, for the Geological Society of America, Columbia. Natal’in, B.A., Sengör, ¸ A.M.C., 1996. Arc-slicing and arc-shaving strike-slip faults in the Palaeozoic collage of the Altaids, Central Asia. Geological Society of America Abstracts with Programs 28 (7), A-67–A-68. Nozaka, T., Liu, Y., 2002. Petrology of the Hegenshan ophiolite and its implication for the tectonic evolution of northern China. Earth and Planetary Science Letters 202, 89–104. Obrutschew, W.A., 1926. Geologie von Sibirien. In: Soergel, W. (Ed.), Fortschritte der Geologie und Paläontologie, No. 15. Gebrüder Borntraeger, Berlin, XI + 572 pp. + 1 p. of errata + 1 folding map and 10 plates in back pocket. Parfenov, M.L., Bulgatov, A.N., Gordienko, I.V., 1995. Terranes and accretionary history of the Transbaikal orogenic belt. International Geology Review 37, 736–751. Parfenov, M.L., Popeko, L.I., Tomurtogoo, O., 1999. Problemy tektoniki Mongolo-Okhotskogo orogennogo poyasa. Tikhookeanskaya Geologiya 18 (5), 24–43. Pearce, J.A., Lippard, S.J., Roberts, S., 1984. Characteristics and tectonic significance of suprasubduction zone ophiolites. Geological Society of London Special Publication 16, 77–94. Peive, A.V., 1969. Okanicheskaya kora geologicheskogo proshlogo. Geotektonika (4), 5–24. Peive, A.V., Mossakovskiy, A.A., 1982. Tektonika Kazakhstana (Tectonics of Kazakhstan), Explanatory note to the tectonic map of Eastern Kazakhstan in scale 1:2,500,000. Nauka, Moscow, p. 138. Peive, A.V., Streis, N.A., Mossakovsky, A.A., Perfiliev, A.S., Ruzhentsev, S.B., Bogdanov, N.A., Burtman, V.S., Makarychev, G.I., Markov, M.S., Suvorov, A.I., 1972. Paleozoidy Evrazii i nekotorye voprosi evolutsii geosinklinalnogo protsessa. Sovyetskaya Geologia (1), 7–25. Penrose field conference participants, 1972. Ophiolites. Geotimes 17 (12), 24–25. Pfänder, J.A., Jochum, K.P., Kozakov, I., Kröner, A., Todt, W., 2002. Coupled evolution of back-arc and island arc-like mafic crust in the late-Neoproterozoic Agardagh Tes-Chem ophiolite, Central

References

723

Asia: evidence from trace element and Sr-Nd-Pb isotope data. Contributions to Mineralogy and Petrology 143, 154–174. Plafker, G., Moore, J.C., Winkler, G.R., 1994. Geology of the southern Alaska margin. In: Plafker, G., Berg, H.C. (Eds.), The Geology of North America, G-1, The Geology of Alaska. Geological Society of America, Boulder, pp. 389–449. Polat, A., Kerrich, R., 1999. Formation of an Archean tectonic melange in the Schreiber-Hemlo greenstone belt, Superior Province, Canada: Implications for Archean subduction-accretion processes. Tectonics 18, 733–755. Polat, A., Kerrich, R., 2001. Magnesian andesites, Nb-enriched basalt-andesites, and adakites from late Archean 2.7 Ga Wawa greenstone belts, Superior Province, Canada: Implications for late Archean subduction zone petrogenetic processes. Contribution to Mineralogy and Petrology 141, 36–52. Polat, A., Hofmann, A.W., Rosing, M., 2002. Boninite-like volcanic rocks in the 3.7–3.8 Ga Isua greenstone belt, West Greenland: Geochemical evidence for intra-oceanic subduction zone processes in the early Earth. Chemical Geology 184, 231–254. Reinhardt, B.M., 1969. On the genesis and emplacement of ophiolites in the Oman Mountains Geosyncline. Schweizerische Mineralogische und Petrographische Mitteilungen 49, 1–30. Ricou, L.-E., 1971. Le croissant ophiolitique péri-arabe: une ceinture de nappes mises en place en Crétace supérieur. Revue de Geographie Physique et de Geologie Dynamique 13, 327–349. Robinson, P.T., Zhou, M.F., Hu, X.F., Reynolds, P., Bai, W.J., Yang, J.S., 1999. Geochemical constraints on the origin of the Hegenshan Ophiolite, Inner Mongolia, China. Journal of Asian Earth Sciences 17, 423–442. Rutland, R.W.R., 1968. A tectonic study of part of the Philippine Fault zone. Quarterly Journal of the Geological Society of London 123, 293–325. Ruzhentsev, S.V., Badarch, G.A., Voznesenskaya, T.A., Sharkova, T.T., 1987. Formatsii i struktura varictsid Yuzhnoi Mongolii. In: Pushcharovsky, Y.M., Khvorova, I.V. (Eds.), Rennegeosinklinalnye Formatsii i Struktury. Nauka, Moscow, pp. 101–136. Ruzhentsev, S.V., Badarch, G., Voznesenskaya, T.A., 1985. Tectonics of the Trans-Altay zone of Mongolia. Geotectonics 19, 276–284. Ruzhentsev, S.V., Pospelov, I.I., 1992. Yuzhno-Mongolskaya varisskaya skladchataya sistema. Geotektonika, 45–62. Saleeby, J., 1977. Fracture zone tectonics, continental margin fragmentation, and emplacement of the Kings-Kaweah ophiolite belt, southwest Sierra Nevada, California. In: Coleman, R.G., Irwin, W.P. (Eds.), North American Ophiolites. State of Oregon, Department of Geology and Mineral Industries. Bulletin 95, 141–159. Saleeby, J., 1981. Ocean floor accretion and volcanoplutonic arc evolution of the Mesozoic Sierra Nevada. In: Ernst, W.G. (Ed.), The Geotectonic Development of California, Rubey Volume I. Prentice–Hall, Englewood Cliffs, NJ, pp. 132–181. Saleeby, J.B., Goodin, S.E., Sharp, W.D., Busby, C.J., 1978. Early Mesozoic paleotectonicpaleogeographic reconstruction of the southern Sierra Nevada region. In: Howell, D.G., McDougall, K.A. (Eds.), Mesozoic Paleogeography of the Western United States. Pacific Coast Paleogeography Symposium 2, The Pacific Section. SEPM, pp. 311–336. Salvador, A. (Ed.), 1994. International Stratigraphic Guide—A Guide to Stratigraphic Classification, Terminology, and Procedure. International Union of Geological Sciences. Geological Society of America, Boulder, xix + 214 pp. Saunders, A.D., Tarney, J., Marsh, N.G., Wood, D.A., 1980. Ophiolites as ocean crust or marginal basin crust: A geochemical approach. In: Panayiotou, A. (Ed.), Ophiolites—Proceedings of the

724

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

International Ophiolite Symposium, Cyprus, 1979. Ministry of Agriculture and Natural Resources Geological Survey Department, Republic of Cyprus, pp. 193–204. Schardt, H., 1893. Sur l’origine des Alpes du Chablais et du Stockhorn, en Savoie et en Suisse. Comptes-Rendus hébdomadaires de l’Academie des Sciences, Paris 117, 707–709, errata on p. 874. Sengör, ¸ A.M.C., 1982a. The classical theories of orogenesis. In: Miyashiro, A., Aki, K., Sengör, ¸ A.M.C. (Eds.), Orogeny. Wiley, Chichester, pp. 1–48. Sengör, ¸ A.M.C., 1982b. Eduard Suess’ relations to the pre-1950 schools of thought in global tectonics. Geologische Rundschau 71, 381–420. Sengör, ¸ A.M.C., 1982c. A note on a late revision of the theory of embryotectonics by Argand himself. Eclogae Geologicae Helvetiae 75, 177–188. Sengör, ¸ A.M.C., 1984. The Cimmeride orogenic system and the tectonics of Euraisa. Geological Society of America Special Paper 195, xi + 82 pp. Sengör, ¸ A.M.C., 1990. Plate tectonics and orogenic research after 25 years: A Tethyan perspective. Earth Science Reviews 27, 1–201. Sengör, ¸ A.M.C., 1992. The Palaeo-Tethyan suture: A line of demarcation between two fundamentally different architectural styles in the structure of Asia. The Island Arc 1, 78–91. Sengör, ¸ A.M.C., 1993. Turkic-type orogeny in the Altaids: Implications for the evolution of continental crust and methodology of regional tectonic analysis (the 34th Bennett Lecture). Transactions of the Leicester Literary and Philosophical Society 87, 37–54. Sengör, ¸ A.M.C., 1998. Die Tethys: vor hundert Jahren und heute. Mitt. Österr. Geol. Ges. 89, 5–176. Sengör, ¸ A.M.C., 2003. The repeated rediscovery of mélanges and its implications for the possibility and the rôle of objective evidence in the scientific enterprise. In: Dilek, Y., Newcomb, S. (Eds.), Ophiolite Concept and the Evolution of Geological Thought. Geological Society of America Special Paper 373, 385–445 + 1 foldout plate. Sengör, ¸ A.M.C., Natal’in, B.A., 1996a. Palaeotectonics of Asia: Fragments of A Synthesis. In: Yin, A., Harrison, M. (Eds.), The Tectonic Evolution of Asia, Rubey Colloquium. Cambridge Univ. Press, Cambridge, pp. 486–640. Sengör, ¸ A.M.C., Natal’in, B.A., 1996b. Turkic-type orogeny and its role in the making of the continental crust. Annual Review of Earth and Planetary Sciences 24, 263–337. Sengör, ¸ A.M.C., Okuro˘gulları, A.H., 1991. The role of accretionary wedges in the growth of continents: Asiatic examples from Argand to plate tectonics. Eclogae Geologicae Helvetiae 84, 535– 597. Sengör, ¸ A.M.C., Natal’in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectonic collage and Palaeozoic crustal growth in Eurasia. Nature 364, 299–307. Sengör, ¸ A.M.C., Niemi, N., McQuarrie, N., Friedrich, A.M., Wernicke, B., 2001. A large, subduction-generated syntaxis in the Gondwanides of southern Gondwana-Land. Geological Society of America Abstracts with Programs 33 (6), A-397. Sieh, K., Natawidjaja, D., 2000. Neotectonics of the Sumatran fault, Indonesia. Journal Geophysical Research 105, 28,295–28,326. Simonian, K.O., Gass, I.G., 1978. The Arakapas fault belt, Cyprus: a fossil transform fault? Geological Society of America Bulletin 89, 1220–1230. Simonov, V.A., Dobretsov, N.L., Buslov, M.M., 1994. Boninitovye serii v strukturakh paleoaziatskogo okeana. Geologiya i Geofizika 35 (7–8), 182–199. Sleep, N.H., 1975. Formation of oceanic crust, some thermal constraints. Journal of Geophysical Research 80, 4037–4042.

References

725

Spiridonov, E.M., 1991. Kaledonskie magmaticheskie kompleksy, nekotorye aspekty evolyutsii zemnoi kory i metallogenii severnogo Kazakhstana. In: Abdulin, A.A. (Ed.), Magmatizm i Rudonosnost Kazakhstana. Gylym, Alma-Ata, pp. 114–123. Staub, R., 1924. Der Bau der Alpen—Versuch einer Synthese. In: Beiträge zur Geologischen Karte der Schweiz, Neue Folge, vol. 52. Lieferung, VI + 272 pp. + 7 plates. Staub, R., 1928. Der Bewegungsmechanismus der Erde dargelegt am Bau der irdischen Gebirgssysteme. Gebrüder Borntraeger, Berlin, VIII + 270 pp. + 1 plate in pocket. Steinmann, G., 1905. Geologische Beobachtungen in den Alpen II. Die SCHARDT’sche Überfaltungstheorie und die geologische Bedeutung der Tiefseeabsätze und der ophiolitischen Massengesteine. Berichten der Naturforschenden Gesellschaft zu Freiburg i.B. 16, 18–67. Steinmann, G., 1927. Die ophiolitischen Zonen in den Mediterranen Kettengebirgen. In: XIVe Congrés Géologique International (1926, Espagne), 4. Fscl. Graficas Reunida S.A., Madrid, pp. 637– 667. Stille, H., 1936. Die Entwicklung des amerikanischen Kordillerensystems in Zeit und Raum. Preussische Akad. Wiss. Phys.-Math. Kl. 15, 134–155. Stille, H., 1939. Bemerkungen betreffend die “sardische” Faltung und den Ausdruck “ophiolitisch”. Zeitschrift der Deutschen Geologischen Gesellschaft 91, 771–773. Stille, H., 1940. Einführung in den Bau Amerikas. Gebrüder Borntraeger, Berlin, XX + 717 pp. Stille, H., 1958. Die Assyntische Tektonik im Geologischen Erdbild. Bh. Geol. Jb. 22, 255 pp. + 3 folded plates. Stöcklin, J., 1968. Structural history and tectonics of Iran: a review. Bulletin of the American Association of Petroleum Geologists 52, 1229–1258. Suess, E., 1888. Das Antlitz der Erde, vol. II. F. Tempsky/G. Freytag, Prag/Wien/Leipzig. IV + 704 pp. Suess, E., 1901. Das Antlitz der Erde, vol. III/1 (Dritter Band. Erste Hälfte). F. Tempsky/G. Freytag, Prag/Wien/Leipzig. IV + 508 pp. Suess, E., 1904. Sur la nature des charriages. Comptes Rendus hébdomadaires de l’Academie des Sciences 139, 714–716. Suess, E., 1909. Das Antlitz der Erde, vol. III/2. F. Tempsky, Wien, IV + 789 pp. Suess, F.E., 1937. Bausteine zu einem System der Tektogenese, I. Periplutonische und enorogene Regionalmetamorphose in ihrer tektogenetischen Bedeutung. Fortschr. Geol. Paläont. 13, 1–86. Taylor, B., Karner, G.D., 1983. On the evolution of marginal basins. Reviews of Geophysics and Space Physics 21, 1727–1741. Termier, P., 1903. Les nappes des Alpes Orientales et la synthèse des Alpes. Bulletin de la Société Géologique de France, Série 4 (3), 711–765 + 2 foldout plates. Tomurtogoo, O., 1997. A new tectonic scheme of the Paleozoides in Mongolia. Geoscientist (3), 12–19. Trümpy, R., 1975. Penninic-Austroalpine boundary in the Swiss Alps: a presumed former continental margin and its problems. American Journal of Science 275-A, 209–238. Trümpy, R., 2003. Trying to understand Alpine sediments—before 1950. Earth Science Reviews 61, 19–42. Van der Voo, R., Degtyarev, K.E., Collins, A., Levashova, N., Bazhenov, M.L., 2002. The Kazakhstan orocline (C. Asia), Part II: Evidence for relative rotations. Geological Society of America Abstracts with Programs 34 (6), 491–492. Volkov, V.V., 1988. Nezhnepaleozoiskii Vulkanizm Tsentralnoi Azii [Lower Paleozoic Magmatism of Central Asia]. Nauka, Novosibirsk, p. 194.

726

Chapter 21: Phanerozoic Analogues of Archaean Oceanic Basement Fragments

Wang, Q., Liu, X., 1986. Paleoplate tectonics between Cathaysia and Angaraland in Inner Mongolia. Tectonics 5, 1073–1088. Wang, Q., Liu, X., 1991. Pre-Jurassic tectonic evolution between Cathaysia and Angaraland. In: Ishii, K., Liu, X., Ichikawa, K., Huang, B. (Eds.), Pre-Jurassic Geology of Inner Mongolia, China. Report of China-Japan Cooperative Research Group, 1987–1989. Osaka City University, Osaka, pp. 127–148. Wang, Z., Sun, S., Li, J.L., Hou, Q.L., Qin, K.H., Xiao, W.J., Hao, J., in press. Paleozoic tectonic evolution of the northern Xinjiang, China: geochemical and geochronological constrains from the ophiolites. Tectonics. Wilson, J.T., 1949. The origin of continents and Precambrian history. Transactions Royal Society of Canada 43 (Ser. 3, Sect. 4), 157–184. Woodcock, N.H., 1986. The role of strike-slip fault systems at plate boundaries. Philosophical Transactions of the Royal Society of London A 317, 13–29. Yakubchuk, A.S., Degtyarev, K.E., 1991. O kharaktere sochleneniya Chingizskogo i Boshchekulskogo napravlenii v kaledonidakh severo-vostoka Tsentralnogo Kazakhstana. Doklady Akademii Nauk SSSR 317, 957–962. Yilmaz, Y., Sengör, ¸ A.M.C., 1985. Paleo-Tethyan ophiolites in northern Turkey: Petrology and tectonic setting. Ofioliti 10, 485–504. Zhang, Z.M., Liou, J.G., Coleman, R.G., 1984. An outline of the plate tectonics of China. Bulletin of the Geological Society of America 95, 295–312. Zonenshain, L.P., 1972. Uchenie o Geosinklinalyakh i Ego Prilozhenie k Tsentralno-Aziatskomu Skladchatomu Poyasu. Nedra, Moscow, p. 240. Zonenshain, L.P., 1973. The evolution of central Asiatic geosynclines through sea-floor spreading. Tectonophysics 19, 213–232. Zonenshain, L.P., Kuzmin, M.I., 1978. Khan-Taishirskii ofiolitovyi kompleks zapadnoi Mongolii i problemy ofiolitov. Geotektonika (1), 19–42. Zorin, Y.A., 1999. Geodynamics of the western part of the Mongolia-Okhotsk collisional belt, TransBaikal region (Russia) and Mongolia. Tectonophysics 306, 33–56.

Precambrian Ophiolites and Related Rocks Edited by Timothy M. Kusky Developments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) © 2004 Elsevier B.V. All rights reserved.

727

Chapter 22

EPILOGUE: WHAT IF ANYTHING HAVE WE LEARNED ABOUT PRECAMBRIAN OPHIOLITES AND EARLY EARTH PROCESSES? TIMOTHY M. KUSKY Department of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63103, USA

The chapters in this book have presented clear, even unequivocal evidence that Precambrian ophiolites are preserved in many Precambrian terranes. Proterozoic examples are abundant, especially in the Arabian Nubian Shield, where ophiolites have been recognized for many years. Archean examples are more controversial but a number of excellent examples of whole, dismembered, and metamorphosed ophiolites are described in this volume. In this brief epilogue, we assess what, if anything, we have learned about the early Earth from the identification of specific sequences as ophiolitic. In addition, we present a new list of criteria to help discriminate between ophiolitic and other sequences. The recognition that many of the allochthonous mafic/ultramafic complexes in Archean and Proterozoic greenstone belts are ophiolites provides researchers with a much longer record of oceanic processes than the record from Phanerozoic ophiolites alone. From this record we are able to deduce that the classical Penrose model (Anonymous, 1972) for the structure of ophiolitic lithosphere is too simplistic to explain the great variations found in ophiolites over this greater sample of time. The Penrose model for ophiolite stratigraphy is too restrictive to explain even present day sea floor and Paleozoic ophiolites, which all show much greater diversity (related to spreading rate, temperature, magma supply, etc.). Since modern environments and young ophiolites rarely conform to this strict definition, it makes little sense for Precambrian ophiolites to be held to this standard for recognition. It is more sensible to allow the diversity of modern ophiolites to be a guide to recognizing older ophiolites and their tectonic settings, and then to try to determine, through comparison, if there are any significant secular changes in ophiolitic structure and stratigraphy with time. With this caveat in mind, the chapters in this book have identified dozens of Precambrian ophiolites that contain an ophiolitic igneous stratigraphy. This basic recognition opens the way for a myriad of other studies on the chemistry, structure, thickness, rheology, biology, and other aspects of ancient oceanic crust and lithosphere that are only beginning to be appreciated. Once this recognition becomes more widespread and accepted, even greater insight to processes on the early Earth will be obtained. DOI: 10.1016/S0166-2635(04)13022-3

728

Chapter 22: Epilogue

1. KOMATIITES, BONINITES, BIF’S, AND PODIFORM CHROMITES It has long been held that komatiites are abundant in Archean greenstone belts and that the Archean oceanic crust may have been dominantly komatiitic, reflecting early higher mantle temperatures. However, komatiites are much less common than many workers originally thought, and they do not necessarily mean much hotter mantle (see Parman and Grove, 2004). There has been a disproportionate number of studies of komatiites from Archean greenstone belts compared to other rock types, because petrologists have focused on the unusual aspects of these rocks, but they typically do not form more than a few percent of any greenstone terrain. However, they do appear to be more abundant in Archean terrains than younger ophiolites (e.g., Alvarado et al., 1997). Boninites are geochemically distinct mafic rocks that have been suggested to be absent from Archean terrains. As reported in several chapters in this volume (see Polat and Kerrich, 2004; Shchipansky et al., 2004; Stern et al., 2004), boninites have now been identified in several ophiolitic Proterozoic and Archean greenstone belts extending back in time to the 3.8 Ga Isua belt, suggesting that these ophiolites formed in environments similar to their modern counterparts. Boninites of Phanerozoic age occur in ophiolites or intraoceanic island arcs, such as the Izu-Bonin-Mariana arc system. These primary liquids are interpreted as second-stage high-temperature, low-pressure melting of a depleted refractory mantle wedge fertilized by fluids and/or melts, above a subduction zone. Precambrian boninitic lavas are likely products of the same conjunction of processes, suggesting that mantle melting processes above subducting slabs was broadly similar in the Archean to that of today. Podiform chromites form very distinctive deposits in many Phanerozoic ophiolites, and have been found in a few places on the modern sea-floor. Podiform chromites form small clusters of typically orbicular and nodular textured chromite in dunite pods, enclosed within mantle harzburgite tectonite. These chromite pods are distinctive, both physically and chemically, from layered chromite of layered ultramafic intrusive complex in continents (such as the Bushveld) and arcs (see Lago et al., 1982; Nicolas and Azri, 1991; Leblanc and Nicolas, 1992; Stowe, 1994; Butcher et al., 1999; Edwards et al., 2000). Until recently, podiform chromites were not known from any Archean greenstone belts, but their documentation in the Zunhua ophiolitic mélange and Dongwanzi ophiolite of North China (Kusky et al., 2004a; Huang et al., 2004) shows clearly not only that these rocks are ophiolitic, but that mantle melting processes in the Archean were similar to those of younger times. We suggest that since podiform chromites are only known from ophiolites, that they are as distinctive for recognizing a rock sequence as an ophiolite as the presence of the entire Penrose sequence. Banded Iron Formations (BIF’s) are a major component of many Archean greenstone terranes, and are described from several of the ophiolitic sequences in this volume. While the origin of BIF’s has been controversial, and there are several different origins (e.g., Fowler et al., 2002; Coward and Ries, 1995; Simonson, 1985), Hofmann and Kusky (2004) have shown how BIF’s in low-grade greenstone terranes may mark sites of regional structural detachment, with iron and sulfide mineralization focused along early shear zones.

2. Transitional Ophiolites

729

Workers in other greenstone terranes, particularly those that are more highly deformed and metamorphosed, should note the relationships at Belingwe, and re-assess whether or not BIF’s in other greenstones and Precambrian ophiolite terranes may mark the sites of major regional detachment and displacement. Several authors (e.g., Bickle et al., 1994; Hamilton, 2003) have noted that some Precambrian greenstone belts show evidence of contamination by continental type material, and have then suggested that this means that they cannot be fragments of oceanic crust and lithosphere. These authors have failed to note that many modern and Phanerozoic ophiolites also show such contamination (e.g., Moores, 2002), invalidating those arguments. Nonetheless, apparent contamination by continental crustal material presents interesting constraints on the origin of these ophiolites. For instance, apparent crustal contamination can mean lavas were derived from unusual mantle, such as an older forearc environment, where subduction-related processes may have depleted the mantle leading to unusual, apparently contaminated geochemical signatures (see Parman and Grove, 2004). Alternatively, some ophiolites may be truly contaminated, having formed near a stretched continental margin. Some ophiolites seem to preserve magmatism near these margins, and some even have subcontinental lithospheric mantle and/or crust preserved. We coin a new term for these ophiolites, and call them transitional ophiolites.

2. TRANSITIONAL OPHIOLITES Several of the ophiolites described in this volume appear to have formed within the transition from rifted continental margins to ocean spreading centers during early stages of ocean opening, then were structurally detached and/or deformed and incorporated into convergent margins during ocean closure. These ophiolites are distinctive from classical Penrose-style ophiolites and others formed in forearc and back arc environments. During early stages of ocean formation, continental crust and lherzolite of the subcontinental mantle is extended forming graben on the surface, and ductile mylonites at depth. Sedimentary basins may form in the graben, and as the extension continues magmatism sometimes affects the rifted margin, either forming volcanic rifted margins, or migrating to a spreading center forming a oceanic spreading center. New asthenospheric mantle upwells along the new ridge, and may intrude beneath the extended continental crust. In some cases, wedges of extended mid-to-lower continental crust overlying mylonitic lherzolitic subcontinental mantle become intruded by numerous dikes and magmas from this new asthenospheric mantle. In this case, magmas may pool both above and below the stretched continental crust, forming mafic/ultramafic cumulates in igneous contact with older continental crust (see Fig. 1 in the Introduction to this volume). Dikes from these magma chambers may then feed a crustal gabbroic magma chamber closer to the surface, which in turn may feed a dike complex and basaltic pillow/massive lava section. If preserved, this unusual sequence forms what we term a “transitional ophiolite”, grading down from subaquatic sediments, to pillow lavas, dikes, sheeted dikes, layered gabbro, dunite and pyroxenite cumulates, then remarkably into stretched, typically mylonitic granitic mylonites, underlain by lherzolite.

730

Chapter 22: Epilogue

The lherzolite tectonic may be underlain by harzburgite tectonite or harzburgite. Recognition of this relationship represents a major advance in understanding some of the ophiolitic complexes described in this volume, and elsewhere. Examples of this type of transitional ophiolite are found in the Proterozoic Jourma complex, and in some of the Slave Province ophiolites (see papers by Peltonen and Kontinen, 2004, and Corcoran et al., 2004). Modern analogs for such transitional ophiolites are found around the Red Sea, including at Tihama Asir, Saudi Arabia, where a 5–10 Ma old transitional ophiolite has a dike complex overlying layered gabbro, which in turn overlies continental crust. Also, on Egypt’s Zabargad Island, oceanic mantle is exposed, and it is likely that the crustal structure near this region preserves transitional ophiolites as well. The main lesson here is that ophiolites may form in many tectonic settings, from extended continental crust, to mid ocean ridges, to forearcs, arcs, back arcs, to triple junctions along convergent margins.

3. PROTEROZOIC OPHIOLITES The formation of the Gondwanan supercontinent at the end of the Precambrian and the dawn of the Phanerozoic represents one of the most fundamental problems being studied in Earth Sciences today. It links many different fields, and there are currently numerous and rapid changes in our understanding of events related to the assembly of Gondwana. One of the most fundamental and most poorly understood aspects of the formation of Gondwana is the timing and geometry of closure of the oceanic basins which separated the continental fragments that amassed to form the Late Proterozoic supercontinent. Final collision between East and West Gondwana most likely occurred during closure of the Mozambique Ocean, forming the East African Orogen including the Arabian-Nubian Shield. Neoproterozoic ophiolite fragments have been recognized as a component part of many nappe complexes associated with sutures in the Arabian-Nubian Shield. The recognition of these ophiolite-decorated sutures played a major role in understanding the formation of the Arabian-Nubian Shield as an amalgam of different arc and microcontinental terranes that collided during the closure of the Mozambique Ocean (Stern, 1994; Kusky et al., 2003), but also contributed to many scientists’ acceptance that plate tectonics extended back in time to 890 Ma, the age of the oldest Arabian ophiolite (see Stern et al., 2004; Johnson et al., 2004). The chemistry of the Arabian Shield ophiolites include both tholeiitic and calc-alkaline varieties, with minor boninites, suggesting that they largely formed in a forearc environment, with extensive partial melting of the mantle. Many of the ArabianNubian shield ophiolites formed over a critical interval of Earth history that saw many changes in the Earth’s biota and climate, yet very few studies have yet been aimed at the sedimentary sequences that overlie these ophiolites, potentially preserving a treasure drove of information about the Neoproterozoic Earth.

4. Archean Ophiolites

731

4. ARCHEAN OPHIOLITES Over the course of several decades, a number of possible partial and dismembered ophiolite sequences have been described from a number of Archean greenstone belts of different ages and locations (e.g., de Wit et al., 1987; de Wit and Ashwal, 1997; Fripp and Jones, 1997; Harper, 1985; Kusky, 1989, 1990, 1991). However, few complete Phanerozoic-like ophiolite sequences have been recognized in Archean greenstone belts, leading some workers to the conclusion that no Archean ophiolites or oceanic crustal fragments are preserved (Bickle et al., 1994; Hamilton, 1998, 2003). These ideas were challenged by the recognition of a complete but partially dismembered Archean ophiolite sequence from the North China Craton (Kusky et al., 2004a), that was later found to be associated with mantle tectonites in mélange beneath the ophiolite (Li et al., 2002). This discovery has important implications for understanding other Archean greenstone belts, many of which contain only part of the typical ophiolite assemblage. With a complete assemblage present in at least one locality, it is more likely that the other reported partial sequences are truly parts of ophiolites, and not representative of some other tectonic setting that was unique in the Precambrian. Some workers have even suggested that the mechanisms of planetary heat loss changed so much with time so that what resembles an ophiolite from the Archean record is actually equivalent to a continental rift in the younger rock record. The Penrose definition of ophiolites (Anonymous, 1972; cf. Brongniart, 1813, 1821) includes “dismembered”, “partial”, and “metamorphosed” varieties, with rock types the same as those that typify Archean greenstone belts. Ophiolite-like relationships have been described for many years from Archean greenstone belts (Hess, 1955), yet many of the examples of partial ophiolites in Archean terrains were questioned, because no complete sequences were found anywhere. If such sequences were found in younger, Phanerozoic mountain belts, the ophiolitic origin for the rock sequence would not likely be questioned. In this book, many such ophiolitic sequences are described, and the authors take the approach of using the same criteria to identify ophiolites in very old rocks as they do in younger orogenic belts. The application of different paradigms to the Archean and Phanerozoic is no longer necessary, although detailed studies are beginning to reveal some differences in the style of older and younger sea floor spreading. Better quantification of these differences and similarities will help constrain geochemical, geodynamic, and thermal modeling of what effects the Archean mantle thermal and melting regime had on the structure of oceanic lithosphere produced in those times. Archean oceanic crust was possibly thicker than Proterozoic and Phanerozoic counterparts, resulting in accretion predominantly of the upper basaltic section of oceanic crust. However, structural repetition and complexities in greenstone belts makes it very difficult to assess original thicknesses, as shown by papers in this book, and in Kusky and Vearncombe (1997). The crustal thickness of Archean oceanic crust may have resembled modern oceanic plateaux (e.g., Kusky and Kidd, 1992; Kusky and Winsky, 1995; Kusky, 1998), but if average oceanic crust was this thick, then they would not be topographically high standing plateaus, and the term plateau would be meaningless. If this were the case, the rheological stratification of the oceanic lithosphere would have been different (Hoffman

732

Chapter 22: Epilogue

and Ranalli, 1988), and complete Phanerozoic-like MORB-type ophiolite sequences would have been very unlikely to be accreted or obducted during Archean orogenies. In contrast, only the upper, pillow lava-dominated sections would likely be accreted. Future research should be directed at considering what the consequences of changes in the style and/or composition of accreted oceanic material has on the structure and composition of the continental crust. For instance, the observation that Archean greenstone belts have an abundance of accreted ophiolitic fragments compared to Phanerozoic orogens suggests that thick, relatively buoyant, young Archean oceanic lithosphere may have had a rheological structure favoring delamination of the uppermost parts during subduction and collisional events (see Hoffman and Ranalli, 1988; Kusky and Polat, 1999). Subcrustal oceanic lithosphere slabs may have been underplated beneath cratons, forming mantle roots (Kusky, 1993). Descriptions of the various Precambrian ophiolites in this volume has shown that many are contained within thrust complexes that include elements formed at different stages of ocean opening and closing, with a strong bias toward convergent margin environments such as suprasubduction zone or arc-related spreading centers, accretionary wedge material, triple-junction related magmas and arc magmas that have migrated through these orogenic collages. Many of these ophiolites are parts of orogenic complexes that have experienced complex tectonic histories, similar to those of material accreted to younger accretionary orogens (Kusky et al., 2004b; Sengör and Natal’in, 2004). As in younger ophiolites, at present we observe a huge variation in inferred crustal thickness of ophiolites, and in the units that are preserved. With present limited data, and the amount of structural complication, it is not yet possible to assess whether or not there has been a demonstrable secular change in the thickness of oceanic crust (e.g., Moores, 2002). However, obtaining better constraints on the thickness and petrological relationships in Precambrian ophiolites remains a high priority for the Earth Sciences, since these are sensitive indicators to the nature of how the Earth lost the extra heat produced during the Precambrian. Since the current range of known ophiolitic thicknesses and internal stratigraphic relationships from Precambrian ophiolites are within the range of those from the Phanerozoic to present regime, we favor the idea that Precambrian ophiolites were not drastically thicker than those of younger times, and that much of the heat from the Precambrian Earth was lost though a greater total ridge length, and faster spreading rates, rather than production of dramatically thicker melt columns. Such relationships would produce a Precambrian Earth dominated by smaller oceanic plates, more triple junction interactions, and a younger average age of subducting lithosphere. Thicker sedimentary piles of graywacke turbidities on subducting plates would lead to fewer mélanges being formed (see Kusky et al., 2004b), and to more low-angle subduction with many ridge subduction events and belts of near-trench magmas intruding the accretionary margins and ophiolites, forming the TTG suite. The smaller oceanic plate size does not necessarily mean that continents were also smaller. We know, for instance, that Precambrian quartzites such as the Mt. Narryer required long rivers on large continents to form the extensive mature sands, and that some Archean strike slip faults (Sleep, 1992; Kusky and Vearncombe, 1997) and Archean passive margin sequences (Kusky and Hudleston, 1999) both had lengths exceeding 1,000 km.

4. Archean Ophiolites

733

Table 1. Criteria for recognition of a rock sequence as an ophiolite Indicator

Importance

Full Penrose sequence diagnostic in order

Status Status in Phanerozoic in Dongwanzi Ophiolites rare, about 10% suggested, needs documentation and verification

Conclusion

not conclusive

Podiform chromites w/nodular textures

diagnostic

about 15%

present

diagnostic

Full sequence dismembered

convincing

about 30–50%

dismembered units present

convincing

3 or 4 of 7 main units present

typical for accepting Phan. Ophiolite

about 80%

6 of 7 units known; dikes still uncertain (age)

convincing

Sheeted dikes

distinctive, nearly diagnostic

about 20–30%

suggested, age needs verification

not conclusive

Mantle tectonites

distinctive

about 20–30%

present

distinctive

Cumulates

present, not distinctive

about 70%

present

supportive

Layered gabbro

typical

about 70%

present

supportive

Pillow lavas

typical, not distinctive

about 85%

present

supportive

Chert, deep water seds

typical

about 85%

present

supportive

Co-magmatic dikes and gabbro

necessary, rare to about 15% observe

present

distinctive

High-T silicate defm. rare, but as inclus. in melt pods distinctive

about 10%

present

distinctive

Basal thrust fault

necessary (except in rare cases), not diag.

about 60%

present

supportive

Dynamothermal aureole

distinctive, almost diagnostic

about 15%

not determined

inconclusive

Sea floor metamor

distinctive

all

present

supportive

Hydrothermal vents black smoker type

distinctive

rare

present

strongly supports (continued on next page)

734

Chapter 22: Epilogue

Table 1. (Continued) Indicator

Importance

Status Status Conclusion in Phanerozoic in Dongwanzi Ophiolites Ophiolites are defined on the basis of field relationships and the overall rock sequence. Many workers have added chemical criteria to the ways to recognize and distinguish between different types of ophiolites. Some of the more common traits are: MORB chem.

distinctive

about 65%

present

distinctive

CA chemistry

common

about 30%

present in some units

inconclusive

Flat REE

distinctive

about 65%

present

distinctive

Boninites

distinctive of SSZ

rare but increasingly recognized

not known

inconclusive

Terranes in which Precambrian ophiolites are found resemble younger accretionary orogens, such as Alaska, the Altaids, or the Philippines, where each of these has long history and many magmatic events (Kusky et al., 2004b; Sengör and Natal’in, 2004; Encarnacion, 2004). Comparative studies between accretionary orogens and Precambrian cratons, orogens, and ophiolites are likely to continue to yield useful insights about how the early Earth operated. 5. IS IT AN OPHIOLITE? Several authors have presented various schemes to purportedly discriminate between ophiolitic and other sequences (e.g., Pearce, 1987; Wood et al., 1979), although most of these are either arbitrary, or based on models of what the authors believe Precambrian ophiolites should have looked like (e.g., Bickle et al., 1994). Here, we present a shamelessly uniformitarian list of criteria that can be used to determine the likelihood of whether or not a partial, dismembered, or complete sequence is ophiolitic, though comparison with betterunderstood Phanerozoic sequences. For comparison, the Dongwanzi ophiolite is compared to Phanerozoic ophiolites, and it stands up well to such comparison, and would clearly be called an ophiolite if it were preserved in a Phanerozoic orogen. Table 1 can be used for other questionable sequences, by replacing the column for the Dongwanzi ophiolite with the sequence in question. REFERENCES Alvarado, G.E., Denyer, P., Sinton, C.W., 1997. The 89 Ma Tortugal komatiitic suite, Costa Rica: implications for a common geological origin of the Caribbean and eastern Pacific region from a mantle plume. Geology 25, 439–442.

References

735

Anonymous, 1972. Ophiolites. Geotimes 17, 24–25. Bickle, M.J., Nisbet, E.G., Martin, 1994. Archean greenstone belts are not oceanic crust. J. Geology 102, 121–138. Brongniart, A., 1813. Essai d’une classification minéralogique des roches mélangées. Journal of Mines 199, 1–48. Brongniart, A., 1821. Sur le gisement ou position relative des ophiolites, euphotides, jaspes, etc., dans quelques parties des Apennines. Ann. Mines Rec. Mém. Exploit Mines 6, 177–238. Butcher, A.R., Pirrie, D., Prichard, H.M., Fisher, P., 1999. Platinum-group mineralization in the Rum layered intrusion, Scottish Hebrides, UK. Journal of the Geological Society of London 156, 213– 216. Corcoran, P.L., Mueller, W.U., Kusky, T.M., 2004. Inferred ophiolites in the Archean Slave Craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 363–404. Coward, M.P., Ries, A.C., 1995. Early Precambrian Processes. Geological Society of London Special Publication 95, p. 295. de Wit, M.J., Ashwal, L.D. (Eds.), 1997. Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, p. 809. de Wit, M.J., Hart, R.A., Hart, R.J., 1987. The Jamestown ophiolite complex, Barberton mountain belt: a section through 3.5 Ga oceanic crust. J. African Earth Sci. 6, 681–730. Edwards, S.J., Pearce, J.A., Freeman, J., 2000. New insights concerning the influence of water during the formation of podiform chromitite. In: Dilek, Y., Moores, E., Elthon, D., Nicolas, A. (Eds.), Ophiolites and Oceanic Crust. Geological Society of America Special Paper 349, 139–147. Encarnacion, J., 2004. Northern Philippine Ophiolites: Modern Analogues to Precambrian Ophiolites. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 615–626. Fowler, C.M.R., Ebinger, C.J., Hawkseworth, C.J., 2002. The Early Earth: Physical, Chemical and Biological Development. Geological Society of London Special Publication 199, p. 360. Fripp, R.E.P., Jones, M.G., 1997. Sheeted intrusions and peridotite-gabbro assemblages in the Yilgarn Craton, western Australia: elements of Archean ophiolites. In: de Wit, M.J., Ashwal, L.D. (Eds.), Greenstone Belts. In: Oxford Monographs on Geology and Geophysics, vol. 35, pp. 422–437. Hamilton, W.B., 1998. Archean tectonics and magmatism. International Geology Reviews 40, 1–39. Hamilton, W.B., 2003. An alternative Earth. GSA Today 13, 4–13. Harper, G.D., 1985. A dismembered Archean ophiolite, Wind River Mountains, Wyoming (USA). Ophioliti 10, 297–306. Hess, H.H., 1955. Serpentinites, orogeny, and epeirogeny. In: Poldervart, A. (Ed.), Crust of the Earth (A Symposium). Geological Society of America Special Paper 62, 391–408. Hoffman, P.F., Ranalli, G., 1988. Oceanic flake tectonics. Geophysical Research Letters 15, 1077– 1080. Hofmann, A., Kusky, T.M., 2004. The Belingwe Greenstone Belt: Ensialic or Oceanic? In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 487–537. Huang, X.N., Li, J.H., Kusky, T.M., Chen, Z., 2004. Microstructures of the Zunhua 2.50 Ga Podiform chromite, North China Craton and implications for the deformation and rheology of the Archean oceanic lithosphere mantle. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 321–337.

736

Chapter 22: Epilogue

Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004. Neoproterozoic ophiolites in the Arabian Shield: Field relations and structure. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162. Kusky, T.M., 1989. Accretion of the Archean Slave Province. Geology 17, 63–67. Kusky, T.M., 1990. Evidence for Archean ocean opening and closing in the southern Slave Province. Tectonics 9, 1533–1563. Kusky, T.M., 1991. Structural development of an Archean orogen, western Point Lake, Northwest Territories. Tectonics 10, 820–841. Kusky, T.M., 1993. Collapse of Archean orogens and the generation of late- to post-kinematic granitoids. Geology 21, 925–928. Kusky, T.M., 1998. Tectonic setting and terrane accretion of the Archean Zimbabwe craton. Geology 26, 163–166. Kusky, T.M., Hudleston, P.J., 1999. Growth and Demise of an Archean carbonate platform, Steep Rock Lake, Ontario, Canada. Canadian Journal of Earth Sciences 36, 1–20. Kusky, T.M., Kidd, W.S.F., 1992. Remnants of an Archean oceanic plateau, Belingwe greenstone belt, Zimbabwe. Geology 20, 43–46. Kusky, T.M., Polat, A., 1999. Growth of Granite-Greenstone Terranes at Convergent Margins and Stabilization of Archean Cratons. In: Marshak, S., van der Pluijm, B. (Eds.), Special Issue on Tectonics of Continental Interiors. Tectonophysics 305, 43–73. Kusky, T.M., Vearncombe, J., 1997. Structure of Archean Greenstone Belts. In: de Wit, M.J., Ashwal, L.D. (Eds.), Tectonic Evolution of Greenstone Belts, Oxford Monograph on Geology and Geophysics, pp. 95–128. Kusky, T.M., Winsky, P.A., 1995. Structural relationships along a greenstone/shallow water shelf contact, Belingwe greenstone belt, Zimbabwe. Tectonics 14, 448–471. Kusky, T.M., Abdelsalam, M., Tucker, R., Stern, R., 2003. Evolution of the East African and Related Orogens, and the Assembly of Gondwana. Special Issue of Precambrian Research 123, 81–344. Kusky, T.M., Li, J.H., Glass, A., Huang, X.N., 2004a. Origin and emplacement of Archean ophiolites of the Central Orogenic belt, North China craton. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 223–274. Kusky, T.M., Ganley, R., Lytwyn, J., Polat, A., 2004b. The Resurrection Peninsula ophiolite, mélange, and accreted flysch belts of southern Alaska as an analog for trench-forearc systems in Precambrian orogens. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 627–674. Lago, B.L, Rabinowicz, M., Nicolas, A., 1982. Podiform chromitite ore bodies: a genetic model. J. Petrol. 23, 103–125. Leblanc, M., Nicolas, A., 1992. Ophiolitic chromitites. International Geological Reviews 34, 653– 686. Li, J.H., Kusky, T.M., Huang, X., 2002. Neoarchean podiform chromitites and harzburgite tectonite in ophiolitic melange, North China Craton: Remnants of Archean oceanic mantle. GSA Today 12 (7), 4–11. Moores, E.M., 2002. Pre-1 Ga (Pre-Rodinian) ophiolites: Their tectonic and environmental implications. Geological Society of America Bulletin 114, 80–95. Nicolas, A., Azri, H.A., 1991. Chromite-rich and chromite-poor ophiolites: the Oman case. In: Peters, Tj., Nicolas, A., Coleman, R.G. (Eds.), Ophiolite Genesis and Evolution of the Oceanic Lithosphere. Kluwer Academic, Boston, pp. 261–274.

References

737

Parman, S.W., Grove, T.L., 2004. Petrology and geochemistry of Barberton komatiites and basaltic komatiites: evidence of Archean fore-arc magmatism. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 539–565. Pearce, J., 1987. An expert system for the tectonic characterization of ancient volcanic rocks. J. Volcanol. Geotherm. Res. 32, 51–65. Peltonen, P., Kontinen, A., 2004. The Jormua ophiolite: A mafic-ultramafic complex from an ancient ocean-continent transition zone. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 35–71. Polat, A., Kerrich, R., 2004. Precambrian arc associations: Boninites, adakites, magnesian andesites, and Nb-enriched basalts. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 567–597. Sengör, A.M.C., Natal’in, B.A., 2004. Phanerozoic analogues of Archean oceanic basement fragments: Altaid ophiolites and ophiorags. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 675– 726. Shchipansky, A.A., Samsonov, A.V., Bibikova, E.V., Babarina, I.I., Konilov, A.N., Krylov, K.A., Slabunov, A.I., Bogina, M.M., 2004. 2.8 Ga Boninite-hosting partial suprasubduction zone ophiolite sequences from the North Karelian greenstone belt, NE Baltic Shield, Russia. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 425–486. Simonson, B.M., 1985. Sedimentologic constraints on the origins of Precambrian iron-formations. Geological Society of America Bulletin 96, 244–252. Sleep, N.H., 1992. Archean plate tectonics: what can be learned from continental geology?. Canadian Journal of Earth Sciences 29, 2066–2071. Stern, R.J., 1994. Arc Assembly and continental collision in the Neoproterozoic East African Orogen: Implications for the assembly of Gondwanaland. Annual Reviews of Earth and Planetary Sciences 22, 319–351. Stern, R.J., Johnson, P.R., Kröner, A., Yibas, B., 2004. Neoproterozoic ophiolites of the ArabianNubian Shield. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and Related Rocks. In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 95–128. Stowe, C.W., 1994. Compositions and tectonic settings of chromite deposits through time. Economic Geology 89, 528–546. Wood, D.A., Joron, J.L., Treuil, M., 1979. A re-appraisal of the use of trace elements to classify between magma series erupted in different tectonic settings. Earth Planet. Sci. Lett. 45, 326–336.

This page intentionally left blank

739

SUBJECT INDEX

A Abitibi, 10, 567, 584 Abitibi greenstone belt, 577, 580 Abitibi terrane, 23 Abitibi-Wawa subprovince, Superior Province, 663 accreted oceanic plateau, 592 accretion, 662, 732 accretionary complex, 714 accretionary lapilli, 520 accretionary orogen, 429, 623, 689, 732, 734 accretionary prism, 619, 628, 647, 656, 661 accretionary wedge, 14, 207, 628, 703, 707 accretionary wedge material, 732 adakite, 23, 448, 477, 568, 569, 581, 592, 663 adakite-high magnesian andesite, 577 adakitic melt, 478 Agardagh Tes-Chem ophiolite, 17, 207–219, 706 Al Amar suture, 147 Al-depleted komatiite, 554 Al-depleted type of komatiitic rocks, 413 Al-undepleted komatiite, 554 Al-undepleted type komatiite, 413 Alaska, 10, 25, 628, 734 Aldan Shield, 21, 405, 406 Aleutian Islands convergent margin, 584 Aleutians, 678 Alfred Wegener, 682 alkaline, 67 Allaqi-Sol Hamid suture, 132 Alpine-type peridotite, 200, 608 Altaid evolution, 699 Altaids, 26, 208, 676, 688–716, 734 Altay Mountains, 688 Altay-Mongol domain, 692 Altay-Sayan sector, 698 amphibole gabbro, 111, 183, 184 amphibolite, 146, 572 amphibolitization, 150 Andes, 584 andesite, 113, 155, 432, 620 andesite-dacite-rhyolite suite, 477

Angara craton, 694 anhydrite, 353 anhydrous melting, 541 anorthosite, 240 antinodular chromite, 249, 250 antinodular texture, 250, 325 Anton complex, 366 Anton terrane, 397 apatite, 185 Arabian-Nubian Shield, 11, 15, 16, 95–123, 129–159, 163–202, 582, 727, 730 arc pluton, 620 arc tholeiite, 577 arc-like chemistry, 647 arc-related spreading center, 732 arc-shaving fault, 714 arc-slicing, 714 arc-trench assemblage, 448 Archaean granite-greenstone terrain, 699 Archaean greenstone belt, 714 Archaean mélange, 715 Archaean tectonic style, 716 Archaean ultramafic and mafic rocks, 714 Archean, 264 Archean greenstone belt, 599, 731 Archean mantle, 23 Archean mantle temperature, 560 Archean plate tectonics, 481, 560, 568, 591, 610, 662, 732 Archean spreading rate, 610 Arizona, 75 assimilation, 215, 647 assimilation of sediments, 666 Atacama fault, 694 Aufbruchszone, 679 augite, 545, 548 Austroalpine nappe system, 679 Azores fracture zone, 678 B back-arc basin, 202, 619, 677 back-arc magma, 217 back-arc tholeiite, 551

740

backstop, 707 backstop to accretionary wedge, 703 Baltic Shield, 22, 425, 426, 567, 579 banded iron formation, 80, 117, 224, 236, 506, 571, 572, 663, 728 Banting Group, 368 Barberton greenstone belt, 8, 23, 539–561, 606 Barrovian-type metamorphism, 442 basalt, 113, 141, 148, 155, 210, 581, 638 basaltic andesite, 638 basaltic komatiite, 541, 544, 551 basaltic sill, 80 basement complex, 84 basement contacts, 363 basement of ensimatic arc, 703 Batomga, 406 Beaulieu River volcanic belt, 379–397 Belingwe belt, 22 Belomorian mobile belt, 428 Beniah Lake fault zone, 364 Besshi-type deposit, 339, 358 Besshi-type sulfide deposit, 355 Besshi-type VMS deposit, 356 Betts Cove ophiolite, 583 Bi’r Tuluhah ophiolite, 143–147 Bi’r Umq ophiolite, 141–143 Bi’r Umq suture, 137 Bi’r Umq-Nakasib suture, 99, 158 BIF, 224, 225, 343, 345, 608, 663 birbirite, 118 black shale, 39 black smoker, 7 black smoker chimney, 20, 339, 349, 357 black smoker model, 357 Bogoin-Boali greenstone belt, 582 boninite, 16, 22, 23, 99, 114, 420, 425, 426, 438, 541, 551, 554, 567–592, 606, 619, 663, 704, 706, 728, 730 boninite series, 429, 434, 442, 457 Boruss mélange belt, 707 Boshchekul-Tarbagatay, 704, 713 brine-pool model, 357 Brongniart, 687 Bulawayan Group, 490, 527 C Calaguas ophiolite, 619 calc-alkaline, 113, 298, 644, 647, 730 calc-alkaline arc association, 580 calc-alkaline association, 577 calc-alkaline basalt, 645

Subject Index

calc-alkaline ophiolite, 647 calc-alkaline series, 195 Cameron and Beaulieu River belts, 366 Cameron and Beaulieu River greenstone belts, 9 Cameron and Beaulieu River volcanic belts, 20 Cameron River, 605 Cameron River belt, 379–397 carbonate alteration, 122, 152 carbonate-sulfide chimney, 351 carbonated, 141 carbonatite, 64 carbonatization, 44, 118, 171, 173, 179 carbonatized serpentinite, 172 Casiguran ophiolite, 619 Central African Republic, 582 Central America, 584 Central Asian mobile belt, 17, 207 Central Orogenic belt, 226, 228, 229, 265, 275, 283, 322, 340, 600 chemical stratigraphy, 656 chert, 4, 80, 116, 141, 142, 210, 225, 571, 572, 608, 637 chert horizon, 488 chert tectonite, 501, 506, 510 chert-BIF, 349 Cheshire Formation, 498, 504–506, 525 Chile Rise, 657 Chilimanzi suite, 517 China, 226 Chingezi gneiss complex, 490, 493 Chingezi suite, 493 chrome number, 58, 106, 200, 216, 252, 257, 258, 412 chromite, 53, 122, 150, 173, 180, 417 chromite vein, 326 chromitite, 248 Chugach accretionary wedge, 661 Chugach terrane, 13 Chugach-Prince William terrane, 25, 627, 628 climate, 662 clinopyroxenite, 41, 146 coherent flysch terrain, 658 collisional suture zone, 703, 713 Coloumb wedge, 661 coloured mélange, 678 conglomerate, 571 contact relationships, 370 continental flood basalt, 520 continental growth, 8, 11, 543, 568, 591, 658, 663, 665, 686, 699 continental rifting, 36, 64

Subject Index

Contwoyto terrane, 397 convergent margin, 12, 664 convergent margin environment, 732 convergent plate boundary, 591 Cordilleran-type, 165 Cordilleran-type ophiolite, 201, 368 critical taper, 661 crustal contamination, 521–523, 528, 558, 561, 729 crustal growth, 14, 543, 569 crustal thickness, 8, 731 crustal thickness of ophiolites, 732 cumulate, 2, 146, 171, 178, 180, 212, 216, 293 cumulate layer, 42, 173 cumulate rocks, 134 cumulate ultramafics, 110 cumulate zone, 410 Cyprus-type sulfide body, 356 D D1 ophiolitic thrusting, 450 dacite, 83, 432 dacitic, 113, 620 dacitic dike, 78 dacitic dyke, 433 dacitic flow, 80 deep water sediments, 7 deep-sea environment, 681 deep-sea sediments, 4 deformation, 714 depleted mantle, 241 Destor-Porcupine-Manneville fault zone, 577 Dibut Bay ophiolite, 619 differentiated flow, 409 differentiation process, 610 dike complex, 394 diorite, 176, 211, 477 diorite-tonalite, 615 diorite-tonalite intrusion, 622 discrimination diagram, 734 dismembered, 7, 714, 731 dismembered ophiolite, 102 disseminated chromite, 53, 55, 252 Djezkazgan-Kirgiz, 708 Dongwanzi ophiolite, 18, 19, 223–267, 275, 276, 283–318, 600 ductile shear zone, 231 Duncan Lake Group, 368 dunite, 110, 134, 146, 172, 179, 210, 216, 240, 244, 248, 289, 293, 321, 412, 606, 608, 632 dunite-peridotite, 409

741

dunitic, 171, 605 dunitic tectonite, 606 dyke, 706 dynamic magmatism, 263 Dzhida, 710 E earliest Precambrian rocks, 684 earliest thrust faulting, 433 early life, 7, 339 early nappe-style deformational event, 450 early shear zone, 728 Earth’s early biosphere, 225 East African Orogen, 11, 96, 99, 164, 730 Eastern Altay, 704, 710 Eastern Block, 227, 228, 275, 283, 340 eclogite, 10, 265, 609, 664 eclogitic residue, 664 Eduard Suess, 679 Émile Argand, 681 emplacement, 265 enriched mid-ocean ridge basalts (EMORB), 47 ensimatic arc, 707 epidosite, 85, 339, 354 epidote-clinozoisite, 238 epidotized, 60 Ethiopia, 99 exhalite, 345 F felsic tuff, 544 Fennoscandian Shield, 38 filter pressing, 55 flexural loading, 519 flower structure, 158 flysch, 12, 26, 267, 630, 634, 637, 648, 657 fold-thrust belt, 229, 542 forearc, 560, 583, 703, 706, 729 forearc evolution, 657, 665 forearc extension, 479 forearc ophiolite, 120 forearc spreading, 706 foreland basin, 229, 267, 519, 525, 526 foreland fold-thrust belt, 228, 232 foreland-thrust belt, 20 Fortescue Group, 521 Fox Island shear zone, 627, 650, 652, 653 fractionation history, 455 fracture zone environment, 678 frontal accretion, 661 Frotet-Evans greenstone belt, 579

742

G gabbro, 2, 39, 47, 50, 77, 111, 134, 141, 146, 148, 152, 155, 173, 174, 183, 238, 240, 246, 294, 295, 444, 610, 635, 706, 707 gabbro sill, 409, 433 gabbrodiorite, 210, 211 gabbroic, 171, 210 gabbroic dike, 58 gabbroic plutonic sequence, 605 gabbroic rocks, 619 gabbroic unit, 443 gabbronorite, 183, 210, 211 garnet amphibolite, 704 geochemical discrimination, 647 geothermal gradient, 7, 24, 568, 583, 591, 703 giant ophiolite nappe, 678 glaciation, 118 gold, 122, 173, 634, 658, 663 Gondwana, 165 Gondwanaland, 96 Gondwanan supercontinent, 730 Gorringe Bank, 678 granite, 84 granodiorite, 477, 633, 664 granodiorite suite, 477 granodioritic magma, 224 granulite, 228, 231, 265 graywacke, 39 Great Dyke, 490, 517 Great Lakes tectonic zone, 580 Greenland, 570 greenstone belt, 731 growth faulting, 357 H Hackett River arc, 397 Halaban ophiolite, 147 Halaban suture, 159 Hamisana shear zone, 165 Han-Taishir, 706 Hans Stille, 682 harzburgite, 2, 39, 104–110, 133, 137, 146, 179, 210, 216, 224, 240, 241, 243, 246–248, 262, 289, 290, 321, 326, 606, 619, 632 harzburgite tectonite, 104, 110, 240, 247, 275, 285, 728 harzburgitic ultramafics, 99 hazelwoodite nickel, 182 heat flow, 426, 592 heat loss, 1, 6, 732 heat production, 6

Subject Index

Hess, 682 Hess’s Precambrian oceans, 686 high degree of mantle melting, 420 high H2 O contents, 550 high P-T mineral, 448 high-Mg andesite, 432 high-pressure granulite, 228, 264 high-pressure granulite belt, 229, 267 high-temperature deformation, 104, 137, 241, 327, 329, 334 high-temperature ductile fabric, 178 high-temperature fabric, 180 high-temperature foliation, 247 high-temperature isoclinal folding, 178 high-temperature metamorphism, 433 high-temperature plastic flow, 19, 321 high-temperature shear strain, 183 high-temperature shearing, 331 high-Ti-ophiolite, 195 history of ophiolite studies, 676 Hoogenoeg Formation, 544, 606 Hoogenoeg magma, 560 hornblende gabbro, 172, 210, 211, 216 hornblende-rich gabbroic, 211 hornblende-rich phase, 146 hornblendite, 41, 247, 290, 293 hornblendite-garnetite cumulate, 35 hot spot, 4 hot subduction, 588, 592 hydrated mantle wedge, 478 hydrothermal alteration, 84, 179 hydrothermal circulation, 339 hydrothermal system, 85 hydrothermal vent, 122, 349 hypabyssal sill, 606 I IAT-MORB, 201 Iceland, 4 imbricate fold-thrust wedge, 433 imbricate stack, 446 immature sediments, 615 Indus suture, 678 inherited zircon, 523 Inner Mongolia, 226 intra-oceanic arc, 577 Iringora ophiolite, 443 Iringora SSZ ophiolite, 451 Iringora Structure, 426, 435, 442 iron formation, 38 ironstone, 494, 499, 501 Ishkeolmes, 704

Subject Index

island arc, 677 island arc magma, 216 island-arc tholeiite, 645 isotropic gabbro, 2, 42, 174, 184 Isua, 567, 584 Isua greenstone belt, 570, 573 Isua terrane, 23 Itogon ophiolite, 620 Ivrea Zone, 679 Izu-Bonin-Mariana arc, 24, 567 Izu-Bonin-Mariana fore-arc, 583 Izu-Bonin-Mariana subduction zone, 568 J Jabal al Uwayjah ophiolite, 155–157 Jabal Ess ophiolite, 130–136 Jabal Tays ophiolite, 152–155 Jalair-Nayman, 713 Jamestown ophiolite, 560 Jamestown ophiolite complex, 606 Japan, 10 Jormua ophiolite, 14 Jormua ophiolite complex, 36 Junggar-Alakol-Turfan system, 698 Junggar-Balkhash, 709 K Kaapvaal craton, 8, 23, 539 Kaapvalley allochthon, 606 Kainuu Schist Belt, 38 Kalgoorlie, 524 Kalgoorlie terrane, 9, 605 Kam Group, 368 Kamuikotan complex, 661 Kapuskasing structural zone, 579 Karelian, 38 Karelian Craton, 15, 35, 36, 38, 66, 428 Karelian province, 428 Kazakhstan-Tien Shan, 692 Kazakhstan-Tien Shan domain, 698 keirogen, 692 keirogenic movement, 699 Kenya, 99 Khangai-Khantey mountains, 692 Khangai-Khantey subduction-accretion complex, 712 Khangay-Khantey, 710 Khantaishir, 706 Khizovaara boninite series, 434, 464 Khizovaara greenstone structure, 429 Khizovaara Structure, 426, 430

743

kimberlite, 67, 609 Kings-Kaweah ophiolite, 711 Kipchak arc, 694, 696 Knight Island ophiolite, 630, 637 Komati Formation, 544, 606 Komatii river, 608 komatiite, 23, 306, 410, 494, 501, 539–561, 577, 728 komatiite alteration, 546 komatiite series, 551 komatiite water contents, 550 komatiitic basalt, 409, 704 Kromberg allochthon, 608, 609 Kukasozero belt, 429 Kula-Farallon ridge, 634, 654 Kurtushiba ophiolite, 707 Kuznetskii Alatau, 704 L Lagonoy ophiolite, 619 Lake Irinozero, 443 lamprophyre, 41, 61 Lapland-Kola province, 426 lava, 35 lava unit, 443 layered gabbro, 2, 50, 110, 143, 171, 174 layered mafic rocks, 444 Leopold Kober, 681 leucogabbro, 184 lherzolite, 5, 110, 137, 146, 293 lherzolitic subcontinental mantle, 729 Liaoxi ophiolitic mélange, 322 Limpopo belt, 490 Limpopo mobile belt, 516 listwaenite, 118, 136, 141, 148, 157 listwanitization, 173 lode gold, 663 low-angle shear zone, 506 low-Ti tholeiite, 551 Luzon volcanic arc, 617 M mafic and ultramafic sills, 420 mafic metavolcanics, 409 mafic mylonite, 363, 371, 379 magma chamber, 6, 610 magma flow structure, 332 magmatic arc front, 689 magmatic arc system, 218 magmatic flow structure, 327, 331 magmatic front, 692 magmatic layering, 77, 325

744

magnesian andesite, 568, 663 magnesite, 179 magnesium number, 54, 113, 199, 216, 456, 458, 546, 548, 549, 569, 581, 583 Maikain-Balkybek ophiolitic suture, 713 Makran, 662 Makran accretionary complex, 699 Manjeri Formation, 498–501, 527, 528 mantle cooling, 561 mantle deformation, 263 mantle diapir, 36 mantle flow, 134, 250 mantle lithosphere, 609 mantle melting process, 728 mantle peridotite, 42, 53, 600, 619 mantle plume, 65, 540 mantle root, 732 mantle tectonite, 39, 43, 53, 146, 356, 731 mantle temperature, 264, 539, 541, 728, 731 mantle wedge, 468, 478, 569, 583, 584, 588, 591 mantle xenolith, 67 Marianas, 678 martitization, 182 Mashaba ultramafic suite, 517 Mashaba-Chibi dyke, 524 Mashaba-Chibi dyke swarm, 524 massive chromitite, 55, 252 massive sulfide, 85, 238 massive sulfide deposit, 343 massive tholeiite, 608 Mazatzal, 90 Mazatzal block, 87, 89 Mazatzal crustal block, 75, 88 Mazatzal orogeny, 86 Mberengwa allochthon, 499, 510, 512, 528, 529 McHugh complex, 628, 658 mélange, 9, 12, 17, 26, 131, 136, 141, 142, 150, 152, 172, 207, 210, 212, 224, 231, 234, 235, 242, 262, 276, 339–341, 443, 446, 447, 450, 516, 517, 600, 605, 628, 657, 658, 662, 706, 708–710, 712, 715, 731, 732 melt channels, 248, 263 melt-rock reaction, 263 melting process, 539 metagabbro, 147, 605 metamorphic sole, 119, 443, 447, 448, 450 metamorphic tectonite, 172, 244 metamorphosed, 731 metazoa, 118 microgabbro, 176, 210, 706 mid-ocean ridge basalt, 645

Subject Index

Middle Marker, 544 migration of magmatic arc fronts, 676 Moho, 2 molasse, 265 Mongol-Okhotsk sector, 692, 699 Mongolia, 207 Mongolia-Okhotsk intracontinental collision, 661 Montalban ophiolite, 619 MORB, 236, 368, 442, 448, 645, 647, 708 MORB-like crust, 619 MORB-like pillow lava, 218 MORB-type ophiolite, 709 Mozambique belt, 164 Mozambique Ocean, 96, 159, 730 Mtshingwe Group, 490, 494–498, 527 mylonite, 506, 510 mylonitized, 146 N N-MORB, 47, 217, 377 N-MORB basalt, 433 Najd fault system, 99 Nankai accretionary wedge, 662 nappe, 65, 95 narrow ocean basin, 215 Nb-enriched basalt, 568 near-trench magma, 732 near-trench magmatism, 13, 634 near-trench pluton, 12, 657 near-trench plutonic rocks, 657 Neoproterozoic, 130 Ngezi Group, 22, 490, 498–530 niobium-enriched basalt, 581 nodular, 55 nodular chromite, 249, 321 nodular chromitite, 248, 250 North China, 275, 340 North China craton, 18, 226, 228, 283, 322, 340, 600, 731 North Karelia schist belt, 36 North Karelian greenstone belt, 22, 425–481, 579 North Karelian terrane, 567 North Korea, 226 North Sayan, 707 O Ob-Zaysan-Surgut, 709 obduction, 7, 99, 265, 732 ocean floor basement remnant, 714 ocean floor remnant, 677 ocean floor tholeiite, 645

Subject Index

ocean floor tholeiitic basalt-komatiite association, 580 ocean-continent transition, 35 oceanic basement remnant, 714 oceanic crust, 677 oceanic mantle, 250, 264, 321 oceanic plate stratigraphy, 10 oceanic plateau, 4, 5, 121, 528, 529, 577, 580, 731 ocelli, 610 OIB, 47 Olekma gneiss-greenstone terrain, 406 olivine, 545 olivine composition, 104 olivine gabbro, 210, 211, 240, 246, 293 olivine spinifex komatiite, 544 Olondo greenstone belt, 21, 405, 407–423 one-way chilling, 4 Onib ophiolite, 169 Onib-Sol Hamed ophiolite-decorated suture, 200 Onib-Sol Hamed suture, 163 Onverwacht allochthon, 606, 608 Opatica, 579 Opatica subprovince, 579 ophiolite, 560, 657 ophiolite conundrum, 21, 364 ophiolite nappe, 703 ophiolite sequence, 3 ophiolite thickness, 609 ophiolites and orogeny, 676 ophiolitic mélange, 242, 265, 340, 678 ophiolitic nappe, 443, 447 ophiolitic suite, 170 ophirag, 688, 708, 710, 712, 714, 715 orbicular chromite, 250, 321 orbicular chromitite, 55, 249 Orca Group, 630 Ordos block, 227 orogenic belt, 676 orogenic complex, 732 orogenic lherzolite, 42, 43, 62 orthopyroxenite, 134 osmium, 276 P Palaeo-Asian Ocean, 207 Pan-African evolution, 699 Pannotia, 96 partial, 731 partial melting of oceanic slab, 478 passive margin, 35, 65 passive margin ophiolite, 36

745

Patterson Lake structural complex, 379 Payson ophiolite, 15 pelagic rocks, 137 pelagic sediments, 116, 177 Penninic ophiolite, 679 Penrose-style, 170 Penrose-type ophiolite, 2, 99 peridotite, 39, 133, 141, 143, 146, 148, 155, 171, 172, 240, 244, 289, 412, 429, 605, 635, 707 peridotite massif, 248 peridotite xenolith, 609 Philippine arc, 89 Philippine fault system, 694 Philippine Sea plate, 616 Philippines, 25, 584, 615–623, 734 picrite, 663 picritic rocks, 433 Pilbara craton, 10 pillow, 35, 80 pillow basalt, 4, 112, 134, 140, 146, 296, 497, 504, 634, 636, 637, 706 pillow lava, 7, 39, 47, 140, 172, 176, 177, 186, 210, 236, 243, 248, 371, 379, 409, 444, 501, 502, 571, 600, 606, 619, 638 pillow structure, 410 pillowed tholeiitic basalt, 429 plagiogranite, 52, 120, 172, 184, 210, 636, 706 plate boundary length, 662 plate tectonics, 225 plateau, 710 platinoid, 182 plume, 561 plume model, 540 plume-arc interaction, 583 plume-arc scenario, 542 podiform chromite, 19, 110, 182, 263, 275, 285, 321, 324, 728 podiform chromitite, 39, 55, 200, 224, 235, 243, 262, 334, 600 Point Lake, 20 Point Lake belt, 366 Point Lake greenstone belt, 8, 605 Point Lake volcanic belt, 371 polyphase deformation, 225 porcelainite, 84 pore fluid pressure, 661 post-orogenic intrusive suite, 623 pre-arc spreading, 677 primitive island arc, 619 Prince William terrane, 630 Purtuniq ophiolite, 14

746

pyroxene spinifex komatiite, 545 pyroxenite, 110, 155, 172, 181, 210, 216, 240, 247, 289, 290, 293, 606, 608, 635 Q Qinling-Dabie Shan orogen, 226 R radiolarite, 680 Raquette Lake Formation, 385 Re-Os, 417 Re-Os model age, 275, 280 recumbent folding, 450 Red Sea, 96, 730 Red Sea Hills, 17, 163, 201 Reliance Formation, 498, 501–504 residual mantle, 216 Resurrection Peninsula ophiolite, 26, 630 rhenium, 276 rhyolite, 345, 478, 663 rhyolitic dyke, 433 ribbon chert, 172, 177 ridge length, 732 ridge subduction, 13, 14, 657, 658, 664, 665 rifted continental margin, 729 rodingite, 44, 48, 60, 119, 147, 150 Rodinia, 96, 159 Russian craton, 694 S Sanak-Baranof belt, 633 Sangilen, 706 sanukitoid, 580, 592, 664 Saudi Arabia, 98, 730 sea-floor alteration, 194 sea-floor hydrothermal alteration, 187, 452 sea-floor hydrothermal vents, 225 sea-floor metamorphism, 44, 150 seamounts, 710 Sebakwe protocraton, 493 Sebakwian Group, 488, 526 secular change, 732 secular change in the melting conditions in subduction zone, 561 sediment contamination, 647 sediment subduction, 207, 215, 561, 588 sedimentary cap, 177 serpentinite, 118, 147, 152, 155 serpentinitic mélange, 706, 711 serpentinization, 179 serpentinized peridotite, 172

Subject Index

Shabani gneiss complex, 490, 493 Shabani ultramafic complex, 517, 524 shallow subduction, 567 Shamvaian Group, 527 shear zone, 8, 16, 130, 136, 140, 143, 173, 433, 489, 513, 516, 527, 650 shearing, 158 sheeted dike, 79, 111, 140, 185, 243, 479, 605, 619, 635, 637 sheeted dike complex, 4, 6, 9, 39, 42, 48, 78, 134, 176, 211, 236, 248, 296, 368, 389, 443, 444, 605, 706 sheeted intrusion, 608 sheeted mafic dyke complex, 172 sheeted mafic-ultramafic intrusion, 606 shoshonite, 577 Siberia, 10 Siberian craton, 406 siliceous unit, 156 siliceous zone, 83 silicification, 44, 173, 339, 343, 506 silicified, 141 sill, 605, 706 sinter vent complex, 346 slab fluid, 561 slab window, 658 Slave Craton, 20, 363–399 Slave Province, 8 Sleepy Dragon complex, 379–397 Sleepy Dragon terrane, 397 slow-spreading type ophiolite, 43 Snowball Earth, 118 Solonker Ocean, 699, 712 source composition, 548 South Gobi, 712 South Mongolian, 712 South Pass, 9 spinifex texture, 550 spreading rate, 610, 732 SSZ ophiolite, 178 staurolite, 435 staurolite-amphibolite, 438 staurolite-bearing amphibolite, 434, 435 staurolite-garnet geothermometer, 442 Steinmann trinity, 680 Stille’s geotectonic cycle, 685 stockwork feeder vein, 351 structural detachment, 728 structural dislocation, 265 structural repetition, 731 subcontinental lithospheric mantle, 35, 43, 67, 664

Subject Index

subducted continental sediments, 218 subducted sediments, 216 subduction back-stepping, 676 subduction erosion, 662 subduction mélange, 628, 662 subduction process, 560 subduction-accretion complex, 12, 580, 676, 692 Subgan granite-greenstone complex, 407 submarine hydrothermal alteration, 452 Sudan, 99, 163, 201 sulfide, 85 sulfide deposit, 339 sulfide mound, 349 sulfidic shear zone, 519 Sunda accretionary prism, 662 superimposed ophiolite, 615 Superior Province, 10, 23, 567, 569 364, 579 suprasubduction belt, 321 suprasubduction complex, 608 suprasubduction zone, 96, 120, 163, 200, 201, 264, 281, 283, 285, 364, 418, 425, 523, 528, 579, 615, 619, 732 suprasubduction zone ophiolite, 298, 317, 419, 426, 479 suprasubduction zone ophiolitic belt, 356 Susunai complex, 662 suture, 98, 143, 265 suture zone, 95, 136, 158, 676, 678 Svecofennian collision, 67 Svecofennian Ocean, 36 Svecofennian orogeny, 428 T Taito ophiolite, 657 tectonic ironstone, 501, 506, 514–516, 518 tectonic settings, 5 tectosphere, 227 Tekturmas, 709 term ophiolite, 687 Tethyan ophiolite terrane, 399 Tethyan-type, 165 Tethyan-type ophiolite, 20, 363, 366 Tethysides, 699 Tharwah ophiolite, 136–141 thermal plume, 540 thin-skinned thrusting, 506 tholeiite, 448 tholeiitic, 85, 113, 195, 377, 394, 644, 730 tholeiitic basalt, 368, 379, 409, 410, 433, 577 tholeiitic igneous rocks, 67 tholeiitic pillow basalt, 608

747

tholeiitic pillow lava, 605 thrust, 450 thrust complex, 732 Tien Shan accretionary complex, 708 Tihama Asir, 730 Tihama Asir, Saudi Arabia, 5 Tokwe gneiss, 493 Tokwe gneiss complex, 488 Tokwe segment, 525 Tokwe terrane, 493 tonalite, 83, 84, 471, 633, 664 tonalite-trondhjemite gneiss (TTG), 231, 406 tonalite-trondhjemite-granodiorite, 13, 25, 658 tonalite-trondhjemite-granodiorite batholith, 569 tonalite-trondhjemite-granodiorite pluton, 580, 627 tonalitic magma, 224 trace element, 298, 299 Trans-Hudson orogen, 567, 582 transform suture, 703 transition zone, 2, 10, 78, 110, 111, 171, 172, 240, 248, 263 transitional continental crust, 397 transitional IAT/MORB, 195 transitional ophiolite, 5, 729 transposition, 265 transpression, 158 trench axis, 656 triple junction, 647 triple-junction interaction, 732 triple-junction related magma, 732 troctolite, 201, 262 trondhjemite, 52, 184, 471, 636, 650, 664 Troodos, 464, 583 Troodos Massif in Cyprus, 677 Troodos ophiolitic lava, 469 TTG, 13, 228, 229, 569, 580, 592, 663 TTG magma, 664 TTG suite, 732 tuff, 649, 651 turbidite, 265, 448, 449, 571, 630, 649 Tuva, 207 Tuva-Mongol arc, 699 Tuva-Mongol continental fragment, 692, 712 Tuva-Mongolian Massif, 208 U ultramafic, 290, 293, 707 ultramafic cumulate, 240, 619 ultramafic layered complex, 606 ultramafic mylonite, 8, 510, 513, 514 ultramafic plutonic rocks, 409

748

ultramafic rocks, 289, 572, 635 ultramafic tectonite, 134, 241 ultramylonite, 147 Umtali line, 530 underplating, 661 uniformitarianism, 731 Ural/Yenisey margin, 694 uralitic amphibole, 184, 186 V Valdez Group, 630, 658 vanished ocean, 676 vent chimney, 348 Ventersdorp, 521 Ventersdorp Group, 521 Vermilion district, 579 viscosity, 6 VMS, 663 VMS deposit, 664 volcaniclastics, 620 volcanogenic massive sulfide, 340, 663 W Wabigoon, 10 Wadi Onib, 17 Wadi Onib mafic-ultramafic complex, 163, 165 Wadi Onib ophiolite, 17, 163–202 Wawa greenstone belt, 580 Wawa subprovince, 569 websterite, 146, 182, 293 wehrlite, 110, 134, 146, 181, 210, 211, 216, 240, 293, 606, 608 wehrlitic transition, 171, 181 West Greenland, 23, 567 Western Block, 226, 228, 275, 283, 322, 340 Western Sayan, 707 wet melting, 541

Subject Index

Wind River Range, Wyoming, 9 Wrangellia composite terrane, 628 Wutai mélange belt, 357 Wutai Mountains, 20, 340 Wutai ophiolitic mélange, 322 Wutai VMS, 354 X xenocryst, 62 xenocrystic zircon, 62, 141, 523 Y Yakutat terrane, 662 Yanbu suture, 132 Yavapai orogeny, 86, 87, 89, 90 Yavapai-Mazatzal orogenic belt, 73, 87 Yellow Sea, 226 Yellowknife, 20 Yellowknife belt, 366 Yellowknife greenstone belt, 605 Yellowknife volcanic belt, 368–379 Yilgarn craton, 9, 524 Yinshan-Yanshan orogen, 226 Z Zabargad Island, 5, 730 Zagros crush belt, 678 Zambales ophiolite, 619 Zambales range, 617 Zeederbergs Formation, 498, 502–504 Zimbabwe, 22 Zimbabwe craton, 488, 516, 526, 527 Zunhua ophiolitic mélange, 275, 281, 322 Zunhua podiform chromite, 19 Zunhua structural belt, 229, 231, 232, 234, 242, 275, 284, 322 Zunhua tectonic zone, 600